Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Exploring the experiences of women living with metastatic breast cancer [MBC]: A systematic review of qualitative evidence

Contributed equally to this work with: Trína Lyons-Rahilly, Pauline Meskell, Eileen Carey, Alice Coffey

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliation Department of Nursing & Health Care Sciences, Munster Technological University, Tralee, Kerry, Ireland

ORCID logo

Roles Conceptualization, Methodology, Supervision, Validation, Visualization, Writing – review & editing

Affiliation Department of Nursing & Midwifery, University of Limerick, Limerick, Ireland

Roles Conceptualization, Project administration, Resources, Supervision, Writing – review & editing

Roles Conceptualization, Visualization, Writing – review & editing

¶ ‡ EM and DO also contributed equally to this work.

Affiliation Oncology Department, HSE Dublin Mid Leinster, Midlands Regional Hospital, Tullamore, Co. Offaly, Ireland

Roles Investigation, Project administration, Resources, Software

Affiliation MTU Kerry Library, Munster Technological University, Tralee, Co Kerry, Ireland

Roles Conceptualization, Investigation, Resources, Supervision, Validation, Writing – review & editing

  • Trína Lyons-Rahilly, 
  • Pauline Meskell, 
  • Eileen Carey, 
  • Elizabeth Meade, 
  • Donal O’ Sullivan, 
  • Alice Coffey

PLOS

  • Published: January 5, 2024
  • https://doi.org/10.1371/journal.pone.0296384
  • Reader Comments

Fig 1

Metastatic breast cancer [MBC] is the leading cause of cancer death in women globally with no cure. Women diagnosed with MBC endure a catastrophic upheaval to multiple aspects of their life and a radically transformed future landscape. Evidence suggests that the provision of care for women living with metastatic breast cancer is inadequate, socially isolating and stigmatising. To date, this topic has received little research attention. To increase understanding of the experiences of women living with MBC, a synthesis of current evidence is required. This paper presents a review of qualitative evidence on women’s experiences of MBC.

A qualitative evidence synthesis [QES] was conducted to synthesise primary qualitative research on the experiences of women living with MBC. Searches were performed of electronic databases Medline, Medline Ovid, PsycINFO, Psych articles, PubMED, CINAHL Complete, Scopus and grey literature databases. The methodological quality of the included studies was appraised using a modified version of the Critical Appraisal Skills Programme [CASP]. Title, abstract, and full-text screening were undertaken. A ‘best fit’ framework approach using the ARC [Adversity, Restoration, Compatibility] framework was used to guide data extraction and synthesis. Confidence in the findings was assessed using the Grading of Recommendations Assessment, Development and Evaluation, Confidence in the Evidence from Reviews of Qualitative research [GRADE-CERQual].

28 papers from 21 research studies containing 478 women’s experiences of living with MBC were deemed suitable for inclusion in this qualitative evidence synthesis. Findings are presented in a new conceptual framework RAAW [adapted from ARC] for women living with MBC under themes: R eality, A dversity, A djustment and W ellbeing. Findings revealed that a diagnosis of MBC impacted every aspect of women’s lives; this is different to a diagnosis of early breast cancer. An overarching theme of lack of support extended across various facets of their lives. A lack of psychological, emotional, and psychosocial support was evident, with a critical finding that models of care were not fit for purpose. Deficits included a lack of information, knowledge, inclusion in shared decision-making and MDT support, specifically the need for palliative care/oncology support access. Some women living with MBC wanted to be identified as having a chronic illness not a life-limiting illness. Culture and socioeconomic standing influenced the availability of various types of support. The impact of treatment and symptoms had an adverse effect on women’s quality of life and affected their ability to adjust.

This review synthesised the qualitative literature on the experiences of women living with MBC. The ARC framework used in the synthesis was adapted to develop a revised conceptual framework titled RAAW to represent the evidence from this review on experiences for women living with MBC; R eality & A dversity: A diagnosis of MBC; A djustment: Living with MBC; W ellbeing: Awareness, meaning, engagement [RAAW; MBC].

Citation: Lyons-Rahilly T, Meskell P, Carey E, Meade E, O’ Sullivan D, Coffey A (2024) Exploring the experiences of women living with metastatic breast cancer [MBC]: A systematic review of qualitative evidence. PLoS ONE 19(1): e0296384. https://doi.org/10.1371/journal.pone.0296384

Editor: Cho Lee Wong, The Chinese University of Hong Kong, HONG KONG

Received: January 7, 2023; Accepted: December 12, 2023; Published: January 5, 2024

Copyright: © 2024 Lyons-Rahilly et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data has been submitted within the paper.

Funding: The author(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

1. Introduction

Breast cancer significantly impacts women’s lives physically, psychologically, and socially; in 2020, there were 685,000 deaths worldwide from breast cancer [ 1 , 2 ]. Breast cancer is referred to as ‘metastatic’ or ‘advanced’ if it cannot be removed with surgery or has travelled to other sites within the body [ 3 ]. Metastatic breast cancer [MBC], including metastases found at diagnosis and recurrence post an early diagnosis of stage I-III, represents the most serious type of breast cancer [ 4 ]. With developments in treatment, women with metastatic breast cancer are living longer; consequently, it is essential to understand the experiences and needs of women with MBC to inform future models of care [ 5 – 7 ].

A qualitative evidence synthesis [QES] is the systematic gathering of primary qualitative data on similar topics of interest [ 8 – 10 ]. This maximises an understanding of health conditions, gleaning a deeper awareness of individuals’ experience of a disease, perceptions of their health condition, and decisions concerning their use of health service delivery [ 11 ]. QES can increase our understanding of cultural communities and has applicability within nursing as it offers a significant opportunity to develop knowledge. The QES process facilitates an uncovering of the deeper meaning of phenomena: it is not just a description of how people feel about a health condition, topic, or treatment but an understanding of ‘why’ they feel and behave the way they do [ 12 ].

1.1. Rationale for the study

Currently, there is a dearth of research into the experiences of women living with metastatic breast cancer [ 13 ]; Major gaps exist in the treatment and management of metastatic breast cancer patients, whilst the death rates due to breast cancer, are expected to almost double over the next 15 years, highlighting the crucial need to address these gaps [ 14 ]. The limited available literature on women living with MBC reveals high symptom burden, stigma, isolation, physical, psychological, psychosocial, and spiritual concerns with added financial pressure along with the associated complexity of decision-making [ 15 – 17 ]. Therefore, more challenging decisions must be made concerning women’s health and future care needs. It is recognised that these women require organised support that is different from women with an early breast cancer diagnosis [ 17 – 19 ]. MBC remains a virtually incurable disease, with a median survival time of two years, with some women living to five years or more [ 1 , 17 ]. Moreover, MBC is associated with a humanistic and financial burden to individuals, families, carers, society and health care systems [ 1 ]. Further research is required on the experiences of different subgroups of women with MBC, particularly those who have lived with MBC long term, from different cultural communities, women living well with MBC, and those receiving different kinds of active or palliative treatment [ 13 ]. With developments in treatment, women with metastatic breast cancer are living longer; therefore, it is essential to understand the experiences and needs of this cohort of women [ 5 – 7 ]. Significant adjustment is required to current models of care, as these are no longer fit for purpose given the increasingly chronic nature of the disease. Women living with MBC engage with palliative care services only at the end-of-life juncture, this is a relatively short period compared to the length of the chronic MBC disease trajectory. It can be increasingly difficult to recognise when decline is occurring. An amalgamated palliative care and oncology approach is needed to support women and oncologists in making difficult decisions, such as when to stop treatment and to plan for end of life [ 20 ]. To date published research is fragmented, it is now timely to undertake a comprehensive synthesis of current evidence, to increase understanding of the experiences of women living with MBC, and to inform future models of care. To the reviewer’s knowledge no qualitative evidence synthesis current or underway has been completed to date.

This study presents a systematic review and synthesis of qualitative evidence of the lived experience of metastatic breast cancer. It was conducted to evaluate and synthesise qualitative evidence, exploring women’s attitudes, perceptions and experiences regarding living with MBC. The method encompasses a comprehensive search for retrieval of qualitative research publications, a critical appraisal of primary qualitative studies, classification of results and synthesis of key findings. This qualitative evidence synthesis facilitates the identification of the issues women encounter living with MBC. This information can be used to inform the development of care trajectories to optimise care. In addition, this qualitative evidence synthesis will contribute significantly to understanding the needs of these women and facilitate the development of policy and practice to support their needs.

This QES aimed to synthesise evidence from principal qualitative studies, exploring women’s attitudes perceptions, and experiences of MBC, and establish an aggregated holistic view of the overarching themes related to this experience [ 21 ].

2.1. The research question is

What are the attitudes, perspectives, and experiences of women living with metastatic breast cancer?

2.2. The objective

The objective of this Qualitative Evidence Synthesis is to synthesise primary qualitative research exploring women’s attitudes, perceptions, and experiences regarding living with metastatic breast cancer.

2.3. The topic of interest

The phenomena of interest in this review are attitudes, perspectives, and experiences of women living with metastatic breast cancer.

A protocol on this QES was developed by applying Cochrane Effective Practice Organisation of Care Group [EPOC] guidelines template. The protocol was registered with Open Science Framework [ 22 ]. The ‘best-fit’ framework approach to synthesis was chosen as the most suitable methodology to undertake this synthesis. S1 Table outlines the seven steps of the ‘best fit’ framework technique undertaken [ 23 ]. Framework synthesis is better used when an a priori framework can be applied to the review therefore, the concepts that drive the framework are secure beforehand [ 10 ]. The ARC [Adversity, Restoration, Compatibility] conceptual framework for individuals living with and beyond cancer [ 24 ] was adopted in this QES [ Fig 1 ]. The ‘best fit’ framework synthesis procedure allows testing, and reinforcement building on an established model for a similar population [ 10 , 25 ].

thumbnail

  • PPT PowerPoint slide
  • PNG larger image
  • TIFF original image

https://doi.org/10.1371/journal.pone.0296384.g001

3.1. Systematic identification of primary qualitative studies

The Spice Framework S2 Table was used to identify the main concepts of the review focus, classify key search terms for the search string, and conduct a search for theoretical frameworks to be used as the ‘best fit’ framework. The inclusion and exclusion criteria identified keywords and search strings were agreed. Criteria for inclusion and exclusion for this search can be reviewed in S3 Table . A summary of the electronic search strings is presented in S4 Table .

3.2. Search strategy

With the assistance of a Librarian [D O’S], a systematic search for primary qualitative research studies was undertaken on Medline, Medline Ovid, PsycINFO, Psych articles, PubMed, CINAHL Complete, Scopus and grey literature databases. This search was undertaken from January 2010 to July 2020 and a further updated search using the same search strategy was undertaken from July 2020 to December 2022. The reference lists of all the included studies and key references [i.e. relevant systematic reviews] were checked for any salient studies not identified in the original search. Using the same keywords, a separate grey literature search was undertaken to identify primary qualitative literature not indexed in the databases listed, this was conducted on the following sources:

  • Open Grey [ www.opengrey.eu ]
  • Grey Literature Report [New York Academy of Medicine; www.greylit.org ]
  • Agency for Healthcare Research and Quality AHRQ [ www.ahrq.gov ]
  • National Institute for Health and Clinical Excellence [NICE; www.nice.org.uk ]: http://www.eldis.org .
  • Bielefeld Academic Search Engine [BASE]

Both comprehensive literature searches have been combined from January 2010 to December 2022 and reported in the PRISMA flow diagram [ Fig 2 ], identified studies were collated and uploaded to EndNote X9 [ 26 ]. The search outputs from of 1978 research papers were imported into Covidence, [ 27 ] a systematic review software package for the management of reviews; 378 duplicates were removed. Utilising this software accommodates filing the imported references accordingly [ 28 ]. The lead reviewer [TLR] the supervisory team [AC, PM, EC], and an Advanced Nurse Practitioner in Oncology [LM] undertook double-blind title abstract screening of all 1600 research papers. 1514 studies were deemed irrelevant. Three study authors were contacted to gain permission to access their research papers, no response was received, so these studies were excluded. Disagreements were resolved between team members. This resulted in a full-text review of 86 research papers. Following the full-text screen review, 28 papers were selected for inclusion. Justification for excluding fifty-eight full-text papers that did not meet the inclusion criteria are presented in [ Fig 2 ] PRISMA. There are 28 papers from 21 research studies included in this QES, details of which are summarised in S5 Table characteristics of included studies. The literature search results are reported using the PRISMA flow diagram [ 29 , 30 ] [ Fig 2 ].

thumbnail

https://doi.org/10.1371/journal.pone.0296384.g002

3.3. Quality appraisal of included studies

A quality appraisal was performed employing an adapted qualitative Critical Appraisal Skills Programme [CASP] checklist/tool [ 31 ], S6 Table . TLR undertook the assessment, assessing methodological limitations using the CASP tool. This facilitated identifying and categorising quality aspects of research conduct and reporting of included studies. This resulted in richer quality papers contributing more to the synthesis as opposed to those of lower quality [ 32 ]. The justification for using this appraisal was not to eliminate poor-quality papers but to decipher the quality of studies relative to their potential influence on findings [ 33 ]. TLR and PM conducted a pilot of the CASP on three studies to test the tool itself and its application. Once pilot evaluations were completed, TLR, PM, AC and EC discussed, and compared findings and a consensus was reached regarding the tool’s application and interpretation [ 34 ].

3.4. Selection of a framework for ‘best fit’ analysis

Selecting a theory/framework for ‘best-fit’ synthesis was conducted using the BeHEMoTh [Behaviour of Interest, Health context, Exclusions & Models or Theories] technique [ 23 ]. This technique presents a viable effective method for identifying a theory for a systematic review [ 23 ]. It is important to note it is not a requirement for the theoretical framework to be an ideal fit for the question [ 35 ].

4. ‘Best Fit’ framework synthesis

The ‘best fit’ framework synthesis method employing the ARC framework was utilised in developing the synthesis for this review [ 36 ]. This process is outlined in seven steps in S1 Table [ 23 ]. This approach can be tailored towards patient experiences of a particular disease trajectory [ 25 ], such as women’s experience of living with MBC. It offers an opportunity to underpin and shape existing published models [ 36 ]. It is better adapted to research with a specific question, where the prime concern is to define and understand what is occurring within a particular setting or patient’s experience of a specific disease trajectory [ 25 ]. This QES comprises of 28 papers from 21 research studies. These papers were examined to identify theoretical conceptual models employed. A search of databases was also conducted using the BeHEMoTh template to review suitable theories/frameworks and to engender a priori framework, to code the evidence against. The Adversity, Restoration, Compatibility [ARC] conceptual framework, for living with and beyond cancer, developed by Le Boutillier et al. [ 24 ], was identified as the most suitable framework to use in this review. Thus, an augmentative deductive process was undertaken, building on this existing model [ 36 ].

4.1. The ARC conceptual framework

The resulting overarching ARC framework [ Fig 1 ] represents Adversity [realising cancer], Restoration [readjusting to life with cancer] and Compatibility [reconciling cancer]. The themes are interlinked because the experience of living with and beyond cancer is not one dimensional. The ARC framework was established from the lived individual experience of the person affected by cancer. It is also a more suitable model than chronic illness models, as it is patient-centred, focusing on living with and beyond cancer [ 24 ]. ARC themes support biographical disruption, coping strategies in managing disease, attitudes to loss of self and go beyond the physical suffering of illness. Furthermore, Le Boutillier et al. stipulate that the framework is patient-centred and valuable in shaping supportive clinical oncology care [ 24 ].

4.2. Data extraction

The ARC framework was populated within a specifically adapted google form, this was devised by one team member PM. TLR became familiar with the 28 papers. PM, AC, EC undertook indexing of four studies to ensure transparency. During the process of indexing TLR read the papers line by line and populated the google form with the data from the 28 primary papers. This was undertaken in a deductive manner, while being mindful of not forcing data that did not ‘fit’ into unsuited categories [ 10 ]. Following this, outstanding data was synthesized inductively using the approach of thematic synthesis [ 33 ] to develop themes until all the data were accounted for [ 37 ]. Therefore, a specific two-phase subsequent method to ‘best fit’ framework synthesis facilitates an audit trail of themes evolving from the framework synthesis and those from the subsequent thematic synthesis [ 37 ]. The ‘best fit’ framework synthesis technique follows seven steps [ 23 ] [ S1 Table ]. The 28 papers included were mapped to the adjacent themes resonating with each framework category. As synthesis progressed, new themes emerged. The framework was adapted to better reflect a conceptual framework specific to women living with MBC. This process allowed for the identification of a gap in knowledge pertaining to the experiences of this cohort of women.

4.3. Data synthesis

The data synthesis of the qualitative studies and evolving findings were continually discussed in supervisory team meetings to interrogate the data and enrich the synthesis. The a priori categories in the identified model were used to guide the synthesis in a series of five stages:

Identifying a thematic framework: An a priori framework, the ARC framework [ 24 ], was utilised for qualitative data extraction to guide the synthesis, the framework was adapted and constructed on the emergent themes from the analysis.

Familiarisation: TLR became familiar with the data by reading the included primary research papers, fluidly moving over and back reviewing the data and the themes identified in the studies. The conceptual themes of the ARC framework were assimilated within an adapted google form.

Indexing: Indexing involved TLR applying the ARC Framework to code each of the papers. This was achieved by TLR extracting data from the research papers and mapping it against the ARC framework. Data coding was undertaken based on the themes identified in the data. Each paper was indexed using the codes associated to the themes of the framework. During this process TLR became immersed within the 28 studies and continuously moved between the data, developing themes and discussing emerging data/themes with AC, PM, EC and LM. All papers were read and reviewed until there was a consensus there were no new emerging themes. TLR presented findings of the data extracted via the ARC framework on google forms on a MIRO diagram to the team [AC, PM, EC, & LM]. In-depth discussion and debate with the team took place and revisions were undertaken.

Charting/Mapping: Charting/mapping involved rearranging data according to the relationships between themes, mapping and interpreting data where the range and nature of reviewed concepts were mapped and associations between themes identified. The review question and aim was maintained in focus throughout this process. The data was sorted by theme and presented in the form of an analysis. The columns rows of the table reflect the studies related to themes and allowed comparison on findings of the research papers across different themes, subthemes and subsequently new themes developed from the evidence in identified studies.

As outlined an a priori framework, the ARC framework [ 24 ] was utilised for qualitative data extraction to guide the synthesis. The framework was adapted based on the emerging themes from the analysis. As further concepts developed during synthesis that did not translate to the prevailing themes, thematic synthesis was undertaken to create new themes that were built/added onto the existing framework. This took time, TLR continuously reviewed emergent themes and concepts in partnership with AC, PM, EC and LM. The framework was adapted accordingly, established from the existing conceptual ARC framework encompassing the new concepts and theories [ 36 ]. This is a non-linear framework, the arrows represent the oscillations within the MBC disease trajectory. The experience of living with MBC does not remain static, there are fluctuations across the framework. This new non-linear framework [ Fig 3 ] is named RAAW : The Conceptual Framework for women living with MBC [adapted from ARC]. R eality & A dversity: A diagnosis of MBC; A djustment: Living with MBC; W ellbeing: Awareness, meaning and engagement [RAAW; MBC]. It is the researcher’s responsibility to ensure that the context of the qualitative primary research data is not misconstrued during the extraction synthesis process; TLR was cognisant of this throughout the process [ 34 ]. Enhancing the transparency in reporting the synthesis of qualitative research [ENTREQ] for this review can be viewed in S8 Table .

thumbnail

R eality & A dversity: A diagnosis of MBC; A djustment: Living with MBC; W ellbeing: Awareness, meaning engagement [RAAW; MBC].

https://doi.org/10.1371/journal.pone.0296384.g003

5.1. Synthesis output

The findings of the synthesis are presented under the following themes.

R eality and A dversity: A diagnosis of MBC,

A djustment: Living with MBC,

W ellbeing: Awareness, meaning and engagement [ RAAW ; MBC].

These represent RAAW: The Conceptual Framework for women living with MBC [adapted from ARC]. R eality & A dversity: A diagnosis of MBC; A djustment: Living with MBC; W ellbeing: Awareness, meaning engagement [RAAW; MBC] [ Fig 3 ].

5.2. Reality and adversity: A diagnosis of MBC

This is one of three main themes divided into four associated subthemes. Adversity for women living with MBC related to the initial shock and fear of cancer recurrence, the reality that they now live with a life-limiting condition.

5.2.1 Life changing impact of a diagnosis of MBC.

From the moment of their initial diagnosis of breast cancer, to the juncture of diagnosis of progressive MBC disease, many women lived for an extended time with the prevailing thoughts and fears that cancer may return; it was always in their consciousness [ 6 , 38 ]. The news that breast cancer had progressed was profoundly shocking and terrifying. Women described being fearful for the future and reported a loss of hope, heightened uncertainty, and reduced control over their lives [ 39 , 40 ]. This was noticeably different from the women’s reactions to previous initial diagnosis with primary breast cancer, as at that point there was a potential cure, whereas with MBC, the focus was not on cure but on extension of life [ 41 – 44 ]. The women revealed that progression of the disease, debilitating symptoms of treatment, the side-effects of previous surgery frequently led to reduced functional ability, psychological distress, physical discomfort and ultimately altered body image [ 41 , 45 – 47 ].

The diagnosis had a damaging impact on their lives and precipitated a sense of loss of the women’s holistic existence, affecting their professional identity. For example, many women had to give up work [ 38 ]. Worries about personal appearance; relationships with partners; role as mothers; bonding with family and friends; loss of purpose, well-being, and interference with the ability to manage activities of daily living were reported as overwhelming [ 41 – 43 , 46 – 50 ].

5.2.2 Influence of culture and socio-economic standing.

Drawing on data from studies conducted in Africa, Malaysia and two from the USA, a new subtheme emerged relating to cultural socio-economic influences on the reality of living with MBC [ 51 – 53 ]. These women had a significant financial burden and a perceived health provider bias related to their lower socio-status, race, and ethnicity. This resulted in women having little trust in the medics caring for them, significantly reducing their quality of life. This has highlighted that equitable comprehensive end-of-life care needs to be available to all, no matter the socio-economic status [ 53 ]. Breast cancer is increasing in developing countries, where women are presenting later with more advanced disease. There is a requirement for these women to be educated on the signs of early detection of breast cancer, this would allow women to be empowered to make decisions on their care [ 48 ].

5.2.3 Lack of information, disease knowledge, disclosure, and support.

When diagnosed with MBC women became active self-managers of their care searching for reliable, relevant information. However, gaining knowledge on their disease was difficult as women had limited access to literature and struggled to get information/supports from medical professionals [ 41 , 54 – 56 ]. Women regularly turned to social media to explore support options and learn more about the disease treatment [ 54 ]. On initial diagnosis there was support for women from a physical point of view, but this did little to meet their psychological needs [ 20 ]. Moreover, the most unmet need centred on a lack of psychological counselling and information [ 40 , 56 ]. Conversations and good communication with healthcare professionals were indicated as being integral to forming a supportive network, this was often lacking, as demonstrated in eight studies [ 40 , 41 , 43 , 47 , 50 , 55 , 57 , 58 ]. Women had negative experiences of being talked about by Health Care Professionals (HCP) while still in the room. They were treated as if they did not understand the medical terminology used and felt unsupported in dialogue about living with this disease [ 55 ]. Additionally, there is a deficiency in accessibility to clinical trials for women with MBC and difficulty in obtaining information on their disease evolution. There is a need to have a specialised MBC nurse whose role could potentially support women with MBC throughout their cancer trajectory [ 38 , 59 ]. Furthermore, there was a lack of conversation from HCPs on aspects of the disease that affected women’s sexuality because of treatments [ 47 ].

There is a clear differentiation between those diagnosed with early breast cancer compared to those living with metastatic breast cancer [ 17 – 19 ]. These two disease trajectories are often intertwined but are, however, very different. This led women to feel forgotten about and accentuated this sense of silence that had already been imposed upon them. Women stated that their experience deserved to be recognised [ 60 ].

In a longitudinal mixed methods study by Reed and Corner, few women appeared to have access to official support services, describing their primary support as family and friends [ 20 ]. For women living with MBC the primary source of meaning in life was family and valued social relationships; women acknowledged that social support played a pivotal role in living well [ 40 , 49 , 51 , 53 , 58 ]. Women with MBC drew mainly on family for support; however, few supportive formal resources are available to families, which often resulted in an additional burden on family members assisting women with MBC [ 40 , 56 ]. Women with MBC were constantly striving for normality through communications with a select group of friends or inner circle; these individuals were confidants and an important social espousal for women with MBC [ 41 , 42 , 55 ]. This was a form of self-management on the women’s part, creating a new selected supportive network. As the disease progressed, the role of the select group of family and friends also had to evolve [ 61 ]. Children were often the ‘silent observers’ of the disease trajectory; roles were reversed with the children looking after mothers; women felt it easier not to talk about their diagnosis to keep an equilibrium in the household [ 41 , 43 ]. They wanted specialised support for their children and partners [ 6 , 47 ]. Being a mother placed women with MBC in the position where the needs of their family and children were often placed before their own [ 57 ].

Furthermore, talking about their diagnosis of MBC was challenging. Women were selective in their dialogue with others. There was a delicate balance between preserving privacy on the aspects of living with the disease and getting support [ 41 ]. Women tended to take control of their illnesses to safeguard themselves by being conservative in conversations about cancer and were cautious with whom they shared information [ 49 , 54 , 57 ]. Because women with MBC looked ‘normal’, there was a heightened lack of awareness and sometimes disbelief that women were living with MBC [ 54 ]. In a cross-sectional qualitative study [ 42 ] and an interpretative phenomenology analysis [IPA] study [ 60 ] women used tactics called ‘strategizing disclosure’. This involved discussing specific elements of their diagnosis and health with family and friends. This led to suffering in silence for women as they could not verbalise their feelings around the disease trajectory. Approaches to dialogue could be categorised into 1] selective on who to tell, 2] selective on what to tell, 3] avoiding negative discussions, 4] personal feelings of anger and resentment.

Women reported that they did not attend a support group because they felt most women in support groups they attended had an early-stage cancer diagnosis rather than MBC, they could not relate to this [ 20 , 55 , 56 ]. Women expressed a need for support groups with women of a similar age and stage of diagnosis [ 47 , 58 ]. This subtheme relates to the paucity of information and discourse in the area of MBC. It includes a lack of medical support, shared decision-making, information/disease knowledge, poor psychological support, an absence of MBC specific support groups for women and highlights the importance of the support of family and friends. Furthermore, there is a need to have a specialised breast care nurse whose role could potentially support women with MBC throughout their cancer disease trajectory [ 38 , 59 ].

5.2.4 The impact of treatment and effect of symptoms.

Treatment was defined as the ultimate suffering [ 4 ], with women with MBC unable to fully articulate the experience. Women had constant concerns about whether the treatment was working, if there were other treatment options and if the side effects of these treatments could potentially affect their activities of daily living [ 38 , 40 , 49 , 57 ]. In three studies, women experiencing MBC choose to have the most aggressive form of treatment for the longevity of life—stopping the treatment was not an option [ 40 , 50 , 51 ]. Women were determined to cope with debilitating side-effects that impacted every system in the body, as it meant prolonging life; their days were planned around this treatment [ 50 ]. Additionally, living on borrowed time caused enormous psychological stress, which included being hyper-alert to symptoms of possible relapse. Women desired continuity of care within the multidisciplinary team; they wanted to meet the same HCP’s, which enabled good collaboration, communication and understanding of treatment planning. Women wanted to reduce the time spent in treatment and attending clinical appointments; significantly, the women in this study wanted to focus on living, not simply surviving [ 40 ].

Women with MBC had various debilitating symptoms from treatment that diminished their quality of life. These included neuropathy, pain, hair loss, sores, fatigue, breaking bones, nausea, chemo brain, constipation, diarrhoea, oral mucositis, anorexia, inability to mobilise, vaginal dryness, chemo-generated menopause and low libido [ 39 , 42 , 47 , 48 , 55 ]. Decreased functional status forced women to modify how they engaged with everyday activities, such as housework, self-care, family life, rest and food preparation [ 46 ]. The essence of coping was complying with the treatment offered [ 51 ]. Aspects of having to continue to work to maintain health insurance, despite feeling unwell, had far-reaching consequences [ 53 ]. Insufficient financial resources sometimes forced a delay in necessary treatment, thus raising the risk of inadequate symptom management and premature mortality [ 53 ]. Psychological loss, grief and anxiety characterised the emotional landscape. Women found the emotional challenges combined with a reduced capacity to make decisions and the realisation that they would be on treatment for the remainder of their lives overwhelming [ 47 , 54 ].

6. ‘Adjustment’—Living with metastatic breast cancer

The second central theme, ‘Adjustment’ to living with MBC, refers to women with MBC adapting to life in the context of living and managing MBC disease. During this phase of the disease trajectory, women adjust to living with a new normal and navigating the intricacies of life that come with a metastatic breast cancer diagnosis.

6.1. Health care professionals [HCP]/patient relationships and shared decision making

The supportive scaffold for women adapting to the new normal of living with MBC was their relationship with the HCP and the wider multidisciplinary team [MDT] [ 62 ]. The importance of clarity in autonomously making shared decisions was paramount, and the need for recognition of others’ expertise within the multidisciplinary team [ 47 ]. There is a discrepancy between various treatment programmes, with the ultimate focus on longevity rather than the quality of life [ 49 , 55 ]. Evidence in two studies depicted the overarching concept of the complex relationship women had with their oncologists and how this relationship impacted women’s healthcare outcomes and decision-making [ 20 , 50 ].

The relationship women had with their Doctors/HCP’s was linked to enabling women to make informed choices, become involved in shared decision-making, and have greater satisfaction in the care they received, thus improving their quality of life [ 39 ]. Fourteen studies referred to this subtheme and focused on the importance of knowledgeable shared decision making for women with MBC; using easily understood language; it underlined the need for an open, honest relationship with doctors/HCPs and the wider MDT in charge of women’s care. These studies further emphasised the importance of educating women on the transitions within the metastatic breast cancer disease trajectory and the necessity of including the palliative care experts and the broader multidisciplinary team to allow these women to make informed autonomous decisions about their care trajectory [ 20 , 40 , 41 , 44 , 47 , 50 , 51 , 53 , 55 , 59 – 63 ].

6.2. Changes to lifestyle

Women made various lifestyle changes due to altered functional status, debilitating symptoms, and side-effects of treatment. Various creative lifestyle changes were adopted, such as painting, mindfulness, and an interest in nutrition. One of the main reasons for a lifestyle change for women with MBC was the exacerbation of treatment and symptom burden, which restricted women’s lifestyle, functional status, and disrupted activities of daily living [ 48 , 49 ]. Furthermore, concentration difficulties negatively affected relationships with others, and changes to their body adversely impacted their self-image [ 39 – 41 , 43 ].

The capacity for women with MBC to utilise strategies to live well was explored in three studies [ 49 , 53 , 61 ]. Findings indicated that women differed in the methods they employed to live well. The most common strategy to live well was to restore a level of normality within their lives; all these women had children [ 49 ]. The second strategy was to reevaluate their lives; they used exercise to improve quality of life, reduced stress, having a positive attitude, and social support were vital in attaining this. However, due to fluctuating health status from treatment symptoms, restrictions to lifestyle led to burden, poor quality of life, and a loss of a sense of purpose to participate fully in social roles.

7. Wellbeing: Awareness, meaning and engagement

The theme wellbeing: awareness, meaning, and engagement refer to women finding a way to face the future with a renewed outlook on their lives, a new sense of personal worth and a sense of empowerment to live well. This theme is constructed on the subthemes of future self-identity and wellbeing, hope and altruism.

7.1. Future self-identity and wellbeing

A realisation that women with MBC were living on reduced time appeared to be the incentive for a significant shift and change in how they lived their everyday lives, particularly time spent with families. Initially, women felt a multitude of emotions at diagnosis of MBC, such as fear, dread and anxiety; However, as time passed, a sense of acceptance allowed the women to gain an appreciation for life; in facing their mortality, time left was perceived as a gift [ 53 ]. This allowed them to gain control of their life and achieve a positive mindset [ 49 ]. Women perceived their prognosis as uncontrollable; they wanted to control what they could, and having a positive mindset played a significant role in achieving a good quality of life [ 45 , 61 ]. This involved adjusting perspectives to a new norm, concentrating on what was most important in life [ 46 ]. Key sources of meaning were children, family, selective close social relationships, spirituality, a new value of life, maintaining normality, and accepting the diagnosis [ 46 , 50 , 54 ]. Women began to live a more existential authentic life where beliefs and ethical values came to the fore and death became more meaningful [ 60 ].

Looking to an uncertain future, women living with MBC wanted to be identified as individuals living with a chronic disease instead of incurable cancer; they rejected the identity of being sick [ 38 , 40 , 49 ]. Planning a future for women with MBC is difficult as the complexity of the holistic experience of this illness leads to a ‘crisis of the presence’ this relates to losing oneself and ones place in the world [ 38 ]. Women with MBC reported losing their sense of femininity and were challenged to live with an altered body image [ 45 , 48 ]. Working formed a significant aspect of their self-identity and allowed the women some normality to their lives [ 54 ]. The establishment of professional roles was a priority for some women and was an integral part of their identity. Women who were unable to work reported a sense of ‘valuelessness’; this, coupled with financial challenges, was difficult [ 42 , 44 , 47 ]. The women lived with an ‘altered mode of being’ they felt a disconnectedness from the world they lived in [ 60 ].

Living a life of engagement, purpose and meaning were integral to the wellbeing of these women [ 49 ]. Aspects of sexual health, mental wellbeing, psychological support were fundamental to maintaining wellness [ 45 , 49 , 63 ]. However, six studies allude to the enormous psychological stress that women with MBC experience, the palpable lack of support is coupled with a lack of research on interventions to help women cope [ 40 – 42 , 47 , 50 , 56 ]. Furthermore, HCPs were not always qualified to support these women from this standpoint [ 59 ]. Yoga, painting, meditation, mindfulness, and the role of complementary therapies [ 42 ] were all considered helpful; however, these were instigated by the women themselves in these studies, not by HCP.

Hope in the midst of living with MBC was evident amongst women in the following studies [ 6 , 20 , 38 , 40 – 42 , 48 – 55 , 61 , 62 ]. Hope was defined differently for many women in eight studies [ 38 , 42 , 48 – 50 , 52 , 61 , 62 ]. In these studies, women sought hope in the treatment they were receiving, in spirituality, a new outlook, living in the present moment, having a focus of short quantifiable goals, support from the multidisciplinary team, complementary therapies, and viewing their disease as a chronic illness, all of these brought hope and assisted the women to cope. Women searched for meaning in their suffering, [ 45 ] and found hope in adversity [ 48 ].

7.3. Altruism

In response to their personal experiences, numerous women felt encouraged to support other women with cancer and improve their understanding [ 53 ]. Altruism included a willingness for women to share their stories and to assist others in benefiting from the knowledge they had gained [ 47 , 51 , 54 , 57 ]. These women promoted breast cancer awareness and helped others in a similar situation [ 51 ].

The women viewed life as living rather than just surviving and engaged in self-management methods that positively impacted their health. Therefore, they felt enabled to educate/inform women going through similar experiences [ 60 ].

8. Confidence in the evidence from reviews of qualitative research [GRADE-CERQual]

GRADE- CERQual allows for clarity and aids in evaluating confidence in the findings from a QES review [ 64 , 65 ]. This approach evaluated the 10 review findings of this QES. These findings were derived from undertaking data synthesis applying the ‘best fit’ framework approach of the 28 papers. GRADE-CERqual is a systematic, robust framework allowing assessment of the confidence level in the findings of a qualitative evidence synthesis [ 65 ]. It is based on four components methodological limitations, coherence, adequacy of data and relevance. Two team members [PM, TLR] assessed the overall confidence in each of the review findings. The goal when assessing methodological limitations is to stipulate concerns when methodological limitations have been detected as significant to reduce the confidence in the finding [ 66 ]. Concerns regarding any component was discussed with the other two team members [AC, EC] and a consensus made. Confidence was judged as high, moderate, or low. The final assessment was based on consensus among the team [TLR, AC, PM, EC]. All findings started as high confidence and were then graded down if there were concerns. After assessing each component in relation to the total confidence in the evidence associated with each finding within this review, the supervisory team and the lead researcher made a judgment. A judgement is made in relation to concerns [ 66 ]. There are ten findings from this review; five were rated as high confidence, four were considered to have moderate confidence, one finding had low confidence. For more in-depth detail the results of the GRADE-CERqual are in S7 Table summary of qualitative findings table.

9. Discussion

The aim of this QES is to review the literature on the experiences of women living with metastatic breast cancer. The comprehensive data search yielded 28 papers from twenty-one research studies; it is apparent that there are clear correlations between the literature reviewed and the ARC framework. However, the evidence across studies indicates various unmet physical, social, and psychological needs and health care disparities for this cohort of women that were not represented within the framework. There is a consensus that the majority of women with MBC are living with a chronic illness; therefore, they will experience various oscillations within the disease journey [ 61 ]. The synthesised evidence from this QES highlights a considerable gap in the knowledge on how women with MBC live their lives daily and what approaches they use to manage their illness journey [ 49 , 55 , 57 ].

In particular a common theme throughout the literature refers to the psychological issues, emotional distress, and a lack of psychosocial support that these women are confronted with on a regular basis, usually cascading from the physical impact of treatment, symptom burden and underlying reality of transience of life [ 40 – 42 , 45 , 50 , 55 , 60 ]. The most challenging of these symptoms was pain, which is interrelated with fatigue, being anxious and depressed, this led to death anxiety, especially when feeling poorly [ 40 , 50 ]. Coping with MBC necessitates relentless adaptation and adjustment [ 40 ]; women identified these transitions as adjustments in emotional, social, and physical wellbeing [ 61 ]. Unfortunately, fluctuations between illness, treatments and recovery undermined their ability to adjust. What is not evident within the literature is how these women overcame and moved through these various transitions throughout the disease trajectory.

These emotional transitions are periods of enormous psychological turbulence [ 40 ]. However, only two studies referred briefly to women’s mental health [ 42 , 56 ] highlighting a gap in research. There is minimal uptake of mental health services amongst women with MBC, and more discussions are required with HCP’s around the mental health of these women [ 42 , 56 ]. Women lacked a collaborative multidisciplinary approach to their care [ 37 ]. They were medically well treated however, the most significant unmet need was psychological support interventions [ 40 ]. Women used creative activities such as painting, art, and yoga as a form of distraction from emotional distress [ 58 ]. Although suggestions were made for women to utilise acceptance commitment therapy, cognitive behaviour therapy [CBT], and cognitive processing theory to enhance psychological wellbeing, there was no elaboration on how these could be undertaken [ 41 , 45 ].

The literature also underscores a lack of supportive care services for these women [ 41 ]. There is an obligation to have a more synchronised delivery of care; this would be in the approach of a person-centred care plan and care services [ 41 ]. Women with MBC were often bewildered as to which multidisciplinary team member managed specific aspects of their care. A limitation in an integrated approach to care management emphasises the need for supportive nursing care services to facilitate the behavioural cognitive transitions experienced by women with MBC [ 46 , 59 ]. This reflects the iterative aspect of the ARC framework where the women’s experience of MBC is not linear; they tend to transition from one phase to the next and back again depending on overarching factors such as treatment, symptoms, diagnosis, psychological wellbeing, time in the disease journey, relationships with HCPs and social relationships. Women with MBC require a person-centred approach to care that reflects their individual needs and challenges; HCPs need to be educated to assess these women individually and understand their perceptions of their symptoms and overarching concerns, all of which are personal to each woman [ 45 , 59 ].

Models of care for women with MBC are no longer fit for purpose [ 20 , 55 ] results from this QES support this. MBC is currently considered a chronic illness rather than a life-limiting disease, therefore arguably there is a requirement for utilising chronic illness models of care that favour self-management of care, for women living with an MBC diagnosis [ 20 ]. Additionally, HCPs need to recognise, identify, and facilitate when a woman is about to go through a significant transition and review/revisit their preferences for managing their care at these junctures. Breast care nurse specialists felt they lacked sufficient expertise, competencies, and time to deliver the degree of care necessary to women with MBC [ 55 , 59 , 67 ]. Despite the passage of time the situation has not improved, and breast care nurse specialists still need further training and education in MBC disease transitions and more clinical time to engage with this cohort of women. Specialist services appear not to be meeting women’s needs. Women with MBC reported unrecognised and unmet supportive care needs, supportive care from specialist nurses was ad hoc and proactive allocation was a rarity [ 59 ]. There is a need for a more assimilated oncology palliative care approach [ 20 ]. However, there appears to be a pre-conceived view that palliative care is only for the terminal phase of the disease. The HCP’s in palliative care services have the expertise, knowledge and skills to improve the quality of life for women when introduced at the correct time within the disease trajectory and in consultation with their oncologist. The relationships these women have with their health care providers significantly impact their overall well-being. Informational needs were often interconnected with psychological needs. Crucial to this is a requirement for shared decision-making imparting disease knowledge from a HCP perspective at timely junctures allowing time for women to make informed decisions of their care; this is a rarity in the care of MBC patients [ 50 ].

9.1. Implications for future research and practice

The findings from this QES demonstrate women with MBC are poorly understood and there are failings in the care of this cohort particularly in their psychological wellbeing and holistic care. What is urgently required from a practice perspective is an open transparent, person-centred care plan, devised via a collaborative approach between medical oncologists and women with MBC. Furthermore, when a woman is diagnosed with MBC an automatic referral for psychological support is necessary, to support women’s mental health during this period. Psychological support must be scaffolded on each individual women’s needs providing strategies to cope with living with a life limiting disease. In order to provide consistency and enhanced care for these women, an identified case worker should be instated to support women with MBC. These findings can be used to inform planning of ongoing care in an integrated manner, so the woman is supported and managed effectively.

Additionally primary quantitative follow up research around MBC is required to determine the pattern of care management across jurisdictions and results explored to determine facilitating factors in the provision of high-quality cost effective and timely care. Findings from this study can be used to inform the development of a questionnaire for this work. Or the development of a nurse led intervention study focusing on the provision of support, education and advocacy for women living with MBC.

10. Limitations of this study

This review has limitations. Firstly, only papers published in the English language were included. Although the search strategy was comprehensive and rigorous it is recognised that in the application of language limits, we are likely to have omitted some relevant international studies related to women’s experiences of MBC. It is also recognised that there are particular challenges in retrieving qualitative research because of poor indexing [ 10 ]. To address this additional search of supplementary sources such as grey literature, reference checking and hand searching were conducted.

11. Conclusion

It is clear from undertaking this QES the most significant unmet needs experienced by women with MBC, is a lack of holistic person-centred care, with no psychological support provided. It is evident that there is a dearth of knowledge pertaining to women’s mental health particularly after their initial diagnosis of MBC and coping with the oscillations of the disease. There is a requirement for Health Care Providers who care for these women to have training and education on supporting their holistic wellbeing. There is limited evidence of input from a multidisciplinary team perspective, and there appears to be no structured care pathway to ensure MBC patients received appropriate care and support [ 20 , 44 , 59 ]. Further research is required to explore the themes from this QES with women currently experiencing MBC and test the RAAW: Conceptual Framework for women living with MBC [adapted from ARC]. It is apparent that further research is urgently required to explore evidence based psychological interventions. The researchers of this QES feel there is merit in exploring a nurse led intervention study focusing on the provision of support, including psychological, education and advocacy for women living with MBC. The results of the Grade-CERQual indicate that there is high to moderate confidence in the majority of findings of this review. The result of this research adds to the available body of knowledge around women’s experiences of living with MBC and can inform developments in relation to integrated care models.

Supporting information

S1 table. the seven-steps of ‘best fit’ framework..

https://doi.org/10.1371/journal.pone.0296384.s001

S2 Table. SPICE framework.

https://doi.org/10.1371/journal.pone.0296384.s002

S3 Table. Inclusion and exclusion criteria.

https://doi.org/10.1371/journal.pone.0296384.s003

S4 Table. Search strings.

https://doi.org/10.1371/journal.pone.0296384.s004

S5 Table. Characteristics of included studies.

https://doi.org/10.1371/journal.pone.0296384.s005

S6 Table. CASP.

https://doi.org/10.1371/journal.pone.0296384.s006

S7 Table. Summary of qualitative findings table.

https://doi.org/10.1371/journal.pone.0296384.s007

S8 Table. Enhancing the transparency in reporting the synthesis of qualitative research [ENTREQ].

https://doi.org/10.1371/journal.pone.0296384.s008

  • View Article
  • Google Scholar
  • PubMed/NCBI
  • 12. Noyes J, Booth A, Cargo M, Flemming K, Harden A, Harris J, et al. Chapter 21: Qualitative evidence. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA [editors]. Cochrane Handbook for Systematic Reviews of Interventions version 6.3 [updated February 2022]. Cochrane, 2022; Available from www.training.cochrane.org/handbook .

Breast Cancer Metastasis: Mechanisms and Therapeutic Implications

Affiliations.

  • 1 Department of Medical and Biological Sciences, The Catholic University of Korea, Bucheon 14662, Korea.
  • 2 Department of Biotechnology, The Catholic University of Korea, Bucheon 14662, Korea.
  • PMID: 35743249
  • PMCID: PMC9224686
  • DOI: 10.3390/ijms23126806

Breast cancer is the most common malignancy in women worldwide. Metastasis is the leading cause of high mortality in most cancers. Although predicting the early stage of breast cancer before metastasis can increase the survival rate, breast cancer is often discovered or diagnosed after metastasis has occurred. In general, breast cancer has a poor prognosis because it starts as a local disease and can spread to lymph nodes or distant organs, contributing to a significant impediment in breast cancer treatment. Metastatic breast cancer cells acquire aggressive characteristics from the tumor microenvironment (TME) through several mechanisms including epithelial-mesenchymal transition (EMT) and epigenetic regulation. Therefore, understanding the nature and mechanism of breast cancer metastasis can facilitate the development of targeted therapeutics focused on metastasis. This review discusses the mechanisms leading to metastasis and the current therapies to improve the early diagnosis and prognosis in patients with metastatic breast cancer.

Keywords: EMT; breast cancer; metastasis; tumor microenvironment.

Publication types

  • Breast Neoplasms* / genetics
  • Epigenesis, Genetic
  • Epithelial-Mesenchymal Transition
  • Lymph Nodes / pathology
  • Neoplasm Metastasis / pathology
  • Neoplasms, Second Primary* / pathology
  • Tumor Microenvironment

Grants and funding

  • M2022B002600003/Korea Ministry of Education

Advertisement

Issue Cover

  • Previous Article
  • Next Article

Introduction

Materials and methods, authors’ disclosures, authors’ contributions, acknowledgments, estrogen receptor mutations as novel targets for immunotherapy in metastatic estrogen receptor–positive breast cancer.

ORCID logo

J. Goldberg, N. Qiao, G. Alatrash, and E.A. Mittendorf contributed equally to this article.

  • Funder(s):  Parker Institute for Cancer Immunotherapy (PICI)
  • Award Id(s): 6325701 / 2018-1922 – PICI Contract #C-00462
  • Principal Award Recipient(s): Elizabeth A.   Mittendorf
  • Funder(s):  Rob and Karen Hale
  • Funder(s):  HMS | Ludwig Center at Harvard (Ludwig Center)

Attribution

  • Split-Screen
  • Article contents
  • Figures & tables
  • Supplementary Data
  • Peer Review
  • Open the PDF for in another window
  • Get Permissions
  • Cite Icon Cite
  • Search Site
  • Version of Record February 22 2024
  • Accepted Manuscript February 9 2024

Jonathan Goldberg , Na Qiao , Jennifer L. Guerriero , Brett Gross , Yagiz Meneksedag , Yoshimi F. Lu , Anne V. Philips , Tasnim Rahman , Funda Meric-Bernstam , Jason Roszik , Ken Chen , Rinath Jeselsohn , Sara M. Tolaney , George E. Peoples , Gheath Alatrash , Elizabeth A. Mittendorf; Estrogen Receptor Mutations as Novel Targets for Immunotherapy in Metastatic Estrogen Receptor–positive Breast Cancer. Cancer Research Communications 1 February 2024; 4 (2): 496–504. https://doi.org/10.1158/2767-9764.CRC-23-0244

Download citation file:

  • Ris (Zotero)
  • Reference Manager

Estrogen receptor–positive (ER + ) breast cancer is not considered immunogenic and, to date, has been proven resistant to immunotherapy. Endocrine therapy remains the cornerstone of treatment for ER + breast cancers. However, constitutively activating mutations in the estrogen receptor alpha ( ESR1 ) gene can emerge during treatment, rendering tumors resistant to endocrine therapy. Although these mutations represent a pathway of resistance, they also represent a potential source of neoepitopes that can be targeted by immunotherapy. In this study, we investigated ESR1 mutations as novel targets for breast cancer immunotherapy. Using machine learning algorithms, we identified ESR1 -derived peptides predicted to form stable complexes with HLA-A*0201. We then validated the binding affinity and stability of the top predicted peptides through in vitro binding and dissociation assays and showed that these peptides bind HLA-A*0201 with high affinity and stability. Using tetramer assays, we confirmed the presence and expansion potential of antigen-specific CTLs from healthy female donors. Finally, using in vitro cytotoxicity assays, we showed the lysis of peptide-pulsed targets and breast cancer cells expressing common ESR1 mutations by expanded antigen-specific CTLs. Ultimately, we identified five peptides derived from the three most common ESR1 mutations (D538G, Y537S, and E380Q) and their associated wild-type peptides, which were the most immunogenic. Overall, these data confirm the immunogenicity of epitopes derived from ESR1 and highlight the potential of these peptides to be targeted by novel immunotherapy strategies.

Estrogen receptor (ESR1) mutations have emerged as a key factor in endocrine therapy resistance. We identified and validated five novel, immunogenic ESR1-derived peptides that could be targeted through vaccine-based immunotherapy.

Approximately 70% of breast cancers express the estrogen receptor (ER), rendering them susceptible to endocrine therapy ( 1, 2 ). Despite the initial response to endocrine therapy, many patients with metastatic disease eventually progress on endocrine therapy ( 3 ). Endocrine resistance evolves through many mechanisms including genetic dysregulation, posttranslational modifications, altered cell signaling, ligand-independent activation of the ER, and decreased sensitivity to antiestrogens. A major mechanism of resistance is mutation of the ERα gene ESR1 , which occurs in approximately 30% of patients after treatment with aromatase inhibitors (AI; refs. 3–7 ). Several ESR1 mutations have been identified, the majority of which occur in the ligand-binding domain ( 6 ). The most common mutations, D538G, Y537S, and E380Q, are associated with poor therapeutic response and overall survival ( 3, 5, 8 ). In addition, preclinical data show that these mutations are drivers of metastases, and thus targeting these mutations could lead to improved outcomes ( 9 ). In this study, we aimed to generate and validate an immune-based approach for targeting these mutations that could lead to improved outcomes and restore endocrine sensitivity.

There is growing interest in using tumor-expressed mutations to inform immunotherapeutic approaches. The total number of mutations present within a tumor [i.e., tumor mutational burden (TMB)] is an emerging biomarker that has been shown to independently predict the response to immunotherapy in multiple cancer types ( 10–12 ). However, the TMB of breast cancer is relatively low, especially in ER-positive (ER + ) breast cancer ( 13, 14 ). However, specific mutations can lead to the generation of neoantigens that are expressed by malignant cells and are critical for distinguishing tumors from normal cells. Tumor-expressed mutations have been exploited as potential targets for T cell–based and vaccine-based approaches. In addition to mutated epitopes, several studies ( 15, 16 ), including our work ( 17–20 ), have shown that non-mutated self-antigens are effective therapeutic targets. Importantly, ESR1 expression is higher in breast cancer tissues than in other malignant and healthy tissues, including healthy breast tissues. Therefore, our interest in ER as a target for immunotherapy stems from the fact that wild-type ESR1 itself can be a source of self-antigens and ESR1 can be mutated, making it a potential source of neoantigens.

Using publicly available machine learning algorithms, we identified novel peptides derived from ESR1 that can present to CD8 + T cells. To validate these candidate peptides, we used in vitro T2 binding assays to measure peptide affinity for HLA-A*0201, and dissociation assays to measure the stability of these complexes on the cell surface. Next, we focused on the ability of high-affinity peptides to induce an endogenous cytolytic CD8 + T-cell response in healthy volunteers. Overall, we identified five novel peptides derived from wild-type and mutant ESR1 with a high affinity for HLA-A*0201 that elicited a strong CD8 + T-cell response. Together, our data support the potential of ESR1 candidate peptides as T-cell targets in vaccine or adoptive T-cell therapy approaches.

Peptide Preparation

ESR1 gene sequencing was performed using the NIH gene database NM_000125.4, and clinically relevant ESR1 mutations were confirmed from previous studies. To identify potential HLA-A*0201 binding peptides within ESR1 mutation sites, the Immune Epitope Database and Analysis Resource (IEDB) at iedb.org was used ( 21 ), which has been used in our previous work and in similar investigations into ESR1 peptides ( 22 ). Candidate peptides with low IC 50 scores (<500 nmol/L) predicted by ANN, SMM, and NetMHCpan algorithms, along with those peptides validated through the overlapping peptide approach, were synthesized at Bio-Synthesis and processed to at least 95% purity by reverse-phase high-performance liquid chromatography and confirmed by mass spectrometry. Influenza peptide (FLU: GILGFVFTL) was used as a positive control, and each mutant peptide's corresponding wild-type sequence was used as a negative control. Lyophilized peptides were dissolved in PBS containing 5% DMSO at a concentration of 1 mg/mL and stored at −80°C.

Cell Lines and Flow Cytometry Analysis

The T2 and MCF7 cell lines were purchased from the ATCC. Prior to use, these cell lines were validated using short tandem repeat DNA fingerprinting at the MD Anderson Cancer Center sequencing facility. MCF7-mutant cell lines were gifted from Dr. Rinath Jeselsohn. After thawing, cells were used for experiments within 5 passages. Mycoplasma testing was performed quarterly using the MycoAlert Kit (Lonza Inc.). All cell lines were maintained in RPMI1640 media supplemented with 10% heat-inactivated FBS (Gemini BioProducts), 100 U/mL penicillin, and 100 µg/mL streptomycin (Gibco-Invitrogen). All cells were cultured at 37°C and 5% CO 2 . Cells were pulsed with or without peptides at designated timepoints, harvested, and stained with a FITC-conjugated anti-HLA-A2 mAb (BB7.2, BD Biosciences). Live/dead Aqua staining (Life Technologies) was used to assess the cell viability. Flow cytometry was performed using a BD LSRFortessa flow cytometer (BD Biosciences), and the data were analyzed using FlowJo software (TreeStar Inc.).

Peptide Binding Assay and Stability Assay

T2 binding assays were performed as described previously ( 23 ). Briefly, to test peptide binding affinity to HLA-A*0201 molecules, T2 cells were incubated with 40 µg/mL of mutant, wild-type, or control peptides for 4 hours at 37°C in RPMI1640 media supplemented with 0.5% FBS. Surface expression of HLA-A*0201 in T2 cells was determined by staining with an anti-HLA-A2 mAb (BB7.2, BD Biosciences). The median fluorescence intensity (MFI) was measured using a FACSCando II (BD Biosciences). The peptide half-life was calculated by linear regression using GraphPad Prism software.

Peptide-specific CTL Expansion and Tetramer Assay

Healthy donor peripheral blood samples were purchased from the local blood bank and used under an MD Anderson Cancer Center Institutional Review Board–approved protocol. Because the samples came from deidentified healthy donors, the Institutional Review Board determined that informed consent was not required. Peripheral blood mononuclear cells (PBMC) were isolated from buffy coats of healthy female HLA-A2 positive donors by Ficoll/Histopaque (Sigma-Aldrich) density gradient centrifugation and split equally into two groups. One group of PBMC was used to assess the pre-expansion frequency of the peptide-specific CTLs. These cells were immediately stained with FITC-conjugated anti-CD3 and APC-H7–conjugated anti-CD8 to select for CD8 + T cells. Pacific blue (PB)-conjugated anti-CD4, anti-CD14, anti-CD16, anti-CD19, and anti-CD56 (BD Biosciences) were used as a dump gate to gate out other immune cell types; PE-conjugated mutant peptide and APC-conjugated wild-type peptide tetramers (Baylor College of Medicine, MHC tetramer core, Houston, TX) were used to enumerate the percentage of peptide-specific CD8 + T cells present in healthy PBMCs. The analysis was performed using flow cytometry.

The second group of donor PBMC was used to assess the expansion potential of the peptide-specific CTLs. To expand peptide-specific CTLs, dendritic cells (DC) were matured from adherent monocytes by the addition of GMCSF (100 ng/mL, Sanoti), IL4 (50 ng/mL, rhIL4, Tonbo Biosciences), and TNFα (25 ng/mL, BioLegend). Lymphocytes from the same donor were separated and cocultured with 40 µg/mL of mutant and wild-type peptides. After 5 days, DCs were harvested and cocultured with autologous CD8 + T cells for an additional 7 days. The cocultured cells were supplemented with IL7 (10 ng/mL; BioLegend) for CTL activation and IL2 (25 ng/mL; R&D Systems) for CTL expansion. After CTL activation and expansion, cells were stained and analyzed as described above.

Peptide-specific CTLs Cytotoxicity Assay

Healthy female donor lymphocytes and DCs were cocultured in 6-well plates supplemented with IL7 and IL2, as described above. On days 12–14, cultured peptide-specific CTLs were harvested and peptide-specific cytotoxicity was assessed using a standard 4-hour Calcein-AM (Sigma-Aldrich) release assay as described previously ( 24 ). Briefly, target cells, including T2 cells pulsed with mutant and wild-type peptides, were labeled with 5 µg/mL calcein-AM for 15 minutes in the dark. After two washes, cells were adjusted to 2 × 10 5 /mL, plated on a Terasaki plate, and cocultured with peptide-specific CTLs at effector:target ratios of 20:1 and 5:1. After 4 hours of incubation, Trypan blue was added to quench the reaction. The fluorescence was measured using a Cytation 3 Imaging Reader (BioTek). The percentage of specific cell lysis was calculated using the following formula:

formula

Statistical Analysis

Statistical analyses were performed using the GraphPad Prism, version 8.0. Data were compared using paired Student t test. A probability of less than 0.05 was considered significant at *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Ethical Compliance

All procedures performed in this study involving human participants were in accordance with the ethical standards of the Institutional and/or National Research Committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Data Availability

The data generated in this study are available upon request from the corresponding author.

Multiple HLA-A*0201 Binding Epitopes are Identified

We hypothesized that ESR1 mutations would generate novel peptides with high binding affinities for HLA class I molecules. To determine which epitopes derived from within ESR1 mutation sites would have a high propensity to bind to HLA-A*0201 (IC 50 < 500 nmol/L; ref. 25 ), IEDB-binding algorithms were used to identify 10 nonameric/decameric peptide sequences containing ESR1 mutations, as well as their associated wild-type peptides ( Supplementary Table S1 ). These peptides had a predicted IC 50 of <500 nmol/L using at least two separate algorithms. We validated multiple ESR1 peptides in vitro , including a peptide derived from the E380Q mutation. However, IEDB-predicted peptides for the two most common mutations, D538G and Y537S, failed downstream biological validation. Given the frequency of D538G and Y537S mutations and their potential clinical value as immunotherapy targets, we next pursued an overlapping peptide approach in which we tested every octameric, nonameric, and decameric permutation of peptides that included D538G and Y537S point mutations. Although IC 50 < 500 nmol/L is the conventional cutoff for predicting avid peptide binding to HLA-A201 and effective T-cell recognition ( 26 ), a number of peptides, including MART1, gp100, and GP2, have been shown to be effective immunotherapy targets despite a predicted IC 50 > 500 nmol/L (refs. 27–29 ; Supplementary Table S2 ). Ultimately, in our analysis, nonameric peptides were predicted to be better binders than octomeric or decameric peptides for both Y537S and D538G. Thus, we proceeded with nonameric overlapping peptides for both Y537S and D538G mutations ( Supplementary Table S3 ). The top three peptides for the Y537S and D538G mutations were selected on the basis of the predicted IC 50 (lowest IC 50 from an individual algorithm) and subsequently validated using in vitro T2 binding assays ( Supplementary Fig. S1 ). By analyzing the overlapping peptides, we established a high-affinity peptide for both D538G and Y537S.

Although our initial in silico approach yielded several high-affinity peptides derived from multiple ESR1 mutations, we focused on targeting mutations that are consistently and frequently identified in clinical settings. Analysis of data derived from the MSK-IMPACT, PALOMA-3, SOFEA, and FERGI trials revealed that 60%–80% of ESR1 mutations involve E380Q, Y537S, and D538G (ref. 30 ; Supplementary Table S4 ). In addition to the mutated peptides, we also included their respective wild-type peptides to serve both as a negative control and a potential immunotherapeutic target, as ESR1 is a highly enriched tumor-associated antigen (TAA) in ER + breast tumors ( Supplementary Fig. S2 ). Because Y537S and D538G are derived from the same wild-type sequence, only one wild-type control was required. Therefore, we ultimately established five novel ESR1-derived peptides for downstream validation: three from the most common ESR1 mutations and two from their associated wild-type sequences ( Table 1 ).

Identification of candidate ESR1-derived peptides

Candidate Peptides Bind to HLA-A*0201 with High Affinity and Stability

To verify the HLA-A*0201 binding affinity of the five candidate peptides, T2 binding assays were performed. In comparison with non-pulsed T2 cell controls, the five peptides showed an average fold change in MFI of 1.5 (range: 1.38–1.69), indicating high-affinity peptide binding to surface HLA-A*0201 on T2 cells ( Fig. 1 ). Importantly, previous studies have identified peptide affinity measured using biological assays as the strongest predictor of a robust CD8 + T-cell response ( 31, 32 ). T2 dissociation assays were used to determine the half-life and dissociation rates of peptides from HLA-A*0201 molecules. The results showed that these five peptides had half-lives ranging from 5.4 to 7.2 hours ( Fig. 2 ; Table 2 ). There was a trend toward a shorter half-life in the wild-type peptides compared with the mutant peptides; however, all peptides showed a half-life of over 5 hours, which has been associated with an increased likelihood of immunogenicity ( 33–35 ). Overall, these data provide biological evidence that the predicted peptides can bind HLA-A*0201 with both high affinity and prolonged dissociation kinetics.

Peptides bind to HLA-A*0201. T2 cells were pulsed with indicated peptide for 4 hours and surface HLA-A*0201 was stained. MFI of pulsed T2 cells were normalized to experimentally matched non-pulsed T2 cells, and data expressed as fold change in MFI. Dashed horizontal line indicates non-pulsed T2 HLA-A*0201. FLU peptide was used as a positive control HLA-A*0201 peptide known to bind with high affinity. Data represent three independent experiments performed in triplicate. Statistical significance was determined via comparison of pulsed MFI versus non-pulsed raw MFI using unpaired Student t test. ***, P < 0.001.

Peptides bind to HLA-A*0201. T2 cells were pulsed with indicated peptide for 4 hours and surface HLA-A*0201 was stained. MFI of pulsed T2 cells were normalized to experimentally matched non-pulsed T2 cells, and data expressed as fold change in MFI . Dashed horizontal line indicates non-pulsed T2 HLA-A*0201. FLU peptide was used as a positive control HLA-A*0201 peptide known to bind with high affinity. Data represent three independent experiments performed in triplicate. Statistical significance was determined via comparison of pulsed MFI versus non-pulsed raw MFI using unpaired Student t test. ***, P < 0.001.

Kinetics of peptide: HLA-A*0201 dissociation. T2 cells were pulsed with indicated peptides for 8–12 hours, washed, and then stained for surface HLA-A*0201 expression at indicated timepoints for wildtype-1 and E380Q (A), wildtype-2 and Y537S (B), wildtype-2 and D538G (C). MFI for each timepoint was normalized to initial staining (time = 0). Linear regression analysis was used to plot lines of best fit for each individual peptide, and half-life for each peptide is shown in Table 2. Data are pooled from three independent experiments performed in triplicate.

Kinetics of peptide: HLA-A*0201 dissociation. T2 cells were pulsed with indicated peptides for 8–12 hours, washed, and then stained for surface HLA-A*0201 expression at indicated timepoints for wildtype-1 and E380Q ( A ), wildtype-2 and Y537S ( B ), wildtype-2 and D538G ( C ). MFI for each timepoint was normalized to initial staining (time = 0). Linear regression analysis was used to plot lines of best fit for each individual peptide, and half-life for each peptide is shown in Table 2 . Data are pooled from three independent experiments performed in triplicate.

Stability of candidate peptides to HLA-A*0201

ESR1 Peptide CTLs Can be Expanded from Healthy Female Donors

Having shown that these peptides can be presented for immune recognition complexed with HLA-A*0201, we next tested whether a CD8 + T-cell response that recognizes these specific peptide–HLA complexes can be detected in the peripheral blood. To confirm the immune recognition ability of the five candidate peptides, peptide-specific tetramer staining was performed on peripheral blood samples from multiple healthy female donors. These data revealed an average frequency of E380Q (0.03%), Y537S (0.18%), and D538G (0.17%)-specific CTLs in the peripheral blood of healthy females ( Fig. 3 ), which is comparable to the precursor frequency of well-established, clinically relevant peptide tumor targets ( 36 ). Next, we aimed to determine whether peptide-specific CTLs can be expanded from healthy female donors. CTL expansion and subsequent tetramer staining revealed that mutant peptide-specific CTLs could be expanded, as shown by the average frequency fold changes (vs. pre-expansion) for E380Q (5.77-fold), Y537S (4.85-fold), and D538G (4.31-fold). Wildtype-1 did not show tetramer expansion, while wildtype-2 showed modest expansion compared with that of the mutant peptides. Overall, 4/5 candidate peptides yielded tetramer expansion, implying the potential of these peptides to elicit CD8 + T-cell immunity.

Peptide-specific expansion of CD8+ T cells. PBMCs from healthy donors were isolated and expanded as described in the Materials and Methods. Tetramer staining was performed pre- and post-expansion to quantify frequencies of peptide-specific CD8+ T cells. A, Wildtype-1, E380Q. B, Wildtype-2, Y537S, D538G. Data represents three independent experiments, from 3 separate donors, performed in triplicate. Statistical significance was determined via comparison of pre-expansion versus post-expansion CD8+tetramer+ population using unpaired Student t test. *, P < 0.05.

Peptide-specific expansion of CD8 + T cells. PBMCs from healthy donors were isolated and expanded as described in the Materials and Methods. Tetramer staining was performed pre- and post-expansion to quantify frequencies of peptide-specific CD8 + T cells. A, Wildtype-1, E380Q. B, Wildtype-2, Y537S, D538G. Data represents three independent experiments, from 3 separate donors, performed in triplicate. Statistical significance was determined via comparison of pre-expansion versus post-expansion CD8 + tetramer + population using unpaired Student t test. *, P < 0.05.

ESR1 Peptide-specific CTLs Lyse ESR1 Peptide Pulsed T2 Cells

After confirming high-affinity binding, long half-lives, and peptide-specific CTLs in peripheral blood, we next aimed to determine whether targets expressing the candidate peptides could be lysed by peptide-specific CTLs. Peptide-specific CTLs were isolated and expanded from healthy female donors. T2 cells were pulsed with peptides as target cells, and peptide-specific CTL-mediated lysis was assessed using a calcein-AM cytotoxicity assay. The data revealed that all peptide-specific CTLs from both mutant and wild-type peptides lysed the T2 target cells pulsed with the corresponding peptide at an increasing effector:target ratio ( Fig. 4 ). At 20:1 effector:target ratio, the E380Q, Y537S, and D538G peptides generated specific cytotoxicities of 28.45%, 30.03%, and 30.48%, respectively. Wildtype 1 and wildtype 2 also showed specific cytotoxicity of 25.99% and 13.27%, respectively. Taken together, these data confirm peptide-specific lysis, further validating the immunogenic potential of these five novel peptides.

Expanded CD8+ T cells demonstrate antigen-specific cytotoxicity. T2 cells were pulsed with corresponding peptides and labeled with calcein-AM. Cytotoxicity was determined by a standard calcein-AM release assay. Wildtype-1 (A), E380Q (B), Wildtype-2 (C), Y537S (D), D538G (E). Non-pulsed T2 cells were used as negative controls. Statistical significance was determined via unpaired Student t test. Data represent the average of three to four experiments from separate healthy female donors run in triplicate. *, P < 0.05; ***, P < 0.001.

Expanded CD8 + T cells demonstrate antigen-specific cytotoxicity. T2 cells were pulsed with corresponding peptides and labeled with calcein-AM. Cytotoxicity was determined by a standard calcein-AM release assay. Wildtype-1 ( A ), E380Q ( B ), Wildtype-2 ( C ), Y537S ( D ), D538G ( E ). Non-pulsed T2 cells were used as negative controls. Statistical significance was determined via unpaired Student t test. Data represent the average of three to four experiments from separate healthy female donors run in triplicate. *, P < 0.05; ***, P < 0.001.

Peptide-specific CTLs Effectively Kill ER + Breast Tumor Cells

We next aimed to identify whether breast cancer cells show similar cytotoxicity to peptide-specific CTLs as we demonstrated for peptide-pulsed T2 cells. Peptide-specific CTLs were isolated and expanded from healthy female donors. MCF7 wild-type cells were pulsed with peptides and served as the target cells. Peptide-specific CTL-mediated lysis was assessed using a calcein-AM cytotoxicity assay. Non-pulsed and CG1-peptide pulsed MCF7 cells were used as negative controls. In addition, mutant MCF7 cells for each of the three mutants, E380Q, Y537S, and D538G, were also evaluated for killing by corresponding mutant peptide-specific CTL ( Fig. 5 ). These data showed increased killing of peptide-pulsed MCF7 cell lines by the appropriate peptide-specific CTLs compared with controls. In addition, our data demonstrate cytotoxicity of mutant ESR1 cell lines, comparable to what was observed with peptide-pulsed wild-type MCF7 cell lines.

Peptide-specific CTLs effectively kill ER+ breast tumor cells. MCF7 cells were pulsed with corresponding peptides and labeled with calcein-AM. Cytotoxicity was determined by a standard calcein-AM release assay. MCF7 cells were pulsed with corresponding peptides. CG1-pulsed MCF7 cells and non-pulsed MCF7 cells were used as negative controls. Statistical significance was determined via one-way ANOVA. Data show percent cytotoxicity at the 10:1 effector:target ratio, and represent the average cytotoxicity of five experiments (run in triplicate) from different healthy female donors. *, P < 0.05; ***, P < 0.001.

Peptide-specific CTLs effectively kill ER + breast tumor cells. MCF7 cells were pulsed with corresponding peptides and labeled with calcein-AM. Cytotoxicity was determined by a standard calcein-AM release assay. MCF7 cells were pulsed with corresponding peptides. CG1-pulsed MCF7 cells and non-pulsed MCF7 cells were used as negative controls. Statistical significance was determined via one-way ANOVA. Data show percent cytotoxicity at the 10:1 effector:target ratio, and represent the average cytotoxicity of five experiments (run in triplicate) from different healthy female donors. *, P < 0.05; ***, P < 0.001.

In this study, we provided evidence that T-cell immunity can be generated against ESR1-derived peptides. We identified multiple candidate peptides using well-defined machine learning algorithms, and subsequently identified and systematically validated five ESR1-derived nonameric peptides that bind with high affinity and form a stable complex with HLA-A*0201. We then detected ESR1 peptide-specific CTLs in the peripheral blood of healthy females and demonstrated their expansion. Finally, using functional assays, we demonstrated that peptide-specific CTLs lysed peptide-pulsed T2 targets as well as MCF7 cell lines. Taken together, these data support further investigation of immunotherapeutic strategies, including vaccines or adoptive T-cell therapy approaches targeting ESR1.

Advances in early screening and extended regimens of endocrine therapy have partly led to improved outcomes in patients with ER + breast cancer. For postmenopausal patients, therapy with AIs for 5–10 years is standard ( 37 ). AIs, such as anastrozole, prevent the synthesis of estrogen and starve the ER of its ligand. Under the selective pressure of AIs, tumors are driven to develop endocrine resistance through multiple pathways, with ESR1 mutations being the most common genetically acquired pathway. In the metastatic setting, approximately 30% of patients who progress on AIs develop one or multiple ESR1 mutations ( 6 ). At least 62 ESR1 mutations have been identified; however, the majority of these mutations are rarely identified and are either inactive or yet to be characterized ( 6 ). Thus, we focused our investigations on the three most common ESR1 mutations (E380Q, Y537S, and D538G), which are robustly characterized and directly linked to poor outcomes ( 3, 8, 38 ). By focusing on the most common mutations, we anticipate that further development of immunotherapeutic strategies targeting these mutations will yield the greatest benefits to patients.

The E380Q, Y537S, and D538G mutations represent the source of neoantigens in the limited mutational landscape of ER + breast tumors. While neoantigen load is predictive of T-cell infiltration in multiple cancers, including breast cancer, Williams and colleagues found that ESR1 mutations did not increase endogenous CD8 + T-cell infiltration. Conversely, ER + tumors with other common mutations, such as TP53 and PIK3CA, show significantly elevated levels of CD8 + T-cell infiltration ( 39 ). One hypothesis for this lack of T-cell response to ESR1 mutations is the poor binding of endogenously processed peptides to MHC class I. Our initial in silico approach failed to identify high-affinity peptides for the Y537S and D538G mutations, which may indicate that this region of the ligand-binding domain has certain qualities that limit antigen presentation. Strong binding to MHC molecules relies on peptide length and specific anchor residues. For HLA-A2, peptides comprised of nine amino acids with leucine at anchor residue 2 confer enhanced binding ( 40 ). Each ESR1-mutated peptide presented here meets both of these criteria, which may, in part, explain their high binding affinity to HLA-A*0201.

As part of our initial experimental design, we aimed to use wild-type peptides as negative controls, from which we compared the corresponding mutant peptides. Interestingly, the two wild-type “controls” displayed similar binding characteristics to that of the mutant peptides and were similarly able to elicit a strong and specific T-cell response. Although neoepitopes, such as ESR1 mutations, are known for their enhanced immunogenicity, we hypothesize that the peptide sequences discovered in this study may have inherent properties, rendering them sufficiently immunogenic, with or without mutations. Moreover, highly overexpressed wild-type proteins can also be TAAs that are recognized by the immune system and have been successfully targeted in multiple cancer types ( 15–18 ). In ER + breast cancer, wild-type ERα is a highly enriched TAA. IHC studies have revealed significantly higher levels of ERα in malignant mammary epithelial cells than in benign mammary epithelial cells. Furthermore, Khan and colleagues found that ER positivity in benign mammary tissues was correlated with an increased risk of breast cancer ( 41, 42 ). On the basis of these data, in addition to mutated ESR1 peptides, we included their corresponding wild-type epitopes as potential immunotherapeutic targets. Importantly, these wild-type peptides had HLA-binding properties and immunogenic potential comparable to those of their respective mutant epitopes.

The selective ER degrader (SERD) fulvestrant, often used as a second-line therapy for metastatic ER + breast cancer, has a weaker binding affinity to mutated ER compared with wild-type receptors ( 38 ). The poor bioavailability of intramuscularly administered fulvestrant further limits the effective targeting of mutant ER. Importantly, newer agents are in development including oral SERDs, which have better absorption and increased bioavailability, enabling enhanced antagonist activity ( 43 ). In addition, there are selective ER modulators and SERDs that have been shown to have a structure that destabilizes mutant ER. ( 44 ) Small-molecule therapeutics, such as mTOR and CDK4/6 inhibitors, have shown some efficacy against ESR1-mutated tumors ( 6, 37 ); however, these therapies, combined with standard therapy, can lead to significant toxicities and disease progression is nearly always inevitable. The phase III PADA-1 trial demonstrated that switching to fulvestrant as the partner endocrine treatment with the CDK 4/6 inhibitor palbociclib was superior to continuing on an AI with palbociclib in the setting of rising ctDNA ESR1 mutations ( 45 ). These data provided evidence that better blockade of ESR mutations during treatment with endocrine therapy plus a CDK 4/6 inhibitor improved outcomes. Moreover, once patients progress with endocrine therapy, they are likely to cycle through other treatments, ultimately progressing to chemotherapy.

Our findings pave the way for the development of ESR1 targeted immunotherapy. One approach could be vaccine development, as we were able to expand CD8 + T cells to target these five novel peptides. Importantly, vaccination against immunogenic epitopes has been shown to successfully generate a CD8 + T-cell response against another cancer relevant protein, HER2. E75, a nanomeric peptide derived from HER2, has been used in multiple clinical trials for breast, ovarian, and gastric cancers, many of which were led by our group. In preclinical development, E75 displayed high affinity and stability for HLA-A*0201 and could be recognized by and prime CTLs ( 46 ). In a phase I study, vaccination with E75 in 14 patients with metastatic HER2 + breast cancer resulted in immunologic responses in the majority of patients ( 47 ). In a subsequent phase I/II trial enrolling 187 patients with high-risk HER2 + breast cancer, E75 vaccination in an adjuvant setting reduced the relative risk of recurrence by 48% ( 48 ). A subsequent multicenter phase III trial (NCT 01479244) targeting patients with HER2-low–expressing cancer (1–2+ by IHC) failed to demonstrate the clinical benefit of E75 vaccination as monotherapy ( 49 ). To explain the negative phase III trial data with the E75 vaccination, we hypothesized that a vaccine comprised of a single-epitope vaccine is insufficient to consistently generate an antitumor response. Therefore, in our current effort we are pursuing a multiepitope vaccine strategy that incorporates the five immunogenic epitopes identified in this study. The goal of such a vaccine strategy would be to educate CD8 + T cells to identify and eliminate circulating tumor cells that may contain ESR1 mutations which could potentially extend the length of time patients respond to endocrine therapy. Importantly, other groups are now investigating peptide vaccines targeting ESR1, including a phase I trial evaluating a vaccine comprised of peptides and an immunoadjuvant (either montanide or GMCSF; NCT04270149).

A second therapeutic approach would be cell-based treatment that utilizes T-cell receptor (TCR)-engineered T cells to target ESR1 peptides, as has been done for gp100 and MART-1 peptides in the setting of metastatic melanoma ( 50 ) and for MAGE-A4 peptide in the setting of esophageal cancer ( 7 ). With the knowledge of peptide–HLA complexes, a third approach could utilize a TCR mimic (TCRm) antibody to target ESR1 peptide–HLA complexes, as was shown preclinically in breast cancer models with E75 peptide targeting TCRm antibody ( 19 ) and in the setting of leukemia targeting WT1 and PR1 peptides with TCRm antibodies ( 51, 52 ).

In conclusion, we show the presence of immunogenic epitopes within the most common mutations of ESR1 as well as the wild-type ESR1 protein. CTL specific for these epitopes can recognize and lyse target cells expressing the mutations. A therapeutic strategy targeting these mutations with immunotherapy such as a multiepitope vaccine warrants further preclinical and clinical investigation.

J.L. Guerriero reports grants and personal fees from Array BioPharma/Pfizer, Duke Street Bio, GlaxoSmithKline; personal fees from AstraZeneca, BD Biosciences, Carisma, Codagenix, Kowa, Kymera, OncoOne; grants from Eli Lilly and Merck outside the submitted work. B. Gross reports other from TScan Therapeutics and Aldevron, LLC outside the submitted work. F. Meric-Bernstam reports personal fees from AbbVie, Aduro BioTech Inc., Alkermes, AstraZeneca, Biovica, Black Diamond, Calibr (a division of Scripps Research), Daiichi Sankyo Co. Ltd. , Dava Oncology, DebioPharm, Ecor1 Capital, eFFECTOR Therapeutics, Eisai, F. Hoffman-La Roche Ltd., FogPharma, Genentech Inc., GT Apeiron, Harbinger Health, IBM Watson, Immunomedics, Incyte, Infinity Pharmaceuticals, Inflection Biosciences, Jackson Laboratory, Karyopharm Therapeutics, Kolon Life Science, LegoChem Bio, Lengo Therapeutics, Loxo Oncology, Menarini Group, Mersana Therapeutics, OnCusp Therapeutics, OrigiMed, PACT Pharma, Parexel International, Pfizer Inc., Protai Bio Ltd, Puma Biotechnology Inc., Samsung Bioepis, Sanofi, Seattle Genetics Inc., Silverback Therapeutics, Spectrum Pharmaceuticals, Tallac Therapeutics, Theratechnologies, Tyra Biosciences, Xencor, Zymeworks, Zentalis; grants from Aileron Therapeutics, Inc. AstraZeneca, Bayer Healthcare Pharmaceutical, Calithera Biosciences Inc., Curis Inc., CytomX Therapeutics Inc., Daiichi Sankyo Co. Ltd., Debiopharm International, eFFECTOR Therapeutics, Genentech Inc., Guardant Health Inc., Klus Pharma, Takeda Pharmaceutical, Novartis, Puma Biotechnology Inc., Taiho Pharmaceutical Co.; and other from European Organization for Research and Treatment of Cancer (EORTC), European Society for Medical Oncology (ESMO), Cholangiocarcinoma Foundation, Dava Oncology outside the submitted work. R. Jeselsohn reports grants and personal fees from Pfizer, Lilly, Novartis, and personal fees from Carick Therapeutics and Genentech outside the submitted work. S.M. Tolaney reports grants, personal fees, and other from Novartis, Pfizer, Merck, Eli Lilly, AstraZeneca, Roche/Genentech, Eisai, Bristol Myers Squibb, Seattle Genetics, Gilead; personal fees and other from Sanofi, CytomX Therapeutics, Daiichi-Sankyo, Ellipses Pharma, 4D Pharma, OncoSec Medical, Inc., BeyondSpring Pharmaceuticals, OncXerna, Zymeworks, Zentalis, Blueprint Medicines, Reveal Genomics, ARC Therapeutics, Infinity Therapeutics, Myovant, Zetagen, Umoja Biopharma, Artios Pharma, Menarini/Stemline, Aadi Biopharma, Bayer, Incyte Corporation, Jazz Pharmaceuticals; grants from Exelexis, NanoString Technologies, and OncoPep outside the submitted work. G.E. Peoples reports personal fees from LumaBridge, Emtora Biosciences, and Elios Therapeutics outside the submitted work. E.A. Mittendorf reports grants from Parker Institute for Cancer Immunotherapy during the conduct of the study; personal fees from AstraZeneca, BioNTech, Exact Sciences, Merck, Moderna, Roche/Genentech; other from Bristol Myers Squibb, Roche/Genentech, Merck Sharp & Dohme, Roche/Genentech, and Gilead outside the submitted work; and receives research funding from Susan Komen for the Cure for which she serves as a scientific advisor and uncompensated participation as a member of the American Society of Clinical Oncology board of directors. No disclosures were reported by the other authors.

J. Goldberg: Data curation, formal analysis, investigation, methodology, writing-original draft, writing-review and editing. N. Qiao: Data curation, formal analysis, investigation, methodology, writing-review and editing. J.L. Guerriero: Data curation, formal analysis, writing-review and editing. B. Gross: Data curation, formal analysis, investigation, methodology, writing-original draft. Y. Meneksedag: Data curation, investigation. Y.F. Lu: Writing-review and editing. A.V. Philips: Writing-review and editing. T. Rahman: Writing-review and editing. F. Meric-Bernstam: Conceptualization, writing-review and editing. J. Roszik: Data curation, investigation. K. Chen: Data curation, investigation. R. Jeselsohn: Validation, writing-review and editing. S.M. Tolaney: Validation, writing-review and editing. G.E. Peoples: Conceptualization, validation, writing-review and editing. G. Alatrash: Conceptualization, formal analysis, supervision, validation, methodology, writing-original draft, project administration, writing-review and editing. E.A. Mittendorf: Conceptualization, resources, formal analysis, supervision, validation, methodology, writing-original draft, project administration, writing-review and editing.

This research was funded by the Parker Institute for Cancer Immunotherapy (grant 6325701/2018-1922—PICI contract no. C-00462; E.A. Mittendorf). E.A. Mittendorf is also supported by the Rob and Karen Hale Distinguished Chair in Surgical Oncology and the Ludwig Center at Harvard.

Note: Supplementary data for this article are available at Cancer Research Communications Online ( https://aacrjournals.org/cancerrescommun/ ).

Supplementary data

Kinetics of Overlapping peptide

Initial in-silico identification of 18 novel peptides with predicted high affinity to HLA-A*0201 (IC50<500 nM)

Expression of ESR1 in various Healthy and Malignant Tissues

Clinically Validated Peptides with IC50>500nM

Overlapping nonameric peptides in-silico prediction

Frequency of most common ESR1 mutations in key clinical studies

Citing articles via

Email alerts, affiliations.

  • Journal Archive
  • Journal Alerts
  • Info for Authors
  • Submit an Article
  • About the Journal
  • Online ISSN 2767-9764

AACR Journals

  • Blood Cancer Discovery
  • Cancer Discovery
  • Cancer Epidemiology, Biomarkers & Prevention
  • Cancer Immunology Research
  • Cancer Prevention Research
  • Cancer Research
  • Cancer Research Communications
  • Clinical Cancer Research
  • Molecular Cancer Research
  • Molecular Cancer Therapeutics
  • Info for Advertisers
  • Information for Institutions/Librarians

metastatic breast cancer research papers

  • Privacy Policy
  • Copyright © 2023 by the American Association for Cancer Research.

This Feature Is Available To Subscribers Only

Sign In or Create an Account

Skip to Content

  • Conquer Cancer
  • ASCO Journals
  • f Cancer.net on Facebook
  • t Cancer.net on Twitter
  • q Cancer.net on YouTube
  • g Cancer.net on Google

Types of Cancer

  • Navigating Cancer Care
  • Coping With Cancer
  • Research and Advocacy
  • Survivorship

Breast Cancer - Metastatic: Latest Research

ON THIS PAGE: You will read about the scientific research being done to learn more about this type of cancer and how to treat it. Use the menu to see other pages.

Doctors are working to learn more about metastatic breast cancer, including ways to prevent it, how to best treat it, and how to provide the best care to people diagnosed with this disease. The following areas of research may include new options for patients through clinical trials. Always talk with your doctor about the best diagnostic and treatment options for you.

Understanding the metastatic breast cancer process and finding ways to prevent and stop it.

Developing ways to best evaluate the genes and proteins contributing to cancer growth in each patient and each metastatic breast cancer to determine the best treatment options.

Testing new drugs and treatment combinations, such as new chemotherapies, hormonal treatments, and targeted therapies.

Studying the use of immunotherapy, which boosts the body’s natural defenses to fight cancer.

Determining the best ways to give hormonal therapy, and how to overcome tumor resistance to hormonal therapy.

Determining if surgical removal of the primary breast tumor can improve survival rates.

Testing various drugs for effectiveness against HER2-positive breast cancer that metastasizes to the brain.

Identifying new and effective treatments for triple-negative breast cancer.

Finding ways to check how well treatments are working against metastatic breast cancer and predict survival rates.

Learning more about the social and emotional factors that may affect patients’ treatment plans and quality of life.

Finding better ways of reducing symptoms and side effects of current metastatic breast cancer treatments to improve comfort and quality of life for patients and their caregivers.

Looking for More About the Latest Research?

If you would like more information about the latest areas of research in metastatic breast cancer, explore these related items that take you outside of this guide:

To find clinical trials specific to your diagnosis, talk with your doctor or search online clinical trial databases .

Visit the Cancer.Net Blog to review news and information about metastatic breast cancer, including research announced at recent scientific meetings or in the American Society of Clinical Oncology's (ASCO's) peer-reviewed journals.

Listen to podcasts from an ASCO expert discussing metastatic breast cancer research from 2021 and 2020 scientific meetings.

Get updates from Cancer.Net delivered right to your inbox. Subscribe to the Inside Cancer.Net email newsletter.

Visit the website of Conquer Cancer, the ASCO Foundation , to find out how to help support cancer research. Please note that this link takes you to a different ASCO website.

The next section in this guide is Palliative and Supportive Care . It describes how the symptoms and side effects of metastatic breast cancer and its treatment can be managed. Use the menu to choose a different section to read in this guide.

Breast Cancer - Metastatic Guide

Cancer.Net Guide Breast Cancer - Metastatic

  • Introduction
  • Risk Factors
  • Symptoms and Signs
  • Types of Treatment
  • About Clinical Trials
  • Latest Research
  • Palliative and Supportive Care
  • Coping with Treatment
  • Living with Metastatic Breast Cancer
  • Questions to Ask the Health Care Team
  • Additional Resources

View All Pages

Timely. Trusted. Compassionate.

Comprehensive information for people with cancer, families, and caregivers, from the American Society of Clinical Oncology (ASCO), the voice of the world's oncology professionals.

Find a Cancer Doctor

metastatic breast cancer Recently Published Documents

Total documents.

  • Latest Documents
  • Most Cited Documents
  • Contributed Authors
  • Related Sources
  • Related Keywords

Real-World Evidence of Trastuzumab, Pertuzumab, and Docetaxel Combination as a First-Line Treatment for Korean Patients with HER2-Positive Metastatic Breast Cancer

Survival in women with de novo metastatic breast cancer: a comparison of real-world evidence from a publicly-funded canadian province and the united states by insurance status.

Metastatic breast cancer (MBC) patient outcomes may vary according to distinct health care payers and different countries. We compared 291 Alberta (AB), Canada and 9429 US patients < 65 with de novo MBC diagnosed from 2010 through 2014. Data were extracted from the provincial Breast Data Mart and from the National Cancer Institute’s SEER program. US patients were divided by insurance status (US privately insured, US Medicaid or US uninsured). Kaplan-Meier and log-rank analyses were used to assess differences in OS and hazard ratios (HR) were estimated using Cox models. Multivariate models were adjusted for age, surgical status, and biomarker profile. No difference in OS was noted between AB and US patients (HR = 0.92 (0.77–1.10), p = 0.365). Median OS was not reached for the US privately insured and AB groups, and was 11 months and 8 months for the US Medicaid and US uninsured groups, respectively. The 3-year OS rates were comparable between US privately insured and AB groups (53.28% (51.95–54.59) and 55.54% (49.49–61.16), respectively). Both groups had improved survival (p < 0.001) relative to the US Medicaid and US uninsured groups [39.32% (37.25–41.37) and 40.53% (36.20–44.81)]. Our study suggests that a universal health care system is not inferior to a private insurance-based model for de novo MBC.

Assessment of Role of Platelet Aggregation in Metastatic Breast Cancer Patients

Background: To assess role of platelet aggregation in metastatic breast cancer patients.Methods:40 cases (Group I) of metastatic breast cancer patients and equal number of healthy control (Group II) subjects were included. Platelet aggregation studies in vitro using ADP and Thrombin were performed using an optical aggregometer. Detection of platelet aggregation was done by Chrono log series 490 dual and four channel optical aggregometer systems.Results:There were 4 subjects in group I and 12 in group II having ADP <60, 26 subjects in group I and 28 in group II with ADP 61-72 and 10 subjects in group I with ADP >72. Low thrombin <58 was seen in 8 in group II, normal thrombin between 61-72 was seen among 11 in group I and 32 in group II and high thrombin >82 among 29 in group I respectively. Amongst patients with normal platelet count, 14 patients had platelet aggregation with ADP in the normal range and 4 patients had platelet aggregation with ADP in the lower range. In patients with high platelet count, 12 showed aggregation in the normal range, and 10 patients showed aggregation in the higher range which was statistically significant (P< 0.05) (Table III, Graph II).Conclusion: Platelet aggregation has an important part to play in the tumor metastasis of breast cancer patients.

Transforming Passive into Active: Multimodal Pheophytin‐Based Carbon Dots Customize Protein Corona to Target Metastatic Breast Cancer

Aiming at a tailored cure for erbb2-positive metastatic breast cancer, multiplatform analysis of primary and metastatic breast tumors from the aurora us network identifies microenvironment and epigenetics drivers of metastasis.

Abstract Patients with metastatic breast cancer (MBC) typically have short survival times and their successful treatment represents one of most challenging aspects of patient care. This poor prognostic behavior is in part due to molecular features including increased tumor cell clonal heterogeneity, multiple drug resistance mechanisms, and alterations of the tumor microenvironment. The AURORA US Metastasis Project was established with the goal to identify molecular features specifically associated with metastasis. We therefore collected and molecularly characterized specimens from 55 metastatic breast cancer (BC) patients representing 51 primary cancers and 102 metastases. The 153 unique tumors were assayed using RNAseq, tumor/germline DNA exomes and low pass whole genome sequencing, and global DNA methylation microarrays. We found intrinsic molecular subtype differences between primary tumors and their matched metastases to be rare in triple negative breast cancer (TNBC)/Basal-like subtype tumors. Conversely, tumor subtype changes were relatively frequent in estrogen receptor positive (ER+) cancers where ~30% of Luminal A cases switched to Luminal B or HER2-enriched (HER2E) subtypes. Clonal evolution studies identified changes in expression subtype coincident with DNA clonality shifts, especially involving HER2 amplification and/or the HER2E expression subtype. We further found evidence for ER-mediated downregulation of genes involved in cell-cell adhesion in metastases. Microenvironment differences varied according to tumor subtype where ER+/Luminal metastases had lower fibroblast and endothelial cell content, while TNBC/Basal-like metastases showed a dramatic decrease in B cells and T cells. In 17% of metastatic tumors, we identified DNA hypermethylation and/or focal DNA deletions near HLA-A that were associated with its’ significantly reduced expression, and with lower immune cell infiltrates. We also identified low immune cell features in brain and liver metastases when compared to other metastatic sites, even within the same patient. These findings have direct implications for the treatment of metastatic breast cancer patients with immune- and HER2-targeting therapies and suggest potential novel therapeutic avenues for the improvement of outcomes for some patients with MBC.

Preliminary Clinical Validation of a Filtration-Based CTC Assay for Tumor Burden and HER2 Status Monitoring in Metastatic Breast Cancer

<b><i>Background:</i></b> Bearing multidimensional tumor-relevant information ranging from genomic alterations to proteomic makeup, circulating tumor cells (CTCs) constitute a promising material for liquid biopsy. The clinical validity of CTCs has been most extensively studied in metastatic breast cancer (MBC). The Cellsearch assay is currently the most widely used, while alternative strategies are pursued. A filtration-based microfluidic device has been described for CTC enrichment, but its clinical relevance remains unknown. <b><i>Methods:</i></b> In this preliminary study, we prospectively enrolled 47 MBC patients and evaluated the performance of the abovementioned CTC assay for tumor burden monitoring and human epidermal growth factor receptor 2 (HER2) status determination. <b><i>Results:</i></b> At baseline, 51.1% patients (24/47) were CTC positive. CTC count and positivity were also significantly higher in samples that accompanied poorer radiographic response evaluations. Serial blood draws suggested that CTC count enabled more accurate monitoring of tumor burden than serum markers carcinoembryonic antigen and cancer antigen 15-3. Also, in contrast to previous reports, CTC-HER2 status was moderately consistent with tumor-HER2 status. CTC-HER2 status assessment was further supported by <i>HER2</i> copy number measurements in select samples. <b><i>Conclusion:</i></b> The preliminary results from this study suggest promise for the interrogated CDC assay in several aspects, including sensitive CTC detection, accurate disease status reflection, and HER2 status determination. More studies are warranted to validate these findings and further characterize the value of CTC assay.

Outcome in patients of hormone receptor (HR) positive (Her 2) negative metastatic breast cancer treated with palbociclib – A real-world experience

Objectives: We present real-world outcome with the use of palbociclib in patients with HR-positive Her2-negative breast cancer treated at single center in India. Material and Methods: We conducted a medical audit of consecutive patients with HR-positive Her2-negative metastatic breast cancer, who were treated with palbociclib at our center between November 2016 and May 2020. Palbociclib was commenced at a dose of 125 mg orally once daily and a schedule of 21 days on therapy followed by 7 days off therapy was followed. Survival analysis included the Kaplan–Meier method using Statistical Package for the Social Sciences software (Version 26). HRs were calculated using Cox proportional hazard regression models and 95% confidence intervals (CIs) for the incidence estimates. Results: A total of 67 female patients were commenced on treatment with palbociclib between November 2016 and May 2020. The median age was 55 years (range 29–78 years). A total of 51 (76%) of these patients were postmenopausal and the remaining 16 were premenopausal. Baseline metastatic disease involved one organ/site in 23 (34%), two organs/sites in 32 (48%), three or more in 12 (18%). Bony metastasis alone was seen in 17 (25%) patients, visceral alone in 30 (45%), and the remaining 20 had both bony and visceral metastases. For these 67 patients, palbociclib was commenced as 1st line systemic therapy in 24 (36%) cases. Amongst the remaining 43 cases, it was 2nd line in 21 (31%); 3rd line and beyond in 22 (33%). Median PFS was 16.1 months (95% CI: 9.6–22.8) and median OS was 20.7 months (95% CI: 14.1–27.3). Median PFS for palbociclib use in first line was 18.7 months (95% CI: 4.6–32.9) while in subsequent lines, it was 13.8 months (95% CI: 9.8–17.9; log-rank P = 0.228). Median OS in patients who received palbociclib in first line was 23.2 months (95 % CI 20.1–26.3) and for those why received it in subsequent lines was 16.3 months (95 % CI: 12.5–20.1; P = 0.069). In total population, best response on imaging was CR in 11 (16%) cases (06 in 1st line setting and 05 in subsequent line setting); PR in 33 (49%); SD in 03; and progressive disease in 20. Median PFS with bone only metastasis: 20.9 months (95 % CI: 5.9–36.0), while with visceral metastasis 16.1 months (95% CI: 9.8–22.5; P = 0.537). Median OS with bone only metastasis: 22.7 months (95% CI: 17.8–27.5), while with visceral metastasis, it was 18.5 months (95% CI: 13.6–23.4; P = 0.314). Conclusion: Palbociclib is a useful addition in the management of HR +ve Her2 –ve breast cancer patients. Its benefit is confirmed in our real-world setting, both in the first and subsequent lines of therapy and the data are on similar lines as the global real-world data on palbociclib effectiveness.

Impact of body mass index on the efficacy of aromatase inhibitors in patients with metastatic breast cancer

Response monitoring in metastatic breast cancer: a comparison of survival times between fdg-pet/ct and ce-ct, export citation format, share document.

Metastatic Cancer Research

FDA has approved the combination of the targeted drugs dabrafenib (Tafinlar) and trametinib (Mekinist) for nearly any type of advanced solid tumor with a specific mutation in the BRAF gene. Data from the NCI-MATCH trial informed the approval.

Cancer often spreads to the lymph nodes, but it has never been clear why. A new study in mice suggests lymph node invasion helps the primary tumor spread, or metastasize, to other organs.

New treatments are helping more people with advanced or metastatic cancer live longer. At a recent NCI conference, survivors and researchers came together to discuss how to better address the needs of those living with metastatic cancer.

In an NCI study, treating mice with engineered immune cells shrank tumors and prevented the cancer from spreading to other parts of the body. The immunotherapy approach shows promise as a potential treatment for metastatic cancer.

A recent study quantified the risk of osteonecrosis of the jaw for patients who take zoledronic acid to manage complications from cancer that has spread to the bone. The study also examined risk factors for osteonecrosis of the jaw in these patients.

For some patients with painful spinal metastases from advanced cancer, a type of precise, high-dose radiation therapy—called stereotactic body radiation therapy (SBRT)—may be a highly effective way to relieve that pain, clinical trial results show.

People with oligometastatic cancer have only a few metastatic tumors. Researchers are studying whether treating these individual tumors directly with surgery or stereotactic body radiation therapy (SBTR or SABR) can help patients live longer or improve their quality of life.

Melanoma cells that pass through the lymphatic system before entering the bloodstream are more resistant to cell death and spread more readily than cells that enter the bloodstream directly. The finding could lead to new treatment approaches.

In people with cancer, the abscopal effect occurs when radiation—or another type of localized therapy—shrinks a targeted tumor but also causes untreated tumors in the body to shrink. Researchers are trying to better understand this phenomenon and take advantage of it to improve cancer therapy.

Brain cells called astrocytes can activate PPAR-gamma, a growth protein in cancer cells that helps them gain a foothold in the brain, a new study shows. The findings suggest that drugs that block PPAR-gamma activity may help treat brain metastases.

FDA has approved entrectinib (Rozlytrek) for the treatment of children and adults with tumors bearing an NTRK gene fusion. The approval also covers adults with non-small cell lung cancer harboring a ROS1 gene fusion.

New findings from a clinical trial suggest that a single dose of radiation therapy may control painful bone metastases as effectively as multiple lower doses of radiation therapy.

The NCI-MATCH precision medicine clinical trial has reached a milestone with the release of results from several study treatment arms. Findings from three arms were released at the 2018 ASCO annual meeting, adding to findings from one arm released in 2017.

Researchers have struggled to develop therapies to treat tumors that have spread to other parts of the body. In a new study, researchers tested whether the experimental drug metarrestin can selectively shrink metastases in mouse models of aggressive pancreatic cancer.

NCI’s Dr. Rosandra Kaplan discusses important trends in metastatic cancer research and new ideas for treating and preventing metastatic cancer.

Researchers have used modified stem cells to deliver a cancer drug selectively to metastatic breast cancer tumors in mice. The stem cells target metastatic tumors by homing in on the stiff environment that typically surrounds them.

Researchers at Memorial Sloan Kettering Cancer Center have reported the results of an initiative to characterize the genetic mutations in tumors from more than 10,000 patients with advanced cancer treated at the center.

In a clinical trial involving patients with metastatic cancer, administration of zoledronic acid every 12 weeks was as effective at preventing skeletal-related events caused by bone metastases as administration every 4 weeks.

Cancer cells may exploit a normal function of neutrophils, the most common form of white blood cell, to help form metastatic tumors, a new study suggests.

Researchers have identified proteins that may regulate the movement of breast cancer cells into and out of bone marrow.

In some patients with cancer that has spread to the brain, whole brain radiation following radiosurgery causes more severe cognitive decline and does not improve survival compared with radiosurgery alone, a new study has found.

Cardiac tumors that originate in the heart itself are extremely rare. The highly specialized and most abundant cell in the heart may explain why the organ is such an inhospitable host to cancer.

  • See us on facebook
  • See us on twitter
  • See us on youtube
  • See us on linkedin
  • See us on instagram

Metastatic breast cancer treatments have aided decline in deaths, Stanford Medicine-led study finds

Treatment of metastatic disease is responsible for nearly one-third of the decrease in annual deaths from breast cancer from 1975 to 2019, according to a Stanford Medicine-led study.

January 17, 2024 - By Krista Conger

breast cancer

Stanford Medicine researchers found that deaths from breast cancer have dropped by more than half since 1975 due to screening and treatment advancements.  Emily Moskal

Deaths from breast cancer dropped 58% between 1975 and 2019 due to a combination of screening mammography and improvements in treatment, according to a new multicenter study led by Stanford Medicine clinicians and biomedical data scientists.

Nearly one-third of the decrease (29%) is due to advances in treating metastatic breast cancer —a form that has spread to other areas of in the body and is known as stage 4 breast cancer or recurrent cancer. Although these advanced cancers are not considered curable, women with metastatic disease are living longer than ever.

The analysis helps cancer researchers assess where to focus future efforts and resources.

“We’ve known that deaths from breast cancer have been decreasing over the past several decades, but it’s been difficult or impossible to quantify which of our interventions have been most successful, and to what extent,” said Jennifer Caswell-Jin , MD, assistant professor of medicine. “This type of study allows us to see which of our efforts are having the most impact and where we still need to improve.”

Caswell-Jin and former research assistant Liyang Sun are co-first authors of the study , which was published Jan. 16 in the Journal of the American Medical Association . Sylvia Plevritis , PhD, professor and chair of biomedical data science, and Allison Kurian , MD, MSc, professor of medicine and of epidemiology and population health, are co-senior authors.

A collaboration

The study was a collaborative effort by a national consortium of researchers called CISNET , or the Cancer Intervention and Surveillance Modeling Network. CISNET was established in 2000 by the National Cancer Institute to understand the impact of cancer surveillance, screening and treatment on incidence and mortality. Doing so requires sophisticated computer algorithms capable of modeling the natural course of the disease and the typical treatment paths of individual patients, then translating that information to population-level data collected by the national Surveillance, Epidemiology, and End Results Program, or SEER registry , from 1975 to 2019.

Jennifer Caswell-Jin

Jennifer Caswell-Jin

The study is the third in a trio of papers from CISNET published since 2005 that assess the relative contributions of regular screening and treatment advances on breast cancer deaths. The previous two papers informed national guidelines and helped cancer researchers focus their efforts on the most intractable problems.

“Twenty years ago, there was a question whether routine screening mammography actually decreased the number of deaths from breast cancer,” Plevritis said. But in 2005, she and other CISNET researchers published a paper in the New England Journal of Medicine that conclusively demonstrated that screening was responsible for anywhere from 28% to 65% (different models came up with varying degrees of impact) of the reduction in mortality by 2000 between 1975 and 2000.

The second paper, published in 2018 in the Journal of the American Medical Association , highlighted the differences in treatment responsiveness and survival outcomes among women with differing breast cancer subtypes from 2000 to 2012 — pinpointing subgroups with poorer survival.

“We found that, while screening still had an important impact, most of the decline in annual deaths was due to improvements in treating early-stage breast cancer based on each cancer’s molecular profile,” Plevritis said.

The current study is the first to explicitly include patients with metastatic breast cancer in its models. The finding that 29% of the decrease in mortality is due to advances in treating metastatic breast cancer both surprised and gratified the researchers.

“Initially, we assumed that treatment of advanced disease was unlikely to make a significant contribution to the declines in mortality we documented in the previous two papers,” Caswell-Jin said. “But our treatments have improved, and it’s clear that they are having a significant impact on annual mortality.”

Computer models agree

The CISNET researchers used four computer models to assess the SEER data from 1975 to 2019 — one developed at Stanford Medicine in the Plevritis Lab, one by researchers at the Dana-Farber Cancer Institute, one at MD Anderson Cancer Center, and another jointly developed by researchers at the University of Wisconsin and Harvard Medical School. The four models came up with remarkably similar estimates for the impact of each intervention: screening mammography, treatment of early-stage (stages 1, 2 or 3) breast cancer and treatment of metastatic breast cancer.

Sylvia Plevritis

Sylvia Plevritis

The models reproduced the decline in mortality in breast cancer known from SEER data, from 48 per 100,000 women dying of breast cancer each year in 1975 to 27 per 100,000 in 2019 — a decrease of about 44%. The models arrived at a larger estimated reduction in mortality of about 58% because the incidence of breast cancer has risen during the same period and more women would have died had screening and treatments not improved.

The models concluded that about 47% of this reduction in mortality is the result of improved treatments for early-stage breast cancer, and about 25% is attributed to screening mammography. The remainder, or about 29%, is due to improvements in treating metastatic disease.

“Designing the new model, which had to account for individuals with non-metastatic cancer who underwent treatment but later progressed to metastatic cancer, and who may have been treated with multiple drugs over the course of their disease, was extremely complex,” Plevritis said. “It took about four years. But it was really satisfying when we were able to validate the model’s behavior and see that all four models from different institutions, which used the new model inputs in different ways, delivered consistent findings. The models not only make sense, but also produce meaningful insights.”

The impact of treating metastatic disease is exemplified by the increases in median survival time after metastasis: Patients diagnosed in 2000 with metastatic disease lived an average of 1.9 years versus an average of 3.2 years for those diagnosed in 2019. Survival time varies by subgroup status, however. Patients with what are known as estrogen receptor-positive and HER2 positive cancers saw an average increase in survival time of 2.5 years. Those with estrogen receptor-positive and HER2-negative cancers lived an average of 1.6 years longer, but those with cancers that are estrogen receptor-negative and HER2-negative lived about 0.5 years longer in 2019 than in 2000.

“It was meaningful as a breast oncologist to spend time with this history and see real progress over the past decades,” Caswell-Jin said. “There is much more work to be done; metastatic breast cancer isn’t yet curable. But it is rewarding to see that advances have made a difference in these numbers,” she added. “Our scientific and clinical work is helping our patients live longer, and I believe deaths from breast cancer will continue to steadily decline as innovation continues to grow.”

Researchers from MD Anderson Cancer Center, the Dana-Farber Cancer Institute, the University of Wisconsin-Madison School of Medicine and Public Health, the National Institutes of Health, the Albert Einstein College of Medicine, Harvard Medical School, Georgetown University, and the Georgetown-Lombardi Institute for Cancer and Aging contributed to the study.

The study was funded by the National Institutes of Health (grants U01CA253911 and U01CA199218).

Krista Conger

About Stanford Medicine

Stanford Medicine is an integrated academic health system comprising the Stanford School of Medicine and adult and pediatric health care delivery systems. Together, they harness the full potential of biomedicine through collaborative research, education and clinical care for patients. For more information, please visit med.stanford.edu .

Artificial intelligence

Exploring ways AI is applied to health care

Stanford Medicine Magazine: AI

U.S. flag

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

  • Publications
  • Account settings
  • Advanced Search
  • Journal List
  • World J Clin Oncol
  • v.13(3); 2022 Mar 24

Breast cancer in India: Present scenario and the challenges ahead

Ravi mehrotra.

Department of Health Research, Ministry of Health and Family Welfare, India Cancer Research Consortium, New Delhi 110001, India

CHIP Foundation, Noida 201301, India. [email protected]

Kavita Yadav

Centre of Social Medicine & Community Health, Jawahar Lal Nehru University, New Delhi 110067, India

Corresponding author: Ravi Mehrotra, MBBS, MD, PhD, Director, Doctor, Department of Health Research, Ministry of Health and Family Welfare, India Cancer Research Consortium, New Delhi 110001, India. [email protected]

Breast cancer is the commonest malignancy among women globally. From being fourth in the list of most common cancers in India during the 1990s, it has now become the first. In this review, we examine the available literature to understand the factors that contributed to the high burden of breast cancer in the country. We also provide the landscape of changes in the field of early diagnosis and the treatment modalities as well as the limitations of the Indian healthcare delivery systems ( e.g. , delayed diagnosis, human resources and funding for treatment). This review also sheds light on the newer interventions and the future of breast cancer management keeping in mind the coronavirus disease 2019 imposed limitations.

Core Tip: This review highlights the progress that has been made in the field of breast cancer management in India over the past few decades, in terms of addressing the various challenges of breast cancer control including diagnostic methods and treatment options. It also highlights the future of breast cancer control strategies with a focus on the coronavirus disease 2019 situation.

INTRODUCTION

Introduction and epidemiology.

Breast cancer (BC) is the commonest malignancy among women globally. It has now surpassed lung cancer as the leading cause of global cancer incidence in 2020, with an estimated 2.3 million new cases, representing 11.7% of all cancer cases[ 1 ]. Epidemiological studies have shown that the global burden of BC is expected to cross almost 2 million by the year 2030[ 2 ]. In India, the incidence has increased significantly, almost by 50%, between 1965 and 1985[ 3 ]. The estimated number of incident cases in India in 2016 was 118000 (95% uncertainty interval, 107000 to 130000), 98.1% of which were females, and the prevalent cases were 526000 (474000 to 574000). Over the last 26 years, the age-standardised incidence rate of BC in females increased by 39.1% (95% uncertainty interval, 5.1 to 85.5) from 1990 to 2016, with the increase observed in every state of the country[ 4 ]. As per the Globocan data 2020, in India, BC accounted for 13.5% (178361) of all cancer cases and 10.6% (90408) of all deaths (Figure ​ (Figure1 1 and Figure ​ Figure2) 2 ) with a cumulative risk of 2.81[ 5 ].

An external file that holds a picture, illustration, etc.
Object name is WJCO-13-209-g001.jpg

World Health Organization: Estimated number of new cancer cases in 2020, worldwide. Image available under Common Creative License[ 68 ] (Available at https://gco.iarc.fr/today/data/factsheets/cancers/20-Breast-fact-sheet.pdf ).

An external file that holds a picture, illustration, etc.
Object name is WJCO-13-209-g002.jpg

World Health Organization Globocan 2020 India. Image available under Common Creative License[ 68 ] (Available at https://gco.iarc.fr/today/data/factsheets/cancers/20-Breast-fact-sheet.pdf ).

Current trends point out that a higher proportion of the disease is occurring at a younger age in Indian women, as compared to the West. The National Cancer Registry Program analysed data from cancer registries for the period from 1988 to 2013 for changes in the incidence of cancer. All population-based cancer registries have shown a significant increase in the trend of BC[ 6 ]. In India in 1990, the cervix was the leading site of cancer followed by BC in the registries of Bangalore (23.0% vs 15.9%), Bhopal (23.2% vs 21.4%), Chennai (28.9% vs 17.7%) and Delhi (21.6% vs 20.3%), while in Mumbai, the breast was the leading site of cancer (24.1% vs 16.0%). By the years 2000-2003, the scenario had changed, and breast had overtaken as the leading site of cancer in all the registries except in the rural registry of Barshi (16.9% vs 36.8%). In the case of BC, a significant increasing trend was observed in Bhopal, Chennai and Delhi registries[ 7 ].

When it comes to the 5-year overall survival, a study reported it to be 95% for stage I patients, 92% for stage II, 70% for stage III and only 21% for stage IV patients[ 8 ]. The survival rate of patients with breast cancer is poor in India as compared to Western countries due to earlier age at onset, late stage of disease at presentation, delayed initiation of definitive management and inadequate/fragmented treatment[ 9 ]. According to the World Cancer Report 2020, the most efficient intervention for BC control is early detection and rapid treatment[ 10 ]. A 2018 systematic review of 20 studies reported that BC treatment costs increased with a higher stage of cancer at diagnosis. Consequently, earlier diagnosis of BC can lower treatment costs[ 11 ].

EARLY DETECTION AND SCREENING PROGRAMMES

Success and failure of screening programs depend on several factors ranging from the presence of proper guidance manuals, development and usage of an appropriate instrument for diagnosis to proper implementation and availability of adequate human resources. Another factor is the efficacy of the screening test in avoiding the risk of false positives and unnecessary biopsies and surgeries[ 12 ]. Organised screening programmes provide screening to an identifiable target population and use multidisciplinary delivery teams, coordinated clinical oversight committees and regular review by a multi-speciality evaluation board to maximise the benefit to the target population[ 13 ]. Screening strategies are moving towards a risk-based approach rather than a broad age-based and sex-based recommendation. To use this risk-based approach, India needs to assess risk factors and incorporate this information into BC screening[ 14 ].

A recent study from Mumbai has reported that clinical breast examination conducted every 2 years by primary health workers significantly downstaged breast cancer at diagnosis and led to a nonsignificant 15% reduction in breast cancer mortality overall (but a significant reduction of nearly 30% in mortality in women aged ≥ 50)[ 15 ]. Mammography sensitivity has been reported to vary from 64% to 90% and specificity from 82% to 93%[ 16 ]. Indian women have more dense breasts, and there is a lack of adequate mammography machines and trained manpower. This may result in false positives and over-diagnosis. Digital mammography uses computer-aided detection software but remains costly. It is due to these reasons that mass-scale routine mammography screening is not a favoured option for a transitioning country like India.

Ultrasonography has an overall sensitivity of 53% to 67% and specificity of 89% to 99%[ 17 , 18 ] and might be particularly helpful in younger women (aged 40 to 49 years). However, the requirement of trained professionals to perform and interpret ultrasound is a major hurdle. Though breast self-examination is not accepted as an early detection method for BC, this technique, if used diligently and skilfully, can serve as a useful adjunct to making the woman aware of her normal breast[ 19 ].

Understanding India-specific differences by utilising genomics may enable the identification of women at high risk of developing cancer, where targeted screening may be cost-effective. There is an urgent need to identify Indian-specific genetic/epigenetic biomarkers. These may have the potential to be used as biomarkers for early detection at the screening stage[ 20 ].

TREATMENT OPTIONS IN PAST AND PRESENT

Management of BC is multidisciplinary and has come a long way. In the past, the widely used treatment option was mastectomy followed by adjuvant chemotherapy for locally advanced BC, triple-negative breast cancer and HER2neu expressing tumours (human epidermal growth factor receptor 2). At present, it includes a loco-regional approach (targeting only the tumour with the help of surgery and radiation therapy) and a systemic therapy approach that targets the entire body. The systemic therapy includes endocrine therapy for hormone receptor-positive disease, chemotherapy, anti-HER2 therapy for HER2 positive disease, bone stabilising agents, polymerase inhibitors for BRCA (breast cancer gene) mutation carriers and, recently, immunotherapy. However, the majority of patients still undergo primary ablative surgical procedures. Gene expression profiling in hormone receptor-positive disease is also a promising option but has financial implications.

From using drugs like cyclophosphamide, methotrexate, etc. in the 1970s for chemotherapy to using their modifications like anthracycline-based combination chemotherapy protocols in the 1980s and 1990s, we have come a long way. Taxanes are the newer additions that show a promising future. The radiation treatment of BC has evolved from 2D to 3D conformal radiotherapy and accelerated partial breast irradiation, aiming to reduce normal tissue toxicity and overall treatment time[ 21 ]. The newer additions, viz. intensity-modulated radiation therapy and deep inspiration breath-hold, are still inaccessible to many. The same is the case with brachytherapy.

The outcomes with triple-negative breast cancer are poor, and the treatment options are mainly restricted to systemic chemotherapy. Immunotherapy, poly adenosine diphosphate-ribose polymerase inhibitors (poly(adenosine diphosphate-ribose) polymerase) and antibody-drug conjugates have the potential to change the current scenario of BC treatment. One important point that should be considered while planning the treatment is that there is a lot of hype regarding the newer drugs that flood the market, however with little or no difference in the survival benefit. Hence it is important to choose wisely.

The field of breast surgery has also evolved from total mastectomy to breast conservation therapy to oncoplastic breast surgery. The rapidly advancing field of oncoplastic breast surgery offers a pragmatic alternative to total mastectomy and breast conservation therapy. It is currently nascent but expected to attain mainstream status in the near future as oncoplastic breast surgery has economic feasibility and cost-effectiveness and is well suited for a low-resource setting such as India[ 22 ].

CHALLENGES IN BC CONTROL

Delay in seeking healthcare.

Continuously increasing BC prevalence is just the tip of the iceberg. There are many underlying problems that contribute to this mounting burden. In India, nearly 60% of BC cases are diagnosed at stage III or IV of the disease[ 23 ]. Most of the patients present to the healthcare facility only when there is a large palpable mass or secondary changes like local skin/chest wall changes are visible. Women tend to ignore the minor symptoms and do not show up at the hospital until it is unbearable, owing to their household responsibilities. Other factors that may contribute to the late presentation include a lack of awareness about the disease, especially in rural areas. This also leads to fewer women performing a self-breast examination, opting for a periodic examination by a healthcare worker or mammography for BC screening, despite it being available for free in a few government hospitals. For those willing to pay for it, mammography is available in private hospitals. This lack of awareness regarding the risk factors and early detection methods of BC is unfortunately even prevalent in 49% of healthcare workers[ 24 ].

The initial manifestation of BC, i.e. a lump, is generally not associated with pain, which further adds to the delay in seeking treatment in 50% to 70% of the cases in rural areas[ 25 ]. Other factors that may influence the early detection and treatment of BC are the presence of a diagnostic/treatment facility in the nearby area, patient’s preference and trust in the healthcare provider, amount of time required for travelling to the service centre and amount and availability of money that can be spent on the treatment. However, this grim situation has slowly started to change due to various awareness campaigns, and now women have slowly started to understand and value their health. Another deterring factor in seeking early care is the stigma of social embarrassment and isolation. Women not only fear death and contagion by cancer but also fear that their and their family’s reputation would suffer if people knew of their cancer diagnosis, including potential difficulties in their daughter’s wedding. It is also a widespread assumption that cancer, especially in the private parts (breast and genitals) is linked to ”bad” and “immoral behaviour”[ 26 ]. The issue of social stigma urgently needs to be addressed through awareness campaigns, as it not only jeopardises early diagnosis but also the treatment-seeking behaviour of women with symptoms of BC.

Delay from the healthcare provider’s side

On average, more than 12 wk of delay is seen in diagnosis and treatment in 23% of patients[ 27 ]. A study examined provider’s delay (defined as the period between the first consultation and diagnosis) and observed that the mean provider delay was 80 d in rural areas and 66 d in urban areas[ 28 ]. More than half of the women were observed to have a delay of more than 90 d in seeking care. The patient-related delay was observed to be 6.1 wk, and the system-related delay was 24.6 wk with a mean total delay of 29.4 wk in treatment. This led to a poor prognosis[ 29 ]. These delays may be attributed partly to the ever-increasing patient load on the health care system and competing priorities.

High attrition rates

After crossing all these hurdles, once the diagnosis is done and treatment is started, there are further hazards. In 1990, India’s facilities for the diagnosis and treatment of cancer were far behind recommendations[ 30 ]. Even today, due to a large variation in the health care standards between regions, the quality of treatment for BC patients varies from pathetic to world-class. Few patients are treated at well-equipped centres in a protocol-based manner, while some are subjected to numerous compromises. Fortunately, BC is curable if detected early, but due to various underlying factors, improper treatment provided locally by non-oncologists without standard oncology expertise may lead to the mismanagement of BC cases. At the same time, patients with advanced, metastatic incurable cancer may require only palliative care are referred to tertiary cancer centres. This leads to the improper use of limited, valuable resources.

Another issue that adds to the high attrition rates/loss to follow-up of BC treatment is an unacceptable out-of-pocket expenditure, which is three times higher for private inpatient cancer care in India[ 30 ]. More than half the patients from low-income households spend > 20% of their annual household expenditure on BC treatment, leading to catastrophic results[ 31 ]. An analysis of three public insurance schemes for anticancer treatment in India published in 2018 revealed inconsistencies in the selection of reimbursed treatments. The reimbursed amount was usually found to be insufficient to cover the total cancer chemotherapy costs, leading to an average budget shortage of up to 43% for BC treatment[ 32 ]. Cancer insurance policies can significantly reduce the financial burden caused by out-of-pocket expenditure and prevent catastrophic health expenditure, distress financing and even bankruptcy. However, a study by Singh et al [ 33 ] to understand the use of health insurance in India reported a lack of awareness regarding the use of these schemes as the key reason for the low penetration of health insurance policies.

Shortage of resources/skewed distribution of available limited resources

Another problem is a shortage of manpower. India has just over 2000 oncologists for 10 million patients, and the number of oncologists is unevenly spread, being lower in semi-urban and rural areas. Although nearly 70% of the Indian population lives in rural areas, about 95% of facilities for cancer treatment exist in the urban areas of the country. There are also regional variations. About 60% of specialist facilities are in Southern and Western India, whereas more than 50% of the population lives in the Northern, Central and Eastern regions, distorting service provision[ 34 ].

There are currently 57 courses for radiotherapy technologists and about 2200 certified radiation technologists in practice in India[ 35 ]. As per a recent study, India has just 10% of the total requirement of 5000 radiation therapy units indicating a shortfall of > 4500 machines. The World Health Organization recommends at least one teletherapy unit per million population, and there is a shortfall of 700 teletherapy units[ 35 ]. If we look at the treatment infrastructure, at least half of patients with cancer will be judged to need radiotherapy at some point. Yet only 26% of the population, living in the Eastern region of India, have immediate access to only 11% of radiotherapy facilities[ 36 ]. Nearly 40% of hospitals in India are not adequately equipped with advanced cancer care equipment. Very few centres in the country provide integrated surgical and chemoradiation for BC. Nearly 75% of the patients in the public sector do not have access to timely radiotherapy[ 37 ].

All of these factors aid in raising the overall cost of BC treatment for the common person. They are forced to make out-of-pocket cancer care payments, as most of the patients have to bear the cost of therapy. The government facilities are inadequate in number to cater to a large number of patients. Thus, patients are required to go for treatment in major cities along with their attendants, resulting in loss of livelihood of both the patient and her attendants. There is an urgent need to establish a larger number of cancer care facilities accessible to those living in rural areas so that the gap of cancer detection and treatment services may be bridged.

RAYS OF HOPE/WAY FORWARD

The integrative approach.

The rising human cost, both social and economic, of BC underscores the need for more holistic, multidimensional approaches that encompass the cancer care continuum including prevention, early detection, treatment, palliation and survivorship. A balanced approach is necessary to integrate traditional medical practices into mainstream oncology practice, starting with a meaningful discussion among all the relevant stakeholders and identifying the areas where the benefits of a complementary approach are beyond doubt[ 38 ]. Exercise has been proven to be an effective, safe and feasible tool in combating the adverse effects of treatment, prevents complications and decreases the risk of BC-specific mortality[ 39 ]. A recent review has reported evidence that diet-related and physical activity-related interventions for the primary prevention of BC are cost-effective[ 40 ]. Such lifestyle modifications need to be included in mainstream treatment planning.

The internet era

Screening programmes in high-income countries that have increased patient participation have done so with high-quality and periodic education programmes with campaigns tailored to the specific cultural context of a community[ 41 ]. This can only be achieved by creating and sustaining the level of awareness among the general population. One effective way of doing that is providing information on BC to the relatives/patients in the hospital setting. However, due to the ever-increasing patient burden, it is not always possible for healthcare providers to give appropriate time and counselling to the people. The distribution of pamphlets with valuable information may be the cause. In recent times, the Internet is being increasingly used as a reliable source to seek health-related information[ 42 , 43 ]. A lot of people search for information online. Having a credible source of information that can be revisited as and when required by the people is always a plus. Keeping these things in mind a website ‘Cancer India’ was developed by the National Institute of Cancer Prevention and Research providing comprehensive information on common cancers in India, including BC, in layman’s language (available in English and Hindi). A recent evaluation by this group reported that the website managed to serve the intended purpose of improving cancer awareness with reasonable success[ 44 ].

Involvement of community health workers

Community participation with the engagement of the health system and local self-government are required for implementing a comprehensive cancer screening strategy. A BC screening program using local volunteers for early detection is feasible in low-income settings, thereby improving survival[ 45 ]. Community health workers can play an important role in the early detection of BC in low- and middle-income countries, with responsibilities including awareness-raising, conducting clinical breast examinations, making referrals and supporting subsequent patient navigation. However, this promise can only be turned into genuine progress if these activities are appropriately supported and sustained. This will involve adopting contextually appropriate early detection initiatives that are embedded within the broader health system where community health workers are appropriately trained, equipped, paid and supported with appropriate links to specialist oncology services. A recent study reported the effectiveness of the Extension for Community Healthcare Outcomes model training program in reaching primary care physicians across the country and improving their knowledge and skills related to screening for breast, oral and cervical cancer[ 46 ].

Early diagnosis

With advancements in molecular diagnostics and therapeutics, newer non-invasive prognostic biomarker tests to detect BC at a very early stage, such as digital breast tomosynthesis and breast biopsy techniques, are becoming available[ 47 ]. Above all, early detection programmes in low- and middle-income countries must make provisions for every individual at risk of BC. This will mean considering the needs of the hardest individual to reach first, so that no woman is left behind in the goal to end unjust and untimely deaths attributable to the leading cause of female mortality in low- and middle-income countries[ 48 ].

Out of pocket expenditure

To address the issue of out-of-pocket expenditure and to reach out to the poorer section of the society, a scheme called Pradhan Mantri Jan Arogya Yojana under the new universal healthcare programme named “Ayushman Bharat” has been recently launched in India. It is the largest health assurance scheme in the world that aims at providing a health cover of Rs. 5 Lakhs (6814 USD) per family per year for secondary and tertiary care hospitalisation to over 10.74 crores (approximately 107 million) poor and vulnerable families (approximately 500 million beneficiaries) that form the bottom 40% of the Indian population[ 49 ]. The Indian government’s efforts to bring anti–HER2 drug under-pricing regulations have enhanced access and improved outcomes. The launch of low-cost T-DM1 (the antibody-drug conjugate trastuzumab emtansine) and anti–HER2 therapy biosimilars is keenly awaited. Linking cancer registry data with Ayushman Bharat, mortality databases and the Hospital Information System could improve cancer registration, follow-up and outcome data[ 50 ].

Newer initiatives

There are several health-tech start-ups that aid in different stages of cancer care. For example, Niramai ® uses machine learning and big data analytics to develop low-cost diagnostics for BC[ 51 ]. Oncostem ® uses multi-marker prognostics tests that aid in personalised treatment[ 52 ], and UE Lifesciences ® uses contactless and radiation-free handheld screening devices[ 53 ] that may come in handy during coronavirus disease 2019 (COVID-19) times. Panacea ® and Mitra Biotech ® [ 54 ] are start-up companies that deal with providing precision therapies. Tumourboard ® and Navya Network ® [ 55 ] provides affordable and precise consultations. There are several drug-patient assistance programs from pharmaceutical companies like Roche ® , Novartis ® , Dr Reddys ® , etc. that help in following patient-centric care from beginning to end. The BC Initiative 2.5 is a global campaign to reduce disparities in BC outcomes and improve access to breast health care worldwide. It is a self-assessment toolkit that can help countries conduct a comprehensive breast health care situational analysis[ 56 ] Apart from these, patient navigation strategies are also rapidly growing and evolving concepts. ‘Kevat’ is the first initiative in India to create a trained task force to facilitate the cancer patient’s journey from entry to the hospital to follow-up. It is a nascent area of speciality in cancer care that is set to target the pressing need of well-trained patient navigators in onco-care[ 57 ].

Apart from these, the World Health Organization, on March 9, 2021, introduced a Global Breast Cancer Initiative to reduce global breast mortality by 2.5% by 2040. The aim is to reduce 2.5 million global deaths, particularly in low-income countries, where the progress to tackle the disease has been relatively slow. An evidence-based technical package will be provided to countries as part of the initiative[ 58 ]. Such initiatives instil faith in the future for better breast cancer management.

COVID-19 challenge

The COVID-19 pandemic has challenged the prioritisation of various diseases in healthcare systems across the globe, and BC patients are no exception. It impacted their access to physicians, medication and surgeries. Nearly 70% of patients could not access life-saving surgeries and treatment. Chemotherapy treatments and follow-ups were postponed due to lockdowns[ 59 , 60 ]. A recent study reported that the average monthly expenditure of cancer patients had increased by 32% during the COVID-19 period while the mean monthly household income was reduced by a quarter. More than two-thirds of the patients had no income during the lockdown, and more than half of the patients met their expenditure by borrowing money. The incremental expenditure coupled with reduced or no income due to the closure of economic activities in the country imposed severe financial stress on patients with BC[ 61 ].

It became of utmost importance to balance between the benefits and risks associated with BC treatment[ 62 ]. There are different treatment modalities and each case is different. The shorter duration of radiotherapy, transient spacing of chemotherapy in metastatic BC setting, if deemed feasible, oral hormone therapy to delay surgery and judicious use of immunomodulators are some recent guidelines in evidence-based practice in this unprecedented crisis[ 63 ]. The core idea is to delay the surgery until the pandemic is over whenever feasible. Additionally, apart from healthcare management, one area that has suffered tremendously due to the infectious outbreak is cancer research[ 64 ]. The estimated decrease in cancer funding in India ranges from 5% to 100%, as many funding agencies have cancelled calls for funding. The private/charity sector is the worst hit, with an estimated decrease of more than 60% of its funding[ 65 ]. A robust health system is a prerequisite for providing the facilities for the treatment of BC that is diagnosed through the early detection programmes, whether through screening or the presence of symptoms[ 66 ]. Of course, it goes without saying that a proper evaluation of the programmes will not only allow improvement of quality of services but also generate valuable evidence on the effectiveness of screening and early diagnosis in the countries ”in transition”[ 67 ].

To conclude, the BC burden is rising at a rate much faster than it was a decade ago. Acknowledging that BC is one of the foremost cancers in India now would be the first step towards making people cognizant of the disease. It is fast developing into a public health crisis, and society’s discomfort to talk about women’s bodies has made the situation even worse. To combat the consequences as a country, better preparedness is essential. A robust awareness campaign and effective implementation of a national cancer screening program are the need of the hour. We also need to stand up and deliver on the healthcare front. The shortage of skilled manpower and infrastructural requirements need to be met, and for this, the total healthcare budget of the country needs to be increased. In the jargon of the challenges of BC control, prioritising the adoption of a preventive approach and early detection would go a long way. Another important aspect is the country’s preparedness for unprecedented events like COVID-19, for which there should be a separate provision to deal with public health disasters. Creating a cadre of trained medical and paramedical professionals, efficient utilisation and timely upgrading of skills of the existing healthcare workforce along with adopting newer technologies would further the cause of BC control.

Conflict-of-interest statement: The authors declare no potential conflict of interest for this article.

Provenance and peer review: Invited article; Externally peer reviewed.

Peer-review model: Single blind

Peer-review started: April 9, 2021

First decision: June 7, 2021

Article in press: March 7, 2022

Specialty type: Oncology

Country/Territory of origin: India

Peer-review report’s scientific quality classification

Grade A (Excellent): A

Grade B (Very good): B

Grade C (Good): 0

Grade D (Fair): 0

Grade E (Poor): 0

P-Reviewer: Serban D, Romania; Wang CY, Taiwan S-Editor: Gong ZM L-Editor: Filipodia P-Editor: Gong ZM

Contributor Information

Ravi Mehrotra, Department of Health Research, Ministry of Health and Family Welfare, India Cancer Research Consortium, New Delhi 110001, India. CHIP Foundation, Noida 201301, India. [email protected] .

Kavita Yadav, Centre of Social Medicine & Community Health, Jawahar Lal Nehru University, New Delhi 110067, India.

  • Search by keyword
  • Search by citation

Page 1 of 104

The prostate-specific membrane antigen holds potential as a vascular target for endogenous radiotherapy with [ 177 Lu]Lu-PSMA-I&T for triple-negative breast cancer

Overexpression of prostate-specific membrane antigen (PSMA) on the vasculature of triple-negative breast cancer (TNBC) presents a promising avenue for targeted endogenous radiotherapy with [ 177 Lu]Lu-PSMA-I&T. Thi...

  • View Full Text

metastatic breast cancer research papers

Metabolic adaptation towards glycolysis supports resistance to neoadjuvant chemotherapy in early triple negative breast cancers

Neoadjuvant chemotherapy (NAC) is the standard of care for patients with early-stage triple negative breast cancers (TNBC). However, more than half of TNBC patients do not achieve a pathological complete respo...

Identification of CD160-TM as a tumor target on triple negative breast cancers: possible therapeutic applications

Despite major therapeutic advances, triple-negative breast cancer (TNBC) still presents a worth prognosis than hormone receptors-positive breast cancers. One major issue relies in the molecular and mutational ...

Contrast-enhanced ultrasound to predict malignant upgrading of atypical ductal hyperplasia

A malignancy might be found at surgery in cases of atypical ductal hyperplasia (ADH) diagnosed via US-guided core needle biopsy (CNB). The objective of this study was to investigate the diagnostic performance ...

MRI-based tumor shrinkage patterns after early neoadjuvant therapy in breast cancer: correlation with molecular subtypes and pathological response after therapy

MRI-based tumor shrinkage patterns (TSP) after neoadjuvant therapy (NAT) have been associated with pathological response. However, the understanding of TSP after early NAT remains limited. We aimed to analyze ...

Are better AI algorithms for breast cancer detection also better at predicting risk? A paired case–control study

There is increasing evidence that artificial intelligence (AI) breast cancer risk evaluation tools using digital mammograms are highly informative for 1–6 years following a negative screening examination. We h...

Prognostic impact of HER2 biomarker levels in trastuzumab-treated early HER2-positive breast cancer

Overexpression of human epidermal growth factor receptor 2 (HER2) caused by HER2 gene amplification is a driver in breast cancer tumorigenesis. We aimed to investigate the prognostic significance of manual sco...

The novel phosphatase NUDT5 is a critical regulator of triple-negative breast cancer growth

The most aggressive form of breast cancer is triple-negative breast cancer (TNBC), which lacks expression of the estrogen receptor (ER) and progesterone receptor (PR), and does not have overexpression of the h...

Low-dose acetylsalicylic acid reduces local inflammation and tissue perfusion in dense breast tissue in postmenopausal women

One major risk factor for breast cancer is high mammographic density. It has been estimated that dense breast tissue contributes to ~ 30% of all breast cancer. Prevention targeting dense breast tissue has the ...

Improving lesion detection in mammograms by leveraging a Cycle-GAN-based lesion remover

The wide heterogeneity in the appearance of breast lesions and normal breast structures can confuse computerized detection algorithms. Our purpose was therefore to develop a Lesion Highlighter (LH) that can impro...

TBCRC 039: a phase II study of preoperative ruxolitinib with or without paclitaxel for triple-negative inflammatory breast cancer

Patients with inflammatory breast cancer (IBC) have overall poor clinical outcomes, with triple-negative IBC (TN-IBC) being associated with the worst survival, warranting the investigation of novel therapies. ...

ADAMTS18 deficiency associates extracellular matrix dysfunction with a higher risk of HER2-positive mammary tumorigenesis and metastasis

Human epidermal growth factor receptor 2 (HER2)-positive breast cancer accounts for about 20% of all breast cancer cases and is correlated with a high relapse rate and poor prognosis. ADAMTS18 is proposed as an i...

Development of a machine learning-based radiomics signature for estimating breast cancer TME phenotypes and predicting anti-PD-1/PD-L1 immunotherapy response

Since breast cancer patients respond diversely to immunotherapy, there is an urgent need to explore novel biomarkers to precisely predict clinical responses and enhance therapeutic efficacy. The purpose of our...

Development and prognostic validation of a three-level NHG-like deep learning-based model for histological grading of breast cancer

Histological grade is a well-known prognostic factor that is routinely assessed in breast tumours. However, manual assessment of Nottingham Histological Grade (NHG) has high inter-assessor and inter-laboratory...

A genome-wide association study of contralateral breast cancer in the Women’s Environmental Cancer and Radiation Epidemiology Study

Contralateral breast cancer (CBC) is the most common second primary cancer diagnosed in breast cancer survivors, yet the understanding of the genetic susceptibility of CBC, particularly with respect to common ...

Differential patterns of reproductive and lifestyle risk factors for breast cancer according to birth cohorts among women in China, Japan and Korea

The birth cohort effect has been suggested to influence the rate of breast cancer incidence and the trends of associated reproductive and lifestyle factors. We conducted a cohort study to determine whether a d...

Increased risk of contralateral breast cancer for BRCA1/2 wild-type, high-risk Korean breast cancer patients: a retrospective cohort study

This study aimed to investigate the contralateral breast cancer (CBC) recurrence rate in Korean breast cancer patients according to their BRCA1/2 germline mutation status, focusing particularly on the CBC recurre...

c-MET-positive circulating tumor cells and cell-free DNA as independent prognostic factors in hormone receptor-positive/HER2-negative metastatic breast cancer

Endocrine therapy resistance in hormone receptor-positive/HER2-negative (HR+/HER2−) breast cancer (BC) is a significant clinical challenge that poses several unmet needs in the management of the disease. This ...

Digital image analysis and machine learning-assisted prediction of neoadjuvant chemotherapy response in triple-negative breast cancer

Pathological complete response (pCR) is associated with favorable prognosis in patients with triple-negative breast cancer (TNBC). However, only 30–40% of TNBC patients treated with neoadjuvant chemotherapy (N...

Evidence of steady-state fibroblast subtypes in the normal human breast as cells-of-origin for perturbed-state fibroblasts in breast cancer

Human breast cancer most frequently originates within a well-defined anatomical structure referred to as the terminal duct lobular unit (TDLU). This structure is endowed with its very own lobular fibroblasts r...

PFKP deubiquitination and stabilization by USP5 activate aerobic glycolysis to promote triple-negative breast cancer progression

Triple-negative breast cancer (TNBC) remains the most challenging subtype of breast cancer and lacks definite treatment targets. Aerobic glycolysis is a hallmark of metabolic reprogramming that contributes to ...

Prognostic value of baseline neutrophil/lymphocyte ratio in HER2-positive metastatic breast cancer: exploratory analysis of data from the CLEOPATRA trial

This study aimed to evaluate the prognostic role of the baseline neutrophil/lymphocyte ratio (NLR) in HER2-positive metastatic breast cancer (MBC) patients treated with trastuzumab/pertuzumab.

Risk-management decision-making data from a community-based sample of racially diverse women at high risk of breast cancer: rationale, methods, and sample characteristics of the Daughter Sister Mother Project survey

To understand the dynamics that limit use of risk-management options by women at high risk of breast cancer, there is a critical need for research that focuses on patient perspectives. Prior research has left ...

Machine learning prediction of pathological complete response and overall survival of breast cancer patients in an underserved inner-city population

Generalizability of predictive models for pathological complete response (pCR) and overall survival (OS) in breast cancer patients requires diverse datasets. This study employed four machine learning models to...

Male with an apparently normal phenotype carrying a BRCA1 exon 20 duplication in trans to a BRCA1 frameshift variant

Reports of dual carriers of pathogenic BRCA1 variants in trans are extremely rare, and so far, most individuals have been associated with a Fanconi Anemia-like phenotype.

Targeting tumor–stromal interactions in triple-negative breast cancer using a human vascularized micro-tumor model

Triple-negative breast cancer (TNBC) is highly aggressive with limited available treatments. Stromal cells in the tumor microenvironment (TME) are crucial in TNBC progression; however, understanding the molecu...

Identification of a Notch transcriptomic signature for breast cancer

Dysregulated Notch signalling contributes to breast cancer development and progression, but validated tools to measure the level of Notch signalling in breast cancer subtypes and in response to systemic therap...

A prospective study on tumour response assessment methods after neoadjuvant endocrine therapy in early oestrogen receptor-positive breast cancer

Neoadjuvant endocrine therapy (NET) in oestrogen receptor-positive (ER+) /HER2-negative (HER2-) breast cancer allows real-time evaluation of drug efficacy as well as investigation of the biological and molecul...

Addition of polygenic risk score to a risk calculator for prediction of breast cancer in US Black women

Previous work in European ancestry populations has shown that adding a polygenic risk score (PRS) to breast cancer risk prediction models based on epidemiologic factors results in better discriminatory perform...

Protein disulfide isomerase family member 4 promotes triple-negative breast cancer tumorigenesis and radiotherapy resistance through JNK pathway

Despite radiotherapy ability to significantly improve treatment outcomes and survival in triple-negative breast cancer (TNBC) patients, acquired resistance to radiotherapy poses a serious clinical challenge. P...

Association between allostatic load and breast cancer risk: a cohort study

Allostatic load (AL) reflects the collective load of chronic stress during lifetime. Previous studies have shown that higher AL is associated with poor clinical outcomes among breast cancer patients. However, ...

HER2 amplification level by in situ hybridization predicts survival outcome in advanced HER2-positive breast cancer treated with pertuzumab, trastuzumab, and docetaxel regardless of HER2 IHC results

The role of HER2 amplification level in predicting the effectiveness of HER2-directed therapies has been established. However, its association with survival outcomes in advanced HER2-positive breast cancer tre...

Effect of histological breast cancer subtypes invasive lobular versus non-special type on survival in early intermediate-to-high-risk breast carcinoma: results from the SUCCESS trials

Invasive lobular breast carcinomas (ILC) have different histological features compared to non-special type carcinomas (NST), but the effect of histological subtypes on survival is controversial. In this study,...

Prevalence of BRCA1 , BRCA2 , and PALB2 genomic alterations among 924 Taiwanese breast cancer assays with tumor-only targeted sequencing: extended data analysis from the VGH-TAYLOR study

The homologous recombination (HR) repair pathway for DNA damage, particularly the BRCA1 and BRCA2 genes, has become a target for cancer therapy, with poly ADP-ribose polymerase (PARP) inhibitors showing significa...

Therapeutic protein PAK restrains the progression of triple negative breast cancer through degrading SREBP-1 mRNA

Triple-negative breast cancer (TNBC) represents the most challenging subtype of breast cancer. Studies have implicated an upregulation of lipid synthesis pathways in the initiation and progression of TNBC. Tar...

Mosquito control exposures and breast cancer risk: analysis of 1071 cases and 2096 controls from the Ghana Breast Health Study

Epidemiologic data on insecticide exposures and breast cancer risk are inconclusive and mostly from high-income countries. Using data from 1071 invasive pathologically confirmed breast cancer cases and 2096 co...

Molecular subtypes of breast cancer predicting clinical benefits of radiotherapy after breast-conserving surgery: a propensity-score-matched cohort study

Based on the molecular expression of cancer cells, molecular subtypes of breast cancer have been applied to classify patients for predicting clinical outcomes and prognosis. However, further evidence is needed...

Key regulator PNPLA8 drives phospholipid reprogramming induced proliferation and migration in triple-negative breast cancer

Triple-negative breast cancer (TNBC) is the most aggressive breast cancer subtype and leads to the poorest patient outcomes despite surgery and chemotherapy treatment. Exploring new molecular mechanisms of TNB...

Development and evaluation of a method to assess breast cancer risk using a longitudinal history of mammographic density: a cohort study

Women with dense breasts have an increased risk of breast cancer. However, breast density is measured with variability, which may reduce the reliability and accuracy of its association with breast cancer risk....

Histologic tumor type as a determinant of survival in hormone receptor-positive, HER2-negative, pT1-3 invasive ductal and lobular breast cancer

The aim of the study was to compare the difference in survival between invasive ductal (IDC) and lobular carcinoma (ILC).

Integrative whole-genome and transcriptome analysis of HER2-amplified metastatic breast cancer

In breast cancer, the advent of anti-HER2 therapies has made HER2+ tumors a highly relevant subgroup. However, the exact characteristics which prohibit clinical response to anti-HER2 therapies and drive diseas...

The BET degrader ZBC260 suppresses stemness and tumorigenesis and promotes differentiation in triple-negative breast cancer by disrupting inflammatory signaling

Breast cancer stem cells (BCSCs) are resistant to standard therapies, facilitate tumor dissemination, and contribute to relapse and progression. Super-enhancers are regulators of stemness, and BET proteins, wh...

Formation of an invasion-permissive matrix requires TGFβ/SNAIL1-regulated alternative splicing of fibronectin

As in most solid cancers, the emergence of cells with oncogenic mutations in the mammary epithelium alters the tissue homeostasis. Some soluble factors, such as TGFβ, potently modify the behavior of healthy st...

PROACTING: predicting pathological complete response to neoadjuvant chemotherapy in breast cancer from routine diagnostic histopathology biopsies with deep learning

Invasive breast cancer patients are increasingly being treated with neoadjuvant chemotherapy; however, only a fraction of the patients respond to it completely. To prevent overtreatment, there is an urgent nee...

AMEERA-4: a randomized, preoperative window-of-opportunity study of amcenestrant versus letrozole in early breast cancer

Window-of-opportunity (WOO) studies provide insights into the clinical activity of new drugs in breast cancer.

Blocking channels to metastasis: targeting sodium transport in breast cancer

The development of therapies that can suppress invasion and prevent metastasis, ‘anti-metastatic drugs’, is an important area of unmet therapeutic need. The new results of a recent open-label, multicentre rand...

High inter-laboratory variability in the assessment of HER2-low breast cancer: a national registry study on 50,714 Danish patients

Considering the recent advancements in the treatment of breast cancer with low expression of human epidermal growth factor receptor 2 (HER2), we aimed to examine inter-laboratory variability in the assessment ...

Multimodal prediction of neoadjuvant treatment outcome by serial FDG PET and MRI in women with locally advanced breast cancer

To investigate combined MRI and 18 F-FDG PET for assessing breast tumor metabolism/perfusion mismatch and predicting pathological response and recurrence-free survival (RFS) in women treated for breast cancer.

A catchment and location-allocation analysis of mammography access in Delaware, US: implications for disparities in geographic access to breast cancer screening

Despite a 40% reduction in breast cancer mortality over the last 30 years, not all groups have benefited equally from these gains. A consistent link between later stage of diagnosis and disparities in breast c...

Alterations to DNA methylation patterns induced by chemotherapy treatment are associated with negative impacts on the olfactory pathway

Exposure to cytotoxic chemotherapy treatment may alter DNA methylation (DNAm) in breast cancer patients.

  • Editorial Board
  • Manuscript editing services
  • Instructions for Editors
  • Sign up for article alerts and news from this journal
  • Collections
  • Follow us on Twitter

Annual Journal Metrics

2022 Citation Impact 7.4 - 2-year Impact Factor 7.4 - 5-year Impact Factor 1.764 - SNIP (Source Normalized Impact per Paper) 2.408 - SJR (SCImago Journal Rank)

2022 Speed 27 days submission to first editorial decision for all manuscripts (Median) 166 days submission to accept (Median)

2022 Usage  2,233,470 downloads 8,627 Altmetric mentions 

  • More about our metrics

Breast Cancer Research

ISSN: 1465-542X

Medical Xpress

A breast cancer drug, susceptible to resistance, can be restored to effectiveness, researchers demonstrate

I n a new paper published in Cancer Research , researchers at MUSC Hollings Cancer Center have shown that targeting a protein called TACC3 (transforming acidic coiled-coil containing protein 3) can restore the effectiveness of the breast cancer drug T-DM1 if the cancer cells have developed resistance.

T-DM1, known by the brand name Kadcyla, is approved to treat women with HER2-positive breast cancer. It was first approved in 2013 for metastatic HER2-positive breast cancer. Its use was later expanded to treat early-stage HER2-positive breast cancer if cancer cells remained after surgery and pre-surgery chemotherapy.

"T-DM1 has been the first and one of the most successful antibody drug conjugates (ADCs) used for breast cancer, but resistance is a major problem," explained Hollings researcher Ozgur Sahin, Ph.D., professor and SmartState Endowed Chair in Lipidomics and Drug Discovery in the Department of Biochemistry and Molecular Biology.

An antibody drug conjugate combines an antibody with a chemotherapy drug—a "payload" delivered directly to the cancer cell.

Although the drug has been in use for 10 years, this new research out of the Sahin Lab shows for the first time a previously unrealized mechanism that causes it to work.

"This is the first study showing that T-DM1 induces immunogenic cell death (ICD) and that in resistance, this immunogenic cell death is gone—lost. By targeting TACC3, we can bring it back and make the drug work again," Sahin said.

"Notably, this really opens a new avenue for studying different ADC drugs with different payloads in the context of immunogenic cell death induction."

ICD is a type of cell death that triggers a response from the immune system—when ICD is activated, the dying cancer cell releases danger-associated molecular patterns, or DAMPs. This helps the infiltration of dendritic cells and T-cells. Activated immune cells pick up on the presence of DAMPs and go to work attacking remaining cancer cells.

Some, but not all, chemotherapy drugs induce ICD.

T-DM1 activates the spindle assembly checkpoint (SAC) in cells, a critical checkpoint in cell division, whose purpose is to ensure that one copy of each chromosome is properly attached and lined up on opposite sides in preparation for division into two identical daughter cells. While T-DM1 can activate SAC in sensitive cells, leading to ICD, this checkpoint is lost in T-DM1-resistant cells.

TACC3 also gets involved in cell division, particularly cancer cell division. There's an abundance of TACC3 in cancer cells, and high levels of this protein are associated with worse outcomes.

In their experiments, the researchers found that when cancer cells become resistant to T-DM1, there are high levels of TACC3 that inhibit the spindle assembly checkpoint process. T-DM1 can't kickstart the spindle assembly checkpoint, and the DAMPs are never released, so the immune cells never spring into action.

Inhibiting TACC3, the researchers found, lets T-DM1 get on with its job.

The co-first authors of the paper, Emre Gedik, Ph.D., and Ozge Saatci, Ph.D., noted that the proposed combination of T-DM1 and TACC3 inhibitors aids in reviving immunogenic cell death in T-DM1-resistant tumors that was otherwise lost so that the anti-cancer immune cells can sneak in.

"These findings represent a key advancement in overcoming drug resistance in HER2-positive breast cancer," Gedik and Saatci wrote.

Sahin noted that the assistance of Shikhar Mehrotra, Ph.D., was invaluable. Mehrotra's research centers on targeting T-cell signaling.

Their combined research points to new possibilities.

"Our data encourages testing the combination of TACC3 inhibitors with other ADCs beyond T-DM1, or even with immune checkpoint blockers, to achieve superior and durable responses," the researchers wrote.

More information: Mustafa Emre Gedik et al, Targeting TACC3 induces immunogenic cell death and enhances T-DM1 response in HER2-positive breast cancer, Cancer Research (2024). DOI: 10.1158/0008-5472.CAN-23-2812 aacrjournals.org/cancerres/art … genic-cell-death-and

Provided by Medical University of South Carolina

Credit: Pixabay/CC0 Public Domain

metastatic breast cancer research papers

Featured News - Current News - Archived News - News Categories

Metro Creative Graphics

'Murder at the Mansion': Breast Cancer Network fundraiser for research

The Breast Cancer Network of WNY (BCN) will present its second annual “Murder at the Mansion” fundraiser on Friday, Feb. 23, at The Mansion on Delaware Avenue. The event is a murder mystery cocktail party that will benefit metastatic breast cancer research.

A press release stated, “BCN has a strong commitment to metastatic breast cancer and the many local women who are living with it. Since 2016, BCN has donated over $75,000 to research and hopes to push that number to $100,000 in 2024.”

BCN Executive Director Rob Jones said, “We see far too many women living with metastatic disease and ultimately lose their lives to it. Metastatic breast cancer causes over 42,000 deaths in the United States every year, and this is our way of trying to make a difference.”

The press release noted, “ ‘Murder at the Mansion’ offers a professional murder mystery performance along with an upscale cocktail party, open bar, and many raffle and auction opportunities.”

Jones said, “ ‘Murder at the Mansion’ is a great evening of fun and friendship, but the best part is uniting everyone in such an important cause.”

Event registration and more details can be found at www.bcnwny.org or by calling 716-706-0060.

The Breast Cancer Network of WNY is a local, independent, community-based organization with a mission to support the well-being of all Western New Yorkers impacted by breast cancer. Founded in 1988 and now located on Delaware Avenue in Buffalo, the organization has grown over the years to become a leading source of support, education, and advocacy for breast cancer survivors in Western New York.

Hometown News

metastatic breast cancer research papers

View All News

Follow Us On

           

metastatic breast cancer research papers

Niagara Frontier Publications

1859 Whitehaven Road, Grand Island, NY 14072 Phone: 716-773-7676 Fax: 716-773-7190

Get an Email Update!

All materials appearing on this Web page and website © copyright 2024 Niagara Frontier Publications. No materials may be reproduced without the express written consent of Niagara Frontier Publications.

metastatic breast cancer research papers

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • View all journals
  • My Account Login
  • Explore content
  • About the journal
  • Publish with us
  • Sign up for alerts
  • Open access
  • Published: 29 August 2019

Deep Learning to Improve Breast Cancer Detection on Screening Mammography

  • Li Shen   ORCID: orcid.org/0000-0002-5190-2851 1 ,
  • Laurie R. Margolies 2 ,
  • Joseph H. Rothstein 3 ,
  • Eugene Fluder 4 ,
  • Russell McBride   ORCID: orcid.org/0000-0002-3890-529X 5 &
  • Weiva Sieh 3  

Scientific Reports volume  9 , Article number:  12495 ( 2019 ) Cite this article

104k Accesses

474 Citations

75 Altmetric

Metrics details

  • Computational science
  • Computer science
  • Predictive markers

The rapid development of deep learning, a family of machine learning techniques, has spurred much interest in its application to medical imaging problems. Here, we develop a deep learning algorithm that can accurately detect breast cancer on screening mammograms using an “end-to-end” training approach that efficiently leverages training datasets with either complete clinical annotation or only the cancer status (label) of the whole image. In this approach, lesion annotations are required only in the initial training stage, and subsequent stages require only image-level labels, eliminating the reliance on rarely available lesion annotations. Our all convolutional network method for classifying screening mammograms attained excellent performance in comparison with previous methods. On an independent test set of digitized film mammograms from the Digital Database for Screening Mammography (CBIS-DDSM), the best single model achieved a per-image AUC of 0.88, and four-model averaging improved the AUC to 0.91 (sensitivity: 86.1%, specificity: 80.1%). On an independent test set of full-field digital mammography (FFDM) images from the INbreast database, the best single model achieved a per-image AUC of 0.95, and four-model averaging improved the AUC to 0.98 (sensitivity: 86.7%, specificity: 96.1%). We also demonstrate that a whole image classifier trained using our end-to-end approach on the CBIS-DDSM digitized film mammograms can be transferred to INbreast FFDM images using only a subset of the INbreast data for fine-tuning and without further reliance on the availability of lesion annotations. These findings show that automatic deep learning methods can be readily trained to attain high accuracy on heterogeneous mammography platforms, and hold tremendous promise for improving clinical tools to reduce false positive and false negative screening mammography results. Code and model available at: https://github.com/lishen/end2end-all-conv .

Introduction

The rapid advancement of machine learning and especially deep learning continues to fuel the medical imaging community’s interest in applying these techniques to improve the accuracy of cancer screening. Breast cancer is the second leading cause of cancer deaths among U.S. women 1 and screening mammography has been found to reduce mortality 2 . Despite the benefits, screening mammography is associated with a high risk of false positives as well as false negatives. The average sensitivity of digital screening mammography in the U.S. is 86.9% and the average specificity is 88.9% 3 . To help radiologists improve the predictive accuracy of screening mammography, computer-assisted detection and diagnosis (CAD) software 4 have been developed and in clinical use since the 1990s. Unfortunately, data suggested that early commercial CAD systems had not led to significant improvement in performance 5 , 6 , 7 and progress stagnated for more than a decade since they were introduced. With the remarkable success of deep learning in visual object recognition and detection, and many other domains 8 , there is much interest in developing deep learning tools to assist radiologists and improve the accuracy of screening mammography 9 , 10 , 11 , 12 , 13 , 14 . Recent studies 15 , 16 have shown that a deep learning based CAD system performed as well as radiologists in standalone mode and improved the radiologists’ performance in support mode.

Detection of subclinical breast cancer on screening mammography is challenging as an image classification task because the tumors themselves occupy only a small portion of the image of the entire breast. For example, a full-field digital mammography (FFDM) image is typically 4000 × 3000 pixels while a potentially cancerous region of interest (ROI) can be as small as 100 × 100 pixels. For this reason, many studies 13 , 17 , 18 , 19 , 20 , 21 have limited their focus to the classification of annotated lesions. Although classifying manually annotated ROIs is an important first step, a fully automated software system must be able to operate on the entire mammogram to provide additional information beyond the known lesions and augment clinical interpretations. If ROI annotations were widely available in mammography databases then established object detection and classification methods such as the region-based convolutional neural network (R-CNN) 22 and its variants 23 , 24 , 25 could be readily applied. However, approaches that require ROI annotations 14 , 26 , 27 , 28 , 29 often cannot be transferred to large mammography databases that lack ROI annotations, which are laborious and costly to assemble. Indeed, few public mammography databases are fully annotated 30 . Other studies 9 , 10 have attempted to train neural networks using whole mammograms without relying on any annotations. However, it is hard to know if such networks were able to locate the clinically significant lesions and base predictions on the corresponding portions of the mammograms. It is well known that deep learning requires large training datasets to be most effective. Thus, it is essential to leverage both the few fully annotated datasets, as well as larger datasets labeled with only the cancer status of each image to improve the accuracy of breast cancer classification algorithms.

Pre-training is a promising method to address the problem of training a classifier when the ideal large and complete training datasets are not available. For example, Hinton et al . 31 used layer-wise pre-training to initialize the weight parameters of a deep belief net (DBN) with three hidden layers and then fine-tuned it for classification. They found that pre-training improved the training speed as well as the accuracy of handwritten digit recognition. Another popular training method is to first train a deep learning model on a large database such as the ImageNet 32 and then fine-tune the model for another task. Although the specific task may not be related to the initial training dataset, the model’s weight parameters are already initialized for recognizing primitive features, such as edges, corners and textures, which can be readily used for a different task. This often saves training time and improves the model’s performance 33 .

In this study, we propose an “end-to-end” approach in which a model to classify local image patches is pre-trained using a fully annotated dataset with ROI information. The patch classifier’s weight parameters are then used to initialize the weight parameters of the whole image classifier, which can be further fine-tuned using datasets without ROI annotations. We used a large public digitized film mammography database with thousands of images to develop the patch and whole image classifiers, and then transferred the whole image classifiers to a smaller public FFDM database with hundreds of images. We evaluated various network designs for constructing the patch and whole image classifiers to attain the best performance. The pipeline required to build a whole image classifier is presented here, as well as the pros and cons of different training strategies.

Converting a classifier from recognizing patches to whole images

To perform classification or segmentation on large complex images, a common strategy involves the use of a classifier in sliding window fashion to recognize local patches on an image to generate a grid of probabilistic outputs. This is followed by another process to summarize the patch classifier’s outputs to give the final classification or segmentation result. Such methods have been used to detect metastatic breast cancer using whole slide images of sentinel lymph node biopsies 34 and to segment neuronal membranes in microscopic images 35 . However, this strategy requires two steps that each needs to be optimized separately. Here, we propose a method to combine the two steps into a single step for training on the whole images (Fig.  1 ). Assume we have an input patch \(X\in {{\rm{IR}}}^{{\rm{p}}\times {\rm{q}}}\) and a patch classifier which is a function f so that \(f(X)\in {{\rm{IR}}}^{{\rm{c}}}\) , where the function’s output satisfies f ( X ) i   ∈  [0, 1] and \({{\rm{\Sigma }}}_{i=1}^{c}f{(X)}_{i}=1\) and c is the number of classes of the patches. Here, c  = 5 and the classes are: benign calcification, malignant calcification, benign mass, malignant mass and background for each patch from a mammogram. Assume the input patch is extracted from an image \(M\in {{\rm{IR}}}^{{\rm{r}}\times {\rm{s}}}\) where p   ≪   r and q   ≪   s . If the function f represents a convolutional neural network (CNN), then f can be applied to M without changing the network parameters so that \(f(M)\in {{\rm{IR}}}^{{\rm{u}}\times {\rm{v}}\times {\rm{c}}}\) , where u  > 1 and v  > 1 depend on the image size and the stride of the patch classifier. This is possible because of the weight sharing and locality properties of a CNN 36 . If the function f represents a different class of neural networks, such as the multilayer perceptron (MLP), then this becomes infeasible because a MLP requires the input to be fixed. Therefore, after changing the input from X to M , we have a u  ×  v grid of probabilistic outputs of c classes (referred to as “heatmap”) instead of a single output of c classes. Hence the heatmap has a size of u  ×  v  ×  c . More layers can then be added on top of the heatmap to transform the outputs and connect with the final classification output of the image. Adding a convolutional layer on top of the patch classifier’s outputs turns the entire patch classifier into a filter and enlarges its receptive field. For example, if the patch classifier has a receptive field of 224 × 224 with a stride = 32, adding a 3 × 3 convolutional layer on top of it increases each side of the receptive field to 224 + (3 − 1) × 32 = 228. Thus, the top layers effectively use the patch classifier to “scan” the whole image, looking for cues of cancerous lesions and extracting higher level features that can finally be used for whole image classification. Using function g to represent the top layers, the whole image classification function can be written as \(h(M)=g(f(M))\in {{\rm{IR}}}^{{\rm{d}}}\) , where d is the number of classes of the whole image. Typically, d  = 2 represents the two classes we want to predict: malignant and nonmalignant (benign or normal).

figure 1

Converting a patch classifier to an end-to-end trainable whole image classifier using an all convolutional design. The function f was first trained on patches and then refined on whole images. We evaluated whether removing the heatmap improved information flow from the bottom layers of the patch classifier to the top convolutional layers in the whole image classifier. The magnifying glass shows an enlarged version of the heatmap. This figure is best viewed in color.

The function h accepts whole images as input and produces labels at the whole image level. Therefore, it is end-to-end trainable, providing two advantages over the two-step approach. First, the entire network can be jointly trained, avoiding sub-optimal solutions from each step; Second, the trained network can be transferred to another dataset without explicit reliance on ROI annotations. Large mammography databases with ROI annotations are rare and expensive to obtain. The largest public database with ROI annotations for digitized film mammograms – DDSM 37 – contains several thousand images with pixel-level annotations, which can be exploited to train a patch classifier f . Once the patch classifier is converted into a whole image classifier h , it can be fine-tuned on other databases using only image-level labels. This approach allows us to significantly reduce the requirement for ROI annotations, and has many applications in medical imaging in addition to breast cancer detection on screening mammograms.

Network design

A modern CNN is typically constructed by stacking convolutional layers on top of the input, followed by one or more fully connected (FC) layers to join with the classification output. Max pooling layers are often used amid convolutional layers to improve translational invariance and to reduce feature map size. In this study, two popular CNN structures are compared: the VGG network 38 and the residual network (Resnet) 39 . Consecutive network layers can be naturally grouped into “blocks” so that the feature map size is reduced (typically by a factor of 2) either at the beginning or at the end of a block but stays the same elsewhere in the block. For example, a “VGG block” is a stack of several 3 × 3 convolutional layers with the same depth followed by a 2 × 2 max pooling layer that reduces the feature map size by a factor of 2. Although other filter sizes can be used, 3 × 3 convolution and 2 × 2 max pooling are widely used, and employed throughout this study unless otherwise stated. Therefore, a VGG block can be represented by the pattern of N  ×  K , where N represents the depth of each convolutional layer and K represents the number of convolutional layers. A “Resnet block” uses stride = 2 in the first convolutional layer instead of 2 × 2 max pooling to reduce feature map size at the beginning of the block, followed by the stacking of several convolutional layers. We use the “bottleneck design 39 ” which consists of repeated units of three convolutional layers that have filter sizes of 1 × 1, 3 × 3 and 1 × 1, respectively. A key feature of the Resnet block is that a shortcut is made between the two ends of each unit so that the features are directly carried over and therefore each unit can focus on learning the “residual” information 39 . Batch normalization (BN) is used in every convolutional layer in the Resnet, which is known to speedup convergence and also has a regularization effect 40 . A Resnet block can be represented by the pattern of [ L  −  M  −  N ] ×  K , where L , M and N represent the depths of the three convolutional layers in a unit and K represents the number of units. Here, the 16-layer VGG network (VGG16) and the 50-layer Resnet (Resnet50) are used as patch classifiers. The original design of the VGG16 38 consisted of five VGG blocks followed by two FC layers. To be consistent with the Resnet50, we replaced the two FC layers with a global average pooling layer which calculates the average activation of each feature map for the output of the last VGG block. For example, if the output of the last VGG block has a size of 7 × 7 × 512 (height × width × channel), after the global average pooling layer the output becomes 512. This output is then connected to the classification output with a FC layer.

A straightforward approach to construct a whole image classifier from a patch classifier involves flattening the heatmap and connecting it to the image’s classification output using FC layers. To increase the model’s translational invariance to the patch classifier’s output, a max pooling layer can be used after the heatmap. Further, a shortcut can be made between the heatmap and the output to make the training easier. The heatmap results directly from the patch classifier’s output which uses the softmax activation:

However, the softmax activation diminishes gradients for large inputs, which may impede gradient flow when it is used in an intermediate layer. Therefore, the rectified linear units (ReLU) can be used instead:

In the following, when we refer to the heatmap in a whole image classifier, the activation is always assumed to be ReLU unless otherwise stated.

We further propose to use convolutional layers as top layers, which preserve spatial information. Two blocks of convolutional layers (VGG or Resnet) can be added on top of the patch classifier layers, followed by a global average pooling layer and then the image’s classification output (Fig.  1 ). Therefore, this design creates an “all convolutional” network for whole image classification. As Fig.  1 shows, the heatmap abruptly reduces the depth of the feature map between the patch classifier layers and the top layers, which may cause information loss in the whole image classification. Therefore, we also evaluated the results when the heatmap is removed entirely from the whole image classifier to allow the top layers to fully utilize the features extracted from the patch classifier.

Computational environment

All experiments in this study were carried out on a Linux workstation equipped with an NVIDIA 8 GB Quadro M4000 GPU card.

Developing patch and whole image classifiers on CBIS-DDSM

Setup and processing of the dataset.

The DDSM 37 contains digitized film mammograms in a lossless-JPEG format that is now obsolete. We used a later version of the database called CBIS-DDSM 41 which contains images that are converted into the standard DICOM format. The dataset which consisted of 2478 mammography images from 1249 women was downloaded from the CBIS-DDSM website, and included both craniocaudal (CC) and mediolateral oblique (MLO) views for most of the exams. Each view was treated as a separate image in this study. We randomly split the CBIS-DDSM dataset 85:15 at the patient level to create independent training and test sets. The training data was further split 90:10 to create an independent validation set. The splits were done in a stratified fashion to maintain the same proportion of cancer cases in the training, validation and test sets. The total numbers of images in the training, validation and testing sets were: 1903, 199 and 376, respectively.

The CBIS-DDSM database contains the pixel-level annotations for the ROIs and their pathologically confirmed labels: benign or malignant. It further labels each ROI as a calcification or mass. Most mammograms contained only one ROI. All mammograms were converted into PNG format and downsized to 1152 × 896 using interpolation; no image cropping was performed. The downsizing was motivated by the limitation of GPU memory size. Two patch datasets were created by sampling image patches from ROIs and background regions. All patches had the same size of 224 × 224, which were large enough to cover most of the ROIs annotated. The first dataset (S1) consisted of sets of patches in which one was centered on the ROI and one is a random background patch from the same image. The second dataset (S10) consisted of 10 patches randomly sampled from around each ROI, with a minimum overlapping ratio of 0.9 with the ROI and inclusion of some background, to more completely capture the potentially informative region; and an equal number of background patches from the same image. All patches were classified into one of the five categories: background, malignant mass, benign mass, malignant calcification and benign calcification.

Network training

Training a whole image classifier was achieved in two steps. The first step was to train a patch classifier. We compared the networks with pre-trained weights using the ImageNet 32 database to those with randomly initialized weights. In a pre-trained network, the bottom layers represent primitive features that tend to be preserved across different tasks, whereas the top layers represent higher-order features that are more related to specific tasks and require further training. Using the same learning rate for all layers may destroy the features that were learned in the bottom layers. To prevent this, a 3-stage training strategy was employed in which the parameter learning is frozen for all but the final layer and progressively unfrozen from the top to the bottom layers, while simultaneously decreasing the learning rate. The 3-stage training strategy on the S10 patch set was as follows:

Set learning rate to 10 −3 and train the last layer for 3 epochs.

Set learning rate to 10 −4 , unfreeze the top layers and train for 10 epochs, where the top layer number is set to 46 for Resnet50 and 11 for VGG16.

Set learning rate to 10 −5 , unfreeze all layers and train for 37 epochs for a total of 50 epochs.

In the above, an epoch was defined as a sweep through the training set. For the S1 patch dataset, the total number of epochs was increased to 200 because it was much smaller and less redundant than the S10 patch dataset. For randomly initialized networks a constant learning rate of 10 −3 was used. Adam 42 was used as the optimizer and the batch size was set to be 32. The sample weights were adjusted within each batch to balance the five classes.

The second step was to train a whole image classifier converted from the patch classifier (Fig.  1 ). A 2-stage training strategy was employed to first train the newly added top layers (i.e. function g ) and then train all layers (i.e. function h ) with a reduced learning rate, which was as follows:

Set learning rate to 10 −4 , weight decay to 0.001 and train the newly added top layers for 30 epochs.

Set learning rate to 10 −5 , weight decay to 0.01 and train all layers for 20 epochs for a total of 50 epochs.

We found that the VGG-based image classifiers showed sign of continuing improvement towards the end of the 50 epochs, while the Resnet-based image classifiers had already converged. To be fair for the VGG-based image classifiers, we continued to train them with 200 additional epochs. Due to GPU memory limits, a batch size of 2 was used.

The average gray scale value of the whole image training set was subtracted from both patch and whole image datasets in training. No other preprocessing was applied. To improve the generalization of final models, data augmentation was performed using the following random transformations: horizontal and vertical flips, rotation in [−25, 25] degrees, zoom in [0.8, 1.2] ratio and intensity shift in [−20, 20] pixel values.

Development of patch classifiers

Table  1 shows the accuracy of the classification of image patches into 5 classes using Resnet50 and VGG16 in the CBIS-DDSM test set. A bootstrapping method with 3000 runs was used to derive the 95% confidence intervals for patch classification accuracy. The S10 set was more difficult to classify than the S1 set because it contained patches sampled from around ROIs, rather than centered on the ROI, that were more difficult to distinguish from background regions. On the S1 set, both randomly initialized and pre-trained Resnet50 classifiers achieved similar accuracy but the pre-trained network converged after half as many epochs as the randomly initialized one. On the S10 set, the pre-trained Resnet50 outperformed the randomly initialized one by a large margin, achieving an accuracy [95% confidence interval (CI)] of 0.89 [0.88, 0.90]. These results showed that pre-training can greatly help network convergence and performance. Therefore, pre-trained networks were used for the rest of the study. The accuracy of the pre-trained VGG16 (0.84 [0.83, 0.85]) on the S10 set was lower than that of the pre-trained Resnet50.

To further characterize performance, confusion matrix analyses were conducted on the Resnet50 and VGG16 patch classifiers in the S10 test set (Fig.  2 ). For both patch classifiers, all five classes were predicted into the correct categories with the highest probability. The background class was easiest, and malignant calcifications hardest to classify. Malignant calcifications were most likely to be misclassified as benign calcification, followed by malignant mass. Benign calcifications were most likely to be misclassified as background, followed by malignant calcification. Malignant masses were most likely to be misclassified as benign masses, while benign masses were most likely to be misclassified as malignant masses or background, depending on the patch classifier.

figure 2

Confusion matrix analysis of 5-class patch classification for Resnet50 ( a ) and VGG16 ( b ) in the S10 test set. The matrices are normalized so that each row sums to one. This figure is best viewed in color.

Converting patch to whole image classifiers

Using pre-trained Resnet50 and VGG16 patch classifiers, we tested several different configurations for the top layers of the whole image classifiers. We also evaluated removal of the heatmap and adding two Resnet or VGG blocks on top of the patch classifier layers, followed by a global average pooling layer and the classification output. Model performance was assessed by computing the per-image AUCs on the independent test set.

Resnet-based networks: To evaluate whether the patch classifiers trained on the S1 and S10 datasets are equally useful for whole image classification, the Resnet50 patch classifiers were used. In the original design of the Resnet50 39 , L  ≡  M , N is four times L and K is 3 or more; the L of the current block is also double of the L of the previous block. However, we found this design to exceed our GPU memory limit when it is used for the top layers of the whole image classifier. In the initial experiments, we used instead the same configuration of [512 − 512 − 2048] × 1 for two Resnet blocks on top of the patch classifier. A bootstrapping method with 3000 runs was used to derive 95% confidence intervals for AUCs and AUC differences. The whole image classifier trained using the S10 set (mean AUC = 0.85) performed much better than that trained using the S1 set (mean AUC = 0.63) (Table  2 ), despite its poorer patch classification accuracy (Table  1 ). The S10 dataset contains more information about the ROIs as well as their adjacent regions and other background regions on the image than the S1 dataset, which allows a patch classifier to extract more features that can be important for whole image classification. To test this hypothesis, we created another patch set (referred to as S1g) with one patch each from the ROI and background but a large patch size of 448 × 448 to include the surrounding area. The patch classification accuracy in S1g was much lower than that in S10 (Table  1 ). However, the image classification accuracy was similar for models trained on S1g and S10 (Table  2 ) with an estimated AUC difference [95% confidence interval] of −0.023 [−0.061, 0.016] supporting the hypothesis that the background regions contain useful information. For the rest of the study, only patch classifiers trained on the S10 dataset were used. Varying the configuration by using two Resnet blocks of [512 − 512 − 1024] × 2 yielded a mean AUC of 0.86, while reducing the depths and K of the two Resnet blocks to: [256 − 256 − 256] × 1 and [128 − 128 − 128] × 1 did not significantly decrease the AUC (Table  2 ). This result showed that the depths of the Resnet blocks were relatively uncorrelated with the performance of the whole image classifiers.

VGG-based networks: We tested whole image classifiers using VGG16 as the patch classifier and VGG blocks as the top layers. BN was used for the VGG blocks on the top except for the VGG16 patch classifier because it is a pre-trained network which cannot be modified. The VGG-based whole image classifiers performed similarly to the Resnet-based ones but took longer to achieve the same performance level (Table  3 ). In contrast to the Resnet, using more convolutional layers and higher depths in VGG blocks led to poorer performance: using two VGG blocks of 256 × 1 and 128 × 1 (mean AUC = 0.85) performed better than two VGG blocks of 512 × 3 (mean AUC = 0.81) with an AUC difference of 0.041 [0.011, 0.071]. Reducing the depths further to 128 and 64 did not further improve the AUC. This result illustrates that controlling model complexity (i.e., #layers and depths) is important for achieving good performance with the VGG-based networks, which are more likely to overfit.

Hybrid networks: We also created two “hybrid” networks by adding the VGG top layers that performed the best (two VGG blocks of 256 × 1 and 128 × 1) on top of the Resnet50 patch classifier; and the Resnet top layers that performed the best (two Resnet blocks of the same configuration of [512 − 512 − 1024] × 2) on top of the VGG16 patch classifier. The two hybrid networks achieved mean AUCs of 0.87 and 0.85, respectively, and were among the best performing models (Tables  2 and 3 ).

Augmented prediction and model averaging: Augmented prediction was implemented by horizontally and vertically flipping an image to obtain four images and taking an average of the four images’ scores. This technique increased the AUC (referred to as A-AUC) for each model by 0.01–0.03 (Tables  2 and 3 ), although only some of the models showed significant increase based on the 95% confidence intervals of AUC differences (Table  S1 ). The four best performing models were combined into an ensemble model by taking the average of their augmented prediction scores. Two of the four best models used Resnet50 and VGG16 as patch classifiers and Resnet and VGG blocks as top layers, respectively (referred to as Resnet-Resnet and VGG-VGG); and the remaining two were hybrid models (referred to as Resnet-VGG and VGG-Resnet). Figure  3a shows the Receiver Operating Characteristic (ROC) curves of the four best models and the ensemble model, which yielded an AUC of 0.91. Because the clinical significance of a false negative (FN) is higher than that of a false positive (FP), a clinically useful system should not have significantly lower sensitivity than the current standard of care. Therefore, we evaluated model performance using a sensitivity of 86% as a benchmark based upon the estimated average for U.S. radiologists 3 , and determined the model specificity to be 80.1% at a similar sensitivity of 86.1%.

figure 3

ROC curves for the four best individual models and ensemble model on the CBIS-DDSM ( a ) and INbreast ( b ) test sets. This figure is best viewed in color.

Saliency map and error analysis: Saliency maps were created using the Resnet-VGG model (Fig.  4a–c ), which showed the gradients of the input image with respect to the cancer class output. We used the guided back-propagation approach 43 that calculates only positive gradients for positive activations. A saliency map illustrates which area of the input image is considered to be responsible for the cancer prediction by a whole image classifier. Figure  4a shows the saliency map of a true positive (TP) image where the identified area is in or close to the malignant ROI. This shows that the image classifier was able to correctly locate the cancerous region on which its decision was based. Figure  4b shows a typical FP image where the identified region is located in a benign ROI that resembles a malignant ROI. Figure  4c shows a typical FN image where the malignant ROI is difficult to discern and no response passes the low cutoff.

figure 4

Saliency maps of TP ( a ), FP ( b ) and FN ( c ) image classifications. The outlines represent the regions of interest annotated by the radiologist, and biopsy-confirmed to contain either malignant (blue) or benign (green) tissue. The red dots represent the gradients of the input image with respect to the cancer class output. The gradients were rescaled to be within [0, 1] and a low cutoff of 0.06 was used to remove background noise. Heatmaps ( d ) of the four non-background classes for input image ( a ). The colors of the heatmaps represent the activation values after ReLU. This figure is best viewed in color.

Combining the CC and MLO views for prediction: Combining the CC and MLO views may increase performance because each view can contain unique information. After removing the samples where only a single view was available, 90% of the test set remained for analysis of both views from each of 169 breasts. We used a simple approach of taking the average score of the two views. A breast-level bootstrapping method was used (3000 runs) to compare two-view vs. single-view AUCs for the four best models above: Resnet-Resnet, Resnet-VGG, VGG-VGG and VGG-Resnet. The mean AUC differences were 0.030 [0.018, 0.042], 0.027 [0.016, 0.037], 0.040 [0.028, 0.051] and 0.048 [0.032, 0.064], respectively. Thus, using two views when available significantly increased the AUCs in comparison to single views for all the models tested here.

Max-pooling, shortcut and FC layers: We tested an alternative design by using the heatmap followed by a max-pooling and two FC layers, including a shortcut between the heatmap and the classification output. The Resnet50 and VGG16 patch classifiers were used. The FC layer sizes were chosen to gradually reduce the layer outputs. When the pooling size increased from 1 × 1 (i.e. no pooling) to 5 × 5, the AUCs did not show significant changes with the exception of pooling size 1 × 1 for the Resnet50 patch classifier, in which case the AUC was significantly lower than the others (Tables  2 and 3 ). The best mean AUC for these models was 0.74, falling short of the performance of the all convolutional models.

Evaluation of the heatmap in all convolutional networks: To test our hypothesis that the heatmap can cause information loss in the whole image classification network, we inserted a heatmap in the Resnet-based whole image classifier with two [512 − 512 − 1024] × 2 blocks as top layers. The heatmap inserted was a 1 × 1 convolutional layer that reduces the number of filters from the previous convolutional layer (2048) to 5, which corresponds to the 5 classes of the patch classifier. To facilitate the back-propagation of gradients, ReLU was used to replace the softmax activation in the heatmap. Figure  4d shows an example heatmap that provides a rough segmentation of the input image; the top layers then use the segmentation to classify the whole image. This model achieved a mean AUC of 0.80 (Table  2 ), which was significantly lower than that of the same classifier without the heatmap with an AUC difference of −0.050 [−0.088, −0.012]. To exclude the possibility that the top layers were overfit due to the shallow depth of the heatmap, another model with reduced complexity using two Resnet blocks of [64 − 64 − 256] × 2 and [128 − 128 − 512] × 2 was tested, which achieved a similar mean AUC of 0.81 (AUC difference of −0.044 [−0.075, −0.012]). These results indicated that removing the heatmap was beneficial to the whole image classification networks.

Comparison to prior two-step approach: Finally, for comparison we tested a previously reported approach 34 that used a probability cutoff to binarize the heatmap into a binary image that represents each pixel as background (0) or ROI (1). This was repeated for each of the 4 foreground classes. We then extracted regional features (such as area, major axis length and mean intensity) from the ROIs of the binary images and trained a random forest classifier (#trees = 500, max depth = 9, min samples split = 300) on the regional features. The Resnet50 patch classifier was used and the softmax activation was used in the heatmap to obtain the probabilities for the 5 classes. Four cutoffs—0.3, 0.5, 0.7 and 0.9 were used to binarize the heatmaps and the regional features were combined. This approach achieved an AUC of 0.73, and was inferior to the all convolutional models.

Transfer learning for whole image classification on INbreast

The INbreast 30 dataset is a public database containing more recently acquired FFDM images. These images have different intensity profiles compared with digitized film mammograms from the CBIS-DDSM, as illustrated by example images from the two databases (Fig.  5 ). Therefore, INbreast provides an excellent opportunity to test the transferability of a whole image classifier across mammography platforms. The INbreast database contains 115 patients and 410 mammograms including both CC and MLO views. We analyzed each view separately like above. The INbreast database includes radiologists’ BI-RADS 44 assessment categories which are defined as follows: 0, incomplete exam; 1, no findings; 2, benign; 3, probably benign; 4, suspicious; 5, highly suggestive of malignancy; and 6, known biopsy-proven cancer. Because the database lacks reliable pathological confirmation of malignancy, we assigned all images with BI-RADS 1 and 2 as negative; BI-RADS 4, 5 and 6 as positive; and excluded 12 patients and 23 images with BI-RADS 3 since this assessment is typically not given at screening. We split the dataset 70:30 into training and test sets at the patient-level while maintaining the same ratio of positive and negative images. The total numbers of images in the training and test sets were 280 from 72 women and 107 from 31 women, respectively. We used the same processing steps on the INbreast images as for the CBIS-DDSM images.

figure 5

Representative examples of a digitized film mammogram from CBIS-DDSM and a digital mammogram from INbreast.

Effectiveness and efficiency of transfer learning

Although the INbreast database contains ROI annotations, they were ignored to test the transferability of the whole image classifier across different mammography platforms and databases. The four best performing models (See Tables  2 and 3 ) were directly fine-tuned on the INbreast training set and evaluated by computing per-image AUCs on the test set. Adam 42 was used as the optimizer and the learning rate was set at 10 −5 . The number of epochs was set at 200 and the weight decay at 0.01. All four models achieved an AUC of 0.95 (Table  4 ). The ensemble model based on averaging the four best models improved the AUC to 0.98 with a corresponding sensitivity of 86.7% and specificity of 96.1% (Fig.  3b ).

We also sought to determine the minimum amount of data required to fine-tune a whole image classifier to a satisfactory level of performance, to guide future studies in minimizing the resource intensive process of obtaining labels. Training subsets with 20, 30, 40, 50 and 60 patients were sampled for fine-tuning and model performance was evaluated using the same test set (Table  4 ). With as few as 20 patients or 79 images, the four models already attained AUCs between 0.87 and 0.92. The AUCs quickly approached the maximum as the training subset size increased. These results suggest that the intensive part of learning is to recognize the shapes and textures of the benign and malignant ROIs and normal tissues, and that adjusting to different intensity profiles found in different mammography datasets may require much less data. Importantly, these results clearly demonstrate that the end-to-end training approach can be successfully used to fine-tune a whole image classifier using additional small training sets with image-level labels, greatly reducing the burden of training set construction for multiple different mammography platforms.

This study shows that accurate classification of screening mammograms can be achieved with a deep learning model trained in an end-to-end fashion that relies on clinical ROI annotations only in the initial stage. Once the whole image classifier is built, it can be fine-tuned using additional datasets that lack ROI annotations, even if the pixel intensity distributions differ as is often the case for datasets assembled from heterogeneous mammography platforms. These findings indicate that deep learning algorithms can improve upon classic commercial CAD systems, such as iCAD SecondLook 1.4 and R2 ImageChecker Cenova 1.0, that are not deep learning based and have been reported to attain an average AUC of 0.72 6 . Our all convolutional networks trained using an end-to-end approach have highly competitive performance and are more generalizable across different mammography platforms compared with previous deep learning methods that have achieved AUCs in the range of 0.65–0.97 on the DDSM and INbreast databases, as well as in-house datasets 12 . Two recent studies reported that a new commercial CAD system, Transpara 1.4.0, attained an AUC of 0.89 when used to support radiologists 16 and 0.84 in standalone mode 15 . This commercial CAD used CNNs trained using the lesion annotations from 9000 mammograms with cancer to generate scores at the patch level; the scores for all detected regions were then combined into a score at the examination level. To our knowledge, the commercial CAD cannot easily be fine-tuned on different mammography datasets without lesion annotations. Our approach has the advantage of requiring only image-level labels for fine-tuning once the whole image classifier is built to facilitate scaling to larger datasets and transferring to new mammography systems as they rapidly evolve.

Two recent studies 45 , 46 developed deep learning based methods for breast cancer classification using film and digital mammograms, which were end-to-end trainable. Both studies used multi-instance learning (MIL) and modified the whole image classifier cost functions to satisfy the MIL criterion. In contrast to our approach, neither study utilized ROI annotations to train the patch classifiers first and the AUCs were lower than reported in this study. We found that the quality of the patch classifiers is critical to the accuracy of the whole image classifiers. This was supported by two lines of evidence. First, the whole image classifier based on the S10 patch set performed far better than the one based on the S1 patch set because the S10 patch set contained more information about the background than the S1 patch set. Second, it took much longer for the VGG16-based whole image classifiers to achieve the same performance as the Resnet50-based classifiers because the VGG16 was less accurate than Resnet50 in patch classification.

We also found that the accuracy of whole image classification was improved by sampling more or larger patches to include neighboring regions around the ROI and additional background regions. However, the computational burden increases linearly with the number or size of patches sampled and the performance gain may quickly diminish. Using larger patches can decrease the signal-to-noise ratio, as indicated by the lower patch classification accuracy using the S1g vs. S10 patch sets. Using larger patches also requires higher GPU memory, which may limit network choices. The saliency map analysis showed that our whole image networks were able to correctly identify the ROIs and use the information therein to predict cancer. It also showed that classification errors typically occurred in difficult cases, such as benign lesions with malignant features, or malignant lesions that were difficult to distinguish from background. Further research is needed to investigate how to sample local patches more efficiently, perhaps by augmenting the training data with difficult cases and focusing on the patches that are more likely to be misclassified. This could help overcome the computational burden of training more accurate classifiers.

Although the VGG-based image classifiers were more prone to overfitting and required longer training, the performance of VGG-based and Resnet-based image classifiers was comparable. The fact that the ensemble model performed better than any of the individual models also suggests that the VGG-based and Resnet-based classifiers can complement each other. Moreover, the VGG16 (without the two FC layers), with 15 million weight parameters, is a much smaller network than the Resnet50, with 24 million weight parameters. Having fewer parameters reduces memory requirements and training time per epoch, which is important when computational resources are limited. The Resnet is a more recently developed deep learning method, which is enhanced by shortcuts and batch normalization, both techniques that may help the network train faster and generalize better. The same techniques can be used in the VGG-based networks as well in future work, which may improve the VGG-based classifiers.

This study had some limitations. Mammograms were downsized to fit the available GPU (8 GB). As more GPU memory becomes available, future studies will be able to train models using larger image sizes, or retain the original image resolution without the need for downsizing. Retaining the full resolution of modern digital mammography images will provide finer details of the ROIs and likely improve performance. Although the CBIS-DDSM dataset included pathological confirmation of all cancer diagnoses, the INbreast dataset did not. Therefore, we used the radiologists’ BI-RADS assessments to assign labels to the images in the INbreast dataset, which has the limitation of reproducing radiologists’ impressions instead of discovering new characteristics of malignant lesions. It would be of interest in future work to include interval breast cancers that were missed by radiologists, to help train algorithms to detect more subtle signs of malignancy that may not be visually apparent. Finally, the CBIS-DDSM and INbreast datasets were not nationally representative samples and performance metrics in these datasets are not directly comparable to national estimates of radiologists’ sensitivity and specificity. Future direct comparisons between algorithms and radiologists will be facilitated by public sharing of the code and greater availability of representative benchmarking datasets.

In conclusion, our study demonstrates that deep learning models trained in an end-to-end fashion can be highly accurate and potentially readily transferable across diverse mammography platforms. Deep learning methods have enormous potential to further improve the accuracy of breast cancer detection on screening mammography as the available training datasets and computational resources expand. Our approach may assist future development of superior CAD systems that could be used to help prioritize the most suspicious cases to be read by a radiologist, or as an automatic second reader after making an initial independent interpretation. Our end-to-end approach can also be applied to other medical imaging problems where ROI annotations are scarce.

Data Availability

A preprint version of this article is available at: https://arxiv.org/abs/1708.09427 .

American Cancer Society. How Common Is Breast Cancer? https://www.cancer.org/cancer/breast-cancer/about/howcommon-is-breast-cancer.html (2018).

Oeffinger, K. C. et al . Breast Cancer Screening for Women at Average Risk: 2015 Guideline Update From the American Cancer Society. JAMA 314 , 1599–1614 (2015).

Article   CAS   Google Scholar  

Lehman, C. D. et al . National Performance Benchmarks for Modern Screening Digital Mammography: Update from the Breast Cancer Surveillance Consortium. Radiol. 283 , 49–58 (2016).

Article   Google Scholar  

Elter, M. & Horsch, A. CADx of mammographic masses and clustered microcalcifications: A review. Med. Phys. 36 , 2052–2068 (2009).

Fenton, J. J. et al . Influence of Computer-Aided Detection on Performance of Screening Mammography. New Engl. J. Medicine 356 , 1399–1409 (2007).

Cole, E. B. et al . Impact of Computer-Aided Detection Systems on Radiologist Accuracy With Digital Mammography. Am. J. Roentgenol. 203 , 909–916 (2014).

Lehman, C. D. et al . Diagnostic Accuracy of Digital Screening Mammography With and Without Computer-Aided Detection. JAMA Intern. Medicine 175 , 1828–1837 (2015).

LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nat. 521 , 436–444 (2015).

Article   ADS   CAS   Google Scholar  

Aboutalib, S. S. et al . Deep Learning to Distinguish Recalled but Benign Mammography Images in Breast Cancer Screening. Clin . Cancer Res . (2018).

Kim, E.-K. et al . Applying Data-driven Imaging Biomarker in Mammography for Breast Cancer Screening: Preliminary Study. Sci. Reports 8 , 2762 (2018).

Article   ADS   Google Scholar  

Hamidinekoo, A., Denton, E., Rampun, A., Honnor, K. & Zwiggelaar, R. Deep learning in mammography and breast histology, an overview and future trends. Med. Image Analysis 47 , 45–67 (2018).

Burt, J. R. et al . Deep learning beyond cats and dogs: Recent advances in diagnosing breast cancer with deep neural networks. The Br . J . Radiol . 20170545 (2018).

Kooi, T. et al . Large scale deep learning for computer aided detection of mammographic lesions. Med. Image Analysis 35 , 303–312 (2017).

Agarwal, R., Diaz, O., Lladó, X., Yap, M. H. & Martí, R. Automatic mass detection in mammograms using deep convolutional neural networks. J. Med. Imaging 6 , 031409 (2019).

Rodriguez-Ruiz, A. et al . Stand-Alone Artificial Intelligence for Breast Cancer Detection in Mammography: Comparison With 101 Radiologists. JNCI: J . Natl . Cancer Inst (2019).

Rodríguez-Ruiz, A. et al . Detection of Breast Cancer with Mammography: Effect of an Artificial Intelligence Support System. Radiol. 290 , 305–314 (2018).

Jamieson, A. R., Drukker, K. & Giger, M. L. Breast image feature learning with adaptive deconvolutional networks. Proc. SPIE 8315, (6–13 (2012).

Google Scholar  

Arevalo, J., González, F. A., Ramos-Pollán, R., Oliveira, J. L. & Lopez, M. A. G. Convolutional neural networks for mammography mass lesion classification. In 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society ( EMBC ), 797–800, https://doi.org/10.1109/EMBC.2015.7318482 (2015).

Arevalo, J., González, F. A., Ramos-Pollán, R., Oliveira, J. L. & Guevara Lopez, M. A. Representation learning for mammography mass lesion classification with convolutional neural networks. Comput. Methods Programs Biomed. 127 , 248–257 (2016).

Lévy, D. & Jain, A. Breast Mass Classification from Mammograms using Deep Convolutional Neural Networks. arXiv preprint arXiv:1612 . 00542 (2016).

Dhungel, N., Carneiro, G. & Bradley, A. P. The Automated Learning of Deep Features for Breast Mass Classification from Mammograms. In Ourselin, S., Joskowicz, L., Sabuncu, M. R., Unal, G. & Wells, W. (eds) Medical Image Computing and Computer-Assisted Intervention – MICCAI 2016: 19th International Conference , Athens , Greece , October 17 – 21 , 2016 , Proceedings , Part II , 106–114 (Springer International Publishing, Cham, 2016).

Girshick, R., Donahue, J., Darrell, T. & Malik, J. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. In Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition , CVPR’ 14, 580–587 (IEEE Computer Society, Washington, DC, USA, 2014).

Girshick, R. Fast r-cnn. In Proceedings of the IEEE International Conference on Computer Vision , 1440–1448 (2015).

Ren, S., He, K., Girshick, R. & Sun, J. Faster R-CNN: Towards real-time object detection with region proposal networks. In Advances in Neural Information Processing Systems , 91–99 (2015).

Dai, J., Li, Y., He, K. & Sun, J. R-FCN: Object Detection via Region-based Fully Convolutional Networks. arXiv:1605 . 06409 [ cs ], 1605.06409 (2016).

Dhungel, N., Carneiro, G. & Bradley, A. P. Automated Mass Detection in Mammograms Using Cascaded Deep Learning and Random Forests. In 2015 International Conference on Digital Image Computing: Techniques and Applications ( DICTA ), 1–8, https://doi.org/10.1109/DICTA.2015.7371234 (2015).

Ertosun, M. G. & Rubin, D. L. Probabilistic visual search for masses within mammography images using deep learning. In 2015 IEEE International Conference on Bioinformatics and Biomedicine ( BIBM ), 1310–1315, https://doi.org/10.1109/BIBM.2015.7359868 (2015).

Akselrod-Ballin, A. et al . A Region Based Convolutional Network for Tumor Detection and Classification in Breast Mammography. In Carneiro, G. et al . (eds) Deep Learning and Data Labeling for Medical Applications: First International Workshop , LABELS 2016 , and Second International Workshop , DLMIA 2016 , Held in Conjunction with MICCAI 2016 , Athens , Greece , October 21 , 2016 , Proceedings , 197–205 (Springer International Publishing, Cham, 2016).

Ribli, D., Horváth, A., Unger, Z., Pollner, P. & Csabai, I. Detecting and classifying lesions in mammograms with Deep Learning. arXiv:1707 . 08401 [ cs ], 1707.08401 (2017).

Moreira, I. C. et al . INbreast: Toward a Full-field Digital Mammographic Database. Acad. Radiol. 19 , 236–248 (2012).

Hinton, G. E., Osindero, S. & Teh, Y.-W. A fast learning algorithm for deep belief nets. Neural Comput. 18 , 1527–1554 (2006).

Article   MathSciNet   Google Scholar  

Russakovsky, O. et al . ImageNet Large Scale Visual Recognition Challenge. Int. J. Comput. Vis. 115 , 211–252 (2015).

Oquab, M., Bottou, L., Laptev, I. & Sivic, J. Learning and Transferring Mid-Level Image Representations using Convolutional Neural Networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 1717–1724 (2014).

Wang, D., Khosla, A., Gargeya, R., Irshad, H. & Beck, A. H. Deep Learning for Identifying Metastatic Breast Cancer. arXiv:1606 . 05718 [ cs , q-bio ], 1606.05718 (2016).

Ciresan, D., Giusti, A., Gambardella, L. M. & Schmidhuber, J. Deep Neural Networks Segment Neuronal Membranes in Electron Microscopy Images. In Pereira, F., Burges, C. J. C., Bottou, L. & Weinberger, K. Q. (eds) Advances in Neural Information Processing Systems 25 , 2843–2851 (Curran Associates, Inc., 2012).

LeCun, Y. et al . Convolutional networks for images, speech, and time series. The handbook brain theory neural networks 3361 , 1995 (1995).

Heath, M., Bowyer, K., Kopans, D., Moore, R. & Kegelmeyer, W. The Digital Database for Screening Mammography. In Yaffe, M. (ed.) Proceedings of the Fifth International Workshop on Digital Mammography , 212–218 (Medical Physics Publishing, 2001).

Simonyan, K. & Zisserman, A. Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv:1409 . 1556 [ cs ], 1409.1556 (2014).

He, K., Zhang, X., Ren, S. & Sun, J. Deep Residual Learning for Image Recognition. arXiv:1512 . 03385 [ cs ], 1512.03385 (2015).

Ioffe, S. & Szegedy, C. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. arXiv:1502 . 03167 [ cs ], 1502.03167 (2015).

Lee, R. S., Gimenez, F., Hoogi, A. & Rubin, D. Curated Breast Imaging Subset of DDSM. The Cancer Imaging Arch (2016).

Kingma, D. P. & Ba, J. Adam: A Method for Stochastic Optimization. arXiv:1412 . 6980 [ cs ], 1412.6980 (2014).

Springenberg, J. T., Dosovitskiy, A., Brox, T. & Riedmiller, M. Striving for Simplicity: The All Convolutional Net. arXiv:1412 . 6806 [ cs ], 1412.6806 (2014).

D’Orsi, C., Sickles, E., Mendelson, E. & Morris, E. ACR BI-RADS R Atlas , Breast Imaging Reporting and Data System , 5th edn (American College of Radiology, Reston, Va, 2013).

Zhu, W., Lou, Q., Vang, Y. S. & Xie, X. Deep Multi-instance Networks with Sparse Label Assignment for Whole Mammogram Classification. arXiv:1705 . 08550 [ cs ], 1705.08550 (2017).

Choukroun, Y. et al . Mammogram Classification and Abnormality Detection from Nonlocal Labels using Deep Multiple Instance Neural Network. In Eurographics Workshop on Visual Computing for Biology and Medicine (2017).

Download references

Acknowledgements

This work was partially supported by the Friedman Brain Institute, the Tisch Cancer Institute (NIH P30CA196521) and a Clinical and Translational Science Award (NIH UL1TR001433). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. This work was supported in part through the computational resources and staff expertise provided by the Department of Scientific Computing at the Icahn School of Medicine at Mount Sinai.

Author information

Authors and affiliations.

Icahn School of Medicine at Mount Sinai (ISMMS), Department of Neuroscience, New York, 10029, USA

ISMMS, Department of Diagnostic, Molecular, and Interventional Radiology, New York, 10029, USA

Laurie R. Margolies

ISMMS, Department of Population Health Science and Policy and Department of Genetics and Genomic Sciences, New York, 10029, USA

Joseph H. Rothstein & Weiva Sieh

ISMMS, Department of Scientific Computing, New York, 10029, USA

Eugene Fluder

ISMMS, Department of Pathology, New York, 10029, USA

Russell McBride

You can also search for this author in PubMed   Google Scholar

Contributions

L.S. developed the algorithm, conceived and conducted the experiments and analyzed the results. L.R.M., J.R., R.M. and W.S. analyzed the results. E.F. helped with the computational resources. All authors contributed to writing and reviewing the manuscript.

Corresponding author

Correspondence to Li Shen .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Shen, L., Margolies, L.R., Rothstein, J.H. et al. Deep Learning to Improve Breast Cancer Detection on Screening Mammography. Sci Rep 9 , 12495 (2019). https://doi.org/10.1038/s41598-019-48995-4

Download citation

Received : 14 May 2019

Accepted : 07 August 2019

Published : 29 August 2019

DOI : https://doi.org/10.1038/s41598-019-48995-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Are better ai algorithms for breast cancer detection also better at predicting risk a paired case–control study.

  • Ruggiero Santeramo
  • Celeste Damiani
  • Adam R. Brentnall

Breast Cancer Research (2024)

Dual view deep learning for enhanced breast cancer screening using mammography

  • Samuel Rahimeto Kebede
  • Fraol Gelana Waldamichael
  • Zelalem Chimdesa Merga

Scientific Reports (2024)

Deep learning system for true- and pseudo-invasion in colorectal polyps

  • Charles Ling

Robustness of Deep Networks for Mammography: Replication Across Public Datasets

  • Osvaldo M. Velarde
  • Clarissa Lin
  • Lucas C. Parra

Journal of Imaging Informatics in Medicine (2024)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

  • Explore articles by subject
  • Guide to authors
  • Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

metastatic breast cancer research papers

IMAGES

  1. (PDF) Metastatic breast cancer

    metastatic breast cancer research papers

  2. (PDF) Assessing Treatment Response in Metastatic Breast Cancer

    metastatic breast cancer research papers

  3. Why PDX Are Needed for Metastatic Breast Cancer Research

    metastatic breast cancer research papers

  4. metastatic breast cancer Archives

    metastatic breast cancer research papers

  5. What You Need to Know About Metastatic Breast Cancer by Breast Cancer

    metastatic breast cancer research papers

  6. SHARE: Metastatic Breast Cancer: Cutting-Edge Research from National

    metastatic breast cancer research papers

VIDEO

  1. 2023 Updates In Metastatic Breast Cancer Treatment

  2. What Is Metastatic Breast Cancer?

  3. Metastatic Breast Cancer: Stories of Trials, Perseverance, and Hope

  4. Metastatic breast cancer: A guide for the newly diagnosed

  5. Newly diagnosed with metastatic breast cancer

  6. Metastatic Breast Cancer: What You Need to Know

COMMENTS

  1. Current and Future Management of HER2-Positive Metastatic Breast Cancer

    Current and Future Management of HER2-Positive Metastatic Breast Cancer Authors: Olga Martínez-Sáez, MD, and Aleix Prat, MD Authors Info & Affiliations Publication: JCO Oncology Practice Volume 17, Number 10 https://doi.org/10.1200/OP.21.00172 PDF Abstract

  2. The lingering mysteries of metastatic recurrence in breast cancer

    With a total of 18,989,634 new cases and 10,052,507 associated deaths estimated to occur in 2020, 1 cancer remains a major health problem worldwide. 2 Metastasis is the dissemination of tumour cells from the primary neoplasm to secondary sites in a multistep process that is often depicted as a simple series of sequential events: 3, 4 escape from...

  3. Exploring the experiences of women living with metastatic breast cancer

    Purpose Metastatic breast cancer [MBC] is the leading cause of cancer death in women globally with no cure. Women diagnosed with MBC endure a catastrophic upheaval to multiple aspects of their life and a radically transformed future landscape. ... The search outputs from of 1978 research papers were imported into Covidence, a systematic review ...

  4. Breast Cancer Metastasis: Mechanisms and Therapeutic Implications

    Metastatic breast cancer cells acquire aggressive characteristics from the tumor microenvironment (TME) through several mechanisms including epithelial-mesenchymal transition (EMT) and epigenetic regulation.

  5. Estrogen Receptor Mutations as Novel Targets for Immunotherapy in

    Estrogen receptor-positive (ER +) breast cancer is not considered immunogenic and, to date, has been proven resistant to immunotherapy.Endocrine therapy remains the cornerstone of treatment for ER + breast cancers. However, constitutively activating mutations in the estrogen receptor alpha (ESR1) gene can emerge during treatment, rendering tumors resistant to endocrine therapy.

  6. Classification and diagnostic prediction of breast cancer metastasis on

    Metastatic Breast Cancer (MBC) is one of the primary causes of cancer-related deaths in women. Despite several limitations, histopathological information about the malignancy is used for...

  7. Precision medicine improves outcomes in metastatic breast cancer

    People with metastatic breast cancer without overexpression of the receptor HER2, which promotes cancer-cell growth, and who had received no more than one line of chemotherapy were selected. A ...

  8. Multiomics in primary and metastatic breast tumors from the ...

    Published: 30 December 2022 Multiomics in primary and metastatic breast tumors from the AURORA US network finds microenvironment and epigenetic drivers of metastasis Susana Garcia-Recio,...

  9. Breast Cancer—Epidemiology, Risk Factors, Classification, Prognostic

    In literature, tumor-associated macrophages (TAMs) are associated with M2 macrophages which promote tumor growth and metastasis. For breast cancer, studies have shown that the density of TAMs is related to hormone receptor status, stage, histologic grade, lymph node metastasis, and vascular invasion [248,249,250,251].

  10. Single-cell transcriptomics provide insight into metastasis-related

    Breast cancer metastasis is a complex, multi-step process, with high cellular heterogeneity between primary and metastatic breast cancer, and more complex interactions between metastatic cancer cells and other cells in the tumor microenvironment. High-resolution single-cell transcriptome sequencing technology can visualize the heterogeneity of malignant and non-malignant cells in the tumor ...

  11. Breast Cancer

    Breast Cancer - Metastatic: Latest Research Request Permissions Breast Cancer - Metastatic: Latest Research Approved by the Cancer.Net Editorial Board, 11/2022 ON THIS PAGE: You will read about the scientific research being done to learn more about this type of cancer and how to treat it. Use the menu to see other pages.

  12. Metastatic Breast Cancer: Review of Emerging Nanotherapeutics

    Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications. ... In metastatic breast cancer ...

  13. metastatic breast cancer Latest Research Papers

    14422 (FIVE YEARS 4520) H-INDEX 171 (FIVE YEARS 33) Latest Documents Most Cited Documents Contributed Authors Related Sources Related Keywords Real-World Evidence of Trastuzumab, Pertuzumab, and Docetaxel Combination as a First-Line Treatment for Korean Patients with HER2-Positive Metastatic Breast Cancer Cancer Research and Treatment

  14. Biomedicines

    Background: Cyclin-dependent kinase 4 and 6 (CDK4 and CDK6) inhibitors have changed the therapeutic management of hormone receptor-positive (HR+) metastatic breast cancer (mBC) by targeting the cell cycle machinery and overcoming endocrine resistance. However, a large number of patients present disease progression due to cancer cells resisting CDK4/6 inhibitors. Our research considers which ...

  15. Breast Cancer Research Articles

    NCI Study Advances Personalized Immunotherapy for Metastatic Breast Cancer Posted: February 1, 2022 NCI researchers have shown that an experimental form of immunotherapy that uses an individual's own tumor-fighting immune cells could potentially be used to treat people with metastatic breast cancer who have exhausted all other treatment options.

  16. Causes of death among people living with metastatic cancer

    The Breast Cancer Research Foundation's recent investment in metastatic breast cancer research 21 aims to improve patients' quality of life and understand risk factors for metastasis, but does ...

  17. Metastatic Cancer Research

    Metastatic Cancer Research Metastatic Cancer Research Dabrafenib-Trametinib Combination Approved for Solid Tumors with BRAF Mutations Posted: July 21, 2022 FDA has approved the combination of the targeted drugs dabrafenib (Tafinlar) and trametinib (Mekinist) for nearly any type of advanced solid tumor with a specific mutation in the BRAF gene.

  18. Metastatic breast cancer treatments have aided decline in deaths

    Caswell-Jin and former research assistant Liyang Sun are co-first ... from CISNET published since 2005 that assess the relative contributions of regular screening and treatment advances on breast cancer deaths. The previous two papers informed national guidelines and helped cancer researchers focus their efforts on the most intractable problems ...

  19. How has the field of metastatic breast cancer in bones evolved over the

    This paper was a research and analysis of the previously published papers, and did not involve any human subjects. Therefore, no informed consent was required. ... The global status of research in breast cancer liver metastasis: a bibliometric and visualized analysis. Bioengineered, 12 (2) (2021), pp. 12246-12262. CrossRef View in Scopus Google ...

  20. Breast cancer in India: Present scenario and the challenges ahead

    As per the Globocan data 2020, in India, BC accounted for 13.5% (178361) of all cancer cases and 10.6% (90408) of all deaths (Figure 1 and Figure 2) with a cumulative risk of 2.81 [ 5 ]. Figure 1 World Health Organization: Estimated number of new cancer cases in 2020, worldwide.

  21. The global status of research in breast cancer liver metastasis: a

    Their latest research suggested that Atezolizumab plus nab-paclitaxel can be used as first-line treatment for metastatic triple-negative breast cancer [Citation 32]. Moreover, vessel co-option may be related to resistance to anti-angiogenic studied by UTMD Anderson Cancer Center and McGill University in 2016 [ Citation 33 ].

  22. Association Between Temporomandibular Joint Disorders and Metastatic

    The authors of the present study speculate that metastasis of breast cancer cells to the temporomandibular joint (TMJ) region mimics classical signs and symptoms of temporomandibular disorders (TMD). The aim of this mapping review was to investigate the association between metastatic breast cancer (MBC) and TMD.

  23. A metastasis map of human cancer cell lines

    Five out of seven brain metastatic breast cancer cell lines contained ... with this paper. ... and by the U.S Department of Defense Breast Cancer Research Program Innovator Award W81XWH-10-1-0016. ...

  24. Articles

    PDF MRI-based tumor shrinkage patterns after early neoadjuvant therapy in breast cancer: correlation with molecular subtypes and pathological response after therapy MRI-based tumor shrinkage patterns (TSP) after neoadjuvant therapy (NAT) have been associated with pathological response.

  25. A breast cancer drug, susceptible to resistance, can be restored to

    I n a new paper published in Cancer Research, ... in 2013 for metastatic HER2-positive breast cancer. Its use was later expanded to treat early-stage HER2-positive breast cancer if cancer cells ...

  26. Molecular principles of metastasis: a hallmark of cancer revisited

    Fig. 1 Overview of the metastatic cascade: The five key steps of metastasis include invasion, intravasation, circulation, extravasation, and colonization Full size image Dissemination and...

  27. 'Murder at the Mansion': Breast Cancer Network fundraiser for research

    The Breast Cancer Network of WNY (BCN) will present its second annual

  28. Deep Learning to Improve Breast Cancer Detection on Screening ...

    1. Set learning rate to 10 −3 and train the last layer for 3 epochs. 2. Set learning rate to 10 −4, unfreeze the top layers and train for 10 epochs, where the top layer number is set to 46 for ...