Emerging Data and Current Challenges for Young, Old, Obese, or Male Patients with Breast Cancer

With improved screening, early detection, and the advent of novel and highly effective therapies, overall rates of breast cancer survival have increased over time (1). However, there are distinct subgroups of individuals with breast cancer for whom improvements have lagged, been inconsistent, or been absent, including the very young, elderly, male, and obese patients, as well as those from racial and ethnic minority groups. These patient groups are significantly underrepresented in the large, prospective, randomized control trials that provide the evidence base for standard treatments and survivorship.

As the number of U.S. breast cancer survivors is expected to reach >4 million by 2026 (2), there is also a growing number of survivors from these patient subgroups who often have unique needs and who are at high risk for disparate outcomes. Active patient engagement, patient advocacy, and far-reaching multidisciplinary and translational collaboration will be critical to the success of productive and informative prospective research in underrepresented groups (Fig. 1). In this article, we highlight some of the major research gaps for these “special” populations (with exception of racial/ethnic subgroups, which are covered by Reeder-Hayes and Anderson in this CCR Focus; ref. 3), with suggestions on how to prioritize future investigation and promote evidence-based, high-quality care, leading to equal outcomes for all patients with breast cancer (Table 1). Our suggestions complement the work of others in this CCR Focus, with other articles focusing on advancing genomics and immunotherapy (4–6).

Approximately 11,000 U.S. women under age 40 are diagnosed with breast cancer annually (7), and breast cancer is the leading cause of cancer-related deaths in young women (1). Because of their relative lack of screening and early detection, young women are more likely to present with higher stage, symptomatic breast cancer. Population-based data have demonstrated that young women are more likely to develop more aggressive subtypes of breast cancer, including triple-negative and HER2-positive breast cancer, and within hormone receptor (HR)-positive cancers, tumors that are higher grade (8, 9).

There is controversy over whether younger women have a worse prognosis compared with other patients who have similar disease characteristics. Recent data suggest that the effect of age varies by tumor subtype (9). In the case of HER2-positive disease, the benefits and outcomes from trastuzumab appear similar for older and younger patients (10). However, in the setting of HR-positive cancers, women age <40 experience nearly 1.5 to 2 times the risk of death compared to those age >40 (9). Further research to evaluate whether there are distinct biological underpinnings for cancers that arise in younger and older women as well as differences in host responses should position us to develop novel interventions and improve outcomes. A notable example is the recent work focused on women diagnosed in the initial years following a pregnancy (“pregnancy-associated breast cancer,” or PABC), a group that appears to have a particularly poor prognosis, even when compared to other young breast cancer patients (11, 12). The particularly higher mortality of these women has been hypothesized to be due to a tumor-promoting microenvironment in the post-partum/lactational, involuting breast. Evolving animal and human data support that mammary involution entails a proinflammatory process including an active period of lymphatic growth and remodeling, with tumors demonstrating high peritumoral lymphatic vessel density, increased incidence of lymph node metastasis (12–15), and possible enrichment of liver and brain metastasis in the postpartum period (16). In addition, epidemiologic data reveal that the molecular phenotype of PABC tends to be more aggressive. However, after controlling for age and family history, some suggest that phenotype does not differ by the time since parturition, indirectly supporting potential host rather than tumoral factors as drivers of poor prognosis, although further work in this area is clearly warranted (17).

Although the biology of HR-positive cancers may be less favorable in younger women, there is also mounting evidence that suboptimal treatment contributes to disparities. Only in the past few years have large-scale clinical trials demonstrated the value of extended endocrine therapy and/or ovarian suppression (OS) with either tamoxifen or aromatase inhibitor (AI) therapy in very young women with HR-positive breast cancer, particularly for those age <35 (18, 19). For example, in the Tamoxifen and Exemestane Trial (TEXT) and the Suppression of Ovarian Function Trial (SOFT), disease-free survival was >87% for all women receiving OS with tamoxifen or exemestane, demonstrating favorable outcomes for young women on study, 42.2% of whom had node-positive disease (19). It is anticipated that judicious use of this strategy in very young and higher risk premenopausal patients will improve outcomes in HR-positive disease. However, several studies have now documented relatively poor adherence to adjuvant hormonal therapy among young women, contributing to worse outcomes (20–22). Although it remains a priority to continue to unravel the multifaceted biological and genetic basis of breast cancer in young patients, we argue that a primary focus of investigation should be improving tolerability and striving for better ways to prevent, detect, and manage poor adherence. How we support adherence will differ according to the underlying motivation; women with side effects may require a different intervention than those struggling with high costs or lacking a full understanding of treatment rationale.

In addition, if quality of life (QOL) cannot be maintained for young patients while on prolonged therapy, they may be less likely to even initiate treatment. Llarena and colleagues have revealed that key factors for non-initiation or nonadherence to adjuvant hormonal therapy in young women are not only symptoms or fear of symptoms but also fertility concerns (23). Research to support young women's ability to accept and continue adjuvant hormonal therapy will be key to improvement in compliance, and a number of initiatives are underway to enhance adherence and improve menopausal symptom management in breast cancer survivors in general. The Pregnancy Outcome and Safety of Interrupting Therapy for Women with Endocrine Responsive Breast Cancer (POSITIVE) trial (NCT02308085) is the first-ever prospective study of interruption of endocrine therapy at 18 to 30 months for up to 2 years, and enrolls women who are interested in pregnancy after breast cancer. This unique cohort study provides an opportunity to test the safety and feasibility of this strategy aimed at optimizing both disease, QOL, and pregnancy outcomes for such patients.

Despite the concerns for “undertreatment” and suboptimal treatment adherence among young women, there has been a paradoxical increase in the use of more aggressive surgery options among this patient population, without improvement in breast cancer–specific survival (24, 25). As the pendulum has swung toward bilateral mastectomies for unilateral cancers in young women, we may have inadvertently moved away from successfully delivering important messages and interventions to support adherence to medical treatment, which do significantly impact outcomes (22). The reasons for the trends in bilateral mastectomies in young women are complex but appear largely related to anxiety and “peace of mind,” regardless of patients' acknowledgment that bilateral surgery does not improve outcomes (26, 27). This disconnect provides an important opportunity to improve upon current surgical decision-making processes and psychosocial support strategies. We need to develop coordinated, multidisciplinary approaches to best support decisions, and avoid a “knee-jerk” expectation and facilitation of bilateral mastectomies. Ongoing efforts to develop decision aids and instructive tools focused on young women may prove useful in educating women about their options, but unified messages from surgical oncology, plastic surgery, and medical oncology will be key to dissemination of recommendations to patients (27).

Prospective trials dedicated to young patients are essential to answering many of the outstanding questions. Promising examples include recently accrued, large, multicenter cohort studies of women age ≤40 being run in the United States (The Young Women's Breast Cancer Study, NCT01468246) and the United Kingdom (the Prospective study of Outcomes in Sporadic versus Hereditary breast cancer, or POSH trial), which are both designed to elucidate unique issues and outcomes in young women (28, 29).

As our population ages, the numbers of older patients with breast cancer will increase (30, 31). Approximately 70,000 U.S. women are diagnosed with breast cancer annually at age ≥70 (7). Despite an inverse relationship between age and the development of biologically more aggressive breast cancer phenotypes, outcomes for older patients with breast cancer are highly variable due not only to a number of biological factors (32, 33) but also to potentially mutable factors. It is at the extremes of age that women have the worst breast cancer–related outcomes; in addition to the youngest patients with breast cancer, women in the oldest age groups also do poorly (34). Those presenting with locally advanced or metastatic disease do especially worse (35), and older age in itself is associated with early mortality in those presenting with metastatic disease (36, 37) and slower improvements in breast cancer outcomes over time compared with younger women (38).

At any stage of disease, many older patients will be undertreated or receive nonstandard care (39–44) as a result of concerns for coexisting medical conditions, treatment toxicity, and frailty. Although some undertreatment is appropriate in the setting of lower risk disease, advanced comorbidity, and patient preference, multiple studies have documented similar disease risk reductions and mortality benefits in younger and older patients receiving chemotherapy for higher risk cancers (45, 46). This is particularly relevant for older patients with higher risk disease (47), respectively, who are at risk for a short time to recurrence. Further, we have prospective data from the Cancer and Leukemia Group B 49907 trial that older women tolerate standard therapy well and benefit from its receipt (48). In addition, similar to younger patients, older women with breast cancer also experience suboptimal adherence to hormonal therapy (20, 49–51), increasing their risk for poor outcomes.

Similar to the youngest patients with breast cancer, older patients are consistently underrepresented in clinical trials (52, 53), limiting our knowledge of short- and long-term toxicity to standard therapies. There have been several urgent calls to increase clinical trial enrollment of older adults (52, 54, 55) and to improve the quality of care we deliver to older patients (56). Although there have been some successes with dedicated trials for older patients (48, 57), we still lack a robust evidence base for this growing patient population. We must engage older patients in our efforts to prioritize their clinical trial enrollment on a national level and need to advance past pooling data from prior studies or using claims-based registry data to answer questions about how to approach treatment. Further, we need to explore ways of improving long-term adherence to hormonal therapy in older patients, which may be different than interventions designed for younger patients.

Recent prospective research in older patients has emphasized the concept of functional age rather than chronological age when making treatment decisions, with multiple studies demonstrating the potential value of incorporating a modified geriatric assessment into clinical practice (58–60). The geriatric assessment considers multiple domains, including functional status, comorbidity, medication review, cognition, psychological status, nutrition, and social functioning and support, and has recently demonstrated feasibility in practice (59). Predictive models, which incorporate clinical variables in addition to information gained from the geriatric assessment, have been developed to help anticipate tolerability and severe toxicity in older patients receiving chemotherapy across all cancers (58, 61). However, optimal strategies for application of these models in the real world remains uncertain. The large, prospective Clinical and Biological Predictors of Chemotherapy Toxicity in Older Adults study (NCT01472094) is designed to address these issues in older patients with breast cancer, and will provide powerful information on clinical and functional factors associated with severe toxicity with therapy in this vulnerable population. As a natural next step, we will then need to further examine the role of treatment modification (e.g., up-front dose reductions and agent modifications) driven by better toxicity prediction based on functional status, disease risk, and survival. Along with this, there is a critical need to better estimate life expectancy and improve accurate ascertainment of cause of death for older patients with breast cancer. Tools are currently available online to help calculate life expectancy (62), although none of these have been validated in patients with cancer and none take into account the risks of treatment. Having this type of information when making treatment decisions will be ever more important as the population ages and the prevalence of chronic disease in older patients increases (62, 63).

Finally, examining how physical training and exercise may improve functional status and mitigate treatment toxicity for older patients is also an active area of research, with early results indicating promise for walking interventions (64–67), Tai Chi (68), and remote interventions (66), with multiple small studies forthcoming. It is likely that early identification and intervention for physical and functional decline will impact outcomes for older breast cancer patients as well, but approaches to date are limited.

Approximately 2,600 cases of male breast cancer occur annually in the United States (69), and despite the growing number male survivors, there is a dearth of evidence on the biological and genomic characteristics of their tumors, or how to best treat and support this distinct group of patients. In men with known germline mutations in the BRCA1 and BRCA2 genes, the cumulative incidence of breast cancer is 1.2% [95% confidence interval (CI), 0.22–2.8%] and 6.8% (95% CI, 3.2–12%), respectively (70), and other risk factors include weight, gynecomastia, and Klinefelter syndrome (71). Up until this point, the vast majority of data on men have come from smaller retrospective studies with limited power to effectively support treatment selection and survivorship issues. There is a suggestion that breast cancer behaves differently in men compared with women, but this is not well understood nor do we know how to act on these differences (72, 73). For example, in a study comparing Oncotype DX scores among men and women, although many similarities were noted in scores overall, the mean expression of estrogen and progesterone receptors were higher in men, as were proliferation genes such as Ki-67 (73). In addition, the differential benefits of varying hormonal therapies in men and women have not been well described. For example, controversy remains regarding the necessity of gonadal suppression when using AIs in men (74).

Given the low numbers of men with breast cancer, global efforts will be required to gain meaningful, prospective evidence for men with breast cancer in a timely fashion. One such collaboration has fortunately begun with the Clinical and Biological Characterization of Male Breast Cancer trial (NCT01101425). This unique study represents a worldwide collaboration to create a prospective registry of men with breast cancer. Rich in its scope with regard to prospective data collection on treatment, outcomes, and QOL, this study is also collecting tumor and blood samples for biological and genomic characterization. Although randomized trials examining treatment interventions in men would be ideal, this has proved more difficult. For example, the SWOG trial S0511 (A Phase II Study of Goserelin Plus Anastrozole for the Treatment of Male Patients With Hormone-Receptor Positive Metastatic or Recurrent Breast Cancer) faced challenges with accrual and closed prematurely, further highlighting the need for a global and multidisciplinary commitment to investigate male breast cancer. Another interesting and important opportunity in the era of social media will be to capitalize on the ability to reach men worldwide through online networks (75, 76). This has become a powerful way to engage patients, especially given that male breast cancer is a rare and potentially stigmatizing condition, and to bring men with the disease together not only to answer important research questions, but to form an online community and support one another.

Obesity is a major health epidemic and is associated with an incremental risk of developing breast cancer with increasing body mass index (BMI), a complex relationship with menopause status, and a 35% to 40% increased risk for cancer recurrence in those with breast cancer, with a concomitant worse survival (77–82). However, emerging evidence suggests a variable relationship of obesity and prognosis by breast cancer subtype, even though these relationships are not fully understood, and data are often inconsistent and complex. For example, although studies consistently show worsened long-term prognosis among obese patients with HR-positive cancers (83, 84), in the setting of triple-negative breast cancer, studies show both negative and positive effects for obesity on outcomes (85–87). In the setting of HER2-positive disease, this relationship may be more reliant on the HR status (88–91). In addition, there has been concern regarding the impact of obesity on response to hormonal therapy, and controversy exists over optimal treatment in the setting of elevated aromatase activity (91). As evidence to this, trials comparing AI and tamoxifen in post- and premenopausal patients have suggested that the relative benefits of AI over tamoxifen (with OS in premenopausal patients) are not apparent in patients who are obese (84, 92), possibly related to the independent contribution of adipose-derived estrogens. The impact of obesity on prognosis and its role in cancer was recently the focus of a dedicated issue in Journal of Clinical Oncology [34(5), December 10, 2016].

As a result of the risk of breast cancer with obesity, there is increasing international focus on the importance of preventing and treating obesity, and the American Society of Clinical Oncology has also published position statements on this topic, with dedicated guidelines, educational materials, and recommendations for weight loss interventions (93, 94). In addition, despite the availability of consensus guidelines to guide chemotherapy dosing (95), which support the use of actual weight-based chemotherapy dosing, many questions remain unanswered regarding guideline implementation, toxicity management, and survivorship strategies for obese patients. Further, comorbid conditions and competing causes of death, such as diabetes, hypertension, and heart disease, result in complicated risk–benefit discussions and potentially difficult treatment decision making.

Although gaining weight has been suggested to impact one's risk of developing breast cancer, it is not clear that weight loss in those with preexisting high BMI lowers risk (96), and the impact of the timing of weight change is still not well understood. However, multiple small studies have shown promising results with regard to the ability to induce weight loss and increased physical activity in breast cancer survivors (97, 98). The recently launched BWEL trial (Breast Cancer Weight Loss Study, NCT02750826) will enroll >3,000 patients with higher risk, early-stage breast cancer who are overweight (BMI of ≥27 mg/m²), randomizing them in the postadjuvant setting to an intensive, 2-year weight loss intervention or a control health education program. The trial will provide more definitive answers to questions about the outcomes associated with weight loss among breast cancer survivors. We will also learn about the feasibility of implementing a large-scale phone-based intervention that should ultimately be generalizable to the special populations discussed in this article.

Further, similar to older patients, obese patients have parallel unknowns when it comes to accurately predicting life expectancy, toxicity to treatments, functional decline, and causes of death. We need better tools to inform treatment discussions and decisions to optimally individualize care in this setting, with careful adaptation to patients' health priorities and follow-up care.

To maximize breast cancer survival and survivorship, we must address the individual needs of each patient. For patients who are part of the special populations described in this article, there is only limited understanding for how we can treat them most effectively for breast cancer while balancing their unique medical and psychosocial concerns. Ongoing and future research is needed, including prospective investigation for all special populations of patients, with national, international, and multidisciplinary collaboration and significant efforts to engage and enroll patients on research studies. Table 2 highlights ongoing clinical trials addressing important questions in the special populations, including the studies discussed in our review. We must let patients know how important it is that we learn from their experiences and empower them to be a part of expanding our knowledge base. It is only through these approaches that we can improve outcomes for all patients with breast cancer.

No potential conflicts of interest were disclosed.

R.A. Freedman was supported by the American Cancer Society (grant no. 125912-MRSG-14-240-01-CPPB) and a Susan G. Komen grant (CCR14298143). A.H. Partridge was supported by a Susan G. Komen grant (SAC100008).