Obesity and Breast Cancer Prognosis: Evidence, Challenges, and Opportunities Sao Jiralerspong and Pamela J. Goodwin

VOLUME 34 NUMBER 35 DECEMBER 10, 2016 JOURNAL OF CLINICAL ONCOLOGY R E V I E W A R T I C L E Obesity and Breast Cancer Prognosis: Evidence, Challenges, and Opportunities Sao Jiralerspong and Pamela J. Goodwin Sao Jiralerspong, Baylor College of Medicine, Houston, TX; and Pamela J. Goodwin, Mount Sinai Hospital and University of Toronto, Toronto, Ontario, Canada. Published online ahead of print at www.jco.org on November 7, 2016. Authors disclosures of potential conflicts of interest are found in the article online at www.jco.org. Author contributions are found at the end of this article. Corresponding author: Pamela J. Goodwin, Mount Sinai Hospital, 1284 600 University Ave, Toronto, Ontario, Canada, M5G 1X5; e-mail: pgoodwin@mtsinai.on.ca. 2016 by American Society of Clinical Oncology 0732-183X/16/3435w-4203w/$20.00 DOI: 10.1200/JCO.2016.68.4480 A B S T R A C T Purpose To summarize the evidence of an association between obesity and breast cancer prognosis. Methods We reviewed the literature regarding overweight and obesity and breast cancer survival outcomes, overall and with regard to breast cancer subtypes, breast cancer therapies, biologic mechanisms, and possible interventions. We summarize our findings and provide clinical management recommendations. Results Obesity is associated with a 35% to 40% increased risk of breast cancer recurrence and death and therefore poorer survival outcomes. This is most clearly established for estrogen receptor positive breast cancer, with the relationship in triple-negative and human epidermal growth factor receptor 2 positive subtypes less well established. A range of biologic mechanisms that may underlie this association has been identified. Weight loss and lifestyle interventions, as well as metformin and other obesity-targeted therapies, are promising avenues that require further study. Conclusion Obesity is associated with inferior survival in breast cancer. Understanding the nature and mechanisms of this effect provides an important opportunity for interventions to improve the diagnosis, treatment, and outcomes of obese patients with breast cancer. J Clin Oncol 34:4203-4216. 2016 by American Society of Clinical Oncology INTRODUCTION Obesity is currently an epidemic, with 69% of adults in the United States and 38% in the world being either overweight or obese. 1-3 Obesity is associated with an altered whole-body physiology and hormonal environment that promote a number of disease states, including diabetes and cardiovascular disease. Obesity is also associated with an increased risk of developing a number of cancers and with poorer survival outcomes for patients with those cancers. 4 Here we review the evidence of the relationship between obesity and breast cancer, with an emphasis on breast cancer prognosis. This review is not intended to be comprehensive but rather to provide an overview and to highlight important recent findings. We emphasize results from larger studies, clinical trials, and meta-analyses from the past 5 to 10 years. The reader is referred to prior reviews for further information. 5,6 We apologize to colleagues whose work could not be included because of space limitations. OBESITY AND BREAST CANCER RISK Although our focus is on breast cancer outcomes, understanding associations of body size with the risk of developing breast cancer sheds light on the body habitus of women at the time of breast cancer diagnosis. It has long been recognized that overweight and obesity (and adult weight gain) are associated with an increased risk of postmenopausal breast cancer, notably hormone receptor positive cancers (risk estimate, 1.82; 95% CI, 1.55 to 2.14 for overweight and obese combined). 7 In contrast, overweight and obesity have been thought to be associated with a lower risk of premenopausal breast cancer, and that remains true for hormone receptor positive cancers (risk estimate, 0.80; 95% CI, 0.70 to 0.92 2016 by American Society of Clinical Oncology 4203

Jiralerspong and Goodwin for overweight and obese combined). 7 However, recent evidence has demonstrated an increased risk of triple-negative breast cancers (TNBCs) in obese premenopausal women (odds ratio, 1.43; 95% CI, 1.23 to 1.65). 8 Physical inactivity, an attribute linked to obesity, has been associated with higher risk of breast cancer regardless of menopausal status. 9 Thus, many women with breast cancer are heavier and more physically inactive than the general population at the time of diagnosis. This underscores the importance of the adverse associations of obesity with breast cancer outcomes that are discussed in this article. OBESITY AND BREAST CANCER PROGNOSIS Several recent meta-analyses have examined the relationship between body mass index (BMI) and survival in large numbers of patients (Table 1). A meta-analysis by Protani et al 10 included 43 studies of patients with breast cancer diagnosed between 1963 and 2005. The pooled hazard ratios (HRs) comparing obese versus nonobese patients were 1.33 (95% CI, 1.21 to 1.47) for overall survival (OS) and 1.33 (95% CI, 1.19 to 1.50) for breast cancer specific survival (BCSS) and were statistically significant. This association was maintained after adjusting for age and stage and regardless of menopausal status, year of diagnosis, or study type (treatment or observational). A subsequent meta-analysis by Niraula et al 11 found that the association of obesity (obese v nonobese) with breast cancer outcomes (OS and BCSS) did not differ by menopausal or hormone receptor status. The most recent meta-analysis, by Chan et al, 12 is also the most extensive, involving 213,075 patients from 82 studies, with large numbers of events (41,477 deaths, including 23,182 resulting from breast cancer). The investigators examined total mortality and breast cancer specific mortality,classifyingbmiat baseline and after diagnosis into four categories (underweight, normal weight, overweight, and obese). The summary relative risks (RRs) for total mortality and breast cancer specific mortality for obese versus normal-weight patients at baseline were 1.41 (95% CI, 1.29 to 1.53) and 1.35 (95% CI, 1.24 to 1.47), respectively. These positive associations were present when BMI was measured 12 months before or 12 months or more after diagnosis. They were also present regardless of menopausal status, with a suggestion of larger effects in premenopausal versus postmenopausal patients (RR, 1.75 v 1.34). The RRs for total mortality and breast cancer specific mortality for overweight versus normal-weight patients at baseline were lower, at 1.07 (95% CI, 1.02 to 1.12) and 1.11 (95% CI, 1.06 to 1.17), respectively, but were still statistically significant. Dose-response meta-analysis revealed a nonlinear J-shaped curve for total mortality, with RRs increasing not only for overweight or obese but also for underweight patients. In contrast, and importantly, this analysis revealed evidence only for a linear relationship for breast cancer specific mortality. Linear models (excluding the small proportion of underweight patients) gave RRs for each 5-kg/m 2 increase in BMI of 1.17 and 1.18 for total mortality and breast cancer specific mortality, respectively. A number of studies have shown that obese patients present with larger tumors and, less consistently, with increased lymph node metastases and higher-grade tumors. 13-15 Although the higher T and N stages may reflect delayed diagnosis in obese women, they, along with higher grade, may also reflect more aggressive biology in obesity, leading to higher risk of recurrence. In support of this, after adjusting for these tumor factors, overweight and obesity remain independent risk factors for decreased BCSS and OS. In a large study of 18,967 patients from Denmark, Ewertz et al 15 demonstrated that overweight and obese patients had an increased risk of distant, but not local, recurrence. The increased risk of distant recurrence was significant only in years 5 to 10, with HRs of 1.42 (95% CI, 1.17 to 1.73) and 1.46 (95% CI, 1.11 to 1.92) for overweight and obese patients, respectively; this translated into an increased risk of breast cancer related death primarily after 10 years. Goodwin et al 16 reported that higher BMI was associated with an increased risk of both early (first 5 years) and late (beyond 5 years) distant recurrences and death. An association of obesity (BMI $ 30 kg/m 2 ) with increased late recurrence and mortality in estrogen receptor positive breast cancer was also reported in 6,295 patients from the After Breast Cancer Pooling Project. 17 Similar patterns have been identified in other studies. 18,19 In summary, meta-analyses involving large numbers of patients have provided evidence that obesity is associated with an increased risk of recurrence and death of approximately 35% to 40%. This seems to be true after adjusting for tumor-related variables and regardless of menopausal or hormone receptor status. Overweight and obesity also seem to be associated with distant and late recurrences. Of interest, in a recent meta-analysis, weight gain after breast cancer diagnosis was associated with a modest increased risk of all-cause mortality but not with increased breast cancer specific mortality. 20 Adverse effects on allcause mortality were greatest when weight gain was greater than 10%, suggesting that BMI at diagnosis may be the most relevant predictor of breast cancer outcome. OBESITY AND BREAST CANCER SUBTYPES There is evidence that the relationship between obesity and breast cancer prognosis varies according to breast cancer subtype. The data for an association between BMI and decreased survival are fairly consistent for ER-positive breast cancer (as summarized in the previous section), with less consistent results reported for the other subtypes. This may in part relate to the approach used to classify breast cancer subtypes, and the current reliance on immunohistochemical staining of ER, progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) to define subtypes may not be optimal. One recent study reported that associations of BMI with worse breast cancer outcomes were independent of PAM50-defined breast cancer subtype (eg, luminal A or B, HER2 enriched, basal-like). 21 ER-Negative and TNBCs The data for an association between BMI and decreased survival in ER-negative or TNBC are inconsistent, with some studies showing an association 11,22-26 and others not 27-36 (Table 2). 4204 2016 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Obesity and Breast Cancer Prognosis Table 1. BMI and Breast Cancer Prognosis: Meta-Analyses Study Year Published Setting Study Design Country Patient Types No. of Patients Time Period Follow-Up Chemotherapy Endocrine Therapy Results Protani 10 2010 NA Meta-analysis of 43 studies Niraula 11 2012 NA Meta-analysis of 21 studies Chan 12 2014 NA Meta-analysis of 82 studies International All subtypes; node International All subtypes; node International All subtypes; node NA 1963-2005 4-14 years NA NA Worse OS (HR, 1.33; 95% CI, 1.21 to 1.47) and BCSS (HR, 1.33; 95% CI, 1.19 to 1.50) in obese v nonobese patients No significant differences by menopausal status, year of diagnosis, or treatment v observational cohort 213,075 1957-2009 $ 5 years for all but eight studies 80,326 NA 5-20 years NA NA Worse OS in overweight or obese patients in ERpositive (HR, 1.31; 95% CI, 1.17 to 1.46) and ER-negative subsets (HR, 1.18; 95% CI, 1.06 to 1.31) Worse OS in overweight or obese patients in premenopausal (HR, 1.23; 95% CI, 1.07 to 1.42) and postmenopausal subsets (HR, 1.15; 95% CI, 1.06 to 1.26) Worse BCSS in overweight or obese patients in ER-positive (HR, 1.36; 95% CI, 1.20 to 1.54) and ER-negative subsets (HR, 1.46; 95% CI, 0.98 to 2.19) Worse BCSS in overweight or obese patients in premenopausal (HR, 1.18; 95% CI, 0.82 to 1.70) and postmenopausal subsets (HR, 1.38; 95% CI, 1.11 to 1.71) No significant differences between ER-positive and ER-negative disease or between pre- and postmenopausal patients for any of the above NA NA Poorer total mortality (HR, 1.41; 95% CI, 1.29 to 1.53) andbreast cancer specific mortality (HR, 1.35; 95% CI, 1.24 to 1.47) for obese v normalweight patients Poorer total mortality (HR, 1.07; 95% CI, 1.02 to 1.12) andbreast cancer specific mortality (HR, 1.11; 95% CI, 1.06 to 1.17) for overweight v normal-weight patients Abbreviations: BCSS, breast cancer specific survival; BMI, body mass index; ER, estrogen receptor; HR, hazard ratio; NA, not applicable; OS, overall survival. www.jco.org 2016 by American Society of Clinical Oncology 4205

Jiralerspong and Goodwin Table 2. BMI and Breast Cancer Subtypes: ER-Negative and TNBCs, and HER2 Positive Study Year Published Setting Study Design Country Patient Types No. of Patients Time Period Follow-Up Chemotherapy Endocrine Therapy Results Comments Niraula 11 2012 NA Meta-analysis of 21 studies Positive for association between BMI and ER-negative or TNBC International All subtypes; node Pajares 22 2013 GEICAM trials Four pooled trials Spain All subtypes; node Fontanella 23 2014 GEPAR neoadjuvant trials Eight pooled trials Germany All subtypes; node Widschwendter 24 2015 SUCCESS-A trial Clinical trial Germany All subtypes; node Hao 25 2015 Fudan University Single-institution Bonsang-Kitzis 26 2015 Institut Curie Single-institution 80,326 NA 5-20 years NA NA Worse OS in overweight or obese patients in ER-positive and ER-negative subsets (HR, 1.31 and 1.18, respectively) Worse BCSS in overweight or obese patients in ERpositive and ER-negative subsets (HR, 1.36 and 1.46, respectively) No significant differences in ER-positive v ER-negative subsets for any of above 5,683 1996-2008 93 months TAC v FAC for high-risk node-negative patients FEC 6 paclitaxel once per week for node-positive patients TAC v FAC for nodepositive patients FAC 6 paclitaxel once per week for node-negative patients Tamoxifen, anastrozole, or switch 8,872 1999-2011 43 months Multiple Tamoxifen, anastrozole, or switch 3,754 2005-2007 65 months FEC, then docetaxel or docetaxel plus gemcitabine Tamoxifen or switch Recurrence, breast cancer mortality, and mortality worse with severe obesity (HR, 1.26, 1.32, and 1.35, respectively) across all subtypes pcr rate decreased with increasing BMI; DFS and OS lower in obese and very obese patients; seen in luminal-like and TNBC subsets For TNBC, mean DFS was significantly worse in obese and very obese (68.0 and 42.3 months, respectively) compared with normal-weight patients (77.7 months) For TNBC, mean OS was significantly worse in obese and very obese (74.5 and 48.0 months, respectively) compared with normal-weight patients (85.3 months) DFS and OS worse in severely obese; only seen in TNBC subset (HR, 3.02 and 3.85, respectively) China TNBC 1,106 2002-2012 45 months NA NA Worse OS but not BCSS in all overweight or obese patients with TNBC (HR, 1.36 and 1.34, respectively) OS and BCSS worse (HR, 2.16 and 2.27, respectively) in premenopausal overweight or obese patients with TNBC France TNBC 326 2002-2012 52 months Anthracycline plus taxane (continued on following page) NA For node-negative patients, worse metastasis-free survival after neoadjuvant chemotherapy for obese (HR, 2.64) v nonobese patients Pretrastuzumab Obese patients received lower chemotherapy doses 4206 2016 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Obesity and Breast Cancer Prognosis Table 2. BMI and Breast Cancer Subtypes: ER-Negative and TNBCs, and HER2 Positive (continued) Study Year Published Setting Study Design Country Patient Types No. of Patients Time Period Follow-Up Chemotherapy Endocrine Therapy Results Comments Negative for association between BMI and ER-negative or TNBC Pan 27 2014 EBCTCG trials Meta-analysis of 21 pooled trials Sparano 28,29 2012, 2015 ECOG trials Three trials (one test, two validation) Cecchini 30 2016 NSABP trials Four trials: B-30, B31, B- 34, and B-38 International All subtypes; node Approximately 80,000 NA NA NA NA Breast cancer specific mortality worse in premenopausal obese patients with ER-positive disease (HR, 1.34); no differences in postmenopausal patients regardless of ER status United States Node positive 4,770 1989-2002 7-9 years AC taxane Tamoxifen or 1,502 switch 613 United States B-30: node positive 5,265 1999-2007 9 years AC tamoxifen v AT v TAC B-31: HER2 positive, node positive DFS, OS, and BCSS worse in obese patients in ER-positive subset (HR, 1.24, 1.37, and 1.40, respectively); values for obese and nonobese patients were same for ERnegative and HER2-positive subsets; confirmed in E5188 trial (ER positive) and E3189 trials (ER negative) B-30: recurrence and OS worse in obese patients overall (HR, 1.17 and 1.18, respectively) and specifically the ERpositive subset (HR, 1.30 and 1.33, respectively) 2,119 8.3 years AC P v AC PH B-31: no differences B-34: early stage 3,323 8.4 years Clodronate v placebo B-34: no differences B-38: node positive 4,892 5.9 years TAC v AC P v AC B-38: recurrence worse in P-Gem obese patients overall (HR, 1.13) and specifically the ER-positive subset (HR, 1.27) Dignam 31 2006 NSABP trials Three trials United States ER negative, node 4,077 1981-1998 NA Surgery 6 MTX FU No differences in recurrence in negative MTX FU v CMF obese patients with CMF v AC ER-negative disease Jiralerspong 32 2013 MD Anderson Cancer Center Dawood 33 2012 MD Anderson Cancer Center Tait 34 2014 Washington University Ademuyiwa 35 2011 Roswell Park Cancer Institute Mowad 36 2013 Louisiana State University Single-institution Single-institution Single-institution Single-institution Single-institution United States All subtypes; node 6,342 1996-2005 5.4 years Anthracycline and taxane based Tamoxifen and anastrozole based No differences in RFS, OS, or BCSS in patients with ERnegative or HER2-positive disease United States TNBC 2,311 1990-2010 39 months NA NA No differences in distant DFS in obese patients United States TNBC 501 2006-2010 40 months NA NA No differences in DFS or OS in obese patients United States TNBC 418 1996-2010 37 months NA NA No differences in RFS or OS in obese patients United States TNBC 183 1998-2011 43 months NA NA No differences in DFS or OS in obese patients Unpublished observations Obese v nonobese Mazzarella 42 2013 European Institute of Oncology Single institution Italy Early-stage HER2-positive breast cancer BMI and Breast Cancer Subtypes: HER2 Positive 1,250 1995-2005 8.2 years Pretrastuzumab Likely tamoxifen (continued on following page) In those with ER-negative, HER2-positive disease, worse OS and distant metastases in obese v normal-weight patients (HR, 1.79 and 2.03, respectively); in those with ER-positive, HER2-positive disease, no differences Pretrastuzumab www.jco.org 2016 by American Society of Clinical Oncology 4207

Jiralerspong and Goodwin Table 2. BMI and Breast Cancer Subtypes: ER-Negative and TNBCs, and HER2 Positive (continued) Study Year Published Setting Study Design Country Patient Types No. of Patients Time Period Follow-Up Chemotherapy Endocrine Therapy Results Comments Crozier 43 2013 NCCTG N9831 trial Clinical trial United States HER2 positive, node positive or high-risk node negative Cecchini 30 2016 NSABP trials Four trials: B-30, B31, B-34, and B-38 Sparano 28,29 2012, 2015 ECOG trials Three trials: E1199 (test), E5188 (validation), E3189 (validation) Jiralerspong 32 2013 MD Anderson Cancer Center Single-institution 3017 2000-2005 5.3 years Arm A: AC 3 4 Tamoxifen or Arm B: AC 3 4 anastrozole paclitaxel trastuzumab (sequential) Arm C: AC 3 4 paclitaxel plus trastuzumab (concurrent) United States B-30: node positive 5,265 1999-2007 9 years AC tamoxifen v AT v TAC B-31: HER2 positive, node positive For all treatment arms combined, DFS worse for overweight and obese compared with normalweight patients (HR, 1.30 and 1.31, respectively); within each arm, no significant differences by BMI B-30: recurrence and OS worse in obese patients overall (HR, 1.17 and 1.18, respectively) and specifically the ERpositive subset (HR, 1.30 and 1.33, respectively) 2,119 8.3 years AC P v AC PH B-31: no differences B-34: early stage 3,323 8.4 years Clodronate v placebo B-34: no differences B-38: node positive 4,892 5.9 years TAC v AC P v AC B-38: recurrence worse in P-Gem obese patients overall (HR, 1.13) and specifically the ER-positive subset (HR, 1.27) United States All trials node positive 1989-2002 DFS, OS, and BCSS worse in E1199: all subtypes 4,770 7.9 years AC taxane Tamoxifen obese patients in ER-positive or switch subset (HR, 1.24, 1.37, and E5188: ER positive, premenopausal 1,502 14.0 years CAF None v goserelin E3189: ER negative 613 14.4 years CAF v 16-week regimen United States All subtypes; node 6,342 1996-2005 5.4 years Anthracycline and taxane based v goserelin plus tamoxifen None Tamoxifen and anastrozole based 1.40, respectively); values for obese and nonobese patients were same for ERnegative and HER2-positive subsets; confirmed in E5188 trial (ER positive) and E3189 trials (ER negative) No differences in RFS, OS, or BCSS in ER negative or HER2 positive Obese v nonobese; pretrastuzumab Abbreviations: AC, doxorubicin and cyclophosphamide; AT, doxorubicin and docetaxel; BCSS, breast cancer specific survival; BMI, body mass index; CAF, cyclophosphamide, doxorubicin, and 5-fluorouracil; CMF, cyclophosphamide, methotrexate, and fluorouracil; DFS, disease-free survival; EBCTCG, Early Breast Cancer Trialists Collaborative Group; ECOG, Eastern Cooperative Oncology Group; ER, estrogen receptor; FAC, fluorouracil, doxorubicin, and cyclophosphamide; FEC, fluorouracil, epirubucin, and cyclophosphamide; FU, fluorouracil; GEICAM, Grupo Español de Investigación en Cáncer de Mama; HER2, human epidermal growth factor receptor 2; HR, hazard ratio; MTX, methotrexate; NA, not applicable; NCCTG, North Central Cancer Treatment Group; NSABP, National Surgical Adjuvant Breast and Bowel Project; OS, overall survival; P, paclitaxel; pcr, pathologic complete response; P-Gem, paclitaxel and gemcitabine; PH, paclitaxel and trastuzumab; RFS, recurrence-free survival; TAC, docetaxel, doxorubicin, and cyclophosphamide; TNBC, triplenegative breast cancer. 4208 2016 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Obesity and Breast Cancer Prognosis For example, a recent study involving 1,106 women with TNBC found high BMI (. 24 v # 24 kg/m 2 ) was associated with lower OS (HR, 1.46; 95% CI, 1.04 to 2.06) but found no difference in BCSS (HR, 1.34; 95% CI, 0.90 to 2.01). Associations for both OS (HR, 2.16; 95% CI, 1.21 to 3.87) and BCSS (HR, 2.27; 95% CI, 1.11 to 4.63) were stronger (HRs. 2) and statistically significant in premenopausal compared with postmenopausal women. 25 In an analysis of pooled German neoadjuvant trials involving 8,872 patients, obese and very obese patients in the TNBC subset had significantly decreased disease-free survival (DFS) and OS. Another recent large study (the SUCCESS-A trial, which randomly assigned women to different adjuvant chemotherapy regimens; N = 3,754) reported high BMI was associated with poorer DFS and OS in TNBC but not other breast cancer subtypes. 24 In contrast, several reports nested in randomized controlled trials of systemic therapy failed to identify prognostic associations of obesity in ERnegative or TNBC (Table 2). There are several potential explanations for these variable results: there may be no true association of obesity with prognosis, or there may be an association, but it may be difficult to detect. There are a number of reasons for the latter possibility. Demonstration of an association may be dependent on study design, adjuvant treatment administered, or specific population studied, making it challenging to identify consistently across heterogeneous studies. The postulated effect sizes of obesity on breast cancer prognosis are modest, and associations may be difficult to observe in small studies. At the molecular level, TNBC is not as common as other breast cancer subtypes and comprises at least seven molecular subtypes, each of which may have differing obesity associations. Heterogeneity in study approaches in terms of definition of subtypes (using full ER, PR, or HER2 status v ER [6 PR] status alone), in handling of BMI in statistical analyses (categorical [with two v $ three categories] v continuous; handling BMI as a continuous variable will enhance power), and in definitions of outcomes (breast cancer or overall mortality) and in treatments received (anthracycline v nonanthracycline chemotherapy) may all contribute to inconsistent findings. Evaluation of obesity associations in systemic therapy trials, particularly those including cardiotoxic drugs, may lead to under-representation of women with obesity-associated metabolic factors (eg, high insulin or glucose, systemic inflammation) that may mediate the associations of obesity with breast cancer outcomes, further reducing power. Nevertheless, the recent positive results highlighted in our article lend support to an association of obesity with worse outcomes in at least a subset of patients with TNBC. Furthermore, animal studies have demonstrated that obesity promotes TNBC progression and therefore provide a compelling biologic argument that this effect is real (Jiralerspong et al, manuscript in preparation). 37-39 Consistent with these observations, in WINS (Women s Intervention Nutrition Study), a low-fat diet leading to modest weight loss was associated with improved OS in patients with ERand/or PR-negative disease (and presumably TNBC). 40,41 A metaanalysis of available studies examining associations of obesity with outcome in TNBC may be of value; it would include larger numbers of patients with TNBC, could compare multiple categories of BMI or model associations across BMI as a continuous variable, and potentially stratify by treatment. HER2-Positive Breast Cancer The relationship between BMI and HER2-positive breast cancer is also unclear (Table 2). 28,30,32,42-43 In the pretrastuzumab era, a study of 1,250 patients reported obese (BMI. 30 kg/m 2 ) patients with ER-negative, HER2-positive disease had worse OS (HR, 1.79; 95% CI, 1.03 to 3.10) and distant metastases (HR, 2.03; 95% CI, 1.13 to 3.63) compared with normal-weight patients, whereas there were no significant obesity associations in patients with ER-positive, HER2-positive disease. 42 In contrast, a study that did not stratify by ER status found no differences in survival by BMI for patients with HER2-positive disease. 27 In the trastuzumab era, the North Central Cancer Treatment Group N9831 trial reported worse DFS (but did not examine OS) in the pooled analysis of all treatment arms combined, in both overweight (HR, 1.30; 95% CI, 1.06 to 1.61) and obese (HR, 1.31; 95% CI, 1.07 to 1.59) versus normal-weight patients. 43 In contrast, the related National Surgical Adjuvant Breast and Bowel Project B-31 trial identified no differences in OS or recurrence by baseline BMI, 30 similar to HER2-subset analyses in other studies. 28,32 Both N9831 and B-31 showed no differences in survival by BMI within any single treatment arm (either nontrastuzumab or trastuzumab containing). Taken together, these data do not provide strong evidence that obesity is associated with worse survival outcomes in patients with HER2-positive disease; it is possible that obesity associations may be more salient in those who do not receive trastuzumab. Larger and/or pooled studies should be conducted to address the relationship between BMI and survival in patients with HER2- positive disease, taking into account ER status and whether these effects are modified by trastuzumab or other HER2-directed therapies. OBESITY AND AROMATASE INHIBITOR THERAPY Because obesity is associated with elevated aromatase activity and serum estrogen levels in postmenopausal women, it is possible that obesity modulates responses to endocrine therapy. The most important evidence comes from clinical trials comparing adjuvant aromatase inhibitor (AI) with tamoxifen therapy (Table 3). 18,19,44-46 In postmenopausal patients, there is evidence of differential effects of anastrozole (v tamoxifen) in overweight or obese versus normal-weight patients. In the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial, the relative benefitof anastrozole over tamoxifen for total and distant recurrences was seen in normal-weight but not obese (BMI. 30 kg/m 2 ) women. 18 In the Austrian Breast and Colorectal Cancer Study Group 6a trial, the benefit of extended adjuvant anastrozole over no further therapy was present in normal-weight but not overweight or obese (combined) women. 45 In premenopausal patients, a similar pattern was seen in the Austrian Breast and Colorectal Cancer Study Group 12 trial, in which anastrozole plus goserelin led to greater risks of recurrence (HR, 1.49; 95% CI, 0.93 to 2.38) and death (HR, 3.03; 95% CI, 1.35 to 6.82) in overweight and obese women (combined) when compared with tamoxifen plus goserelin; DFS and OS were similar in normalweight women receiving either treatment. 46 The effects of BMI on outcomes in TEXT (Tamoxifen and Exemestane Trial) and SOFT www.jco.org 2016 by American Society of Clinical Oncology 4209

Jiralerspong and Goodwin Table 3. BMI and Adjuvant Endocrine Therapy Study Year Published Cooperative Group Study Name Country Patients Menopause Status No. of Patients Time Period Follow-Up Chemotherapy Endocrine Therapy Results Comments Dignam 45 2003 NSABP B-14 United States ER1Node- Pre 42% Post 58% 3385 1982-1988 166 months Tamoxifen v placebo x 5 years Sestak 18 2010 ATAC ATAC International ER1Node 1/- Post 4939 1996-2000 100 months NA Anastrozole v tamoxifen Ewertz 19 2012 IBCSG BIG 1-98 International ER1Node 1/- Post 4760 1998-2003 8.7 years Mainly CMF or anthracycline based. Minimal taxanes. Letrozole v tamoxifen Gnant 46 2013 ABCSG ABCSG-6a Austria ER1Node 1/- Post 634 1990-1995 73 months Additional 3 years of Anastrozole v nil Pfeiler 47 2011 ABCSG ABCSG-12 Austria ER1Node 1/- Pre 1684 1999-2006 63 months NA Goserelin plus (anastrozole v tamoxifen) Tam reduced BC recurrence and mortality, regardless of BMI More total and distant recurrences in BMI. 35 v, 23 (HR, 1.39 and 1.46, respectively) benefit for anastrozole over tamoxifen was lost for obese patients Relative benefit for letrozole over tamoxifen did not differ by BMI DFS and OS benefit of Ana seen in normal weight but not overweight/ obese patients Worse DFS and OS for overweight/obese v normal weight patients on Ana (HR, 1.60 and 2.14 respectively); worse DFS and OS for overweight/obese v normal weight patients on Anastrozole v tamoxifen, (HR, 1.49 and 3.03 respectively) Abbreviations: BC, breast cancer; BMI, body mass index; CMF, cyclophosphamide, methotrexate, and fluorouracil; DFS, disease-free survival; NA, not available; OS, overall survival. 4210 2016 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Obesity and Breast Cancer Prognosis (Suppression of Ovarian Function Trial), which investigated exemestane versus tamoxifen plus ovarian suppression, have not been reported, 47,48 although it has been reported that higher BMI is associated with a higher likelihood of elevated estradiol during treatment. 49 In contrast, in the Breast International Group 1-98 trial, there was no evidence that the benefit of letrozole over tamoxifen differed in postmenopausal obese (BMI $ 30 kg/m 2 )versus normal-weight women, in terms of DFS, OS, or distant recurrence-free interval. 19 In combination with the studies we have already discussed, this suggests that letrozole may be more effective than anastrozole in obese patients. However, an unpublished report of the FACE (Femara versus Anastrozole Clinical Evaluation) trial comparing letrozole versus anastrozole in postmenopausal patients with node-positive disease found no differences in DFS or OS according to BMI. 50 In a recent systematic review, a trend for a negative effect of obesity on AI efficacy in postmenopausal women was identified,butthesizeof the effect was unclear. 51 The potential for differential effects of different AIs was not examined. One hypothesis for the different results of letrozole versus anastrozole is differential suppression of estrogens by the AIs. Indeed, letrozole is known to be a more potent inhibitor of aromatase than anastrozole, 52,53 and one study reported that 3 months of letrozole suppressed plasma estrogen levels to lower absolute levels than anastrozole across all BMI categories. 54 Some studies have shown modestly increased estradiol levels for obese versus normal-weight patients receiving AIs, 54-56 but others have not. 57,58 Whether circulating estrogens (eg, estradiol, estrone, estrone sulfate) or surrogate hormones (eg, follicle-stimulating hormone) can serve as markers for endocrine therapy response in obese patients should be evaluated in prospective trials. POTENTIAL BIOLOGIC MECHANISMS LINKING OBESITY AND BREAST CANCER OUTCOMES There is a strong biologic basis for an association of obesity with poor breast cancer outcomes, as discussed elsewhere in this Journal of Clinical Oncology (JCO) special series and in prior reviews. 4,59-61 Table 4 summarizespointsofinterestinbreast cancer. INTERVENTIONS Although obesity is associated with poor outcomes in women with breast cancer, it is not clear whether postdiagnosis weight loss (resulting from changes in caloric intake, physical activity, or other interventions) will reverse this effect. There are several potential reasons for this. 63 It remains possible that obesity is not causally associated with outcomes; in this case, weight loss will not affect outcomes. If obesity is causally associated with outcomes, biologic effects of obesity may be fixed at diagnosis (ie, obese women may have more proliferative or aggressive tumors, characteristics that may not change with weight loss). Alternatively, the degree of weight loss possible with available lifestyle interventions (typically 5% to 7%) may not be sufficient to alter outcomes. Randomized trials will be needed to differentiate these possibilities. Diet and Physical Activity Interventions based on diet and physical activity in breast cancer are reviewed elsewhere in this JCO special series. Here we discuss two dietary intervention adjuvant trials for which there are long-term survival outcomes. The WINS trial randomly assigned 2,437 patients with breast cancer to a low-fat diet (target 15% fat calories, aggressive dietary counseling, and follow-up) versus control (minimal dietary counseling). 40 In the final analysis, 41 a significant improvement in OS in the ER- and PR-negative subset was seen, with an HR of 0.36 (95% CI, 0.18 to 0.74; P =.003). Nonsignificant improvements were seen in the ER- and PR-positive subset and the entire cohort. The WHEL (Women s Healthy Eating and Living) trial randomly assigned 3,088 patients to a diet high in vegetables, fruit, and fiber and low in fat (15% to 20% fat calories, telephone counseling, cooking classes offered) versus control (diet guidelines, no counseling, fewer cooking classes) up to 4 years postdiagnosis (a design that excluded women with early recurrences). 64 There was no difference in DFS or OS between intervention and control groups. The conflicting results of the WINS and WHEL trials may reflect differences in patient populations (younger and more premenopausal in WHEL), tumor characteristics (larger and more node positive in WHEL), and interventions delivered (low-fat diet alone in WINS v low-fat diet plus high-fruit, -vegetable, and -fiber diet in WHEL), as well as cancer-associated therapies Table 4. Biologic Mechanisms Linking Obesity and Breast Cancer Outcomes 4,59-61 Mechanism In patients with breast cancer, multiple metabolic and adipose tissue factors associated with obesity (eg, insulin, glucose, and adipokines such as, leptin, crown-like structures of inflammatory cells surrounding dying adipocytes in adipose tissue) have been associated with both obesity and poor outcomes Insulin, in particular, has received considerable attention; Goodwin et al 16,62 reported a strong association of fasting insulin with poor breast cancer outcomes in nondiabetic women; this association was subsequently shown to be strongest in first 5 years after diagnosis Because breast cancer cells commonly express insulin receptors (typically the fetal form), insulin can bind to those receptors (which may hybridize with IGF-1 receptors to bind IGF-1 as well) to activate signaling through PI3K/Akt and Ras/MAPK pathways, leading to increased proliferation and reduced apoptosis 4 Higher levels of circulating glucose and leptin have also been associated with obesity and poor breast cancer outcomes White adipose tissue-associated inflammation in the breast has also been postulated to play important role What has not been established (and therefore needs to be studied) is whether any or all of these factors mediate association of obesity with poor breast cancer outcomes, the relative importance of different factors in different breast cancer subtypes, and, most importantly, whether postdiagnosis change in these factors (either as result of weight loss or through pharmacologic interventions such as metformin) will improve breast cancer outcomes in obese women Abbreviations: IGF-1, insulin-like growth factor 1; MAPK, mitogen-activated protein kinase; PI3K, phosphatidylinositol 3-kinase. www.jco.org 2016 by American Society of Clinical Oncology 4211

Jiralerspong and Goodwin (more chemotherapy, less endocrine therapy in WHEL, reflecting more advanced tumor stage). 65 The earlier time of enrollment (all patients enrolled within year 1 after diagnosis in WINS v most patients in years 1 to 4 in WHEL, which may have precluded capture of early events, particularly in ER-negative cancers), larger decrease in percentage of fat calories, and greater adherence achieved (in WINS) might have contributed to the beneficial effects on recurrence rates in WINS. Additionally, a modest decrease in weight (average loss of approximately 6 lbs maintained over 5 years) in WINS but not WHEL may have contributed to the different outcomes. Recently, Chlebowski et al 66 reported the long-term effects of dietary fat reduction on breast cancer risk and outcomes postdiagnosis in the WHI (Women s Health Initiative) randomized prevention trial of dietary fat reduction versus no dietary change. Women in the intervention arm reported lower fat intake, higher fruit, vegetable, and grain intake, and a modest amount of weight loss. The risk of breast cancer was significantly lower in the intervention arm (HR, 0.68), and deaths after breast cancer diagnosis were significantly reduced (HR, 0.65). These results are encouraging because they suggest an intervention that modified diet and is associated with modest weight loss may be beneficial in lowering breast cancer deaths. However, because the intervention was started before breast cancer diagnosis, it cannot be assumed that dietary modification and/or weight loss after breast cancer diagnosis will have similar effects. As reviewed elsewhere in this JCO special series, numerous studies have demonstrated that weight loss, achieved through diet, physical activity, and behavior modification (based on counseling, either in-person or telephone, individual or group), is feasible after breast cancer diagnosis. 67-71 A recent systematic review reported most interventions led to weight loss of 5% or greater at 2 to 18 months, without serious adverse events. 68 What remains unknown is whether weight loss achieved via lifestyle interventions will improve survival and in which patient subsets or breast cancer subtypes, so additional studies are needed. One such study is the BWEL (Breast Cancer Weight Loss) trial, which will randomly assign more than 3,300 overweight or obese patients with HER2- negative disease to a telephone-based weight loss intervention versus health education and examine survival outcomes and correlative science. Similar lifestyle intervention trials are under way in Italy and Germany. The Italian DIANA-5 (Diet and Androgens) trial has randomly assigned patients with a high risk of recurrence (ER negative or ER positive with metabolic syndrome, high insulin or testosterone) to an intervention based on Mediterranean and macrobiotic diet, in association with moderate physical activity, versus control. 72 The German SUCCESS-C trial has randomly assigned pre- and postmenopausal patients with HER2-negative disease to a low-fat diet high in fruits, vegetables, and whole grains, with increased physical activity, versus control. 73 Both trials will examine relevant biomarkers as well as survival outcomes and are due to report in the near future. Several dietary strategies are available for weight loss, based on relative fat, carbohydrate, and protein content. Although this is a rapidly evolving area, current evidence suggests that there are no significant differences in terms of weight loss between diets based on low versus high fat, low versus high carbohydrates, or low versus high protein. 74 This is supported by meta-analyses. 75,76 A Mediterranean diet is also an effective option for weight loss. 77 The impact of these different dietary approaches on breast cancer outcomes remains to be fully evaluated. As discussed in our article, most breast cancer trials to date have used a low-fat diet approach, with the pending DIANA-5 trial using a Mediterranean diet. Metformin and Other Interventions Preclinical and clinical evidence has suggested that the diabetes drug metformin may have antitumor activity in breast cancer, 78-80 leading to a number of clinical trials of metformin in breast cancer. 81-83 The largest of these trials, the ongoing National Cancer Institute of Cancer MA.32 randomized adjuvant trial of metformin versus placebo involving 3,649 women, is nearing completion. Although not a weight loss intervention trial, early analyses showed that metformin improved many of the potential physiologic mediators of obesity effects in breast cancer (eg, insulin, glucose, homeostasis model assessment, leptin, highsensitivity C-reactive protein) independently of baseline insulin or BMI. 84 Furthermore, metformin was associated with modest weight loss (3% difference between metformin and placebo arms at 6 months). As a result, it is anticipated that the findings of MA.32 may shed light on the impact of postdiagnosis metabolic change (including that resulting from weight loss) on breast cancer outcomes. Bariatric surgery and recently approved weight loss medications may promote greater weight loss than is seen with lifestyle interventions, as discussed elsewhere in this JCO special series. The safety and impact of these interventions on breast cancer outcomes are unknown. For example, for the obesity drug liraglutide, there was a nonsignificant numeric difference in the number of new breast cancer cases (drug, 10 v placebo, three), which was postulated to be the result of enhanced detection of existing cancers in women who lost weight; no information was provided on breast cancer survival. 85,86 Longer-term follow-up will be needed for this and other recently approved obesity medications. CLINICAL MANAGEMENT OF OBESE PATIENTS WITH BREAST CANCER Surgery, Chemotherapy, and Endocrine Therapy Clinical management considerations relevant to surgery, chemotherapy, and endocrine therapy in obese patients are summarized in Table 5. Weight and Lifestyle Although obesity is associated with inferior outcomes in breast cancer, it is not clear whether this is the case for all subtypes of breast cancer, nor is it established whether weight loss (via diet, physical activity, surgical procedures, or drug therapies) will reverse this effect. Studies to obtain this information are under way. At least two studies (WINS and WHI) have reported that dietary fat reduction (before or within 1 year of diagnosis) associated with modest weight loss may lead to improved breast cancer outcomes. Until additional data are available, it is reasonable to counsel patients to maintain a healthy weight or lose the 5% to 7% that is 4212 2016 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Obesity and Breast Cancer Prognosis Table 5. Clinical Management of Obese Patients With Breast Cancer Management Surgery Obesity is a risk factor for complications after mastectomy, 87,88 either alone or in combination with immediate breast reconstruction, whether implant based or autologous 87-89 Complications include surgical site complications (eg, infections, wound dehiscence, and prosthetic and flap loss) and systemic medical complications (eg, VTE and pneumonia) Morbidly obese patients (class III; BMI. 40 kg/m 2 ) are at particularly elevated risk for these complications 90 Risk of VTE in obese patients is procedure dependent, increasing from 0.4% to 0.8% to 1.8% for mastectomy to implant to autologous reconstruction 91 Obesity is also a risk factor for lymphedema after breast cancer surgery, with risk estimates up to 5.5-fold higher 92 This information can be incorporated into surgical risk counseling and management; for instance, the incidence of VTE after implant-based reconstruction is approximately 1% and that after autologous reconstruction is approximately 2%, so consideration of prophylactic anticoagulants may be reasonable Chemotherapy Chemotherapy dosing in obese patients has been recently reviewed, and we summarize key points here 93,94 Obese patients with cancer should receive full weight-based chemotherapy doses calculated using their actual body weight, especially when the goal is cure There is no evidence to suggest increased hematologic or nonhematologic toxicity with such dosing In contrast, there is evidence to support poorer survival outcomes (DFS, OS) with underdosing of chemotherapy In the event of dose reductions, considerations for resumption of full weight-based dosing should be the same as those for nonobese patients Fixed dosing (independent of body weight or BSA) should be considered for certain chemotherapy agents (eg, carboplatin) Overweight and obesity have recently been associated with higher risk of cardiotoxicity after treatment with trastuzumab in women with HER2-positive breast cancer 95 Close monitoring and effective management of cardiac risk factors (eg, blood pressure, cholesterol, smoking) should be considered in these patients Endocrine therapy Concerns have been raised about the relationship between BMI and AI adjuvant therapy benefit (see Obesity and Aromatase Inhibitor Therapy) Available evidence suggests that anastrozole may be associated with inferior outcomes (v tamoxifen) in both post- and premenopausal patients who areobese; this was not found to be true for letrozole (v tamoxifen) in postmenopausal obese patients For postmenopausal patients, data from the BIG 1-98 trial support use of letrozole and data from ATAC and ABCSG-6a trials raise concerns regarding use of anastrozole As a result, because of potentially increased efficacy but equivalent tolerability, it is reasonable to use letrozole rather than anastrozole in obese postmenopausal patients For premenopausal patients, the current standard of care from TEXT and SOFT trials is to use exemestane plus ovarian suppression when indicated for aggressive disease; in the absence of further evidence to contrary, we believe this is acceptable for obese patients Additional data directly comparing different AIs in obese patients is pending Weight and lifestyle See Clinical Management of Obese Patients With Breast Cancer in the text Abbreviations: ABCSG, Austrian Breast and Colorectal Cancer Study Group; AI, aromatase inhibitor; ATAC, arimidex, tamoxifen, alone or in combination; BIG, Breast International Group; BMI, body mass index; BSA, body surface area; DFS, disease-free survival; HER2, human epidermal growth factor receptor 2; OS, overall survival; SOFT, Suppression of Ovarian Function Trial; TEXT, Tamoxifen and Exemestane Trial; VTE, venous thromboembolism. expected with lifestyle interventions and to follow current diet and physical activity guidelines for cancer survivors (Table 6). 96,97 Physical activity has been incorporated into these guidelines not only because it can contribute to weight management but also because higher levels of physical activity (pre- or postdiagnosis) have been associated with improved outcomes in observational studies. 98 In the absence of high-level evidence, it should be made clear that the rationale for maintaining a healthy weight is for general (noncancer) health benefits and that benefits in terms of improving breast cancer outcomes are not established and are under evaluation. 63 THE PATH FORWARD As reviewed in our article, a large body of evidence now supports an association between obesity and an increased risk of breast Table 6. ACS/ASCO Obesity Breast Cancer Care Guidelines 96,97 Guideline Obesity It is recommended that clinicians counsel survivors to achieve and maintain healthy weight and if overweight or obese to limit consumption of high-calorie foods and beverages and increase physical activity to promote and maintain weight loss Physical activity It is recommended that clinicians counsel survivors to engage in regular physical activity consistent with the ACS guideline 96 and specifically avoid inactivity and return to normal daily activities as soon as possible after diagnosis, aim for $ 150 minutes of moderate or 75 minutes of vigorous aerobic exercise per week, and include strength training exercises $ 2 days per week and emphasize strength training for women treated with adjuvant chemotherapy or hormone therapy Nutrition It is recommended that clinicians counsel survivors to achieve a dietary pattern that is high in vegetables, fruits, whole grains, and legumes; low in saturated fats; and limited in alcohol consumption Abbreviations: ACS, American Cancer Society. www.jco.org 2016 by American Society of Clinical Oncology 4213

Jiralerspong and Goodwin cancer recurrence and death. However, a number of challenges remain: Better defining the relationship between obesity and prognosis in different breast cancer subtypes, particularly TNBC and HER2-positive breast cancer. Understanding the biologic mechanisms in all subtypes. Evaluating whether weight loss and lifestyle interventions will improve outcomes. Developing obesity-targeted biologic therapies and evaluating whether they will improve outcomes. There are 247,000 new cases of breast cancer in the United States each year and 1.7 million in the world. Approximately 70% of these patients in the United States and 40% in the world are at increased risk for breast cancer recurrence and death because of obesity. The magnitude of this effect (35% to 40% increased risk) suggests that obesity may cancel the benefit of our best breast cancer therapies. Thus, overcoming these challenges represents a tremendous opportunity to improve the care and outcomes of obese patients with breast cancer and to impact public health positively. AUTHORS DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Disclosures provided by the authors are available with this article at www.jco.org. AUTHOR CONTRIBUTIONS Conception and design: All authors Collection and assembly of data: All authors Data analysis and interpretation: All authors Manuscript writing: All authors Final approval of manuscript: All authors Accountable for all aspects of the work: All authors REFERENCES 1. Flegal KM, Carroll MD, Ogden CL, et al: Prevalence and trends in obesity among US adults, 1999-2008. JAMA 303:235-241, 2010 2. Ogden CL, Carroll MD, Kit BK, et al: Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA 311:806-814, 2014 3. Ng M, Fleming T, Robinson M, et al: Global, regional, and national prevalence of overweight and obesity in children and adults during 1980-2013: A systematic analysis for the Global Burden of Disease Study 2013. Lancet 384:766-781, 2014 4. Goodwin PJ, Stambolic V: Impact of the obesity epidemic on cancer. Annu Rev Med 66: 281-296, 2015 5. Chlebowski RT, Aiello E, McTiernan A: Weight loss in breast cancer patient management. J Clin Oncol 20:1128-1143, 2002 6. Parekh N, Chandran U, Bandera EV: Obesity in cancer survival. Annu Rev Nutr 32:311-342, 2012 7. Suzuki R, Orsini N, Saji S, et al: Body weight and incidence of breast cancer defined by estrogen and progesterone receptor status: A meta-analysis. Int J Cancer 124:698-712, 2009 8. Pierobon M, Frankenfeld CL: Obesity as a risk factor for triple-negative breast cancers: A systematic review and meta-analysis. Breast Cancer Res Treat 137:307-314, 2013 9. Wu Y, Zhang D, Kang S: Physical activity and risk of breast cancer: A meta-analysis of prospective studies. Breast Cancer Res Treat 137:869-882, 2013 10. Protani M, Coory M, Martin JH: Effect of obesity on survival of women with breast cancer: Systematic review and meta-analysis. Breast Cancer Res Treat 123:627-635, 2010 11. Niraula S, Ocana A, Ennis M, et al: Body size and breast cancer prognosis in relation to hormone receptor and menopausal status: A meta-analysis. Breast Cancer Res Treat 134:769-781, 2012 12. Chan DS, Vieira AR, Aune D, et al: Body mass index and survival in women with breast cancer: Systematic literature review and meta-analysis of 82 follow-up studies. Ann Oncol 25:1901-1914, 2014 13. Majed B, Moreau T, Senouci K, et al: Is obesity an independent prognosis factor in woman breast cancer? Breast Cancer Res Treat 111:329-342, 2008 14. Chen X, Lu W, Zheng W, et al: Obesity and weight change in relation to breast cancer survival. Breast Cancer Res Treat 122:823-833, 2010 15. Ewertz M, Jensen MB, Gunnarsdóttir KA, et al: Effect of obesity on prognosis after early-stage breast cancer. J Clin Oncol 29:25-31, 2011 16. Goodwin PJ, Ennis M, Pritchard KI, et al: Insulin- and obesity-related variables in early-stage breast cancer: Correlations and time course of prognostic associations. J Clin Oncol 30:164-171, 2012 17. Nechuta S, Chen WY, Cai H, et al: A pooled analysis of post-diagnosis lifestyle factors in association with late estrogen-receptor-positive breast cancer prognosis. Int J Cancer 138:2088-2097, 2016 18. Sestak I, Distler W, Forbes JF, et al: Effect of body mass index on recurrences in tamoxifen and anastrozole treated women: An exploratory analysis from the ATAC trial. J Clin Oncol 28:3411-3415, 2010 19. Ewertz M, Gray KP, Regan MM, et al: Obesity and risk of recurrence or death after adjuvant endocrine therapy with letrozole or tamoxifen in the Breast International Group 1-98 trial. J Clin Oncol 30: 3967-3975, 2012 20. Playdon MC, Bracken MB, Sanft TB, et al: Weight gain after breast cancer diagnosis and all-cause mortality: Systematic review and meta-analysis. J Natl Cancer Inst 107:djv275, 2015 21. Ligibel JA, Cirrincione CT, Liu M, et al: Body mass index, PAM50 subtype, and outcomes in nodepositive breast cancer: CALGB 9741 (Alliance). J Natl Cancer Inst 25:107, 2015 22. Pajares B, Pollán M, Martín M, et al: Obesity and survival in operable breast cancer patients treated with adjuvant anthracyclines and taxanes according to pathological subtypes: A pooled analysis. Breast Cancer Res 15:R105, 2013 23. Fontanella C, Lederer B, Gade S, et al: Impact of body mass index on neoadjuvant treatment outcome: A pooled analysis of eight prospective neoadjuvant breast cancer trials. Breast Cancer Res Treat 150:127-139, 2015 24. Widschwendter P, Friedl TW, Schwentner L, et al: The influence of obesity on survival in early, high-risk breast cancer: Results from the randomized SUCCESS A trial. Breast Cancer Res 17:129, 2015 25. Hao S, Liu Y, Yu KD, et al: Overweight as a prognostic factor for triple-negative breast cancers in Chinese women. PLoS One 10:e0129741, 2015 26. Bonsang-Kitzis H, Chaltier L, Belin L, et al: Beyond axillary lymph node metastasis, BMI and menopausal status are prognostic determinants for triple-negative breast cancer treated by neoadjuvant chemotherapy. PLoS One 10:e0144359, 2015 [Erratum: PLoS One 11:e0159123, 2016] 27. Pan H, Gray RG: Effect of obesity on prognosis in premenopausal ER+ early breast cancer: EBCTCG data on 80,000 patients in 70 trials. J Clin Oncol 32: 5s, 2014 (suppl; abstr 503) 28. Sparano JA, Wang M, Zhao F, et al: Obesity at diagnosis is associated with inferior outcomes in hormone receptor-positive operable breast cancer. Cancer 118:5937-5946, 2012 29. Sparano JA, Zhao F, Martino S, et al: Longterm follow-up of the E1199 phase III trial evaluating the role of taxane and schedule in operable breast cancer. J Clin Oncol 33:2353-2360, 2015 30. Cecchini RS, Swain SM, Costantino JP, et al: Body mass index at diagnosis and breast cancer survival prognosis in clinical trial populations from NRG Oncology/NSABP B-30, B-31, B-34, and B-38. Cancer Epidemiol Biomarkers Prev 25:51-59, 2016 31. Dignam JJ, Wieand K, Johnson KA, et al: Effects of obesity and race on prognosis in lymph nodenegative, estrogen receptor-negative breast cancer. Breast Cancer Res Treat 97:245-254, 2006 32. Jiralerspong S, Kim ES, Dong W, et al: Obesity, diabetes, and survival outcomes in a large cohort of early-stage breast cancer patients. Ann Oncol 24: 2506-2514, 2013 33. Dawood S, Lei X, Litton JK, et al: Impact of body mass index on survival outcome among women with early stage triple-negative breast cancer. Clin Breast Cancer 12:364-372, 2012 34. Tait S, Pacheco JM, Gao F, et al: Body mass index, diabetes, and triple-negative breast cancer prognosis. Breast Cancer Res Treat 146:189-197, 2014 35. Ademuyiwa FO, Groman A, O Connor T, et al: Impact of body mass index on clinical outcomes in 4214 2016 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Obesity and Breast Cancer Prognosis triple-negative breast cancer. Cancer 117:4132-4140, 2011 36. Mowad R, Chu QD, Li BD, et al: Does obesity have an effect on outcomes in triple-negative breast cancer? J Surg Res 184:253-259, 2013 37. Dunlap SM, Chiao LJ, Nogueira L, et al: Dietary energy balance modulates epithelial-tomesenchymal transition and tumor progression in murine claudin-low and basal-like mammary tumor models. Cancer Prev Res (Phila) 5:930-942, 2012 38. Ford NA, Rossi EL, Barnett K, et al: Omega-3- acid ethyl esters block the protumorigenic effects of obesity in mouse models of postmenopausal basallike and claudin-low breast cancer. Cancer Prev Res (Phila) 8:796-806, 2015 39. Sundaram S, Freemerman AJ, Johnson AR, et al: Role of HGF in obesity-associated tumorigenesis: C3(1)-TAg mice as a model for human basal-like breast cancer. Breast Cancer Res Treat 142:489-503, 2013 40. Chlebowski RT, Blackburn GL, Thomson CA, et al: Dietary fat reduction and breast cancer outcome: Interim efficacy results from the Women s Intervention Nutrition Study. J Natl Cancer Inst 98: 1767-1776, 2006 41. Chlebowski RT, Blackburn GL: Final survival analysis from the randomized Women s Intervention Nutrition Study (WINS) evaluating dietary invention as adjuvant breast cancer therapy. Presented at San Antonio Breast Cancer Symposium, San Antonio, TX, December 9-13 2014 42. Mazzarella L, Disalvatore D, Bagnardi V, et al: Obesity increases the incidence of distant metastases in oestrogen receptor-negative human epidermal growth factor receptor 2-positive breast cancer patients. Eur J Cancer 49:3588-3597, 2013 43. Crozier JA, Moreno-Aspitia A, Ballman KV, et al: Effect of body mass index on tumor characteristics and disease-free survival in patients from the HER2-positive adjuvant trastuzumab trial N9831. Cancer 119:2447-2454, 2013 44. Dignam JJ, Wieand K, Johnson KA, et al: Obesity, tamoxifen use, and outcomes in women with estrogen receptor-positive early-stage breast cancer. J Natl Cancer Inst 95:1467-1476, 2003 45. Gnant M, Pfeiler G, Stöger H, et al: The predictive impact of body mass index on the efficacy of extended adjuvant endocrine treatment with anastrozole in postmenopausal patients with breast cancer: An analysis of the randomised ABCSG-6a trial. Br J Cancer 109:589-596, 2013 46. Pfeiler G, Königsberg R, Fesl C, et al: Impact of body mass index on the efficacy of endocrine therapy in premenopausal patients with breast cancer: An analysis of the prospective ABCSG-12 trial. J Clin Oncol 29:2653-2659, 2011 47. Pagani O, Regan MM, Walley BA, et al: Adjuvant exemestane with ovarian suppression in premenopausal breast cancer. N Engl J Med 371: 107-118, 2014 48. Francis PA, Regan MM, Fleming GF, et al: Adjuvant ovarian suppression in premenopausal breast cancer. N Engl J Med 372:436-446, 2015 49. Bellet M, Gray KP, Francis PA, et al: Twelvemonth estrogen levels in premenopausal women with hormone receptor positive breast cancer receiving adjuvant triptorelin plus exemestane or tamoxifen in the Suppression of Ovarian Function Trial (SOFT): The SOFT-EST substudy. J Clin Oncol 34: 1584-1593, 2016 50. O Shaughnessy J, Yardley DA, Burris HA, et al: Randomized phase 3 trial of adjuvant letrozole vs anastrozole in postmenopausal patients with hormone receptor-positive, node-positive early breast cancer: Final efficacy and safety results of the Femara vs Anastrozole Clinical Evaluation (FACE) trial. Presented at San Antonio Breast Cancer Symposium, San Antonio, TX, December 8-12, 2015 51. Ioannides SJ, Barlow PL, Elwood JM, et al: Effect of obesity on aromatase inhibitor efficacy in postmenopausal, hormone receptor-positive breast cancer: A systematic review. Breast Cancer Res Treat 147:237-248, 2014 52. Bhatnagar AS: The discovery and mechanism of action of letrozole. Breast Cancer Res Treat 105: 7-17, 2007 (suppl 1) 53. Lønning PE: The potency and clinical efficacy of aromatase inhibitors across the breast cancer continuum. Ann Oncol 22:503-514, 2011 54. Folkerd EJ, Dixon JM, Renshaw L, et al: Suppression of plasma estrogen levels by letrozole and anastrozole is related to body mass index in patients with breast cancer. J Clin Oncol 30: 2977-2980, 2012 55. Pfeiler G, Königsberg R, Hadji P, et al: Impact of body mass index on estradiol depletion by aromatase inhibitors in postmenopausal women with early breast cancer. Br J Cancer 109:1522-1527, 2013 56. Decker DA, Reynolds RB, Molthrop DC, et al: Obesity and nonadherence correlate with elevated serum estradiol levels in postmenopausal women receiving adjuvant aromatase inhibitor therapy. Breast J 20:553-554, 2014 57. Kyvernitakis I, Knöll D, Struck M, et al: Impact of BMI on serum estradiol and bone turnover markers in postmenopausal women with hormone-sensitive early breast cancer treated with anastrozole. J Cancer Res Clin Oncol 140:159-166, 2014 58. Diorio C, Lemieux J, Provencher L, et al: Aromatase inhibitors in obese breast cancer patients are not associated with increased plasma estradiol levels. Breast Cancer Res Treat 136:573-579, 2012 59. Iyengar NM, Hudis CA, Dannenberg AJ: Obesity and cancer: Llocal and systemic mechanisms. Annu Rev Med 66:297-309, 2015 60. Allott EH, Hursting SD: Obesity and cancer: Mechanistic insights from transdisciplinary studies. Endocr Relat Cancer 22:R365-R386, 2015 61. Park J, Morley TS, Kim M, et al: Obesity and cancer mechanisms underlying tumour progression and recurrence. Nat Rev Endocrinol 10:455-465, 2014 62. Goodwin PJ, Ennis M, Pritchard KI, et al: Fasting insulin and outcome in early-stage breast cancer: Results of a prospective cohort study. J Clin Oncol 20:42-51, 2002 63. Goodwin PJ: Obesity and breast cancer outcomes: How much evidence is needed to change practice? J Clin Oncol 34:646-648, 2016 64. Pierce JP, Natarajan L, Caan BJ, et al: Influence of a diet very high in vegetables, fruit, and fiber and low in fat on prognosis following treatment for breast cancer: The Women s Healthy Eating and Living (WHEL) randomized trial. JAMA 298:289-298, 2007 65. Pierce JP: Diet and breast cancer prognosis: Making sense of the Women s Healthy Eating and Living and Women s Intervention Nutrition Study trials. Curr Opin Obstet Gynecol 21:86-91, 2009 66. Chlebowski R, Aragaki AK, Anderson GL, et al: Low-fat dietary pattern and breast cancer mortality in the Women s Health Initiative (WHI) randomized trial. Presented at American Association for Cancer Research Annual Meeting, New Orleans, LA, April 16-20, 2016 67. Playdon M, Thomas G, Sanft T, et al: Weight loss intervention for breast cancer survivors: A systematic review. Curr Breast Cancer Rep 5:222-246, 2013 68. Reeves MM, Terranova CO, Eakin EG, et al: Weight loss intervention trials in women with breast cancer: A systematic review. Obes Rev 15:749-768, 2014 69. Goodwin PJ, Segal RJ, Vallis M, et al: Randomized trial of a telephone-based weight loss intervention in postmenopausal women with breast cancer receiving letrozole: The LISA trial. J Clin Oncol 32:2231-2239, 2014 70. Rock CL, Flatt SW, Byers TE, et al: Results of the Exercise and Nutrition to Enhance Recovery and Good Health for You (ENERGY) Trial: A behavioral weight loss intervention in overweight or obese breast cancer survivors. J Clin Oncol 33:3169-3176, 2015 71. Harrigan M, Cartmel B, Loftfield E, et al: Randomized trial comparing telephone versus inperson weight loss counseling on body composition and circulating biomarkers in women treated for breast cancer: The Lifestyle, Exercise, and Nutrition (LEAN) study. J Clin Oncol 34:669-676, 2016 72. Villarini A, Pasanisi P, Traina A, et al: Lifestyle and breast cancer recurrences: The DIANA-5 trial. Tumori 98:1-18, 2012 73. Rack B, Andergassen U, Neugebauer J, et al: The German SUCCESS C study: The first European lifestyle study on breast cancer. Breast Care (Basel) 5:395-400, 2010 74. Sacks FM, Bray GA, Carey VJ, et al: Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates. N Engl J Med 360:859-873, 2009 75. Hu T, Mills KT, Yao L, et al: Effects of lowcarbohydrate diets versus low-fat diets on metabolic risk factors: A meta-analysis of randomized controlled clinical trials. Am J Epidemiol 176:S44-S54, 2012 (suppl 7) 76. Johnston BC, Kanters S, Bandayrel K, et al: Comparison of weight loss among named diet programs in overweight and obese adults: A metaanalysis. JAMA 312:923-933, 2014 77. Shai I, Schwarzfuchs D, Henkin Y, et al: Weight loss with a low-carbohydrate, Mediterranean, or low-fat diet. N Engl J Med 359:229-241, 2008 78. Dowling RJ, Niraula S, Stambolic V, et al: Metformin in cancer: Translational challenges. J Mol Endocrinol 48:R31-R43, 2012 79. Jiralerspong S, Palla SL, Giordano SH, et al: Metformin and pathologic complete responses to neoadjuvant chemotherapy in diabetic patients with breast cancer. J Clin Oncol 27:3297-3302, 2009 80. Goodwin PJ, Ligibel JA, Stambolic V: Metformin in breast cancer: Time for action. J Clin Oncol 27:3271-3273, 2009 81. Camacho L, Dasgupta A, Jiralerspong S: Metformin in breast cancer: An evolving mystery. Breast Cancer Res 17:88, 2015 82. Pizzuti L, Vici P, Di Lauro L, et al: Metformin and breast cancer: Basic knowledge in clinical context. Cancer Treat Rev 41:441-447, 2015 83. Lord SR, Patel N, Liu D, et al: Neoadjuvant window studies of metformin and biomarker development for drugs targeting cancer metabolism. J Natl Cancer Inst Monogr 2015:81-86, 2015 84. Goodwin PJ, Parulekar WR, Gelmon KA, et al: Effect of metformin vs placebo on and metabolic factors in NCIC CTG MA.32. J Natl Cancer Inst 4:107, 2015 www.jco.org 2016 by American Society of Clinical Oncology 4215

Jiralerspong and Goodwin 85. Pi-Sunyer X, Astrup A, Fujioka K, et al: A randomized, controlled trial of 3.0 mg of liraglutide in weight management. N Engl J Med 373:11-22, 2015 86. Pi-Sunyer X; SCALE Obesity and Prediabetes Investigators: Liraglutide in weight management. N Engl J Med 373:1781-1782, 2015 87. Fischer JP, Tuggle CT, Au A, et al: A 30-day risk assessment of mastectomy alone compared to immediate breast reconstruction (IBR). J Plast Surg Hand Surg 48:209-215, 2014 88. Jeevan R, Browne JP, Pereira J, et al: Socioeconomic deprivation and inpatient complication rates following mastectomy and breast reconstruction surgery. Br J Surg 102:1064-1070, 2015 89. Fischer JP, Nelson JA, Kovach SJ, et al: Impact of obesity on outcomes in breast reconstruction: Analysis of 15,937 patients from the ACS-NSQIP datasets. J Am Coll Surg 217:656-664, 2013 90. Fischer JP, Cleveland EC, Nelson JA, et al: Breast reconstruction in the morbidly obese patient: Assessment of 30-day complications using the 2005 to 2010 National Surgical Quality Improvement Program data sets. Plast Reconstr Surg 132:750-761, 2013 91. Fischer JP, Wes AM, Tuggle CT, et al: Venous thromboembolism risk in mastectomy and immediate breast reconstruction: Analysis of the2005to2011americancollegeofsurgeons National Surgical Quality Improvement Program data sets. Plast Reconstr Surg 133:263e-273e, 2014 92. DiSipio T, Rye S, Newman B, et al: Incidence of unilateral arm lymphoedema after breast cancer: A systematic review and meta-analysis. Lancet Oncol 14:500-515, 2013 93. Griggs JJ, Mangu PB, Anderson H, et al: Appropriate chemotherapy dosing for obese adult patients with cancer: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol 30: 1553-1561, 2012 94. Lyman GH, Sparreboom A: Chemotherapy dosing in overweight and obese patients with cancer. Nat Rev Clin Oncol 10:451-459, 2013 95. Mantarro S, Rossi M, Bonifazi M, et al: Risk of severe cardiotoxicity following treatment with trastuzumab: A meta-analysis of randomized and cohort studies of 29,000 women with breast cancer. Intern Emerg Med 11:123-140, 2016 96. Rock CL, Doyle C, Demark-Wahnefried W, et al: Nutrition and physical activity guidelines for cancer survivors. CA Cancer J Clin 62:243-274, 2012 97. Runowicz CD, Leach CR, Henry NL, et al: American Cancer Society/American Society of Clinical Oncology breast cancer survivorship care guideline. J Clin Oncol 34:611-635, 2016 98. Zhong S, Jiang T, Ma T, et al: Association between physical activity and mortality in breast cancer: A meta-analysis of cohort studies. Eur J Epidemiol 29:391-404, 2014 nnn 4216 2016 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY