- Email: [email protected]
Obesity, insulin resistance and breast cancer outcomes
The Breast xxx (2015) e1ee4
Contents lists available at ScienceDirect
The Breast journal homepage: www.elsevier.com/brst
Obesity, insulin resistance and breast cancer outcomes Pamela J. Goodwin a, b, * a
Department of Medicine, Division of Medical Oncology and Hematology, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada b Division of Clinical Epidemiology, Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada
a r t i c l e i n f o
a b s t r a c t
Article history: Available online xxx
There is growing evidence that obesity is associated with poor outcomes in early stage breast cancer. This paper addresses four current areas of focus: 1. Is obesity associated with poor outcomes in all biologic subtypes of breast cancer? 2. Does obesity effect AI efficacy or estrogen suppression in the adjuvant setting? 3. What are the potential biologic underpinnings of the obesity-breast cancer association? 4. Are intervention studies warranted? If so, which interventions in which populations? Research is needed to resolve these questions; intervention trials involving lifestyle interventions or targeting the biology postulated to link obesity and cancer are recommended. © 2015 Published by Elsevier Ltd.
Keywords: Breast cancer Adjuvant Obesity Aromatase inhibitors Biology Intervention
Introduction There is growing recognition that obesity is associated with adverse breast cancer outcomes, notably a higher risk of distant recurrence and death. Research since St. Gallen 2013 [1] has focused on four key areas e (1) Is obesity associated with poor outcomes in all biologic subtypes of breast cancer?, (2) Does obesity impact aromatase inhibitor (AI) efficacy or estrogen suppression in the adjuvant setting?, (3) What are the potential biologic underpinnings of the obesity-breast cancer outcome association?, and (4) Are intervention studies warranted? If so, which interventions in which populations? These four areas of research will be discussed below. Is obesity associated with poor outcomes in all biologic subtypes of breast cancer? Body mass index (BMI) at diagnosis has been associated with both breast cancer mortality and overall mortality in research spanning five decades. In a recent meta-analysis [2], a curvilinear association of BMI with outcome was seen e increased risk was present in individuals with BMI under 20 kg/m2 and BMI over 25 kg/m2 at diagnosed e in the latter group, the risk increased with increasing BMI. In a prior meta-analysis [3], obesity was associated
* Mount Sinai Hospital, 1284-600 University Avenue, Toronto, Ontario M5G 1X5, Canada. Tel.: þ1 416 586 8605; fax: þ1 416 586 8659. E-mail address: [email protected].
with adverse outcome (overall or breast cancer specific mortality) in both pre- and postmenopausal women, in those diagnosed both prior to and after 1995, and in observational cohorts as well as “treatment cohorts” (i.e. retrospective analyses nested in randomized trials) e Hazard ratios (HRs) for obese vs. non-obese overall and in these subgroups were in the range of 1.2e1.35. Our group [4] performed a meta-analysis of obesity associations in women with estrogen receptor (ER) positive vs. negative breast cancer e hazard ratios (HRs) for breast cancer specific survival (BCSS) and overall survival (OS) were similarly elevated in both groups (BCSS HRs 1.46, 1.36 respectively). However, some recent studies (e.g. Sparano et al.) [5] have found no evidence of an association of obesity with outcome in hormone receptor negative breast cancer. At least ten subsequent studies have explored this issue with varying results. For example, Pajares et al. [6] reported worse BCSS and OS in women having a BMI over 35 kg/m2 with hormone receptor negative or HER2þ breast cancer enrolled onto a series of randomized trials, all of which involved anthracyclines. Importantly, Pan et al. [7] analyzed associations of BMI with outcome in over 80,000 women (20,000 ER positive premenopausal, 40,000 ER positive postmenopausal, 20,000 ER negative) enrolled onto randomized trials included in the Early Breast Cancer Clinical Trialists Cooperative Group (EBCTCG). In their patient-level meta-analyses, they found an increased risk of breast cancer mortality in heavier premenopausal women with ER positive breast cancer e HR per 10 kg/m2 in BMI 1.23 [95% confidence intervals (CIs) 1.16e1.31, p < 0.00001], regardless of whether the trial tested hormone, chemotherapy or other treatments. In contrast to earlier
http://dx.doi.org/10.1016/j.breast.2015.07.014 0960-9776/© 2015 Published by Elsevier Ltd.
Please cite this article in press as: Goodwin PJ, Obesity, insulin resistance and breast cancer outcomes, The Breast (2015), http://dx.doi.org/ 10.1016/j.breast.2015.07.014
e2
P.J. Goodwin / The Breast xxx (2015) e1ee4
study-level meta-analyses, no significant association was seen in postmenopausal women with ER positive breast cancer (HR 1.03, p ¼ 0.27) or in pre- or postmenopausal women with ER negative breast cancer (HR 1.02, p ¼ 0.39). The basis for these recent discrepancies is not well understood. Because most of the discrepant findings arose from randomized clinical trials (RCTs), one possibility is that women enrolled onto these trials differ from the general population of breast cancer patients with respect to factors, such as metabolic health, that are relevant to the obesity-cancer link. This issue has not been examined in breast cancer patients, however, in the Physicians' Health Study, those who responded to an invitation to be assessed for participation, those willing to be assessed for eligibility, those eligible for randomization, and those who actually accepted randomization had increasingly lower total mortality, cardiovascular mortality and cancer mortality than those who did not meet these criteria. This is evidence of a selection bias (towards enrolment of healthier individuals) occurring during the recruitment process. Furthermore, Kramer et al. [8] have provided evidence that obesity does not necessarily predict metabolic health (defined as no components of the insulin resistance syndrome) e those who are metabolically unhealthy have increased all-cause mortality and cardiovascular events, independent of BMI category (normal weight, overweight or obesity). Thus, it is possible that randomization onto breast cancer clinical trials selects for metabolically healthy individuals, regardless of baseline BMI, either as a direct result of inclusion/exclusion criteria (e.g. a requirement for normal cardiac function in trials of anthracyclines or trastuzumab) or through a more subtle selection of healthy individuals as was seen in the Physicians' Health Study. If so, prognostic associations of BMI may have been obscured. Does obesity effect AI efficacy or estrogen suppression in the adjuvant setting? Total body aromatization increases with increasing BMI; leptin and inflammation (both associated with obesity) may also increase aromatase activity independent of BMI. At St. Gallen 2013, the potential effect of baseline BMI on the relative efficacy of AIs vs. tamoxifen in a series of adjuvant trials was discussed [1]. To summarize, in BIG 1-98 [9] (which compared letrozole to tamoxifen) there was no evidence of a significant treatment by BMI interaction. However, in the Anastrozole, Tamoxifen Alone or in Combination (ATAC) [10] and Australian Breast Cancer Study Group (ABCSG) e12 [11] trials (which compared anastrozole to tamoxifen, in postmenopausal women in the former and in combination with goserelin in premenopausal women in the latter), there was evidence that the relative benefit of anastrozole vs. tamoxifen was greatest when BMI was normal vs. elevated. In the ATAC trial [10], this effect of BMI was not statistically significant. In ABSCG-12 [11], disease-free survival (DFS) and OS were worse in women with BMI over 25 kg/m2 who received anastrozole as opposed to tamoxifen (HR 1.49, 95% CI 0.93e2.0 and HR 3.03, 95%CI 1.35e6.82 respectively). In the TEAM Trial [12] there was no evidence that the relative benefit of exemestane vs. tamoxifen differed across BMI levels. New since St. Gallen 2013 is a report by Gnant et al. [13] which examined the use of extended adjuvant anastrozole vs. no treatment in postmenopausal women e a benefit for anastrozole was seen only in women with baseline BMI <25 kg/m2 [DFS HR 0.48 (95% CI 0.26e0.89), distant DFS HR 0.22 (95% CI 0.05e1.0), and OS HR 0.45 (95% CI 0.19e1.04)]. No benefit was seen in individuals with a BMI over 25 kg/m2; the BMI by treatment benefit interaction approached significance for distant DFS (p ¼ 0.07). Recently, a series of studies [14e18] has examined the extent to which AIs suppress circulating levels of estradiol or estrone
according to BMI levels (see Table 1). Folkerd et al. [14] reported that levels of estrone and estradiol were greater at higher levels of BMI in 44 women receiving anastrozole or letrozole e this was statistically significant in those receiving letrozole (r ¼ 0.35, p ¼ 0.013 for estradiol and r ¼ 0.30, p ¼ 0.035 for estrone sulfate) e a non-significant trend was seen in those receiving anastrozole. Of note, levels of both estradiol and estrone sulfate were lower in those receiving letrozole vs. anastrozole across all BMI categories. Pfeiler et al. [15] found a similar pattern of higher estradiol levels in 28 obese (vs. 40. non-obese) women receiving AIs (60 anastrozole, 8 letrozole), r ¼ 0.35, p ¼ 0.05; importantly, FSH levels were lower in obese women (r ¼ 0.34, p ¼ 0.06), consistent with higher endogenous estrogen levels. Kyvernitakis I et al. [17] studied 70 postmenopausal women receiving adjuvant anastrozole; after 12e24 months of treatment, overweight and obese women has non-significantly higher estradiol concentrations than normal weight women. Elliott et al. [18] studied a mixed group of 64 adjuvant and second line patients (who had progressed on an AI) receiving one of the three approved AIs; BMI and estradiol were higher in metastatic patients. BMI was non-significantly correlated with estradiol (9.3% higher for normal vs. overweight vs. obese, p ¼ 0.06). Finally, Lonning et al. [16] reported that treatment levels of estrone sulfate were significantly correlated with BMI in 64 women receiving letrozole or anastrozole in the adjuvant or metastatic setting (n ¼ 25, r ¼ 0.60, p ¼ 0.001 and n ¼ 12, r ¼ 0.61, p ¼ 0.035 respectively). Similar to results of Folkerd et al., [15] estrone sulfate levels were higher in those receiving anastrozole (vs. letrozole), independent of BMI. BMI was not correlated with on-treatment in vivo aromatization in the metastatic setting in women receiving a range of first, second or third line AIs, nor was it correlated with intratumoral levels of estrogens in the neoadjuvant setting. Although many of the reported BMI associations were not statistically significant, sample sizes were small and most studies reported higher estrogen levels in heavier women. Further research employing larger sample sizes is urgently needed. Although these data suggest estrogen suppression may be less effective in women with higher BMI, there is little evidence that this impacted clinical outcomes in obese women receiving letrozole relative to tamoxifen (BIG 1e98) [9]. There is growing evidence that higher BMI may contribute to reduced efficacy of anastrazole vs. tamoxifen. Because of this, selection of other AIs, notably letrozole, is preferred in overweight or obese women. What are the potential biologic underpinnings of the obesitybreast cancer association? There is an evolving understanding of the complex biology potentially underlying the association between obesity and cancer in general, and breast cancer outcomes in particular. A number of recent review articles have addressed this issue in detail [19e21]. Obesity leads to an expanded and reprogrammed, metabolically active, adipose tissue mass, with increased numbers of preadipocytes and inflammatory cells, higher levels of leptin and free fatty acids, and greater release of cytokines and other inflammatory compounds. These changes result in an altered systemic physiology, leading to insulin resistance (higher insulin levels, dysglycemia, other metabolic changes), higher circulating levels of free fatty acids, lipids, leptin, estrogens and inflammatory markers. This altered systemic physiology can have direct effects on cancer cells, with higher levels of circulating estrogen, insulin and inflammatory factors activating estrogen and insulin/IGF signaling pathways (e.g. PI3K, ras) as well as JAK-STAT, NF-kappa-b and other pathways. The associated dysglycemia may be associated with altered tumor cell metabolism, such as a shift from oxidative phosphorylation to aerobic glycolysis (the Warburg effect). Locally, adipose cells in the
Please cite this article in press as: Goodwin PJ, Obesity, insulin resistance and breast cancer outcomes, The Breast (2015), http://dx.doi.org/ 10.1016/j.breast.2015.07.014
P.J. Goodwin / The Breast xxx (2015) e1ee4
e3
Table 1 Estradiol levels According to BMI in adjuvant aromatase inhibitor patients. Duration
Pre-treatment Folkerd [14] Pfeiler [15] Kyvernitakis [17] Post letrozle Folkerd [14] Post anastrozle Folkerd [14] Kyvernitakis [17] Post any AI Pfeiler [15] Elliot [18]
#
Normal
Overweight
Obese
<25 kg/m2
25e30 kg/m2
>30 kg/m2
Sig.
26
YES NO YES
e e e
44 68 70
20 18.9b 9.7
13.6
39/55a 20.2 15.1
3 MOS
44
1.4
1.4
2.0/2.4a
YES
3 MOS 12 MOS
44 70
2.8 5.0
2.5 8.4
3.2/4.2a 5.4
NO NO
3 MOS 3 MOS
68 45
9.0b 3.0
3.5
12.5 3.3
NO NO
Abbreviations: BMI, body mass index; sig., significant at p < 0.05. a BMI 30e35/>35 kg/m2. b BMI <30 kg/m2.
tumor microenvironment can be associated with increased numbers of inflammatory cells, altered levels of cytokines, adipocines and inflammatory markers, which may promote carcinogenesis, tumor progression, invasion and metastasis. In clinical studies, many of the physiologic factors associated with obesity (hyperinsulinemia, dysglycemia, higher circulating estrogen levels) have been associated with poor breast cancer outcomes. It is likely that the relative importance of different mediators varies across breast cancer subtypes and over time. Higher estrogen levels likely contribute primarily to poor outcome in hormone receptor positive breast cancer. When we examined timerelated associations of obesity and related factors with breast cancer outcome [22] we found that insulin and glucose were associated with outcomes only during the first five years after diagnosis, whereas BMI and the associated adipokine leptin had more constant associations over a longer timeframe. Although there have been important advances in understanding the potential biologic basis for an obesity-cancer link, it remains to be demonstrated whether this link is causal, or whether changes achievable through weight loss or pharmacologic interventions that target this link will lead to improved outcomes. Are intervention studies warranted? If so, which interventions in which populations? In the 50 years since an association of obesity with breast cancer outcomes was first recognized, dozens of observational studies have examined this association. As noted above, meta-analyses synthesizing these data have suggested the risk of breast cancer death is approximately one-third higher in overweight/obese women, although there has been recent controversy regarding a potential differential association in women with hormone receptor positive vs. negative cancers. Despite this considerable body of work, it has not been established whether the association of obesity with breast cancer outcome is causal, or whether higher BMI is a marker for other factors (for example, less intense treatment being offered to obese women, effects of comorbidity). It has also not been demonstrated that modifying obesity (through weight loss or pharmacologic intervention targeting obesity-related physiology) will lead to improved breast cancer outcomes. Several small studies have demonstrated that lifestyle interventions, delivered in person or by telephone, can lead to modest weight loss in breast cancer survivors. Our group recently reported a randomized trial of a telephone-based weight loss intervention [23]. Three hundred and thirty-eight women were
randomized to a telephone-based weight loss intervention vs. an information-based control arm e significantly greater weight loss was seen in the intervention arm with a difference of 5.3% at six months. Unfortunately, the sample size was not sufficiently large to allow an examination of intervention effects on breast cancer events. Two randomized trials have tested dietary interventions in the breast cancer adjuvant setting. In the WINS Study [24], 2437 women were randomized to reduced fat diet vs. control arm during the first year after diagnosis e an approximately 3.2% relative weight loss was observed in the intervention group. With 5 years of median follow-up, relapse-free survival was better in the intervention arm (HR 0.76, 95% CI 0.60e0.98); the benefit was greatest in ER negative cases (HR 0.58, 95% CI 0.37e0.91) and in those with baseline BMI >30 kg/m2 (HR 0.66, 95% CI 0.41e1.0). In the second study (WHEL) [25], Pierce et al. randomized 3088 women within the first four years post-diagnosis to a complex dietary intervention that included reduction in dietary fat, increases in fruits and vegetables, and grain intake. The intervention was associated with a modest weight gain. There was no overall effect on DFS (HR 0.96, 95% CI 0.80e1.14). Although these two studies did not directly test the effect of weight loss on breast cancer outcomes, the divergent results are potentially consistent with the weight loss seen in the WINS Study contributing to improved outcomes. A number of observational studies have provided evidence that pre- or post-diagnosis physical activity is associated with better breast cancer outcomes, regardless of menopausal status or hormone receptor status. As for obesity, no randomized data are available. The McTiernan group investigated effects of caloric restriction, physical activity, or both on potential physiologic mediators of the obesity-cancer link in a randomized trial involving 439 healthy postmenopausal women [26e30]. Diet alone or diet combined with physical activity led to weight loss of 10.8 or 11.9% respectively, while physical activity alone led to 3.3% weight loss. Diet alone or combined with physical activity led to reductions of 20% or more in insulin, Homeostasis Model Assessment (HOMA, an estimate of insulin resistance), free estradiol, highly sensitive C-reactive protein (a marker of inflammation), leptin and IL-6; changes with physical activity alone were considerably lower (<9%, apart from leptin which decreased by 12.7%) [26e30]. If these factors mediate the breast cancer-obesity link, the magnitude of change seen with the diet or diet plus physical activity intervention (but not physical activity alone) is potentially sufficient to impact breast cancer outcomes.
Please cite this article in press as: Goodwin PJ, Obesity, insulin resistance and breast cancer outcomes, The Breast (2015), http://dx.doi.org/ 10.1016/j.breast.2015.07.014
e4
P.J. Goodwin / The Breast xxx (2015) e1ee4
Because it is not clear whether the association of obesity with breast cancer outcome is causal, and because of the possibility of bias and confounding contributing to the results of observational studies (including retrospective analyses nested in randomized trials), randomized interventions trials are needed to provide evidence that will guide clinical practice. Our group is conducting a randomized trial of metformin vs. placebo in 3649 high risk early stage breast cancer patients who have received standard treatment. Accrual was completed in early 2013. We have demonstrated that metformin is associated with weight loss (3% relative at six months), as well as significant improvements in insulin, insulin resistance (HOMA), glucose, leptin and inflammation (hsCRP), with no evidence that the improvements seen in the metformin (vs. placebo) arm differed according to baseline BMI or baseline insulin [31]. Randomized trials of a lifestyle-based intervention, targeting the maximum weight loss that can be achieved with lifestyle interventions (at least 5e7%), are needed to formally test whether the extent of weight loss is feasible will impact breast cancer outcomes. Such trials should include strong embedded correlative components and, ideally, would focus on individuals who are overweight or obese, potentially enriching for those who are metabolically unhealthy. Summary The association of obesity with breast cancer outcome continues to be a major clinical concern. The four areas addressed in this manuscript highlight areas of important research focus and controversy. There are sufficient data to generate the hypothesis that obesity is associated with breast cancer outcome. Emerging evidence raises potential concerns regarding the reduced efficacy of adjuvant anastrozole, in obese breast cancer patients; letrozole should be considered in such women given evidence that its efficacy does not vary by BMI. Data from randomized trials are needed to confirm whether pharmacologic or lifestyle interventions that modify obesity and obesity-associated physiology can lead to improved breast cancer outcomes. At present, it is not clear whether the BMI prognostic association is causal and, if causal, whether such interventions will improve outcomes e it remains possible that interventions will not change physiology to the extent needed to improve outcomes or that the biologic effects of obesity are fixed during carcinogenesis and cannot be reversed subsequently. Conflict of interest statement None declared. References [1] Goodwin PJ. Obesity and endocrine therapy: host factors and breast cancer outcome. Breast 2013;22(Suppl. 2):S44e7. [2] 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 2014;25:1901e14. [3] 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 2010;123:627e35. [4] Niraula S, Ocana A, Ennis M, Goodwin PJ. Body size and breast cancer prognosis in relation to hormone receptor and menopausal status: a meta-analysis. Breast Cancer Res Treat 2012;134:769e81. [5] Sparano JA, Wang M, Zhao F, et al. Obesity at diagnosis is associated with inferior outcomes in hormone receptor-positive operable breast cancer. Cancer 2012;118:5937e46. [6] Pajares B, Pollan M, Martin 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 2013;15:R105.
[7] Pan H, Gray RR, on behalf of the Early Breast Cancer Trialists' Collaborative Group. Effect of obesity in premenopausal ERþ early breast cancer: EBCTCG data on 80,000 patients in 70 trials. J Clin Oncol 2014;32(15 Suppl.):503. [8] Kramer CK, Zinman B, Retnakaran R. Are metabolically healthy overweight and obesity benign conditions? A systematic review and meta-analysis. Ann Intern Med 2013;159:758e69. [9] 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 2012;30:3967e75. [10] Baum M, Budzar AU, Cuzick J, et al. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial. Lancet 2002;359:2131e9. [11] Gnant M, Mlineritsch B, Stoeger H, et al. Adjuvant endocrine therapy plus zoledronic acid in premenopausal women with early-stage breast cancer: 62month follow-up from the ABCSG-12 randomised trial. Lancet Oncol 2011;12: 631e41. [12] Seynaeve C, Hille E, Hasenburg A, et al. The impact of body mass index on the efficacy of adjuvant endocrine therapy in postmenopausal hormone sensitive breast cancer patients: exploratory analysis from the TEAM study. Abstr nr Cancer Res 2010;70(24 Suppl.):S2e3. [13] Gnant M, Pfeiler G, Stoger 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 2013;109:589e96. [14] 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 2012;30:2977e80. [15] Pfeiler G, Konigsberg 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 2013;109:1522e7. [16] Lonning PE, Haynes BP, Dowsett M. Relationship of body mass index with aromatisation and plasma and tissue oestrogen levels in postmenopausal breast cancer patients treated with aromatase inhibitors. Eur J Cancer 2014;50:1055e64. [17] Kyvernitakis I, Knoll 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 2014;140:159e66. [18] Elliott KM, Dent J, Stanczyk FZ, et al. Effects of aromatase inhibitors and body mass index on steroid hormone levels in women with early and advanced breast cancer. Br J Surg 2014;101:939e48. [19] Goodwin PJ, Stambolic V. Impact of the obesity epidemic on cancer. Annu Rev Med 2015;66:281e96. [20] Iyengar NM, Hudis CA, Dannenberg AJ. Obesity and cancer: local and systemic mechanisms. Annu Rev Med 2015;66:297e309. [21] Khandekar MJ, Cohen P, Spiegelman BM. Molecular mechanisms of cancer development in obesity. Nat Rev Cancer 2011;11:886e95. [22] 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 2012;30:164e71. [23] 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 2014;32:2231e9. [24] 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 2006;98:1767e76. [25] 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 2007;298:289e98. [26] Campbell KL, Foster-Schubert KE, Alfano CM, et al. Reduced-calorie dietary weight loss, exercise, and sex hormones in postmenopausal women: randomized controlled trial. J Clin Oncol 2012;30:2314e26. [27] Mason C, Xiao L, Duggan C, et al. Effects of dietary weight loss and exercise on insulin-like growth factor-I and insulin-like growth factor-binding protein-3 in postmenopausal women: a randomized controlled trial. Cancer Epidemiol Biomarkers Prev 2013;22:1457e63. [28] Imayama I, Ulrich CM, Alfano CM, et al. Effects of a caloric restriction weight loss diet and exercise on inflammatory biomarkers in overweight/obese postmenopausal women: a randomized controlled trial. Cancer Res 2012;72: 2314e26. [29] Abbenhardt C, McTiernan A, Alfano CM, et al. Effects of individual and combined dietary weight loss and exercise interventions in postmenopausal women on adiponectin and leptin levels. J Intern Med 2013;274:163e75. [30] Mason C, Foster-Schubert KE, Imayama I, et al. Dietary weight loss and exercise effects on insulin resistance in postmenopausal women. Am J Prev Med 2011;41:366e75. [31] 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 2015;107(3). pii: djv006.
Please cite this article in press as: Goodwin PJ, Obesity, insulin resistance and breast cancer outcomes, The Breast (2015), http://dx.doi.org/ 10.1016/j.breast.2015.07.014
Contents lists available at ScienceDirect
The Breast journal homepage: www.elsevier.com/brst
Obesity, insulin resistance and breast cancer outcomes Pamela J. Goodwin a, b, * a
Department of Medicine, Division of Medical Oncology and Hematology, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada b Division of Clinical Epidemiology, Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada
a r t i c l e i n f o
a b s t r a c t
Article history: Available online xxx
There is growing evidence that obesity is associated with poor outcomes in early stage breast cancer. This paper addresses four current areas of focus: 1. Is obesity associated with poor outcomes in all biologic subtypes of breast cancer? 2. Does obesity effect AI efficacy or estrogen suppression in the adjuvant setting? 3. What are the potential biologic underpinnings of the obesity-breast cancer association? 4. Are intervention studies warranted? If so, which interventions in which populations? Research is needed to resolve these questions; intervention trials involving lifestyle interventions or targeting the biology postulated to link obesity and cancer are recommended. © 2015 Published by Elsevier Ltd.
Keywords: Breast cancer Adjuvant Obesity Aromatase inhibitors Biology Intervention
Introduction There is growing recognition that obesity is associated with adverse breast cancer outcomes, notably a higher risk of distant recurrence and death. Research since St. Gallen 2013 [1] has focused on four key areas e (1) Is obesity associated with poor outcomes in all biologic subtypes of breast cancer?, (2) Does obesity impact aromatase inhibitor (AI) efficacy or estrogen suppression in the adjuvant setting?, (3) What are the potential biologic underpinnings of the obesity-breast cancer outcome association?, and (4) Are intervention studies warranted? If so, which interventions in which populations? These four areas of research will be discussed below. Is obesity associated with poor outcomes in all biologic subtypes of breast cancer? Body mass index (BMI) at diagnosis has been associated with both breast cancer mortality and overall mortality in research spanning five decades. In a recent meta-analysis [2], a curvilinear association of BMI with outcome was seen e increased risk was present in individuals with BMI under 20 kg/m2 and BMI over 25 kg/m2 at diagnosed e in the latter group, the risk increased with increasing BMI. In a prior meta-analysis [3], obesity was associated
* Mount Sinai Hospital, 1284-600 University Avenue, Toronto, Ontario M5G 1X5, Canada. Tel.: þ1 416 586 8605; fax: þ1 416 586 8659. E-mail address: [email protected].
with adverse outcome (overall or breast cancer specific mortality) in both pre- and postmenopausal women, in those diagnosed both prior to and after 1995, and in observational cohorts as well as “treatment cohorts” (i.e. retrospective analyses nested in randomized trials) e Hazard ratios (HRs) for obese vs. non-obese overall and in these subgroups were in the range of 1.2e1.35. Our group [4] performed a meta-analysis of obesity associations in women with estrogen receptor (ER) positive vs. negative breast cancer e hazard ratios (HRs) for breast cancer specific survival (BCSS) and overall survival (OS) were similarly elevated in both groups (BCSS HRs 1.46, 1.36 respectively). However, some recent studies (e.g. Sparano et al.) [5] have found no evidence of an association of obesity with outcome in hormone receptor negative breast cancer. At least ten subsequent studies have explored this issue with varying results. For example, Pajares et al. [6] reported worse BCSS and OS in women having a BMI over 35 kg/m2 with hormone receptor negative or HER2þ breast cancer enrolled onto a series of randomized trials, all of which involved anthracyclines. Importantly, Pan et al. [7] analyzed associations of BMI with outcome in over 80,000 women (20,000 ER positive premenopausal, 40,000 ER positive postmenopausal, 20,000 ER negative) enrolled onto randomized trials included in the Early Breast Cancer Clinical Trialists Cooperative Group (EBCTCG). In their patient-level meta-analyses, they found an increased risk of breast cancer mortality in heavier premenopausal women with ER positive breast cancer e HR per 10 kg/m2 in BMI 1.23 [95% confidence intervals (CIs) 1.16e1.31, p < 0.00001], regardless of whether the trial tested hormone, chemotherapy or other treatments. In contrast to earlier
http://dx.doi.org/10.1016/j.breast.2015.07.014 0960-9776/© 2015 Published by Elsevier Ltd.
Please cite this article in press as: Goodwin PJ, Obesity, insulin resistance and breast cancer outcomes, The Breast (2015), http://dx.doi.org/ 10.1016/j.breast.2015.07.014
e2
P.J. Goodwin / The Breast xxx (2015) e1ee4
study-level meta-analyses, no significant association was seen in postmenopausal women with ER positive breast cancer (HR 1.03, p ¼ 0.27) or in pre- or postmenopausal women with ER negative breast cancer (HR 1.02, p ¼ 0.39). The basis for these recent discrepancies is not well understood. Because most of the discrepant findings arose from randomized clinical trials (RCTs), one possibility is that women enrolled onto these trials differ from the general population of breast cancer patients with respect to factors, such as metabolic health, that are relevant to the obesity-cancer link. This issue has not been examined in breast cancer patients, however, in the Physicians' Health Study, those who responded to an invitation to be assessed for participation, those willing to be assessed for eligibility, those eligible for randomization, and those who actually accepted randomization had increasingly lower total mortality, cardiovascular mortality and cancer mortality than those who did not meet these criteria. This is evidence of a selection bias (towards enrolment of healthier individuals) occurring during the recruitment process. Furthermore, Kramer et al. [8] have provided evidence that obesity does not necessarily predict metabolic health (defined as no components of the insulin resistance syndrome) e those who are metabolically unhealthy have increased all-cause mortality and cardiovascular events, independent of BMI category (normal weight, overweight or obesity). Thus, it is possible that randomization onto breast cancer clinical trials selects for metabolically healthy individuals, regardless of baseline BMI, either as a direct result of inclusion/exclusion criteria (e.g. a requirement for normal cardiac function in trials of anthracyclines or trastuzumab) or through a more subtle selection of healthy individuals as was seen in the Physicians' Health Study. If so, prognostic associations of BMI may have been obscured. Does obesity effect AI efficacy or estrogen suppression in the adjuvant setting? Total body aromatization increases with increasing BMI; leptin and inflammation (both associated with obesity) may also increase aromatase activity independent of BMI. At St. Gallen 2013, the potential effect of baseline BMI on the relative efficacy of AIs vs. tamoxifen in a series of adjuvant trials was discussed [1]. To summarize, in BIG 1-98 [9] (which compared letrozole to tamoxifen) there was no evidence of a significant treatment by BMI interaction. However, in the Anastrozole, Tamoxifen Alone or in Combination (ATAC) [10] and Australian Breast Cancer Study Group (ABCSG) e12 [11] trials (which compared anastrozole to tamoxifen, in postmenopausal women in the former and in combination with goserelin in premenopausal women in the latter), there was evidence that the relative benefit of anastrozole vs. tamoxifen was greatest when BMI was normal vs. elevated. In the ATAC trial [10], this effect of BMI was not statistically significant. In ABSCG-12 [11], disease-free survival (DFS) and OS were worse in women with BMI over 25 kg/m2 who received anastrozole as opposed to tamoxifen (HR 1.49, 95% CI 0.93e2.0 and HR 3.03, 95%CI 1.35e6.82 respectively). In the TEAM Trial [12] there was no evidence that the relative benefit of exemestane vs. tamoxifen differed across BMI levels. New since St. Gallen 2013 is a report by Gnant et al. [13] which examined the use of extended adjuvant anastrozole vs. no treatment in postmenopausal women e a benefit for anastrozole was seen only in women with baseline BMI <25 kg/m2 [DFS HR 0.48 (95% CI 0.26e0.89), distant DFS HR 0.22 (95% CI 0.05e1.0), and OS HR 0.45 (95% CI 0.19e1.04)]. No benefit was seen in individuals with a BMI over 25 kg/m2; the BMI by treatment benefit interaction approached significance for distant DFS (p ¼ 0.07). Recently, a series of studies [14e18] has examined the extent to which AIs suppress circulating levels of estradiol or estrone
according to BMI levels (see Table 1). Folkerd et al. [14] reported that levels of estrone and estradiol were greater at higher levels of BMI in 44 women receiving anastrozole or letrozole e this was statistically significant in those receiving letrozole (r ¼ 0.35, p ¼ 0.013 for estradiol and r ¼ 0.30, p ¼ 0.035 for estrone sulfate) e a non-significant trend was seen in those receiving anastrozole. Of note, levels of both estradiol and estrone sulfate were lower in those receiving letrozole vs. anastrozole across all BMI categories. Pfeiler et al. [15] found a similar pattern of higher estradiol levels in 28 obese (vs. 40. non-obese) women receiving AIs (60 anastrozole, 8 letrozole), r ¼ 0.35, p ¼ 0.05; importantly, FSH levels were lower in obese women (r ¼ 0.34, p ¼ 0.06), consistent with higher endogenous estrogen levels. Kyvernitakis I et al. [17] studied 70 postmenopausal women receiving adjuvant anastrozole; after 12e24 months of treatment, overweight and obese women has non-significantly higher estradiol concentrations than normal weight women. Elliott et al. [18] studied a mixed group of 64 adjuvant and second line patients (who had progressed on an AI) receiving one of the three approved AIs; BMI and estradiol were higher in metastatic patients. BMI was non-significantly correlated with estradiol (9.3% higher for normal vs. overweight vs. obese, p ¼ 0.06). Finally, Lonning et al. [16] reported that treatment levels of estrone sulfate were significantly correlated with BMI in 64 women receiving letrozole or anastrozole in the adjuvant or metastatic setting (n ¼ 25, r ¼ 0.60, p ¼ 0.001 and n ¼ 12, r ¼ 0.61, p ¼ 0.035 respectively). Similar to results of Folkerd et al., [15] estrone sulfate levels were higher in those receiving anastrozole (vs. letrozole), independent of BMI. BMI was not correlated with on-treatment in vivo aromatization in the metastatic setting in women receiving a range of first, second or third line AIs, nor was it correlated with intratumoral levels of estrogens in the neoadjuvant setting. Although many of the reported BMI associations were not statistically significant, sample sizes were small and most studies reported higher estrogen levels in heavier women. Further research employing larger sample sizes is urgently needed. Although these data suggest estrogen suppression may be less effective in women with higher BMI, there is little evidence that this impacted clinical outcomes in obese women receiving letrozole relative to tamoxifen (BIG 1e98) [9]. There is growing evidence that higher BMI may contribute to reduced efficacy of anastrazole vs. tamoxifen. Because of this, selection of other AIs, notably letrozole, is preferred in overweight or obese women. What are the potential biologic underpinnings of the obesitybreast cancer association? There is an evolving understanding of the complex biology potentially underlying the association between obesity and cancer in general, and breast cancer outcomes in particular. A number of recent review articles have addressed this issue in detail [19e21]. Obesity leads to an expanded and reprogrammed, metabolically active, adipose tissue mass, with increased numbers of preadipocytes and inflammatory cells, higher levels of leptin and free fatty acids, and greater release of cytokines and other inflammatory compounds. These changes result in an altered systemic physiology, leading to insulin resistance (higher insulin levels, dysglycemia, other metabolic changes), higher circulating levels of free fatty acids, lipids, leptin, estrogens and inflammatory markers. This altered systemic physiology can have direct effects on cancer cells, with higher levels of circulating estrogen, insulin and inflammatory factors activating estrogen and insulin/IGF signaling pathways (e.g. PI3K, ras) as well as JAK-STAT, NF-kappa-b and other pathways. The associated dysglycemia may be associated with altered tumor cell metabolism, such as a shift from oxidative phosphorylation to aerobic glycolysis (the Warburg effect). Locally, adipose cells in the
Please cite this article in press as: Goodwin PJ, Obesity, insulin resistance and breast cancer outcomes, The Breast (2015), http://dx.doi.org/ 10.1016/j.breast.2015.07.014
P.J. Goodwin / The Breast xxx (2015) e1ee4
e3
Table 1 Estradiol levels According to BMI in adjuvant aromatase inhibitor patients. Duration
Pre-treatment Folkerd [14] Pfeiler [15] Kyvernitakis [17] Post letrozle Folkerd [14] Post anastrozle Folkerd [14] Kyvernitakis [17] Post any AI Pfeiler [15] Elliot [18]
#
Normal
Overweight
Obese
<25 kg/m2
25e30 kg/m2
>30 kg/m2
Sig.
26
YES NO YES
e e e
44 68 70
20 18.9b 9.7
13.6
39/55a 20.2 15.1
3 MOS
44
1.4
1.4
2.0/2.4a
YES
3 MOS 12 MOS
44 70
2.8 5.0
2.5 8.4
3.2/4.2a 5.4
NO NO
3 MOS 3 MOS
68 45
9.0b 3.0
3.5
12.5 3.3
NO NO
Abbreviations: BMI, body mass index; sig., significant at p < 0.05. a BMI 30e35/>35 kg/m2. b BMI <30 kg/m2.
tumor microenvironment can be associated with increased numbers of inflammatory cells, altered levels of cytokines, adipocines and inflammatory markers, which may promote carcinogenesis, tumor progression, invasion and metastasis. In clinical studies, many of the physiologic factors associated with obesity (hyperinsulinemia, dysglycemia, higher circulating estrogen levels) have been associated with poor breast cancer outcomes. It is likely that the relative importance of different mediators varies across breast cancer subtypes and over time. Higher estrogen levels likely contribute primarily to poor outcome in hormone receptor positive breast cancer. When we examined timerelated associations of obesity and related factors with breast cancer outcome [22] we found that insulin and glucose were associated with outcomes only during the first five years after diagnosis, whereas BMI and the associated adipokine leptin had more constant associations over a longer timeframe. Although there have been important advances in understanding the potential biologic basis for an obesity-cancer link, it remains to be demonstrated whether this link is causal, or whether changes achievable through weight loss or pharmacologic interventions that target this link will lead to improved outcomes. Are intervention studies warranted? If so, which interventions in which populations? In the 50 years since an association of obesity with breast cancer outcomes was first recognized, dozens of observational studies have examined this association. As noted above, meta-analyses synthesizing these data have suggested the risk of breast cancer death is approximately one-third higher in overweight/obese women, although there has been recent controversy regarding a potential differential association in women with hormone receptor positive vs. negative cancers. Despite this considerable body of work, it has not been established whether the association of obesity with breast cancer outcome is causal, or whether higher BMI is a marker for other factors (for example, less intense treatment being offered to obese women, effects of comorbidity). It has also not been demonstrated that modifying obesity (through weight loss or pharmacologic intervention targeting obesity-related physiology) will lead to improved breast cancer outcomes. Several small studies have demonstrated that lifestyle interventions, delivered in person or by telephone, can lead to modest weight loss in breast cancer survivors. Our group recently reported a randomized trial of a telephone-based weight loss intervention [23]. Three hundred and thirty-eight women were
randomized to a telephone-based weight loss intervention vs. an information-based control arm e significantly greater weight loss was seen in the intervention arm with a difference of 5.3% at six months. Unfortunately, the sample size was not sufficiently large to allow an examination of intervention effects on breast cancer events. Two randomized trials have tested dietary interventions in the breast cancer adjuvant setting. In the WINS Study [24], 2437 women were randomized to reduced fat diet vs. control arm during the first year after diagnosis e an approximately 3.2% relative weight loss was observed in the intervention group. With 5 years of median follow-up, relapse-free survival was better in the intervention arm (HR 0.76, 95% CI 0.60e0.98); the benefit was greatest in ER negative cases (HR 0.58, 95% CI 0.37e0.91) and in those with baseline BMI >30 kg/m2 (HR 0.66, 95% CI 0.41e1.0). In the second study (WHEL) [25], Pierce et al. randomized 3088 women within the first four years post-diagnosis to a complex dietary intervention that included reduction in dietary fat, increases in fruits and vegetables, and grain intake. The intervention was associated with a modest weight gain. There was no overall effect on DFS (HR 0.96, 95% CI 0.80e1.14). Although these two studies did not directly test the effect of weight loss on breast cancer outcomes, the divergent results are potentially consistent with the weight loss seen in the WINS Study contributing to improved outcomes. A number of observational studies have provided evidence that pre- or post-diagnosis physical activity is associated with better breast cancer outcomes, regardless of menopausal status or hormone receptor status. As for obesity, no randomized data are available. The McTiernan group investigated effects of caloric restriction, physical activity, or both on potential physiologic mediators of the obesity-cancer link in a randomized trial involving 439 healthy postmenopausal women [26e30]. Diet alone or diet combined with physical activity led to weight loss of 10.8 or 11.9% respectively, while physical activity alone led to 3.3% weight loss. Diet alone or combined with physical activity led to reductions of 20% or more in insulin, Homeostasis Model Assessment (HOMA, an estimate of insulin resistance), free estradiol, highly sensitive C-reactive protein (a marker of inflammation), leptin and IL-6; changes with physical activity alone were considerably lower (<9%, apart from leptin which decreased by 12.7%) [26e30]. If these factors mediate the breast cancer-obesity link, the magnitude of change seen with the diet or diet plus physical activity intervention (but not physical activity alone) is potentially sufficient to impact breast cancer outcomes.
Please cite this article in press as: Goodwin PJ, Obesity, insulin resistance and breast cancer outcomes, The Breast (2015), http://dx.doi.org/ 10.1016/j.breast.2015.07.014
e4
P.J. Goodwin / The Breast xxx (2015) e1ee4
Because it is not clear whether the association of obesity with breast cancer outcome is causal, and because of the possibility of bias and confounding contributing to the results of observational studies (including retrospective analyses nested in randomized trials), randomized interventions trials are needed to provide evidence that will guide clinical practice. Our group is conducting a randomized trial of metformin vs. placebo in 3649 high risk early stage breast cancer patients who have received standard treatment. Accrual was completed in early 2013. We have demonstrated that metformin is associated with weight loss (3% relative at six months), as well as significant improvements in insulin, insulin resistance (HOMA), glucose, leptin and inflammation (hsCRP), with no evidence that the improvements seen in the metformin (vs. placebo) arm differed according to baseline BMI or baseline insulin [31]. Randomized trials of a lifestyle-based intervention, targeting the maximum weight loss that can be achieved with lifestyle interventions (at least 5e7%), are needed to formally test whether the extent of weight loss is feasible will impact breast cancer outcomes. Such trials should include strong embedded correlative components and, ideally, would focus on individuals who are overweight or obese, potentially enriching for those who are metabolically unhealthy. Summary The association of obesity with breast cancer outcome continues to be a major clinical concern. The four areas addressed in this manuscript highlight areas of important research focus and controversy. There are sufficient data to generate the hypothesis that obesity is associated with breast cancer outcome. Emerging evidence raises potential concerns regarding the reduced efficacy of adjuvant anastrozole, in obese breast cancer patients; letrozole should be considered in such women given evidence that its efficacy does not vary by BMI. Data from randomized trials are needed to confirm whether pharmacologic or lifestyle interventions that modify obesity and obesity-associated physiology can lead to improved breast cancer outcomes. At present, it is not clear whether the BMI prognostic association is causal and, if causal, whether such interventions will improve outcomes e it remains possible that interventions will not change physiology to the extent needed to improve outcomes or that the biologic effects of obesity are fixed during carcinogenesis and cannot be reversed subsequently. Conflict of interest statement None declared. References [1] Goodwin PJ. Obesity and endocrine therapy: host factors and breast cancer outcome. Breast 2013;22(Suppl. 2):S44e7. [2] 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 2014;25:1901e14. [3] 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 2010;123:627e35. [4] Niraula S, Ocana A, Ennis M, Goodwin PJ. Body size and breast cancer prognosis in relation to hormone receptor and menopausal status: a meta-analysis. Breast Cancer Res Treat 2012;134:769e81. [5] Sparano JA, Wang M, Zhao F, et al. Obesity at diagnosis is associated with inferior outcomes in hormone receptor-positive operable breast cancer. Cancer 2012;118:5937e46. [6] Pajares B, Pollan M, Martin 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 2013;15:R105.
[7] Pan H, Gray RR, on behalf of the Early Breast Cancer Trialists' Collaborative Group. Effect of obesity in premenopausal ERþ early breast cancer: EBCTCG data on 80,000 patients in 70 trials. J Clin Oncol 2014;32(15 Suppl.):503. [8] Kramer CK, Zinman B, Retnakaran R. Are metabolically healthy overweight and obesity benign conditions? A systematic review and meta-analysis. Ann Intern Med 2013;159:758e69. [9] 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 2012;30:3967e75. [10] Baum M, Budzar AU, Cuzick J, et al. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial. Lancet 2002;359:2131e9. [11] Gnant M, Mlineritsch B, Stoeger H, et al. Adjuvant endocrine therapy plus zoledronic acid in premenopausal women with early-stage breast cancer: 62month follow-up from the ABCSG-12 randomised trial. Lancet Oncol 2011;12: 631e41. [12] Seynaeve C, Hille E, Hasenburg A, et al. The impact of body mass index on the efficacy of adjuvant endocrine therapy in postmenopausal hormone sensitive breast cancer patients: exploratory analysis from the TEAM study. Abstr nr Cancer Res 2010;70(24 Suppl.):S2e3. [13] Gnant M, Pfeiler G, Stoger 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 2013;109:589e96. [14] 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 2012;30:2977e80. [15] Pfeiler G, Konigsberg 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 2013;109:1522e7. [16] Lonning PE, Haynes BP, Dowsett M. Relationship of body mass index with aromatisation and plasma and tissue oestrogen levels in postmenopausal breast cancer patients treated with aromatase inhibitors. Eur J Cancer 2014;50:1055e64. [17] Kyvernitakis I, Knoll 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 2014;140:159e66. [18] Elliott KM, Dent J, Stanczyk FZ, et al. Effects of aromatase inhibitors and body mass index on steroid hormone levels in women with early and advanced breast cancer. Br J Surg 2014;101:939e48. [19] Goodwin PJ, Stambolic V. Impact of the obesity epidemic on cancer. Annu Rev Med 2015;66:281e96. [20] Iyengar NM, Hudis CA, Dannenberg AJ. Obesity and cancer: local and systemic mechanisms. Annu Rev Med 2015;66:297e309. [21] Khandekar MJ, Cohen P, Spiegelman BM. Molecular mechanisms of cancer development in obesity. Nat Rev Cancer 2011;11:886e95. [22] 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 2012;30:164e71. [23] 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 2014;32:2231e9. [24] 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 2006;98:1767e76. [25] 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 2007;298:289e98. [26] Campbell KL, Foster-Schubert KE, Alfano CM, et al. Reduced-calorie dietary weight loss, exercise, and sex hormones in postmenopausal women: randomized controlled trial. J Clin Oncol 2012;30:2314e26. [27] Mason C, Xiao L, Duggan C, et al. Effects of dietary weight loss and exercise on insulin-like growth factor-I and insulin-like growth factor-binding protein-3 in postmenopausal women: a randomized controlled trial. Cancer Epidemiol Biomarkers Prev 2013;22:1457e63. [28] Imayama I, Ulrich CM, Alfano CM, et al. Effects of a caloric restriction weight loss diet and exercise on inflammatory biomarkers in overweight/obese postmenopausal women: a randomized controlled trial. Cancer Res 2012;72: 2314e26. [29] Abbenhardt C, McTiernan A, Alfano CM, et al. Effects of individual and combined dietary weight loss and exercise interventions in postmenopausal women on adiponectin and leptin levels. J Intern Med 2013;274:163e75. [30] Mason C, Foster-Schubert KE, Imayama I, et al. Dietary weight loss and exercise effects on insulin resistance in postmenopausal women. Am J Prev Med 2011;41:366e75. [31] 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 2015;107(3). pii: djv006.
Please cite this article in press as: Goodwin PJ, Obesity, insulin resistance and breast cancer outcomes, The Breast (2015), http://dx.doi.org/ 10.1016/j.breast.2015.07.014