- © 2012 by American Society of Clinical Oncology
NEW and RENEW: Building the Case for Weight Loss in Breast Cancer
- Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA
- Samuel Lunenfeld Research Institute, Mount Sinai Hospital; and Princess Margaret Hospital, University of Toronto, Toronto, Ontario, Canada
- Corresponding author: Jennifer A. Ligibel, Dana-Farber Cancer Institute, 450 Brookline Ave, Yawkey 1234, Boston, MA 02215; e-mail: jligibel{at}partners.org
Obesity has been linked to poor outcomes in women with early breast cancer.1–3 A meta-analysis of 43 studies examining the relationship between weight at the time of breast cancer diagnosis and prognosis in women with early-stage breast cancer demonstrated a 33% increase in the risk of breast cancer–related and overall mortality in obese versus nonobese women.1 Observational studies have also shown that physical activity, an important mediator of weight maintenance, is associated with better survival in women with early-stage breast cancer.4–7 Finally, two large-scale randomized trials testing the impact of dietary modification on cancer outcomes in breast cancer survivors have provided indirect evidence that weight loss after diagnosis could lead to lower rates of recurrence.8,9 The WINS (Women's Interventional Nutrition Study) trial reported a 24% reduction in the risk of recurrence in breast cancer survivors randomly assigned to a low-fat diet group, who decreased dietary fat to 20% of calories and lost an average of 6 lbs, as compared with controls.8 In contrast, the WHEL (Women's Healthy Eating and Living) study did not demonstrate a lower risk of recurrence in breast cancer survivors randomly assigned to a low-fat, high fruit and vegetable dietary intervention.9 Intervention participants in this study did not experience weight loss, and although there were a number of other differences between the studies, the weight change in WINS has been suggested as one potential reason for the difference in the outcomes of the two studies.
Despite strong observational evidence that weight and related factors could influence cancer outcomes, there are no data from randomized trials testing the impact of purposeful weight loss after breast cancer diagnosis on the risk of cancer recurrence and mortality. A number of small and moderate-sized weight loss10–13 and physical activity intervention14,15 studies in breast cancer populations have demonstrated the feasibility of lifestyle change and documented additional beneficial effects on quality of life, fitness, and fatigue. To optimally design more definitive studies, interventions that reliably, cost effectively, and durably change lifestyle behaviors in large groups of breast cancer survivors are needed. Enhanced understanding of the impact of lifestyle change on biologic factors implicated in the lifestyle-cancer association will guide study design and strengthen the scientific rationale for such studies.
Two reports in Journal of Clinical Oncology (JCO) provide crucial information for the development of large-scale trials testing the impact of lifestyle change on disease outcomes in patients with cancer. In the first report, Demark-Wahnefried et al16 demonstrate the efficacy of a telephone-based lifestyle intervention in effecting long-lasting changes in diet, physical activity, and weight in 641 survivors of colon, prostate, and breast cancers, providing feasibility data for large-scale, distance-based lifestyle interventions among cancer survivors. In the second report, Campbell et al17 describe the impact of different types of lifestyle interventions on biomarkers linked to breast cancer risk and prognosis in postmenopausal women, providing some of the first data comparing the effect of physical activity and weight loss, with or without a physical activity component, on putative pathways linking lifestyle and breast cancer.
The RENEW (Reach Out to Enhance Wellness) study enrolled 641 survivors of breast, prostate, and colon cancers.18 Key eligibility criteria included age ≥ 65, body mass index > 25 kg/m2, sedentary activity pattern, and cancer diagnosis at least 5 years before enrollment. The study employed a cross-over design; participants were randomly assigned to immediate participation in a 1-year telephone- and print materials–based lifestyle intervention, designed to increase physical activity and improve dietary quality, or to delayed participation after a 1-year control period. The primary end point was change in physical functioning. A previous report demonstrated that intervention-group participants experienced significantly less decline in their functional status, improvements in dietary quality, increased physical activity, and modest weight loss, as compared with control participants.18
The updated report16 from the RENEW study shows that favorable changes in physical activity, dietary behaviors, and weight were maintained in the immediate intervention group 1 year after the discontinuation of the lifestyle intervention. Additionally, the delayed intervention group experienced significant changes in physical functioning, weight, dietary quality, and physical activity during their participation in the lifestyle intervention. These data provide key information on the durability of telephone-based lifestyle coaching. However, there are important caveats, including the small proportion of eligible individuals who opted to participate in the study (approximately 6%), the 25% attrition rate at 2 years, and the fact that all study measures, including weight, were self-reported. Despite these caveats, the efficacy of the RENEW intervention is consistent with that of other large-scale trials of distanced-based lifestyle interventions, including the WHEL study,9 which used a telephone-based intervention in 3,088 women with early-stage breast cancer, and LISA (Lifestyle Intervention Study for Adjuvant Treatment of Early Breast Cancer),19 which used a telephone-based intervention to reduce weight in 338 overweight and obese breast cancer survivors. Together, these trials demonstrate that durable lifestyle change can successfully, and cost effectively, be implemented through a centralized, distance-based approach.
The NEW (Nutrition and Exercise for Women) trial,17 also reported in JCO, evaluated the impact of different lifestyle interventions on biomarkers that have been linked to breast cancer risk and prognosis. The study enrolled 439 postmenopausal, sedentary, overweight women and randomly assigned them to one of four groups: dietary weight loss, exercise alone, dietary weight loss plus exercise, or control. The primary outcome was change in levels of sex steroid hormones, which have been linked to breast cancer risk. Metabolic hormones linked to breast cancer prognosis have previously been reported.20 Women randomly assigned to the weight loss arms experienced significant favorable changes in estrogen, testosterone, sex hormone–binding globulin, insulin, c-peptide, leptin, and adiponectin levels compared with controls. Patients randomly assigned to exercise alone experienced smaller but significant changes in estrogen, testosterone, and leptin levels but not other hormones. Additionally, biomarker change was significantly smaller in the exercise alone group versus the dietary weight loss plus exercise group. However, the degree of weight loss was related to the magnitude of biomarker change in the dietary weight loss alone group, but not in the diet plus exercise group, suggesting that modest weight loss was able to effect the same change in biomarkers as more significant weight loss, if the modest weight loss was combined with exercise.
Given observations that higher levels of estrogen and insulin are associated with poor breast cancer outcomes,21–26 the observed reductions in sex steroids and insulin with weight loss would be consistent with beneficial effects of weight loss on breast cancer outcomes. Although the NEW participants were postmenopausal women at risk of breast cancer and not cancer survivors, the demographics of this population, including their age, weight distribution, and baseline physical activity levels were consistent with postmenopausal survivor populations. Thus, the NEW results17 strengthen the biologic rationale for a lifestyle intervention focusing on weight loss (rather than exercise alone) as a potential therapeutic intervention in breast cancer prevention and possibly in the adjuvant setting as well.
A number of prior studies have demonstrated that weight loss11 and increased exercise27–29 can influence sex steroid levels or metabolic hormones in at-risk or breast cancer populations, but most studies have focused on one type of lifestyle intervention and have not provided information regarding which lifestyle factors have the greatest impact on biologic pathways linked to breast cancer. The NEW results17 are not entirely consistent with other reports in the literature, some of which have found a significant decrease in levels of fasting insulin and other metabolic biomarkers in at-risk women30 and breast cancer survivors27,28 participating in exercise interventions. For example, our group demonstrated a 28% decrease in insulin levels in a group of sedentary, overweight breast cancer survivors participating in a 16-week mixed strength training and aerobic exercise intervention (P = .07)27; similar effects were reported by Irwin et al.28 In contrast, no decreases in insulin levels were reported by Schmitz et al31 or Fairey et al,32 who did not restrict their study populations to overweight and/or inactive survivors. The impact of exercise on metabolic hormones linked to breast cancer prognosis is not entirely clear, but it may vary with the type of exercise intervention employed and the study population.
So where does this leave us? Observational data consistently link obesity to poor outcomes in early breast cancer. The RENEW study16 adds to earlier reports by demonstrating the feasibility and durability of weight loss interventions administered at a distance to cancer survivors. The biomarker data from the NEW trial17 point to weight loss as the most efficacious means of affecting biologic pathways implicated in the lifestyle-cancer association. These reports strengthen the rationale for the conduct of well-designed trials, particularly involving weight loss; such trials should be adequately powered for important cancer outcomes in breast cancer, and they should include embedded biologic correlative research to elucidate key biologic mechanisms. Such trials would be practice changing if beneficial effects are identified and would help further unravel the complex pathways of breast carcinogenesis and tumor progression.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Manuscript writing: All authors
Final approval of manuscript: All authors
