Accumulating evidence suggests that lifestyle behaviors such as physical activity and dietary patterns may impact breast cancer prognosis. Observational data suggest that patients who are overweight and inactive have higher rates of breast cancer recurrence and overall mortality as compared to patients who are leaner and more active.1-4 Interventional studies in breast cancer survivors have demonstrated that changes in diet and physical activity patterns are feasible and are associated with improvements in quality of life, less treatment-related weight gain, and fewer disease- and treatment-related side effects.5 Some preliminary evidence suggests that lifestyle change may also improve prognosis in breast cancer,6 but much work still needs to be done to validate this and determine which types of lifestyle change are most important. This chapter provides an overview of the observational and interventional work in this area, as well as clinical recommendations for breast cancer survivors.

Four recent prospective observational studies have demonstrated that women who are physically active after breast cancer diagnosis have a better prognosis than sedentary women (Table 105-1). The Nurses’ Health Study investigators looked at the relationship between physical activity and rates of breast cancer recurrence and death in a cohort of 2987 women diagnosed with stage I-IIIa breast cancer and demonstrated that patients who engaged in more than 9 metabolic equivalent task (MET) hours/week of physical activity, equivalent to walking at an average pace for 3 h/wk, had a 50% lower risk of breast cancer recurrence, breast cancer death, and all-cause mortality than women who were inactive.3 The decreased risk of recurrence and death associated with exercise was seen in both pre- and postmenopausal patients and was independent of body mass index (BMI). Similar results were seen the Collaborative Women’s Longevity Study, which collected information regarding physical activity behaviors from 4482 women who had been diagnosed with early-stage breast cancer at least 5 years previously.4 Women expending more than 2.8 MET h/wk had lower breast cancer mortality when compared to women with lower levels of activity (hazard ratio [HR] 0.65, 95% CI 0.39-1.08 for 2.8-7.9 MET h/wk; and HR 0.59, 95% CI 0.35-1.01 for 8.0-20.9 MET h/wk). Results were again independent of patient age or BMI.

The Health, Eating, Activity and Lifestyle (HEAL) study also prospectively evaluated the relationship between physical activity and breast cancer outcomes.7 The study collected information regarding activity 3 years after diagnosis from 668 women with early-stage breast cancer. After 6 years median follow-up and 164 deaths, the study demonstrated that women who engaged in at least 9 MET h/wk of activity had a HR for total death of 0.33 (95% CI 0.15-0.73). Although the relatively small number of events limited the power, the study also suggested that women who decreased activity after diagnosis had an increased risk of death (HR 3.95, 95% CI 1.45-10.50) and those who increased activity had a decreased risk of death (HR 0.55, 95% CI 0.22-1.38) as compared to those who maintained a constant level of activity.

The Women’s Healthy Eating and Living (WHEL) study assessed the impact of physical activity, body weight, and dietary patterns upon survival in 1490 patients diagnosed with early-stage breast cancer.8 The study demonstrated a linear trend between mortality and physical activity, with participants who engaged in more than 1320 MET min/wk having a 42% lower risk of death as compared to inactive participants (p for trend = 0.02). There was also a significant relationship between fruit and vegetable intake and mortality (p = 0.02), but no trend toward decreasing mortality with increasing fruit and vegetable intake (p = 0.08). A composite measurement looking at fruit and vegetable intake and physical activity showed that participants who both consumed 5 or more servings of fruits and vegetables per day and completed at least 540 MET min/wk of physical activity week (equivalent to walking at a moderate pace for 3 h/wk) had a mortality of 4.8%, compared to 10.4% to 10.7% in participants who had either (but not both) high physical activity or high fruit/vegetable intake, and 11.5% in participants who had low physical activity and low fruit and vegetable intake. The impact of increased physical activity and fruit/vegetable intake was most significant in women whose tumors were estrogen receptor (ER) positive, although there were fewer events in the ER negative subset.

One study has also examined the relationship between physical activity before breast cancer diagnosis and breast cancer outcomes. Abrahamson and colleagues collected information regarding physical activity behaviors from 1264 premenopausal women with newly diagnosed early-stage breast cancer.9 Participants were asked to recall the average frequency of moderate and vigorous activity that they engaged in at age 13, age 20, and in the year prior to breast cancer diagnosis. The study demonstrated a modest improvement in overall survival in women who engaged in the highest quartile of activity in the year prior to diagnosis as compared to the lowest (HR 0.78, 95% CI 0.56-1.08). The relationship was stronger in women who were overweight or obese at the time of breast cancer diagnosis (HR 0.70, 95% CI 0.49-0.99). There was no relationship seen between physical activity at ages 13 and 20 and breast cancer prognosis.

In aggregate, these studies encompass more than 10,000 breast cancer patients. Despite the fact that each trial has looked at exercise at a different point in time, all 5 studies demonstrate that breast cancer patients who engage in moderate amounts of physical activity have better disease-free and overall survival as compared to inactive patients. However, despite this evidence, studies have suggested that many women become less active after breast cancer diagnosis. The Health, Eating, Activity, and Lifestyle (HEAL) study evaluated changes in physical activity in the year after diagnosis in a population-based cohort of 812 patients with newly diagnosed early-stage breast cancer.10 Patients were asked to specify the frequency and duration of participation in a number of different household and recreational activities in the year prior to diagnosis and in the month prior to the interview. Patients were found to decrease total physical activity levels by approximately 2.0 h/wk (11% of total activity) from pre-diagnosis to post-diagnosis. Obese patients demonstrated greater decreases in recreational physical activity than overweight or normal-weight patients. The greatest decreases in activity were observed in women undergoing both radiation and chemotherapy, but activity was found to decrease even in women treated with surgery alone. A follow-up questionnaire administered 3 years post-diagnosis to cohort members who remained free of disease demonstrated that less than 50% of patients had resumed pre-diagnosis levels of physical activity even 3 years after their diagnosis, demonstrating that declines in physical activity after diagnosis can be long-lasting in a significant proportion of breast cancer patients.11

Physical Activity Intervention Studies

To date, no randomized trial has evaluated the impact of increased physical activity after diagnosis on breast cancer outcomes. However, dozens of interventional studies have looked at the feasibility and potential benefits of exercise both in patients undergoing adjuvant treatment and in breast cancer survivors.5,12-14 Most of these trials have been small and enrolled patients at a single institution, although 3 recent trials enrolled patients across multiple institutions to distance-based exercise interventions, with or without a dietary component.15-17 A full review of these studies is beyond the scope of this chapter, but several recent reviews and a meta-analysis have demonstrated that exercise interventions may safely be implemented in breast cancer patients both during and after therapy.5,12-14 Although results are not completely consistent, studies have also demonstrated that patients who participate in exercise programs experience improvements in quality of life, cancer-related fatigue, and side effects of treatment. Some trials also demonstrated improvements in biomarkers associated with breast cancer risk and prognosis, but again, results were not consistent across trials.18-20

Breast cancer incidence has historically varied widely across the world, with Western countries such as the United States and Europe experiencing relatively high rates of the disease and Eastern countries such as China and Japan experiencing lower rates. Migration studies demonstrate that immigrants from low-risk areas of the world experience higher rates of breast cancer after moving to countries such as the United States, leading investigators to postulate that environmental or behavioral differences must be responsible for these differences in breast cancer risk.21,22 As a result, more than 1400 studies have been done to evaluate the relationship between dietary intake of fat, fruits and vegetables, supplements, and other nutrients and breast cancer risk.23,24

Investigators have also been interested in the relationship between dietary intake and cancer prognosis. Dozens of studies have looked at dietary intakes and breast cancer outcomes.24-26 Investigators have been especially interested in studying the relationship between dietary fat intake and breast cancer prognosis, given the huge differences in fat intake across the world and the large body of work examining the relationship between dietary fat intake and breast cancer risk. A recent review looked at 13 studies evaluating the relationships between fat consumption and breast cancer outcomes,25 and demonstrated that 6 of these studies showed an inverse relationship between fat intake and breast cancer outcomes. However, dietary fat intake is often strongly correlated with body weight, which is a known prognostic factor in early stage breast cancer. Four of the 6 studies showing a relationship between dietary fat intake and breast cancer prognosis thus became nonsignificant when analyses were adjusted for BMI and total caloric intake.

Studies have also looked at prognosis in relation to intake of fruits and vegetables, protein, carbohydrates and moderate amounts of alcohol.25,26 Although several individual studies have shown a relationship between intake of a specific dietary nutrient and breast cancer prognosis, there have generally been no consistent relationships between diet and breast cancer prognosis. One recent paper has suggested that there may be nonlinear relationships between intake of some nutrients and breast cancer prognosis, with patients with either higher or lower than average intake of protein, fat, and other nutrients experiencing a worse prognosis as compared to individuals with more average diets, but these findings require further confirmation.26

Dietary Intervention Studies

Two large-scale, randomized trials have examined the impact of dietary modification on disease outcomes in early-stage breast cancer (Table 105-2). The Women’s Interventional Nutrition Study (WINS) randomized 2437 women with stage I-IIIa breast cancer to a low-fat dietary intervention or usual care control group.27 Eligibility included breast cancer diagnosis within 12 months and a baseline diet including at least 20% of daily calories from fat. Participants randomized to the dietary intervention decreased dietary fat intake and body weight compared to controls. After a median follow-up of 5.6 years and 227 relapse events, disease-free survival was significantly better in patients randomized to the dietary intervention as compared to the controls (HR 0.76, 95% CI 0.60-0.98). With further follow-up, this difference was no longer statistically significant, but an exploratory subgroup analysis at that time demonstrated a significant improvement in survival in ER-negative patients randomized to the intervention group (HR 0.41, p = 0.003).28

The Women’s Healthy Eating and Living (WHEL) study looked at the impact of a low-fat, high fruit and vegetable diet on disease-free survival in 3088 women with stage I-IIIa breast cancer.29 Eligibility criteria included breast cancer diagnosis within 4 years. There were no restrictions on baseline diet. Participants randomized to the dietary intervention significantly increased intake of fruits and vegetables, and decreased percentage of dietary calories from fat. The study was designed so that patients did not reduce calories, so as to avoid any confounding effect of weight loss on dietary change. There were no changes in body weight. After a median follow-up of 7.3 years and 518 relapse events, there was no difference in recurrence rates in the diet and control groups (16.7 vs 16.9%, p = 0.63).

Thus evidence regarding the impact of dietary change on breast cancer outcomes is inconsistent. WINS suggests that a low-fat diet and/or modest weight loss may lead to improvements in prognosis for patients with early-stage breast cancer, perhaps especially for patients with hormone-receptor-negative breast cancer.27 However, the WHEL study29 did not show even a suggestion of improvement in breast cancer outcome in patients participating in an intensive, long-term dietary intervention, and observational evidence has not suggested a link between diet and breast cancer outcomes.25 Further work is needed before evidence-based dietary guidelines can be developed for breast cancer survivors.

Abe and colleagues reported the first study looking at the relationship between body weight and breast cancer recurrence in 1976.30 The study demonstrated that women who were overweight or obese had a 55.6% 5-year survival rate, as compared to 79.9% in leaner women. The study also demonstrated that obese women were more likely to have larger tumors, with higher rates of lymphatic invasion and nodal involvement. Since the initial report by Abe and associates, there have been more than 50 studies examining the relationship between body weight and prognosis.1,2,31 Studies have looked at weight in a variety of different ways. Older studies generally picked a weight cut point without regard to patient height, whereas more recent studies divided patients based on Quetelet score (weight/height) or BMI (kg/m2). Regardless of the method used to categorize excess body weight, the majority of these studies demonstrated a strong association between baseline weight and breast cancer outcomes. Several of these studies demonstrate a quadratic or J-shaped relationship between body weight and prognosis,32,33 suggesting that the risk of adverse prognosis increases exponentially with increasing weight.