Abstract

INTRODUCTION

Breast cancer is the most common form of cancer diagnosed among American women (1). More women are surviving breast cancer than ever before due to early diagnosis and improvements in initial treatments (1). Surviving breast cancer, however, often puts women at an increased risk for chronic diseases due in part to unfavorable changes in body composition (2).

Weight gain is common after a breast cancer diagnosis (3), and is often a result of increases in fat mass without concurrent gains in lean body mass, commonly referred to as sarcopenic obesity (4). Although not conclusive, obesity and weight gain may adversely affect the risk of breast cancer recurrence and death (5,6). Breast cancer survivors are also reported to have lower bone mineral density (BMD) and a significant increased risk for fractures compared to healthy women (7,8). Chemotherapy and endocrine therapy with aromatase inhibitors are associated with significant bone loss, osteoporosis and fractures (9,10). Thus, improving body composition is of extreme importance to breast cancer survivors.

Physical activity has been shown to be effective in the improvement of physical functioning and quality of life in breast cancer survivors (11), and may also be effective in the prevention of weight gain and bone loss. Physical activity has been shown to reduce the risk of primary breast cancer (12), and recurrence and death from breast cancer (13,14,15). The mechanisms thought to confer this reduced risk include certain metabolic and sex hormones (16).

In healthy women, aerobic exercise has been associated with favorable changes in body composition (17,18); however, few trials have examined the effect of exercise on outcomes of body composition, assessed with valid and reliable measures such as dual energy x-ray absorptiometry (DEXA), in breast cancer survivors (19–26). Fewer studies have examined the independent effect of aerobic exercise (i.e., without concomitant changes in diet or participation in resistance training) on body composition (25). Thus, there is currently no evidence to support the use of aerobic exercise, such as brisk walking, the most commonly reported physical activity among breast cancer survivors, to maintain or favorably change body fat, lean body mass (LBM) and BMD in these women.

We investigated the effects of a six-month randomized controlled aerobic exercise intervention vs. usual care on body weight and circumferences, body fat, LBM, BMD, and bone mineral content (BMC) in postmenopausal breast cancer survivors. Secondary aims were to examine the exercise effect stratified by important prognostic and physiologic variables. We hypothesized that exercise would be associated with favorable changes in body composition, and that stage at diagnosis, hormonal therapy, age, and body mass index (BMI) would modify the effect of exercise on body composition.

METHODS

Participants were recruited into the Yale Exercise and Survivorship Study, which is described in detail elsewhere (27). All study procedures were reviewed and approved by the Yale University School of Medicine Human Investigation Committee.

Recruitment

Study staff used the Yale-New Haven Hospital Tumor Registry to obtain the names of Connecticut women diagnosed with breast cancer by any Yale-affiliated physician from March 2004 to January 2006 (27) (Figure 1). Staff contacted each patient’s physician to request permission to contact the participant. An invitation letter was mailed to the participant, followed by a telephone screening questionnaire, baseline and 6-month clinic visit. From 788 screening calls made, 75 (9.5%) women were eligible, interested, and randomized.

Open in a separate window

Figure 1

Flow of participants through the YES Study.

RESULTS

Baseline Characteristics

Baseline demographic and physiologic data in the exercise and usual care groups were similar (Table 1). The average age of study participants was 56 yrs. The majority (84%) of participants were non-Hispanic white. Average time since diagnosis for participants was 3.3 yrs. Women were, on average, obese with BMD within normal limits. However, at baseline, eight women (11%; three exercisers and five usual care participants) had osteopenia (t score < −1.0).

Table 1

Baseline characteristics of randomized participants in the YES Study (N=75).^*

	Exercisers Mean (SD) or %	Usual Care Mean (SD) or %
N	37	38
Age (yrs)	56.5 (9.5)	55.1 (7.7)
Ethnicity
Non-Hispanic White	84%	84%
African-American	16%	11%
Asian/Pacific Islander	0%	3%
Education (%)
High school graduate	16%	22%
Some school after high school	24%	30%
College graduate +	60%	41%
Time since diagnosis (years)	3.6 (2.2)	3.3 (2.6)
Disease stage (%)
In Situ	11%	11%
Stage I	54%	27%
Stage II	27%	46%
Stage IIIA	8%	16%
Treatment (%)
None	5%	14%
Radiation only	41%	24%
Chemotherapy only	19%	19%
Radiation and chemotherapy	35%	43%
Hormone therapy
None	43%	30%
Tamoxifen	30%	30%
Aromatase Inhibitors	27%	40%
Current use of calcium supplements	58%	58%
Current use of vitamin D supplements	8%	23%
Current use of bisphosphonates	13%	3%
Physical Activity Questionnaire (min/wk recreational exercise)	13.0 (24.0)	12.0 (20.0)
Daily Activity Log (min/wk recreational exercise)	30.0 (41.1)	11.3 (24.8)
Pedometer steps/day	5,145 (2,312)	5,342 (2,744)

Open in a separate window

^*No statistically significant differences between exercise and usual care groups at baseline.

Effect of Exercise vs. Usual Care on 6-Month Changes in Body Composition

Mean 6-month changes from baseline in body composition for both groups are shown in Table 2. After 6-months, exercisers experienced favorable changes in body fat and LBM compared with unfavorable changes among women randomized to usual care (p < .05). Results remained statistically significant after excluding the six women taking bisphosphonates at baseline.

Table 2

Six-Month effect of exercise on body composition in the YES Study (mean ± SD).

	Baseline	6 months
	Mean ±SD	Mean ±SD	Percent Change	Pa
Weight (kg)
Exercisers b	81.28 ± 16.98	80.73 ± 16.88	−0.7	0.39
Usual Care c	78.41 ± 20.01	78.51 ± 20.64	0.1
BMI (kg·m−2)
Exercisers	30.57 ± 5.95	30.45 ± 6.00	− 0.4	0.42
Usual Care	29.74 ± 7.27	29.90 ± 7.59	0.4
Percent body fat (%)
Exercisers	41.33 ± 6.46	40.54 ± 6.55	−1.9	0.0022
Usual Care	39.18 ± 5.90	39.60 ± 5.97	1.1
Lean Mass (kg)
Exercisers	44.01 ± 6.62	44.35 ± 6.60	0.8	0.047
Usual Care	43.03 ± 7.19	42.68 ± 7.19	−0.8
BMD (g·cm−2)
Exercisers	1.152 ± 0.109	1.144 ± 0.119	−0.7	0.97
Usual Care	1.120 ± 0.094	1.112 ± 0.101	−0.7
BMC (g·cm)
Exercisers	2258 ± 373	2216 ± 383	−1.8	0.062
Usual Care	2116 ± 305	2120 ± 308	0.2
Waist circumference (cm)
Exercisers	91.13 ± 12.33	89.75 ± 12.63	−1.5	0.57
Usual Care	89.42 ± 15.13	88.58 ± 15.48	−0.8
Hip circumference (cm)
Exercisers	112.96 ± 13.59	111.92 ± 12.56	−0.8	0.21
Usual Care	110.79 ± 15.64	111.11 ± 14.97	0.6

Open in a separate window

^aDifference in body composition from baseline to 6 months in exercisers versus usual care.

^bn, 6 month study = 36

^cn, 6 month study = 33

Effect of Exercise vs. Usual Care on 12-Month Changes in Body Composition

Mean 12-month changes from baseline in body composition for both groups are shown in Table 3. After 12 months, exercisers experienced favorable changes in body fat and BMD compared with unfavorable changes among women randomized to usual care (p < .05). Results remained statistically significant after excluding the six women taking bisphosphonates at baseline.

Table 3

Twelve-month effect of exercise on body composition in the YES Study (mean ± SD).

	Baseline	12 months
	Mean ±SD	Mean ±SD	Percent Change	Pa
Weight (kg)
Exercisers b	81.21 ± 18.51	81.60 ± 18.97	−0.2	0.61
Usual Care c	75.93 ± 17.66	76.58 ± 17.41	0.6
Percent body fat (%)
Exercisers	41.21 ± 6.70	40.02 ± 7.06	−2.4	0.043
Usual Care	39.53 ± 6.14	39.50 ± 6.14	0.8
Lean Mass (kg)
Exercisers	43.92 ± 7.06	44.62 ± 7.65	1.7	0.25
Usual Care	42.30 ± 7.16	42.21 ± 6.74	0.2
BMD (g·cm−2)
Exercisers	1.159 ± 0.112	1.167 ± 0.126	0.2	0.043
Usual Care	1.122 ± 0.098	1.097 ± 0.105	−1.7
BMC (g·cm)
Exercisers	2265 ± 406	2267 ± 410	−0.6	0.13
Usual Care	2088 ± 300	2022 ± 295	−2.2

Open in a separate window

^aDifference in body composition from baseline to 12 months in exercisers versus usual care.

^bn, 12 month study = 26;

^cn, 12 month study = 23

Effect of Exercise on Body Composition Stratified by Potential Effect Modifiers

Hormone therapy modified the effect of exercise on body fat (p for interaction = .060), with greater between-group effect sizes in body fat observed among women not taking any hormone therapy vs. women taking tamoxifen or aromatase inhibitors (Figure 2). Results remained statistically significant after excluding the six women taking bisphosphonates at baseline (4 of whom were not taking any hormonal therapy). Age modified the effect of exercise on LBM (p for interaction = .044) (Figure 3) and BMD (p for interaction = .10) (Figure 4), with greater between-group effect sizes in LBM and BMD observed among younger women vs. older women. Disease stage, previous radiation or chemotherapy, and BMI did not modify the effect of exercise on body composition; however, a statistically significant between-group difference (i.e., exercisers vs. usual care) was observed for body fat, BMD, and BMC among women diagnosed with Stage II–III breast cancer (Figure 2, ​,4,4, and ​and5).5). Statistically significant between-group differences were also observed for body fat and BMC among women with a BMI < 30.

Open in a separate window

Figure 2

Baseline to 6 month % change in body fat stratified by stage, hormone therapy, age, and BMI

* p < .05 difference between exercisers and controls

Sample sizes for stage 0–I (exercisers, n = 24, usual care, n = 12), stage II–IIIA (exercisers, n = 12, usual care, n = 21); no HT (exercisers, n = 15, usual care, n = 10), tamoxifen (exercisers, n = 11, usual care, n = 8), AI (exercisers, n = 10, usual care, n = 15); < 56 yr (exercisers, n = 19, usual care, n = 16), ≥56 yr (exercisers, n = 17, usual care, n = 17; BMI < 30 (exercisers, n = 16, usual care, n = 20), and BMI ≥30 (exercisers, n = 20, usual care, n = 13).

Open in a separate window

Figure 3

Baseline to 6 month % change in LBM stratified by stage, hormone therapy, age, and BMI

* p < .05 difference between exercisers and controls

Sample sizes for stage 0–I (exercisers, n = 24, usual care, n = 12), stage II–IIIA (exercisers, n = 12, usual care, n = 21), no HT (exercisers, n = 15, usual care, n = 10), tamoxifen (exercisers, n = 11, usual care, n = 8), AI (exercisers, n = 10, usual care, n = 15), < 56 yr (exercisers, n = 19, usual care, n = 16), ≥56 yr (exercisers, n = 17, usual care, n = 17, BMI < 30 (exercisers, n = 16, usual care, n = 20), and BMI ≥30 (exercisers, n = 20, usual care, n = 13).

Open in a separate window

Figure 4

Baseline to 6 month % change in BMD stratified by stage, hormone therapy, age, and BMI

* p < .05 difference between exercisers and controls

Sample sizes for stage 0–I (exercisers, n = 24, usual care, n = 12), stage II–IIIA (exercisers, n = 12, usual care, n = 21), no HT (exercisers, n = 15, usual care, n = 10), tamoxifen (exercisers, n = 11, usual care, n = 8), AI (exercisers, n = 10, usual care, n = 15), < 56 yr (exercisers, n = 19, usual care, n = 16), ≥56 yr (exercisers, n = 17, usual care, n = 17, BMI < 30 (exercisers, n = 16, usual care, n = 20), and BMI ≥30 (exercisers, n = 20, usual care, n = 13).

Open in a separate window

Figure 5

Baseline to 6 month % change in BMC stratified by stage, hormone therapy, age, and BMI

* p < .05 difference between exercisers and controls

Sample sizes for stage 0–I (exercisers, n = 24, usual care, n = 12), stage II–IIIA (exercisers, n = 12, usual care, n = 21), no HT (exercisers, n = 15, usual care, n = 10), tamoxifen (exercisers, n = 11, usual care, n = 8), AI (exercisers, n = 10, usual care, n = 15), < 56 yr (exercisers, n = 19, usual care, n = 16), ≥56 yr (exercisers, n = 17, usual care, n = 17, BMI < 30 (exercisers, n = 16, usual care, n = 20), and BMI ≥30 (exercisers, n = 20, usual care, n = 13).

Effect of Exercise on Body Composition Stratified by Adherence

Women who exercised more experienced greater losses in body fat, yet slight increases in BMD compared with women who exercised less (p < .05) (Figure 6).

Open in a separate window

Figure 6

Baseline to 6 month % change in body fat, LBM, and BMD stratified by adherence (exercisers only)

Sample size for < 90 min/wk (n = 10), 90–150 min/wk (n = 15), and ≥ 150 min/wk (n = 12).

DISCUSSION

In our study, aerobic exercise, such as brisk walking, was associated with decreases in body fat, increases in LBM, and a maintenance of BMD in breast cancer survivors. The observation that aerobic exercise accompanied by fat loss was not associated with a significant loss of BMD is important. In postmenopausal women, adipose tissue is the main site of androgen conversion to estrogen by the enzyme aromatase (35). As postmenopausal women lose body fat, their serum estrogen concentrations decrease, which could cause a decrease in BMD. Thus, it is plausible that exercise accompanied by fat loss could have a negative effect on bone health in postmenopausal women. However, in our study, we observed a maintenance of BMD with exercise and subsequent fat loss.

Furthermore, women who were taking aromatase inhibitors, which significantly decrease estrogen concentrations to undetectable levels (36), and randomized to exercise, experienced decreases in body fat, increases in lean mass and maintenance of BMD and BMC, compared to no change in body fat and losses in LBM, BMD, and BMC among women randomized to usual care. These results are encouraging for the many postmenopausal women who are recommended to take aromatase inhibitors to improve their prognosis. Recently published reports show that women taking aromatase inhibitors experience greater bone loss and fractures compared to women taking tamoxifen or placebo (37–39). Aromatase inhibitor-associated bone loss may be distinct from normal postmenopausal bone loss. Estrogen deprivation that occurs during aromatase inhibitor therapy is generally abrupt compared with that occurring postmenopausally, and bone loss may occur at an accelerated rate (~2.6% bone loss/year with aromatase inhibitors compared to ~1% bone loss/year after menopause) (37). Although medications can be used to alleviate bone loss, such as bisphosphonate therapy, these drugs also cause significant side effects, such as gastrointestinal toxicity (37). Our study shows that aerobic exercise attenuates aromatase inhibitor-related bone loss. Thus, aromatase inhibitor-associated bone loss may be preventable, allowing women to benefit from endocrine therapy while also being protected against the increased risk of osteoporosis, fracture, and ultimately breast cancer recurrence or death. However, because of the small sub-group analyses, these findings should be interpreted with caution.

To our knowledge, no other study has examined whether certain prognostic or demographic factors modify the effect of exercise on body composition, and only one study has examined the main effect of aerobic exercise only on body composition assessed using valid and reliable measures such as DEXA (25). In that study, Matthews et al observed no significant changes in body fat or BMD in 10 postmenopausal women enrolled in a 12-week home-based walking intervention vs. usual care.

Three studies have examined the effect of resistance training on body composition in breast cancer survivors who have completed treatment. Schmitz et al. (20) investigated a resistance training program on body composition and observed significant increases in LBM (0.88 for exercisers vs. 0.02 for controls, p < .01) and decreases in body fat (−1.15% for exercisers vs. 0.23%, p = .023) with a twice-weekly yearlong resistance training program in 69 pre- and post-menopausal breast cancer survivors. Their observed between-group effect sizes in LBM and body fat are similar to the effect sizes observed in our study. This finding is promising because some women may prefer to control weight and LBM through aerobic exercise, such as brisk walking, rather than through resistance training. The two other studies were quasi-experimental studies of small sample sizes (< 26 women), but one of the studies observed significant improvements in BMD of the spine and hip (p < .01) with resistance training (22), and the other study reported no change in BMD (26).

Lastly, only one study has examined the effect of exercise on BMD in women with breast cancer receiving adjuvant chemotherapy. Schwartz et al. (19) examined the effects of aerobic or resistance training (therabands) vs. usual care on BMD in 66 pre- and post-menopausal women beginning adjuvant chemotherapy. The average decline in BMD was 6.23% for women randomized to usual care compared to losses of 4.92% and 0.76% among women randomized to resistance training or aerobic exercise.

In our study, stratified analyses also demonstrated greater between-group differences in body composition among women diagnosed with Stage II–III breast cancer compared to women diagnosed with Stage 0–I breast cancer. Women with higher stages of disease are often treated with radiation and chemotherapy, which may adversely affect body fat and BMD. An exercise intervention may, therefore, have stronger effects in women diagnosed with higher stages of disease. Age also modified the effect of exercise on body composition. Younger women may have experienced chemotherapy-induced menopause rather than natural menopause, resulting in greater losses of BMD and LBM during and after treatment for breast cancer. An exercise program may be more advantageous for maintenance or increases in LBM and BMD for younger breast cancer survivors compared to older breast cancer survivors. Lastly, a dose-response effect was observed with greater decreases in body fat, increases in LBM, and maintenance of BMD with higher doses of exercise per week.

The Yale Exercise and Survivorship Study was methodologically strong relative to many studies previously conducted. Our study used a population-based recruitment strategy, randomization to study groups, a physical activity prescription that met national guidelines, detailed measurement of physical activity and adherence, comprehensive assessment of body composition using validated measures, and long study duration. Our sample size is comparable, if not larger, than those of several earlier studies in breast cancer survivors; however it is still small in absolute terms, with implications for compromised statistical power with regard to some of our stratified analyses. An additional limitation of our study is that we did not measure regional bone density. Our sample was also highly-educated and mostly non-Hispanic white. However, the 75 women enrolled in the study did not differ with regard to age, race, or educational status from the non-participants that comprised the majority of women diagnosed with breast cancer at Yale-New Haven Hospital in 2004 and 2005.

Conclusions

Given that a majority of breast cancer survivors are at risk of osteoporosis and fractures (37), improving bone mass is a priority in this population. Weight gain is also common after a diagnosis, and negatively affects prognosis (5,6). In our study, aerobic exercise, such as brisk walking, was associated with favorable changes in body fat and LBM, and maintenance of BMD in postmenopausal breast cancer survivors. Future studies are needed to investigate different doses and types of exercise on body composition in breast cancer survivors and certain subgroups, such as women taking aromatase inhibitors.

Acknowledgments

References