Changes in Weight and Body Composition in Women Receiving Chemotherapy for Breast Cancer

Original Study

Changes in Weight and Body Composition in Women Receiving Chemotherapy for Breast Cancer Mary Jo Nissen, Alice Shapiro, Karen K. Swenson Abstract Purpose: This study aimed to identify predictors of changes in weight and body composition among women receiving chemotherapy for breast cancer. Patients and Methods: Data were from 49 women age 40-54 receiving chemotherapy for breast cancer. Weight, height, and body composition measurements from dual-energy x-ray absorptiometry (DEXA) scanning were completed at baseline (within 1 month of beginning chemotherapy) and 12 months. Caloric intake was assessed from food diaries at baseline, 6 and 12 months, and physical activity was measured by questionnaire at baseline, 3, 6, 9, and 12 months. Results: Baseline body mass index (BMI) was inversely associated with gains in weight (P = .01) and fat mass in torso (P = .006). Women of normal weight gained an average of 4.3 pounds and increased fat mass in torso and arms. Overweight women lost 3.0 pounds, and obese women lost 4.1 pounds, and neither group increased body fat. Decreased physical activity was associated with weight gain (P = .047). Additional predictors of increased fat mass in torso were younger age (P = .023) and treatment with tamoxifen (P = .015). Predictors of loss of bone mineral content included older age (P = .004) and treatment with aromatase inhibitor (P = .024), whereas treatment with bisphosphonate prevented bone loss (P < .0001). Conclusion: Women receiving chemotherapy for breast cancer who are of normal weight at the time of breast cancer diagnosis are more likely to gain weight and body fat during the following year than overweight or obese women. Clinical Breast Cancer, Vol. 11, No. 1, 52-60, 2011; DOI: 10.3816/CBC.2011.n.009 Keywords: Aromatase inhibitor, Bisphosphonate, Dual-energy x-ray absorptiometry, Tamoxifen

Introduction Women receiving chemotherapy for breast cancer undergo a range of physical, emotional, and cognitive changes. Some are transitory,1 but others have lasting effects. One physical change first reported more than 30 years ago is weight gain.2 The significance of weight gain during chemotherapy is highlighted by its potential impact on survival: weight gain following breast cancer diagnosis has been associated with increased rates of recurrence and death.3-6 In addition, weight gain in this population tends to be accompanied by adverse changes in body composition, with increases primarily in fat mass rather than lean mass.7-10 Although some reports suggest that the incidence of weight gain may have diminished somewhat as new chemotherapy regimens with shorter treatment schedules have been implemented,8 it remains an area of concern. Oncology Research Department, Park Nicollet Institute, Minneapolis, MN Submitted: Apr 7, 2010; Revised: Sep 16, 2010; Accepted: Sep 17, 2010 Address for correspondence: Mary Jo Nissen, PhD, Park Nicollet Institute, 3931 Louisiana Ave, FRCC, 1st Floor, St. Louis Park, MN 55426 Fax: 952-993-5235; e-mail: [email protected] 1526-8209/$ - see frontmatter © 2011 Elsevier Inc. All rights reserved.

Because weight gain in particular is a behaviorally modifiable risk factor, there have been efforts to identify predictors of weight gain in breast cancer patients undergoing treatment.3,6-23 Results have been inconsistent. Most studies,3,7,9,11-16 but not all,17 have found that chemotherapy treatment is associated with greater weight gain than local therapy. Several studies reported that premenopausal women gain more weight than postmenopausal women,3,10-13,19 but others have found no effect of menopausal status.14,17,21 Body mass index (BMI) at the time of diagnosis has been reported to be negatively associated,13-17 positively associated,3,20 and not significantly associated with weight gain during treatment.6,8,9,12,18,21 To some extent these inconsistencies are not surprising because the reports span decades (1985-2010) during which the type and duration of chemotherapy treatment for breast cancer has changed, and during which dietary patterns have changed.22 It is important to clarify what factors are predictive of weight gain and changes in body composition with the chemotherapy regimens used today. The possibility that BMI at the time of diagnosis is predictive of weight gain is particularly relevant to efforts to target weight management interventions appropriately, and it raises the question of whether any such association is mediated by differential changes in diet or physical activity among patients who are of normal weight, overweight, or obese. Only a few of the

The summary may include the discussion of investigational and/or unlabeled uses of drugs and/or devices that may not be approved by the FDA. Electronic forwarding or copying is a violation of US and International Copyright Laws. For authorization to photocopy items for internal or personal use, visit www.elsevier.com/permissions.

52

| Clinical Breast Cancer

February 2011

Table 1 Demographic and Disease Characteristics of Participants Characteristic

Table 2A Predictors of Change in Weight and Body Composition From Baseline to 12 Months

Total (N = 49)

Mean Age at Diagnosis, Years (SE)

46.8 (0.5)

Age Range, Years

Predictor Variable

N

Mean

40-54

Tobacco Status

P Value: P Value: SE Univariate Multivariate Analysisa Analysis

Change in Weight, lb BMI at Baselineb

Never smoker

37 (75.5%)

Past smoker

8 (16.3%)

Normal

26

4.3

Current smoker

4 (8.2%)

Overweight

11

–3.0

Obese

12

–4.1

Race White

46 (93.9%)

Asian

3 (6.1%)

Job Status

.010

.043

1.0

–

–

3.7

–

–

3.1

–

–

.047

.041

Change in Blocks Walked From Baseline to 12 Months Increase of ≥ 12

14

–4.3

2.4

–

–

3 (6.1%)

No change to increase of 12

16

2.4

1.7

–

–

9 (18.4%)

Decrease

12

4.2

3.3

–

–

.139

–

18

1.0

2.0

–

–

Full-time

37 (75.5%)

Part-time Unemployed/retired

Age

Cancer Stage I

17 (34.7%)

40-45 years

II or III

32 (65.3%)

46-48 years

16

2.6

2.5

–

–

49-54 years

15

–2.0

2.4

–

–

.061

–

30

2.7

1.7

–

–

Aromatase inhibitor

8

–3.5

4.1

–

–

11

–1.9

2.2

–

–

.649

–

ZO

26

–0.1

2.0

–

–

PA

23

1.4

1.8

–

–

Surgery Type Mastectomy

22 (44.9%)

Lumpectomy

27 (55.1%)

Type of Axillary Dissection

Hormone Treatment Tamoxifen

Sentinel lymph node dissection

30 (61.2%)

None

Axillary lymph node dissection

19 (38.8%)

Intervention

Reconstruction (Breast Implant) Yes

6 (12.2%)

No

43 (87.8%)

Chemotherapy

analysis for each outcome variable included as covariates only those predictor variables that were significant at P < .05 on univariate analyses for the outcome variable. weight: BMI < 25 kg/m2; overweight: 25 kg/m2 ≤ BMI < 30 kg/m2; obese: BMI ≥ 30 kg/m2. Abbreviations: BMI = body mass index; PA = physical activity; ZO = bisphosphonate bNormal

AC (4 cycles)

19 (38.8%)

AC + T (8 cycles)

28 (57.1%)

Other

2 (4.1%)

Hormone Treatment Tamoxifen

30 (61.2%)

Aromatase inhibitor

8 (16.3%)

None

11 (22.5%)

Radiation Therapy Yes

36 (73.5%)

No

13 (26.5%)

Mean Number Calcium/Vitamin D Pills Taken During Study Period (SE)

aMultivariate

583.3 (23.7)

Abbreviations: AC = doxorubicin and cyclophosphamide; T = taxane

previous studies addressing predictors of weight gain also examined body composition. Finally, a hypothesis that has been raised23 but minimally tested24 is that increased dietary calcium intake leads to increased lean mass. This possibility also deserves attention. The purpose of this study is to identify predictors of changes in weight and body composition among women receiving chemo-

therapy for breast cancer. Data were obtained as part of a randomized clinical trial designed to compare the impact of bisphosphonates versus a prescribed physical activity intervention on bone mineral density during breast cancer treatment.25 The study addressed the following questions: (1) What are the predictors of weight change among women during chemotherapy? (2) What are the predictors of changes in body composition among women during chemotherapy? (3) Is calcium intake associated with preservation of lean body mass?

Patients and Methods Patients This study was conducted at Park Nicollet Clinic, an upperMidwestern outpatient cancer center that diagnoses and treats approximately 400 new breast cancer cases annually. Beginning in October 2003, consecutive women diagnosed with breast cancer were invited to participate if they met the eligibility criteria for the study: diagnosed with stage I-III invasive breast cancer, age 40-55, within 24 months of having their last menstrual period, within one month of beginning adjuvant or neoadjuvant chemotherapy, baseline lumbar spine and total hip bone mineral density t score ≥ –2.0

Clinical Breast Cancer February 2011

|

53

Changes in Weight and Body Composition Table 2B Predictors of Change in Weight and Body Composition From Baseline to 12 Months Predictor Variable

N

Mean

SE

Table 2C Predictors of Change in Weight and Body Composition From Baseline to 12 Months

P Value: P Value: Univariate Multivariate Analysisa Analysis

Change in Fat Mass in Torso, g BMI at Baselineb

Predictor Variable

N

Mean

SE

P Value: P Value: Univariate Multivariate Analysisa Analysis

Change in Fat Mass in Legs, g .006

.025

BMI at Baselineb

.359

–

Normal

26 1161.2 279.6

–

–

Normal

26

749.3 164.9

–

–

Overweight

11 –723.8 654.0

–

–

Overweight

11

150.3 508.2

–

–

Obese

12

–

–

Obese

12

303.8 446.0

–

–

.051

–

–84.9 481.0

Change in Blocks Walked From Baseline to 12 Months

.342

–

Change in Blocks Walked From Baseline to 12 Months

Increase of ≥ 12

14

–92.9 534.9

–

–

Increase of ≥ 12

14

–76.3 299.5

–

–

No change to increase of 12

16

443.5 400.7

–

–

No change to increase of 12

16

472.7 224.4

–

–

Decrease

12

986.5 568.9

Decrease

12 1171.6 497.2

Age

–

–

.023

.098

Age

–

–

.019

.216

40-45 years

18

929.0 465.3

–

–

40-45 years

18

531.4 251.6

–

–

46-48 years

16

625.9 357.5

–

–

46-48 years

16

967.8 354.0

–

–

49-54 years

15 –368.3 480.9

49-54 years

15

–17.9 299.0

Hormone Treatment

–

–

.015

.126

Hormone Treatment

–

–

.030

.076

Tamoxifen

30 1016.0 315.7

–

–

Tamoxifen

30

869.7 241.6

–

–

Aromatase inhibitor

8

–

–

Aromatase inhibitor

8

112.6 433.6

–

–

None

11 –625.4 421.8

–

–

None

11 –200.9 184.7

–298.9 694.9

Intervention

Intervention

.708

–

–

.461

–

ZO

26

264.1 380.3

–

–

ZO

26

342.1 240.1

–

–

PA

23

623.7 357.2

–

–

PA

23

690.7 269.7

–

–

aMultivariate

analysis for each outcome variable included as covariates only those predictor variables that were significant at P < .05 on univariate analyses for the outcome variable. weight: BMI < 25 kg/m2; overweight: 25 kg/m2 ≤ BMI < 30 kg/m2; obese: BMI ≥ 30 kg/m2. Abbreviations: BMI = body mass index; PA = physical activity; ZO = bisphosphonate

aMultivariate

bNormal

bNormal

SD, able to read and write English, and signed the consent form. Patients were excluded from the study if they had health conditions that contraindicate exercise participation, previous treatment with bisphosphonate, laboratory evidence of renal or hepatic disease, previous treatment for another cancer, transrectus abdominus muscle (TRAM) flap reconstruction, a positive pregnancy test, or mental illness that precluded the patient from giving informed consent. The protocol and consent form for the study were reviewed and approved by the Park Nicollet Institute Institutional Review Board.

been reported previously.26 Participants in the ZO group received intravenous zoledronic acid (4 mg) every 3 months for 5 treatments. Informed consent was obtained before randomization. All participants were provided with and instructed to take oral supplemental calcium (600 mg/vitamin D 200 IU twice daily). They were instructed to discontinue other dietary supplements except for a B50 complex, if desired. They received standard written guidelines on nutrition and cancer that were general in nature.

analysis for each outcome variable included as covariates only those predictor variables that were significant at P < .05 on univariate analyses for the outcome variable. weight: BMI < 25 kg/m2; overweight: 25 kg/m2 ≤ BMI < 30 kg/m2; obese: BMI ≥ 30 kg/m2. Abbreviations: BMI = body mass index; PA = physical activity; ZO = bisphosphonate

Procedures Study Design This was a longitudinal, observational study to evaluate changes in weight and body composition in a randomized clinical trial evaluating the effects of a physical activity intervention (PA) versus bisphosphonate (ZO) on bone mineral density in women undergoing treatment for breast cancer.25 Participants were randomized to one of two groups. In the PA group, women were assigned to participate in a home-based walking program and were advised to reach a goal of a minimum of 10,000 steps/day, equivalent to approximately 5 miles of walking. Details regarding adherence to this protocol have

54

| Clinical Breast Cancer

February 2011

Participants were instructed on procedures for completing 4-day food intake records for research and were asked to record everything they ate and drank on 3 weekdays and 1 weekend day at baseline (before the first cycle of chemotherapy) and at 6 and 12 months after the start of chemotherapy. All participants had completed chemotherapy within 6 months except for two women who were still receiving taxanes. Dietary records were reviewed by a research dietitian and participants were queried if clarification was needed. Nutritionist Pro software (Version 2.2, 2004, First Data Bank, Inc.) was used for nutrient analysis.

Mary Jo Nissen et al Table 2D Predictors of Change in Weight and Body Composition From Baseline to 12 Months Predictor Variable

N

Mean

SE

Table 2E Predictors of Change in Weight and Body Composition From Baseline to 12 Months

P Value: P Value: Univariate Multivariate Analysisa Analysis

Change in Fat Mass in Arms, g

Predictor Variable

N

Mean

SE

P Value: P Value: Univariate Multivariate Analysisa Analysis

Change in Lean Mass in Torso, g

BMI at Baselineb

.008

.008

BMI at Baselineb

.633

–

Normal

26

274.4 125.7

–

–

Normal

26 –405.0 172.4

–

–

Overweight

11 –223.5 289.1

–

–

Overweight

11

176.4 451.7

–

–

Obese

12 –946.2 503.0

–

–

Obese

12 –536.5 369.9

–

–

.199

–

Change in Blocks Walked From Baseline to 12 Months

.152

–

Change in Blocks Walked From Baseline to 12 Months

Increase of ≥ 12

14 –246.9 159.3

–

–

Increase of ≥ 12

14 –590.4 221.3

–

–

No change to increase of 12

16 –470.5 404.9

–

–

No change to increase of 12

16

137.5 346.3

–

–

Decrease

12

–

–

Decrease

12 –466.7 338.3

–

–

423.7 307.5

Age

Age

.971

–

.217

–

40-45 years

18 –162.7 197.1

–

–

40-45 years

18

195.1

–

–

46-48 years

16 –161.7 418.6

–

–

46-48 years

16 –649.1 281.3

–

–

49-54 years

15

–77.5 230.9

–

–

49-54 years

15 –359.8 368.5

–

–

30

14.9

Hormone Treatment Tamoxifen

Hormone Treatment

.559

–

251.9

–

–

Tamoxifen

.436

–

30 –404.2 191.7

–

–

–

–

–

–

Aromatase inhibitor

8

–283.0 263.5

–

–

Aromatase inhibitor

8

None

11 –441.9 196.0

–

–

None

11

Intervention

42.0

–455.9 352.7 67.9

450.4

0.972

–

Intervention

.218

–

ZO

26 –140.1 142.3

–

–

ZO

26 –121.2 272.7

–

–

PA

23 –131.9 320.7

–

–

PA

23 –516.3 160.3

–

–

aMultivariate

analysis for each outcome variable included as covariates only those predictor variables that were significant at P < .05 on univariate analyses for the outcome variable. weight: BMI < 25 kg/m2; overweight: 25 kg/m2 ≤ BMI < 30 kg/m2; obese: BMI ≥ 30 kg/m2. Abbreviations: BMI = body mass index; PA = physical activity; ZO = bisphosphonate

aMultivariate

bNormal

bNormal

Body height and weight were measured at baseline and 12 months using a wall-mounted stadiometer and a balance beam scale following standardized methods, with participants wearing a clinic gown. Body composition measurements were obtained at baseline and 12 months with a Delphi W-dual-energy x-ray absorptiometry scan (DEXA). The machine was calibrated before each scan by use of a body composition block. Participants were instructed to lie still in a supine position for approximately 15 minutes for the body composition scanning, which generated measurement of bone mass, fat mass, lean mass, and percent body fat in each arm, leg, and the trunk. All of the DEXA scans were performed by the same technician and read by the same rheumatologist, who were blinded to each participant’s group assignment. To assess physical activity, participants completed the Paffenbarger Physical Activity Questionnaire (PPAQ),27 a selfadministered survey that assesses the amount and intensity of physical activity and has been validated in previous studies.28 The first question on the PPAQ is “How many city blocks or their equivalent do you normally walk each day?” The PPAQ does not have a total score. Validation studies support using walking mea-

sures (such as number of city blocks), stair climbing, and sports or recreational participation to measure clinically important physical activity.27,29,30 In this study, distance walked each day, estimated in city blocks, was selected as the most appropriate measure of physical activity. The PPAQ was completed at baseline and at 3, 6, 9, and 12 months following baseline. Demographic data were obtained from patient surveys, and clinical data were obtained from chart review.

analysis for each outcome variable included as covariates only those predictor variables that were significant at P < .05 on univariate analyses for the outcome variable. weight: BMI < 25 kg/m2; overweight: 25 kg/m2 ≤ BMI < 30 kg/m2; obese: BMI ≥ 30 kg/m2. Abbreviations: BMI = body mass index; PA = physical activity; ZO = bisphosphonate

Statistical Analysis Analyses of body composition measures were based on fat mass, lean mass, and bone mineral content (BMC), each expressed in grams, in arms, legs, and torso. We analyzed arms, legs, and torso separately in order to identify variables that were associated with change in body composition restricted to particular area(s). Other measures such as total body fat or percent body fat were considered to be redundant with these measures and so were not included. In addition, measures of total body fat and percent body fat have the potential to be affected differentially by type of breast surgery.24

Clinical Breast Cancer February 2011

|

55

Changes in Weight and Body Composition Table 2F Predictors of Change in Weight and Body Composition From Baseline to 12 Months Predictor Variable

N

Mean

SE

Table 2G Predictors of Change in Weight and Body Composition From Baseline to 12 Months

P Value: P Value: Univariate Multivariate Analysisa Analysis

Change in Lean Mass in Legs, g

N

Mean

SE

P Value: P Value: Univariate Multivariate Analysisa Analysis

Change in Lean Mass in Arms, g

BMI at Baselineb

.400

–

BMI at Baselineb

.018

.018

Normal

26 –116.3 140.3

–

–

Normal

26

74.0

–

–

Overweight

11 –153.0 215.0

–

–

Overweight

11 –196.0 138.8

–

–

Obese

12

–

–

Obese

12 –328.7 146.7

–

–

.605

–

80.7

313.5

Change in Blocks Walked From Baseline to 12 Months

.048

.048

47.7

Change in Blocks Walked From Baseline to 12 Months

Increase of ≥ 12

14 –400.4 167.4

–

–

Increase of ≥ 12

14

62.2

–

–

No change to increase of 12

16 –113.4 132.6

–

–

No change to increase of 12

16 –204.9 131.8

–

–

Decrease

12

328.5 290.8

–

–

Decrease

12

–

–

.504

–

.198

–

40-45 years

18

–17.0 175.4

–

–

40-45 years

18 –184.1 91.3

–

–

46-48 years

16

59.9

242.6

–

–

46-48 years

16 –139.4 122.2

–

–

49-54 years

15 –292.7 177.9

–

–

49-54 years

15

45.5

123.8

–

–

.529

–

30

–94.4

87.8

–

–

Age

Hormone Treatment

–80.2

–27.1 179.9

Age

Hormone Treatment

.848

–

30 –102.2 161.4

–

–

Tamoxifen

Aromatase inhibitor

8

–

–

Aromatase inhibitor

8

–237.2 179.9

–

–

None

11

48.5

–

–

None

11

–11.8

–

–

.796

–

Intervention

.452

–

ZO

26

–97.9 159.5

–

–

ZO

26 –153.3 67.0

–

–

PA

23

–51.8 170.4

–

–

PA

23

–

–

Tamoxifen

–150.7 193.4 238.3

Intervention

aMultivariate

aMultivariate

bNormal

bNormal

analysis for each outcome variable included as covariates only those predictor variables that were significant at P < .05 on univariate analyses for the outcome variable. weight: BMI < 25 kg/m2; overweight: 25 kg/m2 ≤ BMI < 30 kg/m2; obese: BMI ≥ 30 kg/m2. Abbreviations: BMI = body mass index; PA = physical activity; ZO = bisphosphonate

Participants’ report of number of city blocks walked per day was used as an indicator of level of physical activity. Amount of change in the number of blocks walked from baseline to 12 months, amount of change in daily caloric intake from baseline to 12 months, and age at baseline were categorized by tertiles. Number of blocks walked at either baseline or 12 months (and thus change in blocks walked) was missing for seven participants. BMI was categorized as normal weight (BMI < 25 kg/m2), overweight (25 kg/m2 ≤ BMI < 30 kg/m2), or obese (BMI ≥ 30 kg/m2). No participant was underweight (BMI < 18.5 kg/m2). At baseline all participants were within 24 months of their last menstrual period; 35 (71.4%) were having regular menses, seven (14.3%) had irregular menses or had been amenorrheic for six or fewer months, and 7 (14.3%) had been amenorrheic for seven or more months at baseline. Age and menopausal status as indicated by these three categories were strongly associated (χ2 = 18.3; P = .0001). Because there were only a few women who were postmenopausal at baseline and they were most likely only recently postmenopausal, only age and not menopausal status was included in analyses. By 12 months following baseline, all participants were amenorrheic.

56

Predictor Variable

| Clinical Breast Cancer

February 2011

98.4

–38.0 115.4

analysis for each outcome variable included as covariates only those predictor variables that were significant at P < .05 on univariate analyses for the outcome variable. weight: BMI < 25 kg/m2; overweight: 25 kg/m2 ≤ BMI < 30 kg/m2; obese: BMI ≥ 30 kg/m2. Abbreviations: BMI = body mass index; PA = physical activity; ZO = bisphosphonate

Age and hormone treatment were not associated with each other in this sample (χ2 = 2.4; P = .479), and both were included in analyses. Analysis of variance (ANOVA) was used to identify predictors of change from baseline to 12 months in total body weight and change from baseline to 12 months in fat mass, lean mass, and bone mineral content in arms, legs, and torso. Each outcome measure was thus a continuous variable that could range from negative to positive values. For each outcome measure, univariate analyses were first conducted to assess associations with the following potential predictor variables, all of which were treated categorically: BMI at baseline, age, type of chemotherapy, hormone treatment, type of axillary dissection, type of breast surgery, breast implant reconstruction, change from baseline to 12 months in caloric intake, and change from baseline to 12 months in number of blocks walked. Variables that were significant at P < .05 on univariate analysis were included in multivariate analysis of variance for that outcome measure. Student t test was used to determine whether change in weight from baseline was significant. Mixed-model linear regression on repeated measures (PROC MIXED) with compound symmetric

Mary Jo Nissen et al Table 2H Predictors of Change in Weight and Body Composition From Baseline to 12 Months Predictor Variable

N

Mean

SE

Table 2I

P Value: P Value: Univariate Multivariate Analysisa Analysis

Change in Bone Mineral Content in Torso, g BMI at Baselineb

Predictors of Change in Weight and Body Composition From Baseline to 12 Months

Predictor Variable

N

Mean

SE

P Value: P Value: Univariate Multivariate Analysisa Analysis

Change in Bone Mineral Content in Legs, g

.250

–

BMI at Baselineb

.559

–

Normal

26

–7.8

12.2

–

–

Normal

26

–4.8

5.2

–

–

Overweight

11

–31.8

17.1

–

–

Overweight

11

–2.0

8.5

–

–

Obese

12

–34.4

15.1

–

–

Obese

12

7.1

6.8

–

–

.414

–

Change in Blocks Walked From Baseline to 12 Months

.510

–

Change in Blocks Walked From Baseline to 12 Months

Increase of ≥ 12

14

–24.0

14.8

–

–

Increase of ≥ 12

14

–7.3

10.3

–

–

No change to increase of 12

16

–4.4

17.8

–

–

No change to increase of 12

16

–0.8

4.9

–

–

Decrease

12

–29.3

14.3

–

–

Decrease

12

7.6

6.7

–

–

.690

–

.004

.0007

40-45 years

18

–7.2

17.0

–

–

40-45 years

18

10.6

4.7

–

–

46-48 years

16

–20.9

11.3

–

–

46-48 years

16

2.7

5.4

–

–

49-54 years

15

–33.4

14.1

–

–

49-54 years

15

–19.7

7.3

–

–

.568

–

.024

.066

30

–16.8

11.9

–

–

Tamoxifen

30

3.8

3.9

–

–

Aromatase inhibitor

8

–31.6

19.4

–

–

Aromatase inhibitor

8

–24.7

11.5

–

–

None

11

–19.0

14.1

–

–

None

11

2.1

8.1

–

–

< .0001

< .0001

.0003

< .0001

ZO

26

10.0

10.4

–

–

ZO

26

10.9

3.5

–

–

PA

23

–53.3

9.7

–

–

PA

23

–14.9

5.8

–

–

Age

Hormone Treatment Tamoxifen

Intervention

Age

Hormone Treatment

Intervention

aMultivariate

analysis for each outcome variable included as covariates only those predictor variables that were significant at P < .05 on univariate analyses for the outcome variable. weight: BMI < 25 kg/m2; overweight: 25 kg/m2 ≤ BMI < 30 kg/m2; obese: BMI ≥ 30 kg/m2. Abbreviations: BMI = body mass index; PA = physical activity; ZO = bisphosphonate

aMultivariate

bNormal

bNormal

covariance structure was used to analyze caloric intake and physical activity across time. All tests were two-sided and were performed using SAS software (Version 9, 2004; SAS Institute; Cary, NC). All tests were performed using a significance level of P < .05.

the present analyses. Eleven women were excluded due to missing dietary data. The final sample size was 49. Demographic and clinical characteristics of the sample are summarized in Table 1.

analysis for each outcome variable included as covariates only those predictor variables that were significant at P < .05 on univariate analyses for the outcome variable. weight: BMI < 25 kg/m2; overweight: 25 kg/m2 ≤ BMI < 30 kg/m2; obese: BMI ≥ 30 kg/m2. Abbreviations: BMI = body mass index; PA = physical activity; ZO = bisphosphonate

Predictors of Change in Weight and Body Composition

Results Sample Characteristics During the 33-month recruitment period, 107 women were identified as eligible and were invited to participate in this study. Seventy-two women (62%) consented and were enrolled. A total of ten women dropped out of the study: seven because the demands of their treatments hindered their ability to participate in the PA intervention, two because of concerns about the potential side effects of zoledronic acid, and one who left the system to seek treatment at another institution. Two women had undergone bariatric surgery before their breast cancer diagnosis; their data are not included in the present analyses. Finally, any participant who did not provide at least two days of dietary data at each of the three measurement periods (baseline, 6 months, and 12 months) was excluded from

Univariate analyses identified several variables that were significantly associated with at least one outcome measure: BMI at baseline, change in blocks walked, age, hormone treatment, and intervention group. Mean change from baseline in each outcome variable by each of these predictor variables is shown in Table 2, together with the P value of the univariate analysis of variance assessing the association between the predictor variable and the outcome variable. Multivariate analysis of variance on each outcome variable included only those predictor variables that were significant at P < .05 on univariate analysis. P values for each of those variables when included in the multivariate analysis are included in Table 2. Surgical variables (type of breast surgery and axillary dissection, reconstruction), type of chemotherapy, and change from baseline in caloric intake did not affect any outcome significantly and those data are not shown.

Clinical Breast Cancer February 2011

|

57

Changes in Weight and Body Composition Table 2J Predictors of Change in Weight and Body Composition From Baseline to 12 Months Predictor Variable

N

Mean

SE

P Value: P Value: Univariate Multivariate Analysisa Analysis

Change in Bone Mineral Content in Arms, g BMI at Baselineb

.037

.037

Normal

26

5.3

2.8

–

–

Predictors of Change in Body Composition

Overweight

11

–7.6

3.3

–

–

Obese

12

–1.8

4.6

–

–

.210

–

Change in Fat Mass. Baseline BMI was associated with change in fat mass in torso (P = .006) and arms (P = .008). Women of normal weight at baseline showed increases in fat mass in torso and arms, whereas overweight and obese women did not. Age (P = .023) and hormone treatment (P = .015) were also associated with change in fat mass in torso. Younger women but not older women gained fat mass in torso. Women who took tamoxifen following chemotherapy gained fat mass in torso whereas those who took aromatase inhibitors or had no hormone therapy following chemotherapy did not. When BMI, age, and hormone therapy were entered together in a multivariate ANOVA, only BMI at baseline remained significantly associated with change in fat mass in torso (P = .025). On univariate analysis, hormone treatment (P = .03) and age (P = .019) were also associated with change in fat mass in legs, but when entered together in a multivariate analysis neither remained significant.

Change in Blocks Walked From Baseline to 12 Months Increase of ≥ 12

14

–5.4

3.3

–

–

No change to increase of 12

16

2.9

3.5

–

–

Decrease

12

4.7

5.9

–

–

.907

–

40-45 years

18

0.9

3.0

–

–

46-48 years

16

1.2

3.5

–

–

49-54 years

15

–0.1

4.7

–

–

.064

–

30

4.4

2.9

–

–

Aromatase inhibitor

8

–3.9

5.2

–

–

None

11

–6.2

2.4

–

–

.904

–

ZO

26

0.6

3.0

–

–

PA

23

0.8

3.0

–

–

Age

Hormone Treatment Tamoxifen

Intervention

aMultivariate

analysis for each outcome variable included as covariates only those predictor variables that were significant at P < .05 on univariate analyses for the outcome variable. weight: BMI < 25 kg/m2; overweight: 25 kg/m2 ≤ BMI < 30 kg/m2; obese: BMI ≥ 30 kg/m2. Abbreviations: BMI = body mass index; PA = physical activity; ZO = bisphosphonate bNormal

Weight Change. There was a wide range in weight change, from –20 lbs to 25 lbs. BMI at baseline (P = .010) and change in blocks walked (P = .047) were significantly associated with weight change. Women of normal weight (n = 26) gained an average of 4.3 lbs (P = .0002), overweight women (n = 11) lost an average of 3.0 pounds (P = .439), and obese women (n = 12) lost an average of 4.1 pounds (P = .220; Table 2). Taking a change of at least 5% of baseline weight as a clinically significant weight change, and using that threshold to categorize participants as showing a significant gain, loss, or stable weight, results showed that 13 women had a significant weight gain and of those 13, ten had normal BMI at baseline. Regarding change in walking, women who reported an increase in walking of at least 12 blocks from baseline to 12 months (n = 14) lost an average of 4.3 pounds; those whose walking decreased (n = 12) gained an average of 4.2 pounds; and those in the middle tertile (no change to an increase of 12 blocks, n = 16) gained an average of 2.4 pounds (Table 2). When both BMI and change in walking were entered into a multivariate ANOVA, both variables remained significant predictors of weight change (P = .04 for each).

58

Physical activity (blocks walked per day) and caloric intake for normal, overweight, and obese women from baseline to 12 months are shown in Figure 1. It appears that overweight and obese women, but not women of normal weight, tended to increase their physical activity over the 12 months following initiation of chemotherapy. However, neither BMI nor month of assessment, nor their interaction, was significantly associated with physical activity or caloric intake.

| Clinical Breast Cancer

February 2011

Change in Lean Mass. No variables were predictive of change in lean mass in torso. An increase in number of blocks walked was associated with a decrease in lean mass in legs (P = .048), and greater BMI at baseline was associated with a decrease in lean mass in arms (P = .018). Change in Bone Mineral Content. Intervention group was predictive of change in bone mineral content in torso (P < .0001) and legs (P = .0003): Women in the physical activity group lost bone mineral content, whereas those in the zoledronic acid group did not. Other predictors of change in bone mineral content in legs were age (P = .004) and hormone treatment (P = .024). Older women lost bone mineral content in legs whereas younger women did not. Women who took aromatase inhibitors lost bone mineral content in legs; those who took tamoxifen or had no hormone therapy did not. When age, hormone treatment, and intervention group were entered together in a multivariate ANOVA, both intervention group (P = .0004) and age (P = .0007) remained significantly associated with change in leg bone mineral content. The only variable significantly associated with change in bone mineral content in arms was BMI at baseline (P = .037). Women of normal weight at baseline gained bone mineral content in arms whereas overweight and obese women did not. Neither dietary calcium intake at baseline, 6 months, or 12 months, nor change in dietary calcium intake from baseline to 12 months, was associated with change in lean mass.

Discussion Women receiving chemotherapy who are of normal weight at the time of their breast cancer diagnosis are more likely to gain

Mary Jo Nissen et al Figure 1A Physical Activity From Baseline (Beginning of Chemotherapy) to 12 Months in Breast Cancer Patients Who Are Normal Weight (BMI < 25 kg/m2), Overweight (25 kg/m2 ≤ BMI < 30 kg/m2), and Obese (BMI ≥ 30 kg/m2)

Figure 1B Caloric Intake at Baseline (Beginning of Chemotherapy), 6, and 12 Months Following Baseline in Breast Cancer Patients Who Are Normal Weight (BMI < 25 kg/m2), Overweight (25 kg/m2 ≤ BMI < 30 kg/m2), and Obese (BMI ≥ 30 kg/m2)

35

2000 1800

30

Calories Per Day

Blocks Walked Per Day

1600 25 20 15 10

1400 1200 1000 800 600 400

5

200

0

0 Baseline

Month 3

Month 6

Month 9

Month 12

Baseline

Normal Overweight Obese

weight during the following year than women who are overweight or obese. Among women who experienced a clinically significant weight gain, 77% were of normal weight at the time of diagnosis. This result contrasts with findings in the general population, which report that those with higher baseline BMI are more likely to gain weight.31-34 Thus, although our study did not have a control group of women not undergoing breast cancer treatment, the pattern we found is unlikely to reflect what would occur in such a group and instead seems relevant to the experiences of breast cancer patients. Weight gain among women of normal weight was reflected in increases in fat mass and loss of lean mass, whereas overweight and obese women did not show this pattern, suggesting that the development of sarcopenic obesity7 (increase of fat mass and loss of lean mass) is particularly likely for women of normal weight. Evidence of the association of weight gain with breast cancer recurrence has highlighted the importance of interventions to prevent weight gain following diagnosis.35 Our results indicate that it would be wise to recommend those interventions to women who are of normal weight, since they are most likely to experience weight gain. This approach may seem counterintuitive, since women who are overweight and obese have a more apparent problem with weight control. Our results stress the importance of increasing awareness of the higher likelihood of weight gain and increase in fat mass among those of normal weight. Even though we used a very simple measure of physical activity, our results suggest that physical activity is a stronger mediator of weight change during chemotherapy than dietary intake. This agrees with other reports regarding the importance of physical activity for women with breast cancer7,9,36,37 and suggests that interventions among women with normal weight at diagnosis might profitably focus on maintaining and increasing physical activity. Although age and hormone treatment did not affect weight change, they did affect body composition. Women who were

Month 6

Month 12

Normal Overweight Obese

younger at diagnosis and those who took tamoxifen following chemotherapy gained more fat mass in the torso. Those who were older at diagnosis and those who took aromatase inhibitors following chemotherapy lost more BMC in the legs. A counterintuitive finding was that women who increased walking showed a significant loss of lean leg mass. Women who increased walking lost body weight, and that was reflected in a tendency to lose lean, fat, and bone mass. Only the loss in lean leg mass was statistically significant, however. Strengths of this study include the availability of weight, body composition, dietary intake, and physical activity data—a set of variables that are rarely available on the same sample of women undergoing chemotherapy for breast cancer. The limitations of the study stem from the homogeneity of demographic and treatment variables and the relatively narrow range of patient age, which restrict the ability to generalize our findings to breast cancer patients who are older, nonwhite, or do not receive chemotherapy. Confirmation of our findings in a broader population will be important. Although our study suggests that differential change in physical activity contributes to the differential changes in weight and body composition shown by women of different baseline BMI, this result was not statistically significant in our relatively small sample. Future studies should confirm this result and address the question of why overweight and obese women might be more likely than women of normal weight to increase activity following a breast cancer diagnosis.

Conclusion Women receiving chemotherapy for breast cancer who are of normal weight at the time of breast cancer diagnosis are more likely to gain weight and body fat during the following year than overweight or obese women. Women of normal weight should receive particular education regarding the importance of maintaining their healthy weight and increasing exercise during the first year after diagnosis.

Clinical Breast Cancer February 2011

|

59

Changes in Weight and Body Composition Acknowledgments The authors thank Elsie Anderson, RN; John Schousboe, MD, MPH; and Joseph Leach, MD, for their assistance with the study. Funding was provided by a grant from Park Nicollet Foundation and by Novartis Pharmaceutical Corporation.

17. 18. 19. 20.

Disclosures The authors have no relevant relationships to disclose.

References 1. Ganz PA, Desmond KA, Leedham B, et al. Quality of life in long-term, diseasefree survivors of breast cancer: a follow-up study. J Natl Cancer Inst 2002; 94:3949. 2. Dixon JK, Moritz DA, Baker FL. Breast cancer and weight gain: an unexpected finding. Oncol Nurs Forum 1978; 5:5-7. 3. Camoriano JK, Loprinzi CL, Ingle JN, et al. Weight change in women treated with adjuvant therapy or observed following mastectomy for node-positive breast cancer. J Clin Oncol 1990; 8:1327-34. 4. Chlebowski RT, Weiner JM, Reynolds R, et al. Long-term survival following relapse after 5-FU but not CMF adjuvant breast cancer therapy. Breast Cancer Res Treat 1986; 7:23-9. 5. Chlebowski RT, Aiello E, McTiernan A. Weight loss in breast cancer patient management. J Clin Oncol 2002; 20:1128-43. 6. Kroenke CH, Chen WY, Rosner B, et al. Weight, weight gain, and survival after breast cancer diagnosis. J Clin Oncol 2005; 23:1370-8. 7. Demark-Wahnefried W, Peterson BL, Winer EP, et al. Changes in weight, body composition, and factors influencing energy balance among premenopausal breast cancer patients receiving adjuvant chemotherapy. J Clin Oncol 2001; 19:2381-9. 8. Ingram C, Brown JK. Patterns of weight and body composition change in premenopausal women with early stage breast cancer: has weight gain been overestimated? Cancer Nurs 2004; 27:483-90. 9. Irwin ML, McTiernan A, Baumgartner RN, et al. Changes in body fat and weight after a breast cancer diagnosis: influence of demographic, prognostic, and lifestyle factors. J Clin Oncol 2005; 23:774-82. 10. Freedman RJ, Aziz N, Albanes D, et al. Weight and body composition changes during and after adjuvant chemotherapy in women with breast cancer. J Clin Endocrinol Metab 2004; 89:2248-53. 11. Goodwin PJ, Ennis J, Pritchard KI. Adjuvant treatment and onset of menopause predicts weight gain after breast cancer diagnosis. J Clin Oncol 1999; 17:120-9. 12. Heideman WH, Russell NS, Gundy C, et al. The frequency, magnitude and timing of post-diagnosis body weight gain in Dutch breast cancer survivors. Eur J Cancer 2009; 45:119-26. 13. Caan BJ, Kwan ML, Hartzell G, et al. Pre-diagnosis body mass index, post-diagnosis weight change, and prognosis among women with early stage breast cancer. Cancer Causes Control 2008; 19:1319-28. 14. Rock CL, Flatt SW, Newman V, et al. Factors associated with weight gain in women after diagnosis of breast cancer. J Am Diet Assoc 1999; 99:1212-8. 15. Makari-Judson G, Judson CH, Mertens WC. Longitudinal patterns of weight gain after breast cancer diagnosis: observations beyond the first year. Breast J 2007; 13:258-65. 16. Saquib N, Flatt SW, Natarajan L. Weight gain and recovery of pre-cancer weight

60

| Clinical Breast Cancer

February 2011

21. 22.

23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.

after breast cancer treatments: evidence from the women’s healthy eating and living (WHEL) study. Breast Cancer Res Treat 2007; 105:177-86. Han H-S, Lee K-W, Kim JH, et al. Weight changes after adjuvant treatment in Korean women with early breast cancer. Breast Cancer Res Treat 2009; 114:147-53. Heasman KZ, Sutherland HJ, Campbell JA, et al. Weight gain during adjuvant chemotherapy for breast cancer. Breast Cancer Res Treat 1985; 5:195-200. Hoskin PJ, Ashley S, Yarnold JR. Weight gain after primary surgery for breast cancer – effect of tamoxifen. Breast Cancer Res Treat 1992; 22:129-32. Loprinzi CL, Athmann LM, Kardinal CG, et al. Randomized trial of dietician counseling to try to prevent weight gain associated with breast cancer adjuvant chemotherapy. Oncology 1996; 53:228-32. Tredan O, Bajard A, Meunier A, et al. Body weight changes in women receiving adjuvant chemotherapy for breast cancer: A French prospective study. Clin Nutr 2010; 29:187-91. U.S. Department of Agriculture, Agricultural Research Service, Beltsville Human Nutrition Research Center, Food Surveys Research Group (Beltsville, MD) and U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics (Hyattsville, MD). What We Eat in America, NHANES 1987-1988; 1994-1996; 2003-2004. Zemel MB. Role of calcium and dietary products in energy partitioning and weight management. Am J Clin Nutr 2004; 79:907s-12s. Demark-Wahnefried W, Case LD, Blackwell K. Results of a diet/exercise feasibility trial to prevent adverse body composition change in breast cancer patients on adjuvant chemotherapy. Clin Breast Cancer 2008; 8:70-9. Swenson KK, Nissen MJ, Anderson E, et al. Effects of exercise versus bisphosphonates on bone mineral density in breast cancer patients receiving chemotherapy. J Support Oncol 2009; 7:101-7. Swenson KK, Nissen MJ, Henly SJ. Physical activity in women receiving chemotherapy for breast cancer: Adherence to a walking intervention protocol. Oncol Nurs Forum 2010; 37:321-30. Paffenbarger RS, Blair SN, Lee IM, et al. Measurement of physical activity to assess health effects in free-living populations. Med Sci Sports Exerc 1993; 25:60-70. Cauley JA, LaPorte RE, Sandler RB, et al. Comparison of methods to measure physical activity in postmenopausal women. Am J Clin Nutr 1987; 45:14-22. Paffenbarger RS Jr, Kampert JB, Lee IM, et al. Changes in physical activity and other lifeway patterns influencing longevity. Med Sci Sports Exerc 1994; 26:857-65. Winters-Hart CS, Brach JS, Storti KL, et al. Validity of a questionnaire to assess historical physical activity in older women. Med Sci Sports Exerc 2004; 36:2082-7. Sternfeld B, Wang H, Quesenberry CP, et al. Physical activity and changes in weight and waist circumference in midlife women: Findings from the study of women’s health across the nation. Am J Epidemiol 2004; 160:912-22. Williamson DF, Kahn HS, Remington PL, et al. The 10-year incidence of overweight and major weight gain in US adults. Arch Intern Med 1990; 150:665-72. Shah M, Hannah PJ, Jeffery RW. Secular trend in body mass index in the adult population of three communities from the upper mid-western part of the USA: the Minnesota Heart Health Program. Int J Obes 1991; 15:499-503. Lewis CE, Smith DE, Wallace DD, et al. Seven-year trends in body weight and associations with lifestyle and behavioral characteristics in black and white young adults: the CARDIA study. Am J Public Health 1997; 87:635-42. Chlebowski RT. Obesity and early-stage breast cancer. J Clin Oncol 2005; 23:1345-7. Demark-Wahnefried W, Hars V, Conaway MR, et al. Reduced rates of metabolism and decreased physical activity in breast cancer patients receiving adjuvant chemotherapy. Am J Clin Nutr 1997; 65:1495-501. Irwin ML, Crumley D, Mctiernan A, et al. Physical activity levels before and after a diagnosis of breast carcinoma. Cancer 2003; 97:1746-57.