- Click here for explanation of classification scheme.

To determine the reasons for weight gain in women with early breast cancer receiving adjuvant chemotherapy by assessing changes in body composition and energy balance (REE, physical activity, energy intake) during chemotherapy and in the 6 months after chemotherapy.

• Women with newly diagnosed invasive breast cancer
• Breast surgery (mastectomy or wide local excision)
• No endocrine abnormalities (diabetes mellitus, hyper/hypothyroidism)
• No psychiatric morbidity

Not discussed

Recruitment

• Over the course of a year, 44 patients were recruited by telephone within 3 weeks of breast surgery (mastectomy or wide local excision) and immediately prior to commencing adjuvant chemotherapy. 
• Subjects were assigned to groups depending on their desire to undergo metabolic measurements.

 

Design Time series, concurrent comparison group

 

 Blinding used  None

 

Intervention

• Medical intervention
  • All women received adjuvant chemotherapy following mastectomy or wide local excision
  • Those undergoing wide excision also received 3 weeks of radiotherapy following chemotherapy
  • Chemotherapy regimens
    • 5-FU, epirubicin, cyclophosphamide: 6 x 28 day cycles
    • Cyclophosphamide, methotrexate, 5-FU: 6 x 28 day cycles
  • Ondansetron was used as an antiemetic rather than the usual 2-3 days of steroids
• Groups
  • Non-monitored control group
    • No metabolic testing performed
    • Weight change over the year was documented to control for the intensive metabolic monitoring in the intervention group which might depress energy intake and attenuate weight gain.
  • Study group
    • Assessed every 3 months throughout the year from the commencement of chemotherapy
    • Measurements included:
      • anthropometry, body mass and composition determination
      • Total body potassium
      • Bioelectrical impedance
      • Energy (dietary) intake assessment
      • Energy expenditure
      • Physical activity assessment
  • Healthy volunteers
    • 21 women matched for age (± 2 yr) and weight (± 5 kg) were recruited from hospital staff as controls for the REE measurement
• Nutrition education
  • No nutritional counseling was given other than general healthy eating advice.

 

Statistical Analysis

• The independent t-test was used to compare change in weight over the year between adjuvant patients in the study and in the non-monitored control group.
• REE in adjuvant patients was compared with 21 healthy women matched for age and weight using the independent t-test.
• Significance was accepted at 0.05 level of probability.
• Changes over time in the adjuvant group were determined from paired t-tests between baseline and the end of chemotherapy, and baseline and 1 year. Bonferroni adjustments were made to allow for the multiple testing.
• No statistical analyses were performed on measurements taken at mid chemotherapy and 3 months postchemotherapy.

 

Timing of Measurements

	Baseline	Prior to 3rd chemotherapy cycle	1 mo after final chemotherapy	9 mo from start of chemotherapy	12 mo from start of chemotherapy
Height	x
Weight	x	x	x	x	x
BMI	x	x	x	x	x
Waist circumference	x	x	x	x	x
Hip circumference	x	x	x	x	x
Skinfolds	x	x	x	x	x
Total body potassium	x	x			x
Bioelectric impedance	x	x	x	x	x
Dietary intake	x	x	x	x	x
REE	x	x	x	x	x
Physical activity	x	x	x	x	x

 Also at baseline:

• Weight change over the 6 months prior to diagnosis of breast cancer and weight change from diagnosis, were estimated from recalled weights.

Dependent Variables

• Primary
  • Weight
    • Measured using a beam balance/stadiometer to the nearest 0.1 kg
  • Total body fat
    • Determined from body weight and skin fold measurements using the equations by Durnin and Womersly.
    • Skinfolds were measured at the biceps, triceps, subscapular, supra-iliac and abdominal sites with Harpenden calipers by a single investigator. Each measurement was repeated 3 times and the mean value was calculated.
    • Coefficient of variation for the determination of fat by skinfold measurements on different occasions was 1.1%.
  • Fat free mass (FFM)
    • Determined from body weight and skin fold measurements using the equations by Durnin and Womersly.
    • Coefficient of variation for the determination of FFM by skinfold measurements on different occasions was 0.08%.
  • Waist circumference
    • Measured by a single investigator with a tape measure to the nearest 1.0 mm.
  • Hip circumference
    • Measured by a single investigator with a tape measure to the nearest 1.0 mm.
  • Abdominal skinfold (see measurement description under total body fat)
  • REE
    • Determined by indirect calorimetry using a calibrated open circuit ventilated hood system (Deltatrac Metabolic Monitor MBM 100).
    • Manufactured maintained the instrument and periodic alcohol burns gave results all within 1% of predicted.
    • Measured under controlled conditions: following an overnight fast; abstention from caffeine for previous 12 hr, smoking for previous 24 hr; measurements were made a minimum of 15 min once steady state was reached; energy expenditure calculated with Weir equatio
  • Energy intake
    • Determined from 4 day weighed diaries completed before each assessment, in the week prior to the next chemotherapy cycle to avoid the acute effects of the treatment on dietary intake.
    • Records were reviewed with subjects to check for completeness and to determine unspecified portion sizes
  • Physical activity index (PAI)
    • During the week prior to each assessment, physical activity was quantified using the self-reported patient questionnaire of the Framingham study with records the numbers of hours spent asleep, at rest, work, and leisure.
    • Clarified questionnaire with subject and checked for completeness at each visit.
    • Hours at each level of activity were weighted by relative enrgy expenditure for activities and were summed to give a PAI.
• Secondary
  • Bioelectrical impedance
    • Measure of total body water, FFM and fat mass
    • Measurement conditions: apyrexial, fasted for 12 hr, and supine for 10 min prior to measurement
  • Total body potassium (TBK)
    • Measure of lean body mass
    • Relative changes in ⁴⁰K were determined.
    • Sensitivity of the detectors and instrumental stability were determined after each measurement by performing a static count of potassium standard.
    • The measured values were corrected for changes in sensitivity of the scanner and any change in body size of the patient between measurements using a correction factor.
    • Precision of the measurement was 2% (coefficient of variation).

  

Independent Variables

 

Control Variables

 

 

Initial N:

• Study group: 21
• Non-monitored controls: 23  

Attrition (final N):

• Study group: 17
  • 3 withdrew because of work and family commitments
  • 1 developed metastatic disease and die
• Non-monitored controls: 21
  • 1 developed metastatic disease and withdrew
  • 1 received follow up at a different hospital

Subject Characteristeristics

	Study Group (N=17)	Controls (N=21)
Age (yr)	46.1 (9.5)*	45.0 (7.4)*
Smoker	21%	27%
Premenopausal	76%	86%
Chemotherapy induced menopause	70%	63%
Change in wt in the 2-4 wks since diagnosis (kg)	0.0 (-3.6-5.7)**	0.0 (-4.4-6.5)**
Change in wt in the 6 months prior to diagnosis	0.0 (-9.1-7.0)**	0.0 (-10-6.5)**
Weight (kg)	70.3 (11.3)*	67.0 (12.2)*
Height (m)	1.60 (0.07)*	1.62 (0.08)*
BMI	27.1 (4.1)*	25.3 (3.7)*
Tumor size <5 cm:>5 cm	16:1	19.3
WLE:mastectomy	9:8	14:8
grade of tumor 2:3	7:10	8:14
Chemotherapy regimen	12 FECb	15 FEC
	5 CMFa	7 CMF

* Mean (SD)

** Median (range)

^aCMF 6 x 28 day cycles; CMF administered day 1 and 8 of each cycle

^bFEC 6 x 28 day cycles; FEC administered day 1, F & C only administered day 8

C=cyclophosphamide; M=methotrexate; F=fluouracil; E=epirubicin

WLE=wide local excision

Age: see table

Ethnicity: not discussed

Other relevant demographics: see table

Anthropometrics see table

Location: South Manchester University Hospitals NHS Trust, Manchester, UK

 

 

 

Variables	Study Group Mean ± SD	Non-Monitored Control group Mean ± SD	Statistical Significance of Group Difference
Weight change over the year (kg)	5.0±3.9	2.6±4.2	P=0.11

1.  Refutes a priori hypothesis that intensive metabolic investigations would limit weight gain in the study group.

Prechemotherapy and change in body mass and indices of body composition over the year in women receiving adjuvant chemotherapy

	Prechemotherapy Mean (SD)	Immediately Postchemotherapy Mean (95% CI)	1 year Mean (95% CI
Weight (kg)	70.3 (11.3)	3.3 (1.4-5.3)*	5.0 (3.0-6.9)**
Body fat (%)	35.9 (5.8)	3.6 (2.2-4.9)**	6.2 (4.9-7.4)**
Total fat (kg)	25.8 (7.6)	4.0 (2.3-5.8)**	7.1 (4.8-9.5)**
FFM (kg)	44.5 (4.6)	-0.5 (-1.6-0.6)	-1.7 (-0.4 to -3.0)*
Total body K counts	908 (114)	47 (-12-111)	108 (21-196)
Bioelectrical impedance	567 (61)	-14 (-32 to -3)	-27 (-8 to -47)
Abdominal skinfold (mm)	38.0 (15)	11 (6-17)**	16 (11-22)**
Waist (cm)	84.8 (10.5)	3.5 (1.4-5.6)*	5.1 (2.8-7.4)**
Hip (cm)	104.8 (8.7)	2.9 (-0.7-5.0)	3.8 (1.1-6.5)

*Significant differences from prechemotherapy; P<0.05 with Bonferroni adjustment

**Significant differences from prechemotherapy; P<0.01 with Bonferroni adjustment

1. Weight increased 3.3 kg from baseline weight during adjuvant chemotherapy (P<0.05) and continued to increase in the 6 months postchemotherapy so that at 1 year subjects had gained 5.0 kg from baseline (P<0.001).
2. There was a significant increased in body fat during adjuvant chemotherapy (4.0 kg; P<0.01) and in the 6 months postchemotherapy so that at 1 year subjects had gained 7.1 kg of body fat (P<0.01).
3. Abdominal fat stores continued to increase in the postchemotherapy period so that at 1 year subjects had experienced significant increased in both waist (5.1 cm; P<0.01) and abdominal skinfold (16 mm; P<0.01).
4. There appeared to be an increase in hip circumference both during and in the 6 months postadjuvant treatment, but these changes were not statistically significant.
5. There was not significant change in FFM (determined from skinfolds) over the course of chemotherapy (P=0.89), but there was a decline in FFM in the 6 months posttreatment; 1.72±2.5 (P<0.05).
6. There was no significant change in total body potassium over the course of chemotherapy (P=0.93) or in the 6 months posttreatment (P=0.34)
7. There was no consistent pattern of change of impedance over the study. None of the changes were statistically significant.

 

Prechemotherapy and change in energy balance over the year in women receiving adjuvant chemotherapy

	Prechemotherapy Mean (SD)	Immediately Post-chemotherapy Mean (95% CI)	1 year Mean (95% CI)
REE (kJ)	5893 (504)	-151* -363 to -60	-8.3 (-253-237)
REE/kg FFM (kJ/kg)	134.0 (13.7)	-1.8 (-7.2-3.5)	5 (-2-12)
Energy intake (kJ)	8091 (1628)	6 -863-875)	-207 (-1201-786)
PAI	30.7 (2.6)	-0.5  (-2.1-1.1)	2.1 (-0.2-4.4)
Energy cost of physical activity (kJ)	1606 (844)	-190 (-654-274)	523* (-99-1146)

 *Significant differences from prechemotherapy P<0.05 with Bonferroni adjustment

1. Subjects who were to receive adjuvant chemotherapy had a significantly greater REE than the healthy age and weight matched control subjects. Mean (SD) REE in the adjuvant group was 100.5±8.0% Harris Benedict compared to 94.5±8.4% Harris Benedict in the 21 age and weight matched volunteers (P=0.05).
2. REE was 151 kJ lower at the end of chemotherapy (p<0.05) and returned to prechemotherapy values at 1 year (P=0.94).
3. There was no difference in REE/kg FFM at the end of chemotherapy (p=0.49) or at 1 year (P=0.15).
4. There appeared to be a decrease in energy intake at 1 year (P=0.3).
5. The PAI appeared to be lower by the end of chemotherapy (P=0.47) and to increase by the end fo the year (P=0.23).
6. The energy cost of physical activity appeared to decrease at the end of chemotherapy (P=0.58), but was significantly higher at the end of the year (P<0.05). The increase at 1 year was a function of a slight increase in activity level from baseline, but mainly reflected the increased energy cost of exercise at the higher body weight.

Other Findings

 

Treatment with adjuvant chemotherapy causes gain of body fat because of reduced energy expenditure, and the failure of women to reduce thier energy intake to compensate for the decline in energy requirement during and in the 6 months posttreatment.

Industry:

Scientific Hospital Supplies Other:

Whether subjects entered the study or non-monitored control groups depended largely on proximity to the hospital and the ability to attend regularly. Are the groups comparable?

Investigators didn't discuss how weight measurements were obtained throughout the year for the non-monitored control group.

Small sample size with no sample size calculation.

This study mainly compared changes in the adjuvant chemotherapy group before chemotherapy and during the first year. The other groups (non-monitored control group and the healthy volunteers) played a limited role in the study.