Obesity and risk of recurrence or death after adjuvant endocrine therapy with letrozole or tamoxifen in the breast international group 1-98 trial

Abstract

Purpose: To examine the association of baseline body mass index (BMI) with the risk of recurrence or death in postmenopausal women with early-stage breast cancer receiving adjuvant tamoxifen or letrozole in the Breast International Group (BIG) 1-98 trial at 8.7 years of median follow-up.

Patients and methods: This report analyzes 4,760 patients with breast cancer randomly assigned to 5 years of monotherapy with letrozole or tamoxifen in the BIG 1-98 trial with available information on BMI at randomization. Multivariable Cox modeling assessed the association of BMI with disease-free survival, overall survival (OS), breast cancer-free interval, and distant recurrence-free interval and tested for treatment-by-BMI interaction. Median follow-up was 8.7 years.

Results: Seventeen percent of patients have died. Obese patients (BMI ≥ 30 kg/m(2)) had slightly poorer OS (hazard ratio [HR] = 1.19; 95% CI, 0.99 to 1.44) than patients with normal BMI (< 25 kg/m(2)), whereas no trend in OS was observed in overweight (BMI 25 to < 30 kg/m(2)) versus normal-weight patients (HR = 1.02; 95% CI, 0.86 to 1.20). Treatment-by-BMI interactions were not statistically significant. The HRs for OS comparing obese versus normal BMI were HR = 1.22 (95% CI, 0.93 to 1.60) and HR = 1.18 (95% CI, 0.91 to 1.52) in the letrozole and tamoxifen groups, respectively.

Conclusion: There was no evidence that the benefit of letrozole over tamoxifen differed according to patients' BMI.

Fig 1.

CONSORT diagram showing the analytic cohort of 4,760 patients enrolled in the BIG 1-98 clinical trial.

Fig 2.

(A) Overall survival according the three body mass index (BMI) groups is shown in the Kaplan-Meier curve; P value tests for heterogeneity among the three groups with 2 df. (B) Cumulative incidence of breast cancer recurrence and (C) distant recurrence comparing normal, overweight, and obese BMI groups. Comparisons of BMI groups using Gray's test P value stratified by chemotherapy use and random assignment option.

Fig 3.

Hazard ratios (HRs), 95% CIs, and P values from multivariate Cox models comparing (A) obese, body mass index (BMI) ≥ 30 kg/m² and (B) overweight, BMI 25 to less than 30 kg/m², with normal-weight patients, BMI less than 25 kg/m², adjusted for patient/disease characteristics and stratified by chemotherapy use and randomization option. P values are for 1-df tests of treatment-by-BMI interaction from multivariate Cox model that adjusted for prognostic factors and stratified by chemotherapy use and random assignment option. The size of the boxes is inversely proportional to the SE of the HR.

Fig 4.

Hazard ratios (HRs) and 95% CIs from multivariate Cox models comparing letrozole versus tamoxifen according to body mass index (BMI) groups, adjusted for patient/disease characteristics and stratified by chemotherapy use and random assignment option. The size of the boxes is inversely proportional to the SE of the hazard ratio.

Fig 5.

Subpopulation treatment effect pattern plots showing (A) 8-year overall survival according to randomly assigned treatment group across a continuum of baseline body mass index (BMI) subpopulations and (B) overall survival hazard ratio (solid line) with 95% confidence bands (dashed lines) comparing letrozole versus tamoxifen across a continuum of baseline BMI subpopulations.