What question did this study set out to answer?

Explore the impact of a weight loss intervention on weight trajectories in women with stage II-III breast cancer.

February 19, 2026

Abstract PD8-01: Weight Trajectories in the Breast Cancer Weight Loss (BWEL) Trial (Alliance)

Q: What does this research mean for the field?

A telephone-based weight loss intervention reduces weight by 6.1% at 40 weeks and maintains a 4.4% loss at 104 weeks in breast cancer patients with a BMI ≥27. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Key Result

A telephone-based weight loss intervention reduced weight by 6.1% at 40 weeks and maintained a 4.4% loss at 104 weeks in breast cancer patients with BMI≥27.

Key Points

Explore the impact of a weight loss intervention on weight trajectories in women with stage II-III breast cancer.
Conducted a phase 3 randomized trial with women having a BMI ≥27 kg/m2.
Participants received either weight loss intervention with health education or health education alone.
Collected weight and height data at baseline, 6, and 24 months using Fitbits and self-reports.
Applied linear mixed-effect models and principal component analysis to assess weight trajectories.
Mean weight decreased from 91.4 kg at baseline to 85.6 kg by week 40, with a peak loss of 6.1%.
Distinct weight trajectories identified: 67.6% had 6.6% peak loss, 10% had 17.5% loss, and 22.4% gained weight.
Factors influencing weight loss included education level and menopausal status.

Structured PICO

Does a telephone-based weight loss intervention reduce weight in women with stage II-III breast cancer and BMI ≥27 kg/m2?

Population

3,180 women with stage II-III HER2-negative breast cancer within 16 months of diagnosis, BMI ≥27 kg/m2, who completed chemotherapy and radiation (if given). Analyzed cohort includes 1,508 intervention participants with Fitbit weight records (mean age 53.2, mean BMI 34.4).

Intervention

Telephone-based weight loss intervention (WLI) focused on caloric restriction and exercise, plus health education (HE), delivered via 42 phone calls over a 2-year period, along with a Fitbit wireless scale.

Comparator

Health education (HE) alone.

Outcome

Trajectories of weight change over the 104-week weight loss intervention period.surrogate

A 2-year telephone-based weight loss intervention in breast cancer survivors achieved a peak weight loss of 6.1% at 40 weeks, demonstrating the feasibility of using digital health tools to track outcomes and identifying distinct patient response trajectories.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Abstract Background: Obesity is a poor prognostic factor in early breast cancer (BC). The BWEL trial (Alliance A011401; NCT02750826) is a phase 3 randomized trial testing the impact of a weight loss intervention (WLI) on invasive disease-free survival in women with stage II-III BC and a body mass index (BMI) ≥27 kg/m2. Here we report trajectories of weight change in BWEL WLI participants (pts). Methods: Eligible pts were within 16 months of diagnosis of stage II-III HER2-negative BC, had completed chemotherapy (chemo) and radiation (if given), and were randomized 1:1 to a telephone-based WLI, focused on caloric restriction and exercise, plus health education (HE) or to an HE-alone control group. WLI pts were assigned a trained coach at the BWEL call center and received 42 phone calls over the 2-year (yr) WLI period. Weight and height at baseline, 6 and 24 months were collected by study staff. All WLI pts were provided with a Fitbit wireless scale. Fitbit weight records (FWRs) were automatically imported to the BWEL portal for pts with wireless access and self-reported in the portal for pts without wireless access. FWRs were averaged (avg) by week (wk) from the WLI start date up to 104 wks. Baseline pt factors were self-reported; treatment history was abstracted from the medical record. Linear mixed-effect models were used to evaluate trajectories of weight change over the 104-wk WLI. Principal component analysis and multinomial models were used to examine distinct clusters of weight trajectories. Results: 3180 pts were enrolled from 637 sites between 8/2016 and 2/2021. FWRs were available from 1508 of 1591 WLI pts, with a total of 366,294 FWRs collected during the 2-yr WLI period. In pts with FWRs, mean baseline BMI was 34.4 ± 5.6 kg/m2, mean age was 53.2 ± 10.3 yrs, 10.8% self-identified as Non-Hispanic (NH) Black and 6.1% as Hispanic, 81.1% received chemo and 56.9% were postmenopausal. No statistically significant differences in baseline factors were observed between pts with and without FWRs. Intraclass correlation coefficient of baseline weights collected by study staff and by Fitbit scales was 0.99. In the WLI cohort, avg weekly weight (kg) decreased from 91.4 (95% CI: 90.6-92.3) at baseline to 85.6 (95% CI: 84.8-86.5) by wk 40, corresponding to a peak loss of 6.1% (95% CI: 5.7-6.4) of baseline weight. Weight remained stable until wk 52 and then gradually increased to 87.0 kg (95% CI: 86.2-87.9) at wk 104, corresponding to a 4.4% (95% CI: 4.0-4.8) loss at wk 104 vs. baseline. Principal component analysis identified 3 distinct weight trajectories: 67.6% of pts experienced peak weight loss of 6.6% (95% CI: 6.3-6.9) of baseline weight, 10.0% experienced peak loss of 17.5% (95% CI: 16.8-18.2) of baseline weight, and 22.4% did not achieve weight loss, gaining on avg 2.2% (95% CI: 1.7-2.6) of baseline weight. Pts with college (relative risk ratio RRR=1.41, 95% CI: 1.03-1.92) and post-graduate education (RRR= 1.58, 95% CI: 1.13-2.20) and those who were premenopausal (RRR=1.70, 95% CI: 1.28-2.26) were more likely to be in the cluster that did not lose weight. Pts who were NH White (RRR=3.60, 95% CI: 1.95-6.68), postmenopausal (RRR=1.94, 95% CI: 1.28-2.95), had no alcohol use (RRR=1.49, 95% CI: 1.03-2.16), had not received chemo (RRR=1.76, 95% CI: 1.15-2.72), and had 3 chronic conditions (RRR: 2.12, 95% CI: 1.23-3.66) were more likely to be in the cluster with the highest weight loss. Conclusion: Longitudinal weight data from BWEL demonstrate peak weight loss of 6.1% in the WLI cohort at 40 wks, with distinct trajectories indicating differential responses to the WLI. This work demonstrates the feasibility of using digital health tools to track outcomes in the setting of a large-scale weight loss intervention study and informs future precision behavioral interventions in oncology. Support: U10CA180821, U10CA180882, UG1CA189823; U10CA180863. https://acknowledgments.alliancefound.org. Citation Format: C. Cao, K. Ballman, L. McCall, C. Alfano, V. Bernstein, T. Crane, L. Delahanty, L. Frank, P. Goodwin, O. Hahn, D. Hershman, J. Hopkins, M. Irwin, E. Mayer, L. Minasian, L. Nebeling, M. Neuhouser, E. Paskett, P. Spears, V. Stearns, C. Thomson, T. Wadden, A. Weiss, J. White, C. Hudis, E. Winer, A. Partridge, L. Carey, J. Ligibel. Weight Trajectories in the Breast Cancer Weight Loss (BWEL) Trial (Alliance) abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PD8-01.

Bookmark