Purpose: Inflammatory and metabolic markers have been associated with prognosis in breast cancer survivors. We examined changes in prognostic biomarkers (hs-CRP and HOMA2-IR) following a weight loss intervention for breast cancer survivors and examined associations between lifestyle behaviours – measures of weight, physical activity, diet, and circadian rhythm disrupting (CRD) behaviours – with these biomarkers.
Methods: Female breast cancer survivors (n=159; 18-75 years; 25-45 kg/m2; stage I-III) were recruited to participate in a randomised controlled trial of a 12-month behaviour change (diet and physical activity) weight loss intervention versus usual care. Behaviours and biomarkers were measured at 6-monthly time points (80.5% retention at 18-months). Linear mixed-effect models were used to examine changes of biomarkers over time and intervention effects. Tobit and linear regression models were used to test for associations of behaviours with hs-CRP and HOMA2-IR respectively at study baseline and the effects of change in behaviours with change in biomarkers (12 months-baseline). Models were adjusted for confounders identified using Directed Acyclic Graphs.
Results: Statistically significant and meaningful (10% difference) improvements from baseline to 12-months were observed for both biomarkers but were not sustained at 18-months for hs-CRP for women in both study arms combined. The intervention did not lead to any significant differences between groups for either biomarker. At baseline, BMI (β=0.23, 95% CI: 0.12 to 0.34) and moderate-to-vigorous physical activity (β=-0.05, -0.08 to -0.01) were associated with hs-CRP, and BMI (β=0.07, 0.02 to 0.12), physical activity (β=-0.02, -0.03 to -0.00), and eating ≥3.5 to <6 times/day (versus ≥6 to ≤7; β=0.63, 0.09 to 1.16) with HOMA2-IR (all p <0.05). No association was observed between energy intake, sleep, and meal timing with the biomarkers. At 12-months, changes in behaviours were not associated with changes in biomarkers.
Conclusions: While lifestyle behaviours were associated with prognostic biomarkers in breast cancer survivors, neither the weight loss intervention nor behaviour change explained improvements in either biomarker.