Abstract 2232: Association between sleep characteristics and cancer survival: Findings from the UK Biobank study

Abstract Background Sleep characteristics have been associated with patient outcomes after cancer. We aimed to appraise the causal relationship of chronotype, sleep duration, insomnia, and snoring with cancer survival. Methods This large prospective cohort study enrolled 38350 cancer survivors aged ≥40 years from the UK Biobank between 2004 and 2022. Habitual sleep characteristics were obtained from the self-reported baseline questionnaire, including chronotype, sleep duration, insomnia, and snoring. Cox proportional hazards regression models were applied to evaluate the association between sleep characteristics and cancer survival, then one sample Mendelian randomization (MR) analyses were further used to validate these causal relationships. In particular, restricted cubic splines (RCS) and non-linear MR analyses were applied to assess the possible nonlinear association between sleep duration and cancer survival. Results Among 38350 study participants (mean [SD] age, 59.8 [7.3] years; 24150 [63.0%] males; 35534 [92.7%] White individuals), 12671 (33.1%) reported evening preference chronotype; 9050 (23.6%) reported short and 3773 (9.8%) reported long sleep duration;12274 (32.0%) reported snoring; and 30426 (79.4%) reported experiencing insomnia sometimes or usually. During up to 18 years of follow-up, 6274 deaths occurred. Multivariable regression models showed that evening preference (HR = 1.09, 95% CI: 1.03 to 1.15), short (HR = 1.14, 95% CI: 1.07 to 1.21) and long sleep duration (HR = 1.37, 95% CI: 1.28 to 1.48), insomnia (HR = 1.14, 95% CI: 1.09 to 1.21) and snoring (HR = 0.83, 95% CI: 0.78 to 0.87) were associated with death among pan to cancer survivors. Restricted cubic splines suggested nonlinear associations between sleep duration and death after cancer (nonlinear P < 0.001). In the one-sample MR, genetically predicted sleep duration (HR = 0.71, 95% CI: 0.56 to 0.90) and snoring (HR = 1.18, 95% CI: 1.05 to 1.34) demonstrated a reversed cancer survival pattern compared to multivariable regressions. A nonlinear U-shape association between sleep duration and cancer survival was observed in restricted cubic spline analyses (P < 0.001) but not evident when evaluating by nonlinear MR (P > 0.05). Conclusions Evening preference chronotype, insomnia, snoring, and short sleep duration are likely to be causally associated with a higher risk of death among cancer survivors, whilst long sleep duration does not appear to be a causal factor. Citation Format: Jiajing Che, Jiali Lv, Keyu Pan, Tiantian Sun, Chao Cao, Shengxu Li, Tao Zhang, Fuzhong Xue, Lin Yang. Association between sleep characteristics and cancer survival: Findings from the UK Biobank study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2232.