SenMayo Based Aging Signature (Sen-Age) in cancer survivors: The Health and Retirement Study (HRS)

Abstract The SenMayo gene set identifies senescent cells, one of the hallmarks of aging, especially among those with cancer. In the HRS cohort, we constructed a SenMayo based aging signature (Sen-Age) among those without cancer (controls) and examined its association with 4-year mortality in cancer survivors. In the 2016 wave, HRS measured 125 SenMayo genes in blood samples. We applied Lasso regression to participants without cancer in 2016 (N = 3076) and trained SenMayo genes against chronological age; this resulted in a subset of 97 genes with which we created Sen-Age. This Sen-Age signature demonstrated a correlation of 0.46 with chronological age in 609 cancer survivors. We then examined the associations of Sen-Age with 4-year mortality (N = 112) in cancer survivors using Cox Proportional Hazards regression model after adjusting for age, sex, and race. Sen-Age was associated with increased risk of mortality in cancer survivors (HR: 1.47 [1.17–1.85] per 1 year, p = 0.0008). This association remained robust (HR: 1.31 [1.04–1.66], p = 0.02) upon additional adjustment for education level, smoking status, BMI, inflammation and comorbidities. Moreover, cancer survivors have higher Sen-Age score (adj mean: -0.44 ) than non-cancer controls (adj mean: -0.83). Overall, our findings highlight the potential of the Sen-Age signature as a biomarker of aging and mortality in cancer survivors.