Proteomic aging clock (PAC) is cross-sectionally associated with frailty in cancer survivors: The Atherosclerosis Risk in Communities (ARIC) study

Background: Accelerated aging induced by cancer and its treatment contributes to a higher prevalence of frailty in cancer survivors than individuals without cancer. Frailty in cancer survivors poses significant challenges for healthcare providers, as it increases the risk of adverse health outcomes and leads to high rates of hospitalization. However, current frailty assessments often require in-person evaluations in clinics, which can be time-consuming and difficult to perform, particularly in older cancer survivors. Therefore, a biomarker that could predict frailty is needed to facilitate risk stratification in this population. Our previous study suggested that PACs could capture accelerated aging in cancer survivors; however, no previous studies have tested PACs' associations with frailty in cancer survivors. This study examined the cross-sectional associations of a previously validated PAC (Wang et al. [2025]) with frailty in cancer survivors in the ARIC study. Methods: ARIC is an ongoing cohort of White and Black men and women initiated in 1987. At Visit 5 (2011-13), 5,000 plasma proteins were measured using SomaScan in 3,699 participants without a history of cancer (cancer-free) and 806 cancer survivors, all aged 66-90. We previously created a PAC in 67% of randomly selected cancer-free participants and validated it internally in ARIC and externally in another large cohort. We calculated age acceleration after regressing PAC on chronological age (PAC-accel). At Visit 5, ARIC assessed frailty using the cumulative frailty index (FI) and the Fried Frailty Phenotype (FFP). We examined the cross-sectional associations of PAC-accel with frailty in cancer survivors (after cancer diagnosis), using linear regression for FI and logistic regression for FFP (Frail & Pre-frail vs. Robust). All associations were adjusted for chronological age, sex, race, education, BMI, smoking status, diabetes, cardiovascular disease, and eGFR. Results: In our study, cancer survivors had a mean FI (SD) of 0.21 (0.10), and 57.8% of them were either frail or pre-frail. Considering the most common cancers (lung, colorectal, breast, prostate), both mean FI and the proportion of frail & pre-frail cancer survivors were highest among lung cancer survivors (0.28 (0.10) and 80%). In all cancer survivors, PAC-accel (per 5 years) was cross-sectionally associated with both FI (difference = 0.04, 95% CI 0.03-0.05) and FFP (OR = 2.56, 95% CI 1.77-3.70). Limited sample size precluded examining associations with frailty by cancer type. Conclusion: Our findings suggest a cross-sectional association of PAC with frailty. Our next step is to examine whether PAC is associated with future frailty risk. This research will help determine whether PAC holds promise as a tool for frailty risk stratification among older cancer survivors in clinical settings. Funding: NHLBI, NCI, NPCR