Organ-specific proteomic aging clocks (PACs) and mortality in cancer survivors.

Abstract

12112 Background: Our study suggested that plasma conventional PACs can reflect, on average, accelerated aging in cancer survivors (Wang et al., JNCI 2026), but may overlook organ-specific variations. Organ-specific PACs may better predict CVD and cancer mortality in this group. Methods: We compared the associations of one conventional and 11 organ-specific PACs (Oh et al., Nature 2023) with mortality in 806 cancer survivors (aged 66-90) in the Atherosclerosis Risk in Communities (ARIC) study, 2+ yrs post-diagnosis (up to 12-yr follow-up). ARIC measured 4,712 plasma proteins by SomaScan. We computed PACs (conventional, adipose, artery, brain, heart, immune, intestine, kidney, liver, lung, muscle, pancreas) and their deviations from chronological age (age gap) after regressing PAC on age via LOWESS model. Multivariable Cox proportional hazards regression estimated hazard ratios (HRs) for mortality associated with age gap in survivors of any cancer (455 deaths: 127 CVD, 137 cancer deaths) and the most common cancers (165 breast (BrCa), 255 prostate (PrCa), and 78 colorectal cancer (CRC)). Results: In BrCa survivors, all age gaps (except intestine) were higher for those diagnosed 2-10 vs. >10 yrs before the start of follow-up. The pattern was opposite for all age gaps in CRC survivors (except heart and lung gaps). There was no specific pattern across age gaps in PrCa survivors. Further, in all cancer survivors, all age gaps, but artery and kidney, were associated with all-cause mortality (all p<0.05), with highest HRs for conventional, heart, and liver gaps (Table 1). Only heart age gap was associated with CVD mortality (HR per 1 SD=1.29, p<0.05), with no associations of any age gaps with cancer mortality. In BrCa survivors, conventional (HR=1.81) and 4 organ age gaps (brain, heart, immune, intestine; HRs = 1.45–1.68) were associated with all-cause mortality (all p<0.05). In PrCa survivors, conventional, heart, and liver age gaps were associated with all-cause mortality [all HRs=1.35, p<0.05]. In CRC survivors, conventional, heart, and immune age gaps were positively, but non-significantly, associated with all-cause mortality [HRs = 1.48–1.56], likely limited by sample size. Conclusions: Conventional and heart PACs demonstrated consistent associations with all-cause mortality beyond chronological age and lifestyle factors in total and individual cancer survivors. Heart PAC was most strongly associated with CVD mortality. Further work is needed to replicate this analysis and to test these age gaps as a predictor for mortality. Funding: NHLBI, NCI, NPCR. Age gaps and all-cause mortality in cancer survivors; ARIC (2011-23). Age gap HR (95% CI) per 1 SD Age gap HR (95% CI) per 1 SD Conventional 1.48 (1.32, 1.66) Intestine 1.13 (1.01, 1.26) Adipose 1.13 (1.00, 1.27) Liver 1.42 (1.26, 1.60) Brain 1.19 (1.05, 1.34) Lung 1.14 (1.02, 1.26) Heart 1.41 (1.26, 1.59) Muscle 1.19 (1.06, 1.33) Immune 1.24 (1.10, 1.40) Pancreas 1.33 (1.18, 1.49)