Accelerated epigenetic aging and risk of chemotoxicity in older adults with early breast cancer.

person Jingran Ji City of Hope, Duarte, CA info_outline Jingran Ji, Can-Lan Sun, Alexis Ann LeVee, Mina S. Sedrak

Authors person Jingran Ji City of Hope, Duarte, CA info_outline Jingran Ji, Can-Lan Sun, Alexis Ann LeVee, Mina S. Sedrak Organizations City of Hope, Duarte, CA, City of Hope National Medical Center, Duarte, CA, City of Hope Comprehensive Cancer Center, Duarte, CA, UCLA Health Jonsson Comprehensive Cancer Center, Los Angeles, CA Abstract Disclosures Research Funding Conquer Cancer, the ASCO Foundation National Institute of Aging, National Cancer Institute Background: Older adults with breast cancer have considerable heterogeneity in their risk of chemotoxicity; two people with the same chronological age have different risks of chemotoxicity. This heterogeneity in chemotoxicity may be due to older adults having variable rates of biological aging. Some older adults age faster biologically compared with their chronological age (i.e., accelerated epigenetic aging [AEA]), while others do not. AEA is a blood biomarker of aging, but its utility as a biomarker of chemotoxicity is unknown. Here, we evaluated the association between AEA and risk of chemotoxicity in older adults with early breast cancer. Methods: In a prospective multicenter study of 397 women age >65 with stage I-III breast cancer treated with neo/adjuvant chemo, we extracted genomic DNA from blood to estimate the participant’s rate of biological aging at baseline (prior to chemotherapy). Using two established epigenetic clocks (AgeAccelGrim and DunedinPACE), we classified individuals as either having AEA or not (dichotomized as a yes/no variable) for each clock. Our primary endpoint was grade 3+ chemotoxicity (yes/no, yes defined as any grade 3+ toxicity attributed to chemo). Using multivariable logistic regression, we examined the association between AEA and grade 3+ chemotoxicity, adjusting for sociodemographic, geriatric, and clinical covariates. Results: The median (range) pretreatment chronological age was 70 (65-85), AgeAccelGrim was -0.88 (-10.4-15.7), and DunedinPACE was 1.03 (0.69-1.45). Of the 397 women, 31 (8%) had AEA per AgeAccelGrim, 63 (16%) had AEA per DunedinPACE, and 184 (46%) experienced grade 3+ chemotoxicity. Women who had evidence of AEA as measured by AgeAccelGrim (≥6.83 years, adjusted OR = 2.73, 95% CI 1.17-6.41, p = 0.02) or DunedinPACE (≥1.17 biological years per chronological year, adjusted OR = 2.20, 95% CI 1.18-4.10, p = 0.01) had greater odds of having grade 3+ chemotoxicity compared to women without AEA (AgeAccelGrim <6.83 or DunedinPACE <1.17, respectively). Conclusions: In this cohort of older women with early breast cancer, those with pre-chemo AEA had increased odds of chemotoxicity compared to women without any evidence of AEA. Future research is needed to examine whether measures of biological age can be translated to the clinical care of older patients with breast cancer. Clinical trial information: NCT01472094. Grade 3+ Chemotoxicity Multivariable Analysis* Yes No Total OR (95% CI) p-value n=184 n=213 N=397 AEA by AgeAccelGrim (years) <6.83 163 (45%) 203 (56%) 366 1.00 ≥6.83 21 (68%) 10 (32%) 31 2.73 (1.17-6.41) 0.021 AEA by DunedinPACE (biological years per chronological year) <1.17 145 (43%) 189 (57%) 334 1.00 ≥1.17 39 (62%) 24 (38%) 63 2.20 (1.18-4.10) 0.013 *Adjusted for age, BMI, race/ethnicity, education, geriatric assessment, and no. of comorbidities.

Clinical