In:
Cancer Research, American Association for Cancer Research (AACR), Vol. 83, No. 7_Supplement ( 2023-04-04), p. 1185-1185
Abstract:
Chronic lymphocytic leukemia (CLL) is strongly associated with mosaic chromosomal alterations (mCAs), a type of clonal hematopoiesis characterized by large structural chromosomal changes (e.g., gains, losses and copy neutral loss of heterozygosity). We aim to identify specific chromosomal regions that, when impacted by mCAs, are most predictive of CLL risk and evaluate their predictive utility in models that include established CLL risk factors. We utilized genotype array data from 436,784 participants in the UK Biobank (UKBB) and 35,382 participants in the Prostate, Lung, Colorectal, and Ovarian (PLCO) screening trial to identify mCAs. UKBB was split (90/10) into training and test sets, and PLCO served as an external validation set. We developed sequential regression models to predict risk of CLL, with a baseline model consisting of age, age-squared, sex, smoking status, and ancestry. Additional models evaluated mCAs, a CLL polygenic risk score (PRS), and blood cell counts. Area under the receiver operating characteristic curve (AUC) was used to evaluate the predictive utility for both 5-year and 10-year CLL risk. Any detectable mCA was associated with incident CLL (Hazards Ratio (HR)=26, 95% Confidence interval (CI)=21-31), and mCAs on each individual chromosome were associated with increased CLL risk. mCAs on segments of chromosomes 13, 12, 22, 14, and 11 had the largest effect size (e.g., chr 13q14 HR=199, 95%CI=155-257). The CLL prediction model, including baseline factors, CLL PRS, and any autosomal mCA status, showed strong predictive utility in independent UKBB (AUC=0.88) and PLCO (AUC=0.88) samples. Restricting to a subset of mCAs strongly associated with CLL did not improve performance over including any detectable autosomal mCA. After the inclusion of blood cell traits, predictive utility further increased in UKBB (AUC=0.91; no count data available in PLCO). We observed the strongest CLL predictive utility in 5-year risk compared to 10-year risk across all models. mCAs genome-wide are associated with CLL risk, with variable effects across chromosomes and specific chromosomal regions. Autosomal mCAs are strong independent predictors of CLL risk and substantially add to predictive utility over standard clinical measures. The predictive utility of autosomal mCAs is strongest within the first 1-5 years of sample collection. Evaluation of mCAs could provide added utility for risk stratification in populations at high risk of CLL. Citation Format: Aubrey K. Hubbard, Alexander DePaulis, Sruthi Srinivasan, Ian D. Buller, Shu-Hong Lin, Weiyin Zhou, Stephen J. Chanock, Neal D. Freedman, Derek W. Brown, Mitchell J. Machiela. Mosaic chromosomal alterations are independent predictors of chronic lymphocytic leukemia risk [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 1185.
Type of Medium:
Online Resource
ISSN:
1538-7445
DOI:
10.1158/1538-7445.AM2023-1185
Language:
English
Publisher:
American Association for Cancer Research (AACR)
Publication Date:
2023
detail.hit.zdb_id:
2036785-5
detail.hit.zdb_id:
1432-1
detail.hit.zdb_id:
410466-3
Permalink
