AUTHORS
Antonio Jose Jose Cabrera Serrano, PhD, Dennis P. Robinson2, Sameer A. Parikh, MBBS3, Kari G Rabe, MS4, Cristine Allmer, BS5, Paul J. Hampel, MD6, Nicola Camp, PhD7, Esteban Braggio, PhD8, James R. Cerhan, MD, PhD9, Lara Sucheston-Campbell, PhD10, CRuCIAL Consortium11, Juan Sainz12, Susan L. Slager, PhD13, Rosalie Griffin, PhD14
AFFILIATIONS
1 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
2 Mayo Clinic, Rochester, Minnesota, United States
3 Mayo Clinic, Rochester, Minnesota, United States
4 Mayo Clinic, Rochester, Minnesota, United States
5 Mayo Clinic, Rochester, Minnesota, United States
6 Mayo Clinic, Rochester, Minnesota, United States
7 University of Utah, Salt Lake City, Utah, United States, University of Utah, Salt Lake City, Utah, United States
8 Mayo Clinic, Phoenix, Arizona, United States
9 Mayo Clinic, Rochester, Minnesota, United States, Mayo Clinic, Rochester, Minnesota, United States
10 Wayne State University and the Barbara Ann Karmanos Cancer Institute, Detroit, Michigan, United States
11 Centre for Genomics and Oncological Research: Pfizer / University of Granada / Andalusian Regional Government, Granada, Granada, Spain
12 Centre for Genomics and Oncological Research: Pfizer / University of Granada / Andalusian Regional Government, Granada, Granada, Spain, Instituto de Investigación Biosanitaria IBs.Granada, Granada, Granada, Spain, University of Granada, Granada, Granada, Spain
13 Mayo Clinic, Rochester, Minnesota, United States, Mayo Clinic, Rochester, Minnesota, United States
14 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States, The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
ABSTRACT
Introduction
Chronic lymphocytic leukemia (CLL) is an incurable neoplasm of B-cell lymphocytes. A key marker of CLL heterogeneity and prognosis is the immunoglobulin heavy chain variable (IGHV) gene. Mutational status of IGHV remains stable throughout disease progression and reflects differences in the cellular origin of CLL. Patients with unmutated IGHV (U-CLL) have shorter progression-free and overall survival than those with mutated IGHV (M-CLL). While inherited genetics is a major factor in CLL risk overall, supported by 44 inherited susceptibility variants, its relationship with IGHV mutation status is unknown. Here, we investigate inherited risk of CLL by IGHV mutation status.
Methods
We included CLL patients and unaffected controls from two studies: the Mayo Clinic (1,665 controls; 2,062 cases) and the CRuCIAL Consortium (1,062 controls; 541 cases). IGHV mutational status was obtained from electronic medical records. Genotyping data from whole blood DNA was used to conduct genome-wide association studies (GWAS), adjusting for age, sex, and population structure, within each study. Meta-analysis with fixed effects was performed across studies. For each study, we evaluated case-case (U-CLL vs M-CLL) and case-control comparisons (U-CLL vs controls and M-CLL vs controls) and loci that were genome-wide significant (p<5x10-8) were reported. We also specifically evaluated the known 44 susceptibility variants. Secondary analyses included: conjunctional false discovery rate (conjFDR) to identify loci jointly associated across CLL IGHV-subtypes, and local genetic correlation analysis to validate shared and subtype-specific candidate loci. The CRuCIAL cohort was additionally used to evaluate the HLA region (using single-marker, omnibus, conditional haplotype and HLA-interaction analysis) and candidate loci in disease progression (Cox regression of time to first treatment, TTFT).
Results
We first conducted case-case GWAS. The Mayo Clinic cohort included 1,081 M-CLL and 981 U-CLL. Median age was 62 years and 67% were male. The CRuCIAL Consortium included 337 M-CLL and 204 U-CLL. Median age was 66 years and 62% were male. Meta-analysis revealed a novel locus at HLA-DRA, which was significantly enriched in U-CLL compared to M-CLL (rs3135344-C, OR=1.57, 95%CI 1.36-1.82, P=9.41x10-10). HLA typing in CRuCIAL found HLA-DRB1*15 was enriched in U-CLL compared to M-CLL (OR=2.65, 95%CI 1.59-4.41, P=1.77x10-4, PFDR=0.028). Two amino acid positions were associated in case-case analyses: HLA-DRB1 position 133 (PFDR=0.045) and HLA-DQA1 position 175 (PFDR=0.035).
We also examined the 44 established CLL susceptibility loci in the case-case meta-GWAS. M-CLL was significantly associated with IRF8 (OR=0.81, 95%CI 0.71-0.93, P=0.003, PFDR=0.042), and U-CLL was associated with both CXXC1 (OR=1.30, 95%CI 1.12-1.50, P=4.09x10-4, PFDR=0.009) and IRF4 (OR=1.26, 95%CI 1.11-1.43, P=3.52x10-4,PFDR=0.009). The remaining loci were not significantly different by IGHV subtype after FDR correction.
Next, we evaluated case-control GWAS. The Mayo Clinic cohort included 1,665 controls (mean age 61 years and 50% male) and the CRuCIAL Consortium included 1,062 controls (mean age 50 years and 41% male). Meta-analysis of M-CLL vs controls found the HLA-DRA locus was associated with decreased risk of M-CLL (OR=0.70, 95% CI 0.61-0.80, P=9.99x10-8). Known CLL loci at GRAMD1B and DRAIC were genome-wide significant in M-CLL.
Meta-analysis of U-CLL vs controls found no association at the HLA-DRA locus (OR=0.98, 95% CI 0.85-1.12,P=0.733). Known CLL loci at GRAMD1B, DRAIC, and IRF4 were genome-wide significant in U-CLL. We additional identified a novel signal near IRF4 (rs12203592-T, OR=1.64, 95%CI 1.38-1.95, P=3.19x10-8) associated with U-CLL risk.
Finally, we evaluated if the candidate IGHV-subtype loci were associated with disease progression. Among U-CLL cases, the established CLL susceptibility variant at IRF8 was modestly associated with longer TTFT (HR=0.64, 95%CI 0.42-0.97, P=0.036), however the association was not significant after multiple testing correction.
Conclusion
This study provides the first genome-wide characterization of inherited risk of IGHV mutational status in CLL. We found that while IGHV-subtypes share a germline susceptibility background, loci at HLA, IRF4, and CXXC1 increased risk of developing U-CLL while IRF8 increases risk of developing M-CLL. HLA class II genes are potential contributors of the IGHV mutational status differentiation in CLL.
DISCLOSURES
Antonio Jose Jose Cabrera Serrano, PhD: No relevant conflicts of interest to declare.