Pan-cancer analysis of genomic scar patterns caused by homologous repair deficiency (HRD)

Abstract

Homologous repair deficiency (HRD) is present in many cancer types at variable prevalence and can indicate response to platinum-based chemotherapy and PARP inhibition. We developed a tumor classification system based on the loss of function of genes in the homologous recombination repair (HRR) pathway. To this end, somatic and germline alterations in BRCA1/2 and 140 other HRR genes were included and assessed for the impact on gene function. Additionally, information on the allelic hit type and on BRCA1 promoter hypermethylation was included. The HRDsum score including LOH, LST, and TAI was calculated for 8847 tumors of the TCGA cohort starting from genotyping data and for the subcohort of ovarian cancer also starting from WES data. Pan-cancer, deleterious BRCA1/2 alterations were detected in 4% of the tumors, while 18% of the tumors were HRD-positive (HRDsum ≥ 42). Across 33 cancer types, both BRCA1/2 alterations and HRD-positivity were most prevalent in ovarian cancer (20% and 69%). Pan-cancer, tumors with biallelic deleterious alterations in BRCA1/2 were separated strongly from tumors without relevant alterations (AUC = 0.89), while separation for tumors with monoallelic deleterious BRCA1/2 alterations was weak (AUC = 0.53). Tumors with biallelic deleterious alterations in other HHR genes were separated moderately from tumors without relevant alterations (AUC = 0.63), while separation for tumors with such monoallelic alterations was weaker (AUC = 0.57). In ovarian cancer, HRDsum scores calculated from WES data correlated strongly with HRDsum scores calculated from genotyping data (R = 0.87) and were slightly (4%) higher. We comprehensively analyzed HRD scores and their association with mutations in HRR genes in common cancer types. Our study identifies important parameters influencing HRD measurement and argues for an integration of HRDsum score with specific mutational profiles.

Prediction of homologous recombination deficiency from cancer gene expression data

“…28 Our penalized logistic regression model predicted the HRD status with gene expression data in various cancer types with a 0.93 test set AUROC. This prediction performance is similar to the HRD status prediction previously achieved using deleterious BRCA1/2, 29 suggesting that an HRD prediction model based on penalized logistic regression using gene expression data can be applied to patient selection for treatment with PARP inhibitors. This model is preliminary and should be validated in a clinic with prospectively collected cancer tissues, but this study confirms the possibility of HRD prediction using gene expression data.…”

Homologous Recombination Repair Gene Mutations to Predict Olaparib Plus Bevacizumab Efficacy in the First-Line Ovarian Cancer PAOLA-1/ENGOT-ov25 Trial

“…It is possible that HRR mechanisms beyond the established BRCA-mediated HRR, or beyond HRRm detected by the gene panels tested, are present in first-line OC. For example, BRCA1 hypermethylation and RAD51C hypermethylation are present in approximately 11% and 3% of OC, respectively, 14 , 15 BRCA1 hypermethylation is also associated with high levels of genomic instability, 8 , 16 and BRCA1 hypermethylation and RAD51C hypermethylation have been reported to be associated with PARP inhibitor response. 17 , 18 However, epigenetic changes may be more easily reversed than genetic mutations; a recent analysis found that 24% of ovarian tumors lost BRCA1 hypermethylation during treatment.…”

“…This is consistent with recent pan-tumor analyses that suggest that biallelic loss of BRCA1 / BRCA2 greatly distinguishes tumors in terms of HRD, with biallelic loss of other HRR genes having less of an impact. 16 The European experts consensus statement on HRD testing in first-line OC noted a high level of agreement that if non-BRCA tumor mutation status is examined, RAD51C , RAD51D , BRIP1 , PALB2 , and BARD1 could be considered. 31 The utility of testing for ATM , CHEK1 , CHEK2 , ATR , BLM , and CDK12 mutations in OC is yet to be established.…”

Integrative Genomic Tests in Clinical Oncology

“…Furthermore, the existing multigene NGS assays have not been sufficiently validated for the detection of large gene rearrangements (LGRs), i.e., deletions or duplications of gross portions of the genes [ 36 , 37 ]. Some tumors in BRCA1/2 germline mutation carriers still retain the functional copy of the involved gene; this is particularly characteristic for non-breast-ovarian cancer types [ 38 , 39 ]. Although current treatment standards for breast, ovarian, pancreatic and prostate cancers allow the administration of PARP inhibitors solely on the results of a germline test, ideal patient management would imply the subsequent analysis of tumor tissue aimed to prove the inactivation of the remaining BRCA1/2 allele.…”

“…Microhomology-mediated end joining (MMEJ) is a compensatory DNA repair pathway that is activated in HR-deficient tumors and results in a significant accumulation of deletions with flanking homology [ 67 ]. This feature is particularly characteristic of the HRD phenotype; however, its analysis currently requires whole-genome sequencing (WGS) of tumor DNA [ 39 , 64 ]. There is a freely available computing tool, “Classifier of HOmologous Recombination Deficiency” (CHORD), which is aimed at the identification of HRD status through the interpretation of WGS data [ 41 ].…”