Homologous recombination deficiency (HRD) testing landscape: clinical applications and technical validation for routine diagnostics

Abstract

The use of poly(ADP-ribose) polymerase inhibitors (PARPi) revolutionized the treatment of BRCA-mutated cancers. Identifying patients exhibiting homologous recombination deficiency (HRD) has been proved useful to predict PARPi efficacy. However, obtaining HRD status remains an arduous task due to its evolution over the time. This causes HRD status to become obsolete when obtained from genomic scars, rendering PARPi ineffective for these patients. Only two HRD tests are currently FDA-approved, both based on genomic scars detection and BRCA mutations testing. Nevertheless, new technologies for obtaining an increasingly reliable HRD status continue to evolve. Application of these tests in clinical practice is an additional challenge due to the need for lower costs and shorter time to results delay.In this review, we describe the currently available methods for HRD testing, including the methodologies and corresponding tests for assessing HRD status, and discuss the clinical routine application of these tests and their technical validation.

Keywords: BRCA mutations testing; Genomic scars analysis; HRD status; Homologous recombination deficiency; PARP inhibitors.

Fig. 1

Poly (ADP-ribose) polymerase (PARP) inhibitors mechanism of action in homologous recombination (HR) proficient and deficient cells. When the DNA is damaged, leading to the formation of single-strand breaks (SSBs), the base excision repair (BER) pathway repairs these SSBs, mostly through the involvement of PARP enzymes. If PARP enzymes are trapped or inhibited by PARP inhibitors, SSBs are not repaired, resulting in the formation of double-strand breaks (DSBs) in DNA. These DSBs are preferentially repaired via the error-free DNA repair pathway, i.e. the HR repair pathway, which utilises the MRE11-RAD50-NBS1 (MRN) complex to recognize DSBs and BRCA or RAD51 proteins to repair the DNA breaks, thereby allowing the cell to survive. However, this repair pathway can be deficient, e.g. due to pathogenic mutations of BRCA genes, forcing the cell to use the more error-prone non-homologous end joining (NHEJ) DNA repair pathway. This low-fidelity repair pathway leads to an accumulation of a genome instability and finally to cell death. Created using Biorender (https://www.biorender.com/)

Fig. 2

Detection and estimation of homologous recombination (HR) deficiency (HRD) or proficiency (HRP) status through causes and consequences of HRD and HR efficiency. The causes of HRD are observed at genotypic and epigenetic levels and are obtained through the analysis of HR genes, including BRCA1, BRCA2, PALB2 or RAD51 paralogs. In this way, gene mutational status and promoter methylation are studied. HR efficiency can be estimated through a functional test, which allows the identification of the formation of RAD51 foci. The consequences of HRD are observed at the phenotypic level and are determined through the analysis of mutational signatures and genomic scars. With regard to mutational signatures, the single base signature 3 (SBS3) is associated with HRD. With respect to genomic scars, three signatures of chromosomal abnormalities are related to HRD: the loss of heterozygosity (LOH), large-scale transitions (LSTs), and telomeric allelic imbalance (TAI). The unweighted sum of the three signatures is referred to as the genomic instability score (GIS). HRR, homologous recombination repair; DSBs, double-strand breaks. Created using Biorender (https://www.biorender.com/)

Fig. 3

Illustration of methods and corresponding tests for assessing homologous recombination deficiency (HRD) status using next-generation sequencing (NGS) data. The causes of HRD can be assessed by a targeted approach using specific panels of HR-related genes, whole-exome sequencing (WES), or whole-genome sequencing (WGS). These approaches can also identify the consequences of HRD too, in the same way as the shallow WGS (sWGS). HRR, homologous recombination repair. Created using Biorender (https://www.biorender.com/)