Role of homologous recombination repair deficiency testing in ovarian cancer and its clinical implications

Dr. Min Han Tan
Specialist in Medical Cardiology
Lucence Group
Dr. Song Ling Poon
ACT Genomics
Prof. Winnie Yeo
Department of Clinical Oncology
Chinese University of Hong Kong
08 Dec 2021
Role of homologous recombination repair deficiency testing in ovarian cancer and its clinical implications
Homologous recombination repair deficiency (HRD) testing informs treatment decisions in ovarian cancer (OC) and predicts the magnitude of benefit from poly(ADP-ribose) polymerase (PARP) inhibitors, such as niraparib. At a webinar, Professor Winnie Yeo of the Department of Clinical Oncology, Chinese University of Hong Kong, discussed the role of HRD testing in OC management, while Dr Min-Han Tan of Lucence Group and Dr Song-Ling Poon of ACT Genomics reviewed the HRD score algorithms and testing platforms currently available in Hong Kong.

HRD: A milestone in PARP inhibitor development
Understanding of PARP inhibitors’ mechanism of action has increased over the past decade. [Ann Oncol 2019;30:1437-1447] “As a result, their use has extended from patients with germline or somatic BRCA mutations [BRCAm] to other types of HRD, beyond BRCA mutations,” said Poon. 

In addition to BRCA1/2, multiple other homologous recombination repair genes are involved in the DNA repair process. When any of these genes acquire mutations, the repair process becomes compromised and HRD ensues. “Therefore, HRD has emerged as a biomarker that captures more patients who may benefit from PARP inhibitor treatment than BRCA1/2 mutation testing alone,” Poon continued. [Cancer Discov 2015;5:1137-1154]

Niraparib monotherapy: PFS benefit in both HRD and HRP disease
Mutations in homologous recombination repair genes are found in about half of high-grade serous ovarian cancer (HGSOC) cases, leading to a distinct phenotype associated with enhanced sensitivity to PARP inhibitors. [Cancer Discov 2015;5:1137-1154; Ann Oncol 2019;30:1437-1447]

In the randomized, double-blind, phase III PRIMA trial, 733 patients with newly diagnosed advanced OC were randomized to receive niraparib (200 mg or 300 mg once daily based on the individualized dose approach; n=487; median age, 62 years) or placebo (n=246; median age, 62 years). Half (50.9 percent) of the patients had tumours with HRD based on the Myriad myChoice test. [N Engl J Med 2019;381:2391-2402]

Results showed that niraparib monotherapy reduced the risk of disease progression or death by 60 percent vs placebo (median progression-free survival [PFS], 22.1 months vs 10.9 months; hazard ratio [HR], 0.40; 95 percent confidence interval [CI], 0.27 to 0.62) in the HRD–BRCAm subgroup, and by 50 percent (median PFS, 19.6 months vs 8.2 months; HR, 0.50; 95 percent CI, 0.31 to 0.83) in the HRD–BRCA wild-type (BRCAwt) subgroup. (Figure)

In patients with homologous-recombination proficient (HRP) tumours (ie, HRD-negative and BRCAwt), niraparib reduced the risk of disease progression or death by 32 percent (median PFS, 8.1 months vs 5.4 months; HR, 0.68; 95 percent CI, 0.49 to 0.94). (Figure)

HK-ZAI-033mo Zejula

“The data suggested that patients with HRD tumours had a higher response, longer PFS and better outcomes with PARP inhibitors” said Yeo.

In the PAOLA-1 trial, however, no significant PFS benefit was seen with olaparib vs placebo in patients with HRD-negative or unknown status. “For these patients, using olaparib [300 mg twice daily] plus bevacizumab does not provide additional benefit vs placebo plus bevacizumab,” said Yeo. The median PFS was 16.9 months vs 16 months in the olaparib vs placebo group (HR, 0.92; 95 percent CI, 0.72 to 1.17). [N Engl J Med 2019;381:2416-2428]

The benefit [of PARP inhibitors] appeared to be the greatest in patients with HRD tumours with BRCAm, followed by HRD with BRCAwt tumours. Patients with HRP tumours had the least benefit. In the PRIMA study, niraparib was beneficial even in patients with HRP tumours,” reported Yeo. [N Engl J Med 2019;381:2391-2402; N Engl J Med 2019;381:2416-2428]

HRD score algorithms

The Myriad myChoice assay was approved by the US FDA for determination of HRD status in OC patients to predict PARP inhibitor response. The test not only detects BRCA1/2 variants, but also determines the Genomic Instability Score (GIS). [Br J Cancer 2021;125:7-14]

The GIS is an algorithmic measurement of the accumulation of HRD-related genomic damage. It is calculated from three components that reflect different types of tumour genome rearrangements, including loss of heterozygosity (LOH), telomeric allelic imbalances (TAIs), and large-scale state transitions (LSTs). [Br J Cancer 2021;125:7-14]. “The more extensive the genome damage, the greater the link to HRD,” explained Tan.

Apart from the Myriad myChoice test, several other assays can also be applied to determine HRD status and may have potential for clinical use. [J Pers Med 2021;11:612]

High concordance of TissueBRCA-HRD assay with Myriad myChoice 

The TissueBRCA-HRD assay determines HRD status by analyzing tissue BRCA1/2 mutations based on the LUCENCE in-house database and providing the tissue Genomic Scar Score (GSS) calculated by AmoyDx. [Tan MH, 2021; J Pers Med 2021;11:612]

To measure genomic instability, the GSS evaluated >20,000 single nucleotide polymorphisms (SNPs) across the human genome for copy number variations. Three variables including the length of copy number (LCN), type of copy number (TCN) and site of copy number (SCN) are analyzed in the genomic instability testing.

The GSS is a continuous score of 0–100, with a threshold of 50. A GSS ≥50 indicates HRD positivity, while <50 indicates HRD negativity. “A minority of patients [<10 percent] will have a borderline score, posing a challenge for oncologists to interpret the result. More support will be needed in these cases when making a treatment decision,” said Tan.

The performance of AmoyDx assay, along with other assays (ie, OncoScan, CytoSNP), has been validated by an independent German pathology laboratory. “Reassuringly, high concordance was still observed although these tests use slightly different approaches to analyze genomic instability,” commented Tan. An interim analysis of 47 samples showed that GSS developed by AmoyDx had a high concordance (86 percent) with the Myriad myChoice test by independent comparison. [Weichert W, et al, ESMO 2021, abstract 758P; Tan MH, 2021]

ACTHRD: Compatible with FDA-approved test 

The ACTHRD next-generation sequencing (NGS)—based assay determines HRD status by evaluating BRCA1/2 mutations and genomic LOH (gLOH) data. [; Poon SL, 2021]

To determine gLOH status, a proprietary algorithm evaluates almost 9,000 SNPs across the genome, using a LOH fraction instead of segment concept. In other words, the methodology encompasses continuous and noncontinuous LOH SNPs, leading to a more stable LOH score.

The ACTHRD test was compared with the Myriad myChoice test using 46 samples. Out of these samples, 32 were classified as HRD-positive while 12 and 11 were classified as HRD-negative by Myriad myChoice and ACTHRD, respectively. ACTHRD and My Choice were unable to classify one sample each due to insufficient tumour purity. “Although we are using different algorithms for HRD calculation, the concordance between LOH score and GIS is very high [≥ 95 percent],” reported Poon. “The high concordance is also observed in BRCA1 and BRCA2 mutation status, including the detection of large genomic rearrangements in these genes.” 

According to Poon, the gold-standard for HRD testing remains to be determined. Moving forward, the US FDA and several diagnostic and pharmaceutical companies are participating in the Friends of Cancer Research (FoCR)–led HRD harmonization project to better define HRD and parameters that contribute to the determination of HRD status (completed) and achieve assay alignment and clinical contextualization (ongoing). [] “The harmonization project will highlight assays that are less concordant,” said Tan.

Expanding the role of HRD testing

The GSS algorithm showed a good relationship of PFS and HRD-positive status in patients treated with PARP inhibitors. However, Asian data are lacking. Poon suggested that the local test platforms should showcase that good outcomes with the use of PARP inhibitors (ie, PFS, overall survival [OS]) are associated with HRD status in large-scale Asian cohort(s).”

While HRD testing now focuses on HGSOCs, they may also have a role in other histological subtypes of OC. [Ann Oncol 2020;31:1606-1622]

“Across the mutational landscape of five different types of OC, the endometrioid type may be a candidate for HRD testing. Some retrospective analyses have evaluated the role of HRD testing in this type of OC, but we may have to wait for large-scale trials,” said Poon.

 “For clear-cell ovarian cancer [CCOC], the mutational landscape is quite different, especially when CCNE1 amplification is mutually exclusive with the homologous recombination repair–associated mutations. In this case, the benefit of HRD testing in CCOC may not be as high as what we see in HGSOCs,” suggested Poon.

HRD testing and re-testing is relevant throughout the course of the disease. After having originally tested positive for HRD, some tumours undergo BRCA reversion, whereby they develop HRP and acquire resistance to PARP inhibitors. [Ann Oncol 2020;31:1606-1622; BMC Cancer 2021;21:1154] According to Tan and Poon, since BRCA reversion can be picked up on NGS assay, it is highly recommended that the patients take the HRD test again after commencing PARP inhibitor treatment, using the latest biopsy sample. 

Editor's Recommendations
Related Diseases