Beyond BRCA1/2: Homologous Recombination Repair Genetic Profile in a Large Cohort of Apulian Ovarian Cancers

Abstract

Background: Pathogenic variants in homologous recombination repair (HRR) genes other than BRCA1/2 have been associated with a high risk of ovarian cancer (OC). In current clinical practice, genetic testing is generally limited to BRCA1/2. Herein, we investigated the mutational status of both BRCA1/2 and 5 HRR genes in 69 unselected OC, evaluating the advantage of multigene panel testing in everyday clinical practice.

Methods: We analyzed 69 epithelial OC samples using an NGS custom multigene panel of the 5 HRR pathways genes, beyond the genetic screening routine of BRCA1/2 testing.

Results: Overall, 19 pathogenic variants (27.5%) were detected. The majority (21.7%) of patients displayed a deleterious mutation in BRCA1/2, whereas 5.8% harbored a pathogenic variant in one of the HRR genes. Additionally, there were 14 (20.3%) uncertain significant variants (VUS). The assessment of germline mutational status showed that a small number of variants (five) were not detected in the corresponding blood sample. Notably, we detected one BRIP1 and four BRCA1/2 deleterious variants in the low-grade serous and endometrioid histology OC, respectively.

Conclusion: We demonstrate that using a multigene panel beyond BRCA1/2 improves the diagnostic yield in OC testing, and it could produce clinically relevant results.

Keywords: BRCA1/2; HHR genes; PARPi; ovarian cancer; target resequencing.

Figure 1

Overall distribution of deleterious and uncertain significance variants in our cohort.

Figure 2

Lolliplots showing germline and somatic variant spectrum (both pathogenic and VUS) throughout the whole protein sequences of BRCA1 and BRCA2. The vertical black scale bar represents the length (amino acids) of the protein sequence. The horizontal bar represents the number of patients affected by the specific mutation. Each lolliplot represents a variant identified in this study. The blue lolliplot identifies a shorts indels mutation (deletion or duplication). The red lolliplot indicate a missense variant, while the green ones a nonsense mutation.

Figure 3

Lolliplots showing germline and somatic variant spectrum (both pathogenic and VUS) throughout the whole protein sequences of BARD1, BRIP1, and PALB2. The vertical black scale bar represents the length (amino acids) of the protein sequence. The horizontal bar represents the number of patients affected by the specific mutation. Each lolliplot represents a variant identified in this study. The blue lolliplot identifies a shorts indels mutation (deletion or duplication). The red lolliplot indicate a missense variant, while the green ones a nonsense mutation.

Figure 4

Lolliplots showing germline and somatic variant spectrum (both pathogenic and VUS) throughout the whole protein sequences of RAD51C and RAD51D. The vertical black scale bar represents the length (amino acids) of the protein sequence. The horizontal bar represents the number of patients affected by the specific mutation. Each lolliplot represents a variant identified in this study. The blue lolliplot identifies a shorts indels mutation (deletion or duplication). The red lolliplot indicate a missense variant, while the green ones a nonsense mutation.

Figure 5

Proportion of pathogenic variants and VUS in testing only BRCA1/2 genes and retesting all the OC samples for the five HR genes. The rate of VUS detection is increased.

Figure 6

Distribution of somatic and germline variant detected in our study: Most somatic variants (pathogenic and VUS) were also found in the paired blood sample.

Figure 7

Distribution of pathogenic and uncertain significance variants according to the ovarian cancer histology of our cohort.