Combined Tumor-Based BRCA1/2 and TP53 Mutation Testing in Ovarian Cancer

Somatic/germline BRCA1/2 mutations (m)/(likely) pathogenic variants (PV) (s/gBRCAm) remain the best predictive biomarker for PARP inhibitor efficacy. As >95% of high-grade serous ovarian cancers (HGSOC) have a somatic TP53m, combined tumor-based BRCA1/2 (tBRCA) and TP53 mutation testing (tBRCA/TP53m) may improve the quality of results in somatic BRCAm identification and interpretation of the ‘second hit’ event, i.e., loss of heterozygosity (LOH). A total of 237 patients with HGSOC underwent tBRCA/TP53m testing. The ratio of allelic fractions (AFs) for tBRCA/TP53m was calculated to estimate the proportion of cells carrying BRCAm and to infer LOH. Among the 142/237 gBRCA results, 16.2% demonstrated a pathogenic/deleterious variant (DEL) gBRCA1/2m. Among the 195 contributive tumor samples, 43 DEL of tBRCAm (22.1%) were identified (23 gBRCAm and 20 sBRCAm) with LOH identified in 37/41 conclusive samples. The median AF of TP53m was 0.52 (0.01–0.93), confirming huge variability in tumor cellularity. Initially, three samples were considered as wild type with <10% cellularity. However, additional testing detected a very low AF (<0.05) in both BRCA1/2m and TP53m, thus reidentifying them as sBRCA1/2m. Combined tBRCA/TP53m testing is rapid, sensitive, and identifies somatic and germline BRCA1/2m. AF TP53m is essential for interpreting sBRCA1/2m in low-cellularity samples and provides indirect evidence for LOH as the ‘second hit’ of BRCA1/2-related tumorigenesis.


Introduction
Germline mutations (m)/(likely) pathogenic variants in BRCA1 or BRCA2 (BRCA1/2m) genes are well-established causes of breast and ovarian cancer genetic predisposition, leading to deficiency in the homologous recombination repair pathway (HRD), where BRCA1 and BRCA2 are involved in the efficient reparation of DNA double-strand breaks [1]. It is currently established that hereditary predispositions are present in approximately 25% of ovarian cancer cases [2]. Based on the concept of synthetic lethality, by which cell death results from the inactivation of two genes when inactivation of either gene alone is nonlethal [3], poly (ADP-ribose) polymerase (PARP) inhibitors have been developed to inhibit the reparation of DNA single-strand breaks, showing improvement of survival in high-grade serous ovarian cancers (HGSOCs) bearing BRCA1/2 mutations [4][5][6]. It is noteworthy that PARP inhibitors have also contributed to a significant improvement of survival rates in patients with wild-type ovarian cancer, yet still with less efficacy than in patients with BRCA1/2m ovarian cancer [4][5][6].
Approximately 50% of HGSOCs are shown in The Cancer Genome Atlas (TCGA) molecular analysis to harbor HRD deficiency, including somatic BRCA1/2m (sBRCA1/2m) and alterations in other genes essential for the homologous recombination repair pathway such as ATM, ATR, and RAD51C/D [7]. It has been shown that tumor testing is efficient in identifying patients with BRCA1/2m, showing high concordance with germline mutation sequencing [8]. Thus, identifying BRCA1/2 germline and somatic mutations is now essential in routine clinical practice to propose a PARP inhibitor to patients at first relapse, as this is the best predictive biomarker for PARP inhibitor efficacy. With the recent positive results of the SOLO1 phase III trial, it has become increasingly urgent to have BRCA1/2m rapid testing results for all patients with newly diagnosed HGSOCs in order to select patient for PARP maintenance after platinum-based first-line therapy [9].
Approximately 95% of HGSOCs have a clonal somatic TP53 mutation (TP53m) [7]. Combined tBRCA/TP53m testing may provide the advantage of rapid results in comparison to gBRCA1/2 mutation testing via oncogenetic counseling. This approach may also be useful to interpret sBRCAm in low-cellularity samples and provide indirect evidence of the second hit event at the tumor level, such as the loss of heterozygosity (LOH). Evidence suggests that LOH may be a useful biomarker to predict primary resistance to DNA-damaging agents in BRCA1/2m carriers [10]. Recent reports of LOH analysis in the BRCA1/2 locus confirmed a proportion of loss of the wild-type (WT) allele in ovarian tumors ranging from 75% to 93% [10][11][12].
At Gustave Roussy (Villejuif, France), every patient with a new diagnosis of HGSOC (and fallopian or peritoneal carcinoma) is referred to a genetic consultation for counseling and germline testing, along with tBRCA1/2 mutation testing using next-generation sequencing (NGS) via a dedicated academic platform. This study compares the performance of combined tBRCA/TP53m testing to germline testing of ovarian cancer patients seen at Gustave Roussy.

Results
From 1 January 2016 to 1 May 2018, 237 patients with HGSOCs underwent tBRCA/TP53m testing by NGS ( Figure 1). These patients were also assigned to a dedicated genetic consultation for gBRCA1/2 testing, but, for some of them, germline testing results were pending.
Baseline characteristics of the cohort are summarized in Table 1. The median age of patients was 62 years old (IQR 56-68). Most tumors were HGSOCs with stage III or IV at diagnosis. gBRCA1/2m status was available for 189 (79.7%) patients, while it was either still pending or not available for 48 (20.3%) patients ( Figure 1). Of these 189 with available status, 27 (14.3%) gBRCA1m and 12 (6.3%) gBRCA2m were identified. tBRCA1/2 testing was performed on the 237 cases with a median testing turn-around time of 37 days (IQR 26.0-52.0 days). Analysis was non-contributive for 41 (17.3%) samples. Reasons for non-contributive samples were mainly poor tumor cellularity and sample quality (Wilcoxon rank-sum test, p < 0.001). Heterogeneity of tumoral cellularity was observed among all samples (mean tumoral cellularity of 62%; 3-100%). There was no difference between non-contributive or contributive tumor samples regarding proportions of tumor samples from untreated versus post-neoadjuvant chemotherapy samples (χ2 test, p = 0.69). Furthermore, no significant differences were observed between samples collected from biopsies or debulking surgical samples (χ2 test, p = 0.37). Baseline characteristics of the cohort are summarized in Table 1. The median age of patients was 62 years old (IQR 56-68). Most tumors were HGSOCs with stage III or IV at diagnosis.    All 39 (N = 39/39) known germline mutations were identifiable with tumor-based testing, including one large-scale BRCA1 rearrangement.
With tumor-based testing, four additional BRCA1/2 mutations were also identified, and 124 were cases that were confirmed as BRCA1/2 germline WT.
With tumor-based testing, four additional BRCA1/2 mutations were also identified, and 124 were cases that were confirmed as BRCA1/2 germline WT.

Discussion
It now seems clearly established that for every patient with newly diagnosed HGSOC, the mutational status of BRCA1/2 should be determined at diagnosis. In the context of the recently published results of phase III SOLO1, it also now seems mandatory to obtain the BRCA1/2 status as soon as possible to propose a PARP inhibitor in maintenance treatment for patients with stage III-IV in complete response after initial debulking surgery followed by first-line platinum-based chemotherapy [13]. Furthermore, FDA and EMA have recently given their favorable approval to PARP inhibitors regardless of BRCA1/2 status. However, the information remains crucial as the magnitude of the benefit from maintenance PARP inhibitors in first-line treatment varies greatly for BRCA1/2m versus BRCA1/2 WT patients. Germline BRCA1/2 testing can be more complex to organize as access to genetic counseling is required. Starting the analysis by tumoral BRCA1/2 screening can facilitate access to results since tumoral samples can be directly analyzed without any prior genetic counseling.
The first advantage of tumor-based BRCA1/2 testing is that the testing turn-around time is significantly reduced, with a median of 37 days observed, making it suitable for clinical use in practice.
Secondly, tumor-based BRCA1/2 testing is as sensitive as blood-based testing for germline variants as we could identify all the known BRCA1/2 germline mutations, including a large rearrangement. Additional sBRCA1/2m were also identified, providing additional information about factors such as LOH presence.
The results were consistent with previously published studies regarding the efficiency of tBRCA1/2 testing in clinical practice. As an example, the PAOLA-1 study showed rates of non-contributive samples of around 15% to 18% using academic platforms [14]. The number of non-contributiveness samples seems to be high, which could be related more to older material than current practice. It is also important to reject any low-quality sample to avoid the risk of a false negative in the result.
Another noteworthy point is that assessment of TP53 mutational status, along with BRCA1/2, seems to be a good quality control for validating the tumor cellularity of samples, and it is essential for good interpretation of the results. Moreover, with combined tumorbased BRCA1/2 and TP53 testing, we could also validate the presence of somatic BRCA1/2 mutations in samples with a low cellularity.
A number of studies confirmed that PARP inhibitors are also effective in some patients with BRCAwt HGSOC [6,[14][15][16]. Whether mutations in HRD pathway genes, such as RAD51C/D or PALB2 [17], or the methylation in BRCA1 or RAD51C promoters can identify HRD tumors that would benefit from PARP inhibitors is worthy of investigation [18,19]. The position of TP53 mutational status and its allelic fraction could also be an important marker to correctly interpret those results.
Finally, tumor-based testing at progression could be particularly valuable for uncovering acquired resistance mechanisms to PARP inhibitors, such as secondary reversion mutations in BRCA1/2 or RAD51 genes resulting in restoration of homologous recombination function [20].

Materials and Methods
The authors reported all consecutive cases of HGSOCs with tumor-based BRCA1/2m testing that were treated at Gustave Roussy (Villejuif, France) from 1 January 2016 to 1 May 2018. All patients with HGSOC were referred to a dedicated genetic consultation to determine germline BRCA1/2 (gBRCA1/2) mutational status. Tumor-based BRCA1/2 testing was prospectively performed using NGS panels (SureSeq Ovarian Cancer Panel (Oxford Gene Technology-7 genes)) and a customized SureSelect XT HS homemade panel (Agilent Technology-12 genes).
Tumor samples used for BRCA1/2 testing were either samples available at diagnosis or at relapse and were collected either from biopsies at diagnostic laparoscopies or samples from upfront or interval debulking surgery. Pre-treatment samples were preferred. The AF ratio for BRCA1/2 and TP53 mutations was calculated to estimate the proportion of cells carrying the BRCA1/2 mutation and to detect the presence of LOH. A tumor sample was said to have LOH if the BRCA1/2 variant allelic fraction was greater than 60%. For those samples whose BRCA1/2 allelic frequency was below 50%, the authors concluded that there was an LOH only if the BRCA1/2 allelic fraction was similar to that of TP53 mutation.
Univariate analysis, Wilcoxon rank-sum test, Fisher's exact test, and the χ2 test were used for comparisons of patient characteristics and mutational status when appropriate. A two-sided p-value <0.05 was considered statistically significant for all analyses.

Conclusions
In conclusion, combined tumor-based BRCA1/2 and TP53 testing is sensitive for the identification of both somatic and germline BRCA1/2 mutations and feasible in routine practice with an acceptable turn-around time. Additionally, the TP53 AF provides useful information regarding sample tumor cellularity and LOH that can help better identify sBRCA1/2m in low-cellularity samples.