BRCA Mutations, Homologous DNA Repair Deficiency, Tumor Mutational Burden, and Response to Immune Checkpoint Inhibition in Recurrent Ovarian Cancer

Abstract

PURPOSE Homologous DNA repair–deficient (HRD) ovarian cancers (OCs), including those with BRCA1/2 mutations, have higher levels of genetic instability, potentially resulting in higher immunogenicity, and have been suggested to respond better to immune checkpoint inhibitors (ICIs) than homologous DNA repair–proficient OCs. However, clinical evidence is lacking. The study aimed to evaluate the associations between BRCA1/2 mutations, HRD, and other genomic parameters and response to ICIs and survival in OC. METHODS This is a single-institution retrospective analysis of women with recurrent OC treated with ICIs. BRCA1/2 mutation status and clinicopathologic variables were abstracted from the medical records. Targeted and whole-exome sequencing data available for a subset of patients were used to assess tumor mutational burden (TMB), HRD, and fraction of genome altered (FGA). ICI response was defined as lack of disease progression for ≥ 24 weeks. Associations of BRCA1/2 status and genomic alterations with progression-free survival (PFS) and overall survival (OS) were determined using Cox proportional hazards models. RESULTS Of the 143 women treated with ICIs, 134 had known BRCA1/2 mutation status. Deleterious germline or somatic BRCA1/2 mutations were present in 31 women (24%). There was no association between presence of BRCA1/2 mutations and response ( P = .796) or survival. Genomic analysis in 73 women found no association between TMB ( P = .344) or HRD ( P = .222) and response, PFS, or OS. There were also no significant differences in somatic genetic alterations between responders and nonresponders. High FGA was associated with an improvement in PFS ( P = .014) and OS ( P = .01). CONCLUSION TMB, BRCA1/2 mutations, and HRD are not associated with response or survival, cautioning against their use as selection criteria for ICI in recurrent OC. FGA should be investigated further as a biomarker of response to immunotherapy in OC.

Geometric Network Analysis Provides Prognostic Information in Patients with High Grade Serous Carcinoma of the Ovary Treated with Immune Checkpoint Inhibitors

“…Network curvature, formulated in this manner as a consistent information passing measure, thus appears to e↵ectively capture global gene signaling dysregulation, and furthermore functions to identify key contributors to signaling dysregulation. Establishing total curvature as a useful clinical biomarker, possibly in combination with FGA (also proposed as a potential biomarker in ovarian cancer [12]), will require larger datasets in order to further quantify and validate these results. None of the other authors report a potential COI.…”

“…We believe this is compelling evidence that network dysregulation, as measured by curvature, has the potential to provide critical insight for analyzing immune response. More samples are needed to verify this result but it is interesting to note that further investigation into FGA as a potential biomarker for survival in HGSOC has been proposed [12]. Linear regression plots on the HGS cohort (n=45) are shown in Figure 6.…”

“…Next, regarding the data characteristics, we used DNA gene CNA data from a subset of 69 women with recurrent OC who received immunotherapy from a previously published series [12]. The subtypes of ovarian cancer are in fact quite di↵erent diseases, originating in di↵erent cell types and being caused by distinct mutations with diverse outcomes, and should therefore be analyzed separately [19].…”

“…OS was defined from the start of immunotherapy treatment until either death or last follow-up [12]. Survival curves for each parameter were plotted according to the KM estimator, shown in Figure 5 along with the corresponding log-rank p-values (total curvature: p = 0.00047; TMB: p = 0.03153; LST: p = 0.42865; FGA: p = 0.19568).…”

“…Biomarkers of response to immune checkpoint blockade in HGSOC remain largely unknown. Unlike non-small cell lung cancers and melanomas that exhibit increased immunogenicity due to high tumor mutational burden (TMB) [7,8,9,10,11], HGSOCs exhibit low TMB [12]. In virtually all cases, HGSOCs are a disorder of loss of function gene mutations (TP53) leading to CNAs, and subsequently resulting in over-expressed copy number in multiple genes including oncogenes (e.g., K-RAS, c-MYC, cyclin E and AKT protein kinase) commonly due to aneuploidy [13,14].…”