Abstract NTOC-112: GENOMIC SCARS AND CLINICAL RESPONSE TO COMBINATION THERAPY OF PARP AND ANGIOGENESIS INHIBITORS IN OVARIAN CANCER

BACKGROUND: Genomic instability, frequently resulting in chromosomal allelic deletion with allelic imbalance (AI)/loss of heterozygosity (LOH), is characteristic of high-grade serous ovarian cancer (HGSOC). Frequent allelic deletion is thought to arise from deficiency in DNA repair by homologous recombination (HR) resulting in the so called “genomic scars” of HR deficiency. Quantification of AI/LOH events in the tumor genome has previously been shown to predict response to therapy using platinum compounds. Recently, PARP inhibitors have proved useful in treating a sub-set of patients with HGSOC, particularly tumors harboring BRCA1/2 mutations. Combination with an angiogenesis inhibitor significantly improved the outcome. This study explores the potential of using AI/LOH scores to predict clinical response of HGSOC to PARP inhibition alone or in combination with an angiogenesis inhibitor. MATERIALS AND METHODS: Molecular inversion probe array data were generated using tumors from a sub-set of patients (n=37) enrolled in a clinical trial comparing the PARP inhibitor Olaparib to the combination of Olaparib with the anti-angiogenic agent Cediranib (NCT01116648). AI/LOH regions were identified using an ASCAT based algorithm. Markers of genomic instability associated with DNA repair deficiency were scored. These quantify AI regions (NAI), telomeric AI (NtAI), large scale transition (LST), fraction of LOH (FLOH), and HRD-LOH. dChip was used for copy number analysis. The best overall response to therapy was determined using the RECIST 1.1 criteria for complete and partial response (CR, n = 3 and PR, n = 18), and stable disease without objective response (SD, n = 16). RESULTS: A high tumor NAI-score was positively correlated with the degree of clinical response to therapy (either olaparib alone or in combination with cediranib) (Chi-square test for trend, p = 0.036). This association remains statistically significant in the subgroup carrying BRCA mutations (n = 22, Chi-square test for trend, p = 0.0488). In this limited sample, the objective response rate of high NAI tumors to the combination therapy was high (7 out of 8), especially in patient carrying wild-type BRCA1/2 genes (2 out of 2, p = 0.045). The results suggest NAI may be a potential genomic marker for response to the therapy combining PARP and angiogenesis inhibitors. However, no significant association was observed between the degree of objective response and scores of other genomic measurements NtAI, LST, or HRD-LOH. SUMMARY: High NAI-score was associated with objective response to olaparib, alone or in combination with cediranib, supporting NAI as a candidate of genomic marker for predicting response to PARP inhibitor-based therapy in HGSOC. A larger cohort would be required to further evaluate predictive value of NAI for response to the combinational therapy. Citation Format: Zhigang C. Wang, Nicolai Juul Birkbak, William T. Barry, Thomas M. Roberts, Eric P. Winer, J Dirk Iglehart, Ursula A. Matulonis, S. Percy Ivy, and Joyce F. Liu. GENOMIC SCARS AND CLINICAL RESPONSE TO COMBINATION THERAPY OF PARP AND ANGIOGENESIS INHIBITORS IN OVARIAN CANCER [abstract]. In: Proceedings of the 11th Biennial Ovarian Cancer Research Symposium; Sep 12-13, 2016; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2017;23(11 Suppl):Abstract nr NTOC-112.