Abstract 3695: Mapping of molecular landscape underlying drug resistance and recurrence in glioblastoma: Paired analysis of primary and recurrent tumors

Abstract

Abstract Background: Glioblastoma (GBM) is the most frequent and the most malignant brain tumor in adults. Despite aggressive therapy, overall survival of GBM patients remains poor due to intrinsic or acquired resistance to combined radiation and chemotherapy. Underlying biology behind GBM progression after first line therapy is mostly unknown and thus, there are only limited targeted therapeutic options for the 2nd line treatment. The aim of this study was to investigate genetic landscape in paired samples of primary versus recurrent tumors and to validate prognostic and predictive value of the most significant differences in independent cohort of GBMs. Patients and methods: 43 paired tumor samples were obtained during the 1st surgery followed by radiochemotherapy and after tumor recurrence during the 2nd palliative surgery. 24/43 paired samples were analyzed for whole genome copy number variations (CNV), 23/43 by fluorescent in situ hybridization for determination of EGFR, p53, RB1, MDM2, CDKN2A, 1p, 19q and 10p statuses. Methylation-specific PCR was used for analysis of MGMT promoter methylation and competitive amplification of differentially melting amplicons PCR for IDH1/2 mutations detection. The impact of most relevant alterations identified in paired GBMs was then validated in independent cohort of unpaired 104 tumors. Results: The most significant differences induced by chemoradiotherapy and/or GBM progression in primary versus recurrent tumors were EGFR amplification (19/24 primary tumors, 16/24 recurrent tumors), CDKN2A loss (18/24 primary tumors, 16/24 recurrent tumors), MDM2 gain (4/24 primary tumors), PDGFRA gain (3/24 recurrent tumors) and GSTT1 loss in primary tumors (7/24) and GSTM1 gain in recurrent tumors (7/24). Univariate analysis of overall survival for significant copy number alterations shown significantly worse prognosis for patients with GSTT1 deletion (HR = 5.356, CI = 1.161-24.701, p = 0.031) in primary tumor and trend for better prognosis in patients with CDKN2A deletion (HR = 0.391, CI = 0.124-1.229, p = 0.108) and GSTM1 amplification (HR = 0.341, CI = 0.090-1.288, p = 0.112) in recurrent tumors. Significantly worse prognosis for GBM patients with aberrations of GSTT1 gene was also confirmed in independent validation cohort of unpaired GBMs in multivariate analysis adjusted for age and Karnofsky score (gain: HR = 1.95, CI = 1.116-3.260, p = 0.011; loss: HR = 2.15, CI = 1.185-3.900, p = 0.012). Conclusion: Molecular genetic profiling of 24 recurrent GBM showed association between recurrence and status of glutathione S-transferases genes. Significance of GSTT1 gene aberration was validated in independent group of glioblastomas and indicates involvement of redox enzymes in GBM chemoradiation response and tumor progression. Acknowledgment: The financial support of grants TE02000058, IGA UP LF_2017_013 is gratefully acknowledged. Citation Format: Magdalena Houdova Megova, Zuzana Sporikova, Radek Trojanec, Ondrej Kalita, Jana Vrbkova, Jiri Drabek, Vladimira Koudelakova, Marian Hajduch. Mapping of molecular landscape underlying drug resistance and recurrence in glioblastoma: Paired analysis of primary and recurrent tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3695.