Abstract 482: Integrated genetic and epigenetic analysis defines novel molecular clusters in rhabdomyosarcoma

Rhabdomyosarcoma (RMS) is the most common soft-tissue sarcoma in childhood. Recent sequencing studies have revealed a number of recurrent mutational targets of RMS including multiple components of the FGFR4/RAS/AKT pathway. However, due to the limited number of the mutational targets, our knowledge about molecular mechanisms of RMS is still incomplete. Moreover, relatively low mutation numbers in RMS suggests involvement of other mechanisms, such as epigenetic alterations, which have not been addressed in the previous studies. To address these issues, we conducted an integrated genetic/epigenetic study using whole exome sequencing (WES) and transcriptome sequencing (WTS), SNP array and methylation array. We first performed WES of 16 paired tumors, of which 8 tumors were analyzed by WTS. We first sequenced the exome of 16 paired tumors/normal samples, of which 3 were also analyzed for 2 relapsed and one metastatic samples. Among 690 candidate somatic changes detected by our pipeline, 604 in 531 genes (88%) were validated by deep sequencing and included 512 missense, 31 nonsense, and 21 splice site single nucleotide variants (SNVs) as well as 32 frameshift and 8 in-frame indels. Although novel recurrent fusion genes were not detected by WTS, we identified 18 recurrent mutated genes by whole exome sequencing. Among these, 14 putative driver genes were selected for subsequently targeted deep sequencing in the entire cohort of 60 RMS cases. Most frequently mutated genes were TP53, followed by NRAS, FGFR4, BCOR, and ARID1A. FGFR4/RAS/AKT pathway genes were mutated as many as 40% of RMS cases. In addition, CN changes involving FGFR4/RAS/AKT pathway genes as well as cell cycle regulators, such as CN gains in 2p24.3 (MYCN), 12q15 (FRS2), loss of 9p21 (CDKN2A/B) and 17p13.2 (TP53) were also recurrently detected. Most of these FGFR4/RAS/AKT pathway alterations, except for GAB1, were predominantly found in ERMS, suggesting that deregulated FGFR4/RAS/AKT signaling plays an important role in ERMS pathogenesis. We also performed methylation array in 50 cases. Based on the DNA methylation patterns, we identified 2 sample clusters, that exhibit remarkable correlation with histological phenotypes (named as cluster E and A). Cluster E was composed almost exclusively of ERMS, whereas all ARMS cases were grouped into cluster A. Accordingly, all tumors positive for PAX3 or 7-FOXO1 fusions were grouped into cluster A. Cluster E has been divided into 2 distinct signatures (named as E1 and E2), characterized by high frequencies of multiple chromosomal copy number changes, compared to cluster A, including gains of chromosomes 2, 8, and 12, and a higher frequency of FGFR4/RAS/AKT pathway mutations, the latter being especially prominent in the E2 cluster (46% in E1 and 73% in E2 vs. 18% in A). Our integrated analyses provide a framework for the identification of genetic/epigenetic alterations underlying pathogenesis of RMS. Citation Format: Masafumi Seki, Riki Nishimura, Kenichi Yoshida, Teppei Shimamura, Yuichi Shiraishi, Yusuke Sato, Motohiro Kato, Kenichi Chiba, HIroko Tanaka, Genta Nagae, Yusuke Okuno, Hajime Hosoi, Yukichi Tanaka, Hajime Ohkita, Mitsuru Miyashi, Tomoaki Taguchi, Katsuyoshi Koh, Ryoji Hanada, Akira Oka, Satoru Miyano, Hiroyuki Aburatani, Yasuhide Hayashi, Seishi Ogawa, Junko Takita. Integrated genetic and epigenetic analysis defines novel molecular clusters in rhabdomyosarcoma. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 482. doi:10.1158/1538-7445.AM2015-482