Intratumoral KIT mutational heterogeneity and recurrent KIT/ PDGFRA mutations in KIT/PDGFRA wild-type gastrointestinal stromal tumors

// Jing Gao 1, * , Jian Li 1, * , Yanyan Li 1 , Zhongwu Li 2 , Jifang Gong 1 , Jian Wu 3 , Na Liu 3 , Bin Dong 2 , Changsong Qi 1 , Jie Li 1 , Lin Shen 1 1 Department of Gastrointestinal Oncology, Key laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Peking University Cancer Hospital and Institute, Beijing, China 2 Department of Pathology, Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Peking University Cancer Hospital and Institute, Beijing, China 3 MyGenostics Inc. Beijing, China * These authors contributed equally to this work Correspondence to: Lin Shen, email: lin100@medmail.com.cn Keywords: wild-type GISTs, KIT/PDGFRA mutation, intratumoral heterogeneity, next-generation sequencing, imatinib Received: October 15, 2015      Accepted: January 24, 2016      Published: February 02, 2016 ABSTRACT Objective: Gastrointestinal stromal tumors (GISTs) with no mutations in exons 9, 11, 13, and 17 of the KIT gene and exons 12, and 18 of the PDGFRA gene were defined as KIT/PDGFRA wild-type and they accounted for ~15–20% of GISTs. However, some KIT/PDGFRA wild-type GISTs with KIT mutations in other exons were occasionally reported. We therefore assessed GISTs to understand the whole genomic genotypes of KIT or PDGFRA genes in KIT/PDGFRA wild-type GISTs. Methods: A cohort of 185 KIT/PDGFRA wild-type GISTs from 1,080 cases was retrospectively assessed. Thirty-nine patients were excluded due to insufficiency of genomic DNA data or failure of library preparation, and 146 patients were analyzed by targeted next-generation sequencing (NGS) followed by validation. Results: For hot spots in KIT and PDGFRA genes, 23 out of 146 KIT/PDGFRA wild-type cases carried mutations according to NGS; there were 19 KIT mutations and 4 PDGFRA mutations, and these were exclusive. Intratumoral KIT mutational heterogeneity was observed in 4 of 19 samples which potentially triggered mechanisms of polyclonal evolution and metastasis and drug sensitivity. Eleven patients treated with imatinib were evaluable for clinical response, and 2 of 3 patients with KIT mutations achieved partial response (PR), while only 1 of 8 patients without KIT mutations reached PR. Conclusion: NGS had the potential property to identify partial mutant tumors from a subset of GISTs regarded as KIT/PDGFRA wild-type tumors using Sanger sequencing, and provided a better understanding of KIT/PDGFRA genotypes as well as identified patients eligible for imatinib therapy.