Abstract
INTRODUCTION: Gastrointestinal stromal tumors (GISTs) originate from mesenchymal precursor cells of the gastrointestinal wall and account for approximately 3% of all gastrointestinal malignancies. In adults, these tumors most commonly harbor mutually exclusive activating mutations in KIT, PDGFRA, BRAF, or SDH-family genes. KIT and PDGFRA mutations are well-established predictive biomarkers for response to tyrosine kinase inhibitors (TKIs). The advent of Next Generation Sequencing (NGS) has accelerated the identification of novel genetic alterations, improving disease characterization, providing prognostic and predictive information, and enabling detection of rare germline variants. METHODS: We conducted a retrospective analysis of 31 patients with GIST to evaluate the implementation of NGS testing and the interpretation of pathogenicity in real-life clinical practice. Identified mutations were compared with publicly available databases, and rare KIT exon 11 variants underwent computational modeling to assess their impact on protein conformation and imatinib interaction. Germline testing was performed when indicated. RESULTS: In addition to common primary and secondary mutations in KIT and PDGFRA, three patients carried rare KIT exon 11 variants not previously classified in public databases: p.Gln556del (upstream of codons 557/559) and p.Asn566_Pro573del and p.Val569_Leu576_del (downstream), all located within the juxtamembrane (JM) domain. One patient harbored an SDHB c.287-1G⟩C alteration resulting in aberrant splicing, also detected in the germline. The structural modeling predicted that all rare KIT exon 11 variants affected protein conformation and influenced imatinib interaction. Clinically, all three variants were associated with response to imatinib and met ACMG-AMP criteria for pathogenicity as well as ASCO-CAP tier 1 classification. CONCLUSIONS: Our findings highlight the clinical relevance of integrating NGS into routine GIST management, particularly for the identification and interpretation of rare genetic variants. The study underscores the importance of data sharing and collective variant annotation to support accurate molecular classification, prognostic assessment, and therapeutic decision-making for individual patients.