Exploration of drug resistance in advanced gastrointestinal stromal tumor.

Abstract

662 Background: Gastrointestinal stromal tumors (GISTs) are prone to multi-drug resistance after drug treatment, but tissue samples from advanced patients with drug resistance are hard to obtain, impeding the study of the mechanism of drug resistance. In this study, plasma samples from TKI-resistant GIST patients were collected for next generation sequencing (NGS) to detect circulating tumor DNA (ctDNA), which were then used to generate mutation profiles leading to TKI resistance. Methods: A total of 82 GIST patients who acquired TKI resistance were enrolled from Apr. 2016 to Jun. 2021. Plasma ctDNA samples after drug resistance were collected for NGS analysis of 425 genes. Results: Among 82 patients, 54 received monotherapy with imatinib, 24 with 2 TKIs (21 with imatinib and subsequent sunitinib, 2 with imatinib subsequent regorafenib, and 1 with imatinib subsequent anlotinib), and 4 with 3 TKIs (imatinib subsequent sunitinib and regorafenib). Fifty patients (61%) were ctDNA-positive and their mutation profiles were further analyzed. Detection rate of ctDNA in imatinib-resistant patients was 57% (31/54). Among these 31 patients, 12 (39%) showed KIT gene mutation (8 with the known drug-resistant KIT mutations, 6 of these patients exhibited single-drug resistance while 2 showed multi-drug resistance due to KIT mutations mainly in Exon 13 V654A, Exons 17 Y823D, D820Y, N822Y, and N822K). The remaining 19/31 cases (61%) could be attributed to other gene mutations, with 47% due to cancer-associated mutations or other downstream signaling pathways (such as, cell cycle, TP53, RAS and PI3K). Among patients resistant to imatinib followed by other TKIs, ctDNA detection rate was 68% (19/28). Among these 19 patients, 10 (53%) showed KIT gene mutation (including 5 with the known drug-resistant KIT mutations, 4 of these patients exhibited single-drug resistance, while 1 showed multi-drug resistance due to mutations mainly in Exon 13 V654A, Exons 17 D820A, and D820G). The remaining 9/19 cases (47%) could be attributed to other gene mutations, with 44% due to cancer-associated mutations or other in downstream signaling pathways (such as Wnt, cell cycle, TP53, RAS and PI3K). Results indicated a growing trend of mutations leading to resistance among GIST patients treated with multiple TKIs (baseline vs. imatinib resistance vs. multiple TKIs resistance, 3.95% vs. 25.81% vs. 26.32%). Conclusions: Advanced GIST patients who received one or more TKI(s) may be resistant to single or multiple drugs due to complex mutations. NGS can be used to detect ctDNA without the need to obtain tumor tissue samples. This approach facilitates the monitoring of drug resistance in patients with advanced tumors, especially those with multiple metastases. NGS allows detection of mutations at multiple secondary sites and helps to overcome tumor heterogeneity, ctDNA-based NGS analysis will play an increasingly important role in assessment of drug resistance in GIST patients.