PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
January 14, 2026Journal of Clinical Oncology0 citations

Outcomes of novel tyrosine kinase inhibitors in rare GIST subtypes: A clinico-genomic analysis from the Flatiron Health–Foundation Medicine database.

View Full Paper
ABAhmad BasharatMMMuhammad Shaheer MannanAKAtif N. Khan

Key Points

  • This research aims to evaluate the effectiveness of new tyrosine kinase inhibitors in rare GIST subtypes using a clinico-genomic approach.
  • Retrospective cohort study of adult patients from the Flatiron Health–Foundation Medicine database (2016–2024)
  • Inclusion of advanced GIST patients who received at least one TKI beyond imatinib
  • Patients classified into rare subtypes: KIT/PDGFRA wild-type, SDH-deficient, NF1-associated
  • Primary endpoint was overall survival (OS); secondary endpoints included progression-free survival (PFS) and treatment patterns
  • Kaplan–Meier and Cox regression analyses applied to assess outcomes.
  • Among 824 GIST patients, 142 had rare subtypes: 61 SDH-deficient, 47 KIT/PDGFRA wild-type, 34 NF1-associated
  • Median OS after imatinib failure was 27.3 months for KIT/PDGFRA-mutant, 19.4 for SDH-deficient, 15.2 for KIT/PDGFRA wild-type, and 13.6 for NF1-associated
  • Regorafenib and ripretinib improved OS compared to sunitinib in SDH-deficient cases (median OS 22.1 vs. 14.2 months, p = 0.03)
  • Ripretinib provided the longest PFS in KIT/PDGFRA wild-type tumors (median 7.9 months)
  • Avapritinib showed limited benefit in SDH-deficient tumors but activity in select PDGFRA-exon18 rare variants.

Abstract

838 Background: Gastrointestinal stromal tumors (GIST) with KIT or PDGFRA mutations respond well to imatinib, but rare molecular subtypes—including KIT/PDGFRA wild-type, succinate dehydrogenase (SDH)–deficient, and NF1-associated GIST—have limited therapeutic options. The efficacy of newer TKIs such as regorafenib, ripretinib, and avapritinib in these populations remains poorly described in real-world settings. Methods: We performed a retrospective cohort study using the Flatiron Health–Foundation Medicine clinico-genomic database (2016–2024). Adult patients with advanced GIST who underwent comprehensive genomic profiling and received at least one TKI beyond imatinib were included. Patients were stratified into rare subgroups (KIT/PDGFRA wild- type, SDH-deficient, NF1-associated) and compared with KIT/PDGFRA-mutant GIST. Primary endpoint was overall survival (OS) from initiation of post-imatinib therapy. Secondary endpoints included progression-free survival (PFS) and treatment sequence patterns. Kaplan–Meier and Cox regression analyses were applied. Results: Among 824 GIST patients with genomic annotation, 142 (17%) had rare subtypes: 61 SDH-deficient, 47 KIT/PDGFRA wild-type, and 34 NF1-associated. Median OS after imatinib failure was 27.3 months for KIT/PDGFRA-mutant, 19.4 months for SDH- deficient, 15.2 months for KIT/PDGFRA wild-type, and 13.6 months for NF1-associated tumors. Use of regorafenib and ripretinib improved outcomes compared with sunitinib alone in SDH-deficient cases (median OS 22.1 vs. 14.2 months, p = 0.03). In KIT/PDGFRA wild-type tumors, ripretinib conferred the longest PFS (median 7.9 months) compared with regorafenib (5.1) or sunitinib (4.4). Avapritinib demonstrated activity in select PDGFRA-exon18 rare variants but limited benefit in SDH-deficient tumors. Conclusions: In this real-world, genomically annotated cohort, newer TKIs provided clinically meaningful benefit in rare GIST subtypes, particularly SDH-deficient and KIT/PDGFRA wild-type tumors, though outcomes remained inferior to classical KIT/PDGFRA-mutant GIST. These findings support the integration of comprehensive genomic profiling to optimize sequencing of TKIs beyond imatinib in rare GIST populations.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Basharat et al. (2026) studied this question.

synapsesocial.com/papers/6966f30613bf7a6f02c0096ehttps://doi.org/10.1200/jco.2026.44.2_suppl.838
Ask AI
Helpful
Bookmark
Share
View Full Paper