(P 175) KIT AND DOG-1 EXPRESSING GASTROINTESTINAL STROMAL TUMORS HARBORING ETV6-NTRK3 FUSIONS DO EXIST, CASE OF NEAR COMPLETE PATHOLOGIC RESPONSE FOLLOWING LAROTRECTINIB AND REVIEW OF THE LITERATURE

Objective: Previous reports describe cases of ETV6-NTRK3 fusion positive gastrointestinal stromal tumors (GISTs) in so-called quadruple-negative GIST lacking KIT, PDGFRA, RAS-pathway, or SDHx genomic alterations, but some investigators have questioned the rigor of these reports and the existence of NTRK rearrangements in GIST. This study aims to: 1) resolve the question of whether or not NTRK gene rearrangements truly exist in GIST; 2) report findings of a newly identified case involving NTRK rearrangements in GIST; and 3) review the relevant literature.

Methods: Under an IRB-approved protocol, we describe a patient diagnosed with biopsy-proven GIST that underwent genomic and transcriptomic CLIA-certified laboratory testing, molecularly matched therapy and surgical resection. We then performed a literature review of the existing cases of GIST similar to our case.

Results: We report a case of a 74-year-old female with recurrent GIST following resection of a high-risk gastric tumor followed by 55 months of adjuvant imatinib. After recurrence, biopsy revealed a mixed epithelioid and spindled GIST with immunohistochemical (IHC) expression of KIT (CD117) and DOG-1 (Figure 1A). Imatinib therapy was initiated but she developed progression, therefore molecular testing was performed. RNA sequencing of the new biopsy identified an in-frame fusion of ETV6 (exon 4) with NTRK3 (exon 14; 94.6% allelic frequency in the NTRK3 domain). Treatment with larotrectinib, a pan-NTRK inhibitor, was initiated and continued for 8 months (Figure 1B). This resulted in tumor shrinkage in all 5 lesions (range: 4.2% – 77%) with a partial radiologic response by mRECIST v1.1 (Figure 1C). The patient underwent surgical cytoreduction, which demonstrated a near complete pathologic response in all tumors resected (Figure 1D). Given the question as to whether or not NTRK fusion in KIT/DOG-1 expressing GIST truly exist, we performed a review of PubMed and identified 17 reported cases of GIST that also reported an NTRK fusion (Table 1, positive IHC staining in green). Five additional studies reported GIST with KIT and DOG-1 expression by IHC, wild-type KIT/PDGFRA by genomic testing, and an ETV6-NTRK3 fusion. Seven additional ETV6-NTRK3 fusion-positive cases reported KIT IHC expression without DOG-1 IHC expression (N=6) or DOG-1 IHC expression without KIT IHC expression(N=1).

Conclusion: Our findings confirm the existence of ETV6-NTRK3 fusion GIST as a true disease entity, and also suggest that these imatinib-resistant GIST may be exquisitely sensitive to TRK inhibitors. To our knowledge, this is the first reported case of ETV6-NTRK3 fusion- in GIST. Testing for gene fusions should be included in the molecular genetic subtyping for KIT/DOG-1 expressing GIST.