1977P - A registry-based analysis of the projected genomic landscape among unclassified KIT/PDGFRA wildtype mutations in patients with gastrointestinal stromal tumors

Background

Gastrointestinal stromal tumors (GIST) are characterized by multiple subtypes with distinct biological features. Here, we report the frequency of classified mutations and the projected frequency of unclassified KIT/PDGFRA Wildtype (KIT/PDGFRA WT) mutations in GIST patients that received mutational testing.

Methods

This data was supplied by the Life Raft Group (LRG) Registry. Of 2465 patients, 55.3% (n=1363) had mutational testing (Tested) and 44.7% (n=1102) did not (Untested). We used the distribution of classified KIT/PDGFRA WT that had further testing to estimate the frequency in unclassified KIT/PDGFRA WT. By assuming the distribution of unclassified Wildtype patients is similar to the classified patients, we projected the entire cohort by multiplying the percentage of each classified KIT/PDGFRA WT mutation to the total number of unclassified KIT/PDGFRA WT patients (n=101).

Results

Tested Patients Out of 1363 Tested patients, the distribution was: KIT 77.8% (n= 1060), PDGFRA 7.6% (n= 104), SDHx 4.8% (n= 66), NF1 1.5% (n= 20), BRAF (includes 1 fusion) 0.6% (n= 8), ETV6-NTRK3 fusion 0.2% (n= 2), PTCH1 0.1%(n=1), NTRK 0.1%(n=1). KIT/PDGFRA WT Approximately half (49.2%, n= 98) of KIT/PDGFRA WT patients had further testing (classified), e.g., SDHx, NF1, BRAF, etc. The other half (50.8%, n=101) had no further or inconclusive testing (unclassified), which are recorded as “Wildtype” (WT) in the registry. Table displays the projected number of patients in unclassified WT and the overall projection of all Tested patients. Table: 1977P

Known and projected mutation distribution

Conclusions

By using the methods described, we were able to project the frequency of mutations in LRG registry patients and identified a higher frequency of unclassified mutations. A well-defined mutation can provide important information to optimize clinical decision-making, including diagnosis, prognosis, treatment selection, and monitoring treatment response.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.