(P 169) FRENCH LONG TERM REGISTRY WITH LONGITUDINAL FOLLOW UP OF PDGFRA D842V GIST PATIENTS (AVIATOR2020)

Objective: While imatinib has little or no activity in patients (pt) with metastatic GIST harboring the D842V exon 18 PDGFRA mutation, results from the phase I single-arm NAVIGATOR study (NCT02508532) showed that avapritinib (ava) had significant efficacy (86% overall response rate, ORR). These results lead to a French temporary use authorization (ATU) starting on September 2020 and granted by the National Agency for Safety of Medicines and Health Products (ANSM), allowing the early access to ava in France while waiting for Market Authorization Approval (AMM). In 2021, the Transparency Committee of the National Authority for Health (HAS) issued a conditional favourable opinion for reimbursement, based primarily on the results of an exhaustive real world registry identifying all pt with unresectable or metastatic D842V GIST treated with ava in France.
The objective of this registry was to perform a long-term follow up of PDGFRA D842V mutated GIST pt in order to compare the survival of pt who have been treated with ava versus pt who have not and to collect safety data.

Methods: AVIATOR2020 (NCT04927260) is a multicenter, open-label, ambispective non randomized study enrolling all pt with a diagnosis of advanced/metastatic GIST harbouring a D842V exon 18 PDGFRA mutation. The registry was implemented from 2010 to the end of the conditional marketing authorisation. It specifically described pt and their disease characteristics, the therapeutic strategy and the occurrence of adverse events (AEs). Quality of life (QoL) and cognitive function were investigated using FACT-G, FACT-Cog and MoCA questionnaires. Data from the electronic health record (EHR) were collected.

Results: From 2010 to 2024, 25 pt from 7 expert centres were enrolled with a median age at initial diagnosis of 59 (range: 32-89). 18 pt (72%) were retrospectively included while the other 7 pt (28%) were prospectively included and able to complete QoL questionnaires. 56% (14/25) were treated with ava whatever the treatment line. Population and tumor characteristics were not statistically different between those treated by ava and those not. The ORR with ava was 71% (including one complete response) versus 10.6% with imatinib, 12.5% with sunitinib and 0% for other TKIs. 6 pt (43%) received ava in front line. The median progression free survival (mPFS) in 1st line was 7.4 months (CI95%=2.5-14.6) if pt were not treated with ava and not reached (NR) in pt treated with ava (HR 0.33). Three pt received ava in 2nd line and were still under treatment at the time of analysis. The mPFS in 2nd line was 5.0 months (CI95%=2.9-8.6) for pt not treated with ava and NR in the group of pt treated with ava (p=0.04). The 6 pt who received ava in lines 3rd-6th line were gathered in one single group and 4/6 were still under treatment at 18 months while all pt without ava progressed before 12 months.
After a median follow-up of 103 months (range: 5.8-164) from metastatic diagnosis, the median overall survival (mOS) was 53.8 months in the ava group versus 41.8 months in pt never treated by ava (HR 0.8). The mOS from L1 and L2 were 50.1 and 41.5 months in pt treated with ava versus 32.2 and 10.5 months in pt not treated with ava (HR 0.7 and 0.2).
Among pt treated with ava, at least one AE was reported for 6 pt (43%) including 3 pt (21%) with a grade ≥3 AE. Temporary interruption of ava due to toxicity occured in 6/14 pt (43%). Of the 7 patients prospectively included, all completed at least one patient-reported outcome questionnaire (i.e., the FACT-G or the FACT-Cog) and 6 had at least one MoCA available. Among them, two had decreases in total FACT-G score over time while MoCA remained stable.

Conclusion: Real world data confirms the efficacy of ava for advanced/metastatic D842V exon 18 PDGFRA mutant GIST. The descriptive analysis of safety data showed that more treatment interruptions occured in ava treated pt but this observation could be explained by the prospective data entry of AEs which can induce a bias in the evaluation. Further evaluation of long-term ava safety following dose modification is warranted.