PS958 IMPROVING THE SAFETY OF CAR-T CELL THERAPY BY CONTROLLING CRS-RELATED COAGULOPATHY

Abstract

Background: TRK fusion cancer results from gene fusions involving NTRK1, NTRK2 or NTRK3. Larotrectinib, the first selective TRK inhibitor, has demonstrated an overall response rate (ORR) of 75% in patients with TRK fusion solid tumors (Drilon et al., NEJM 2018). Here, we report the clinical activity of larotrectinib in 2 patients with Ph-negative B-cell precursor ALL (BCP-ALL) treated via an expanded access program to patients with TRK fusion cancers (NCT03025360). Aims: To determine the clinical activity of larotrectinib in BCP-ALL Methods: Patients with TRK fusion BCP-ALL who were unable to access larotrectinib through a clinical trial were eligible. Patients received the recommended phase 2 dose of larotrectinib (100 mg or 100 mg/m2, max 100 mg/dose) orally BID continuously. Intrathecal (IT) chemotherapy was allowed at investigator's discretion. Serum and cerebrospinal fluid (CSF) were collected for pharmacokinetic (PK) analysis if investigator felt it was safe to do so. Results: As of Feb 1, 2019, two patients, a 6-year old male from Germany and a 61-year old male from the United States were treated. Both patients had an ETV6-NTRK3 fusion. The pediatric patient was started on larotrectinib in first relapse early post-allo SCT (ocular and CNS involvement, MRD 10–3 in the bone marrow). He received one week of IT/MTX, one dose of systemic thiotepa (50 mg/m2) and was started on larotrectinib. CNS blasts cleared within 5 days of IT therapy and retinal infiltrations gradually resolved. PCR-MRD in the bone marrow and in the CSF became negative on day +45/+53, respectively, and continued to be negative until CAR-T cell therapy was performed after 106 days of treatment with larotrectinib and IT therapy with alternating MTX/Etoposide. PK analysis in this patient showed larotrectinib plasma AUC and CSF concentrations in line with previously published results. There were no dose modifications or discontinuations due to adverse events. The adult patient had received 6 prior therapies, including CAR-T, blinotumomab, and inotuzumab and received larotrectinib monotherapy in 6th relapse. There was no history of CNS involvement. At drug start, ECOG was 4 with lymphadenopathy and leukemia cutis. After 4 days of treatment, circulating peripheral blasts had cleared and bone marrow showed <1% blasts. Lymphadenopathy and leukemia cutis resolved. ECOG remained 4 but the patient was able to work with physical therapy and return home. While on larotrectinib, no evidence of relapse was identified, but pancytopenia persisted which was attributed to the multiple prior treatments. The patient succumbed to overwhelming infection. Total duration of larotrectinib therapy was 42 days. Summary/Conclusion: The 2 patients treated through an expanded access program provide insight that TRK fusion hematologic malignancies may be sensitive to larotrectinib. Given these encouraging results, further research is warranted including the appropriate timing of targeted therapies in the therapy of kinase-fusion positive BCP-ALL and other hematologic malignancies.