Phase 1 Study of JNJ-64619178, a Protein Arginine Methyltransferase 5 Inhibitor, in Patients with Lower-Risk Myelodysplastic Syndromes

Background

Myelodysplastic syndromes (MDS) are characterized by frequent mutations in RNA splicing factor genes. Protein arginine methyltransferase 5 (PRMT5) regulates the activity of the splicing machinery through methylation of key spliceosome proteins. PRMT5 inhibitors have demonstrated preferential killing of acute myeloid leukemia cells with splicing factor mutations. JNJ-64619178 is a potent, selective, oral PRMT5 inhibitor that causes accumulation of splicing abnormalities in preclinical models. It is hypothesized that inhibition of PRMT5 by JNJ-64619178 may target splicing factor gene mutant MDS clones, leading to recovery of normal hematopoiesis. Here we present Part 2 of a phase 1 study of JNJ-64619178 evaluating the safety, pharmacokinetics (PK), pharmacodynamics (PD), and preliminary clinical activity of JNJ-64619178 in adult patients (pts) with lower risk MDS (International Prognostic Scoring System [IPSS] score Low or Intermediate-1) who are red blood cell (RBC) transfusion-dependent and relapsed or refractory to erythropoiesis-stimulating agent (ESA) treatment.

Methods

Dosing was initiated at 2 mg, 14 days on/7 days off (intermittent) on a 21-day cycle, a tolerable dose from the ongoing dose escalation in solid tumors and lymphomas in Part 1. Confirmation of tolerability followed a modified 3+3 design with de-escalation allowed based on the totality of the data. Enrollment was expanded at the selected, tolerable dose level to further evaluate safety, PK, PD, and preliminary clinical activity.

Results

As of 29 May 2021, 21 pts were enrolled. Median age was 71 (range 52-85). Revised IPSS score was Very Low (10%), Low (71%) or Intermediate (19%). Pts had a median of 1 prior line of therapy post-ESA (range 0-4), including lenalidomide (33%), hypomethylating agents (29%), and luspatercept (19%). SF3B1 mutations were identified in 71% of pts. Median treatment duration was 3.1 months (range 0.03-6.37). No dose-limiting toxicities were observed. While the 2 mg intermittent starting dose was tolerable, lower dose schedules of 1 mg intermittent and 0.5 mg continuous (once daily [QD]) dosing were evaluated to reduce myelosuppression, and enrollment was expanded at 0.5 mg QD (n=13 pts).

Fifteen pts (71%) experienced ≥1 treatment emergent adverse event (TEAE) that was considered related to the study agent (TRAE). TRAE in >10% of pts were thrombocytopenia (52%), neutropenia (48%), anemia (19%), and dysgeusia (14%). Grade 3 or higher TRAE were reported for 57% of pts. Grade 3 or higher TRAE worsening from baseline in >1 pt were neutropenia (29%); anemia and thrombocytopenia (19.0% each); and diarrhea (9%). TEAEs contributed to treatment interruption, dose reduction, and treatment discontinuation in 57%, 14%, and 19% of pts, respectively. Anemia, thrombocytopenia, and neutropenia were the only TRAE leading to treatment modification in >1 pt. Hematologic toxicities were dose-dependent and reversible. Grade ≥3 thrombocytopenia and neutropenia were observed, respectively, in 67% and 50% of pts at 2 mg intermittent versus 0% and 31% at 0.5 mg QD. Serious AE considered possibly related to JNJ-64619178 by the investigator included 1 pt with Grade 3 anemia and 1 pt with Grade 1 atrial fibrillation, Grade 3 cardiac troponin I increase, and Grade 5 cardiac failure (>30 days from last dose). One other death of unknown cause was not considered related to treatment.

Comparison of the multiple dose C _trough across all dose levels suggests a dose-dependent increase in JNJ-64619178 plasma concentration. Robust target engagement, as measured by plasma symmetric dimethylarginine, was achieved at all dose levels. Serial analysis of variant allele frequency of clonal mutations in a subset of pts did not show significant reductions from baseline in peripheral blood or bone marrow.

No pts experienced objective response or hematologic improvement according to International Working Group criteria, RBC transfusion independence, or meaningful reduction in transfusion requirements.

Conclusion

JNJ-64619178 demonstrated primarily hematologic toxicity in pts with transfusion-dependent lower-risk MDS, which was manageable at the selected expansion dose. Despite robust target engagement, clinical activity was not observed, and enrollment was stopped. The role of PRMT5 in MDS and the differential impact of PRMT5 inhibition on normal and malignant hematopoiesis require further study.