Phase I Study of TQB3455, a Selective Inhibitor of IDH2 with Favorable Safety and More Potent Efficacy to IDH2-R172K Mutation, for Acute Myeloid Leukemia and Myelodysplastic Syndrome

Introduction: Approximately 15% of patients with acute myeloid leukemia (AML) and 5% of patients with myelodysplastic syndrome(MDS) harbor IDH2 mutations (mIDH2). Enasidenib (ENA), the first selective IDH2 inhibitor, was associated with impressive response rates in relapsed/refractory (R/R) AML and was approved by FDA in 2017. TQB3455, a novel selective inhibitor of IDH2, different from ENA, showed stronger inhibitory activity against IDH2-R172K in vitro, and also demonstrated potent antileukemic activity in humans. The current analysis is the first to evaluate the safety and clinical efficacy of TQB3455 monotherapy in pts with mIDH2-positive AML or MDS.

Methods: This multicenter, phase 1, open-label, dose-escalation/expansion study enrolled AML or MDS pts with mIDH2, who were RR or not candidate for standard treatment, estimated survival ≥12w and ECOG performance status＜2. Prior allogeneic hematopoietic stem cell transplantation is prohibited.

TQB3455 is administered continuously as oral tablets QD in 28-day cycles. Dose levels are 25mg, 50mg, 75mg, 100mg, 150mg, and 300mg.Dose escalation followed a modified 3+3 design. Once the recommended phase 2 dose (RP2D) is determined, subjects will be enrolled into expansion cohort. Primary objectives include safety, tolerability, maximum tolerated dose (MTD) and dose-limiting toxicity (DLT). Safety was assessed by treatment-emergent adverse event (TEAE) reporting and TEAEs were graded per CTCAE version 5.0. DLT was defined as any Grade ≥4 hematologic toxicity or Grade ≥3 non-hematologic toxicity（alopecia excluded）occurring within the first 28 days after initiation of treatment. Secondary objectives include pharmacokinetics (PK), preliminary efficacy per International Working Group (IWG) response criteria for AML and MDS. Fisher's exact test was used to assess the association between categorical variables.

Results: As of May 31, 38 pts were enrolled in the study and had received at least one dose of TQB3455, 35 with AML and 3 with MDS. Baseline characteristics are detailed in Table 1. Pts received daily TQB3455 doses of 25mg(n=1), 50mg(n=1), 75mg(n=3), 100mg(n=3), 150mg(n=7), 300mg(n=1). When the dose escalated to 300mg, the patient experienced a DLT of pneumonia, then the dose was switched back to 150mg. The 150mg cohort was expanded to 7 evaluable patients and 1 patient experienced a DLT of differentiation syndrome. Therefore, the RP2D of TQB3455 was identified as 100 mg daily. 22 additional pts were enrolled on this dose during the expansion phase.

Safety analysis was performed in all patients. The most frequent Grade 3-4 TEAEs in all regimens were 25mg(n=1, 100%), 50mg(n=1, 100%), 75mg(n=2, 66.67%), 100mg(n=13, 68.42%), 150mg(n=7, 87.50%) and 300mg(n=1, 100%). The most frequent TEAEs were pancytopenia(Table 2). 24 pts discontinued treatment, primarily due to progressive disease or relapse (n=17, 44.74%), adverse events (n=1, 2.63%), withdrawal by patient (n=3, 7.89%) or death (n=2, 5.26%).

Efficacy was analyzed only in patients with AML. Three pts was not evaluable for response due to withdrawal by patients before the first efficacy evaluation. Of the 32 evaluable pts, overall response rate (ORR) was 40.63 %, including 12 pts (37.5%) in complete remission(CR) or CR with incomplete hematologic recovery (CRi), 1 pts (3.13%) in partial remission (PR). TQB3455 shows more powerful inhibitory effects on IDH2-R172K rather than IDH2-R140Q/W mutation (ORR, 66.67% vs 25.00%, p=0.03; CR, 41.67% vs 15.00%, p=0.12). The presence of FLT3 and/or RAS co-mutations at study entry was significantly associated with lack of response (ORR, 10% vs 45.46%; CR, 10% vs 31.82%). Median duration of response was 9.17 months (95% CI 7.39-not reached). After a median follow up of 5.49 months, the median overall survival (mOS) was 7.43 months (95% CI 5.55- not reached).

Conclusion: TQB3455 is well-tolerated for myeloid malignancy at the R2PD of 100 mg qd. It also yield high rates of clinical response in pts with mIDH2AML. Patients with IDH2-R172K mutation fared much better than IDH2-R140Q/W with regard to response, this is in line with outcome of TQB3455 in vitro but different from previous reports of ENA. These encouraging findings have prompted further studies of TQB3455 combination therapy in patients with mIDH2 AML and MDS.

No relevant conflicts of interest to declare.