Phase 1 Study of TAK-659, an Investigational Reversible Dual SYK/FLT-3 Inhibitor, in Patients (Pts) with Lymphoma: Updated Results from Dose-Escalation and Expansion Cohorts

Background

Spleen tyrosine kinase (SYK) is a nonreceptor kinase with a key role in B-cell receptor signaling-driven tumorigenesis and a key regulator of FMS-like tyrosine kinase 3 (FLT-3). TAK-659 is an investigational, oral, reversible, and potent dual inhibitor of SYK and FLT-3 that has demonstrated inhibition of cell proliferation in vitro and dose-dependent tumor growth inhibition in lymphoma xenograft models. This first-in-human, phase 1, dose-escalation and expansion study was conducted to determine the safety, tolerability, and potential efficacy of TAK-659 in pts with solid tumors and lymphomas. The MTD/RP2D of TAK-659 was determined as 100 mg QD (Petrich et al. Blood 2015; 126, 2693). Here we report updated data for the lymphoma pts from the dose-escalation and expansion phases.

Methods

Adult pts with relapsed and/or refractory lymphoma received oral TAK-659 60-120 mg (dose escalation) or 100 mg (expansion) QD in 28-d cycles. Adverse events (AEs) were assessed using NCI-CTCAE v4.03. Plasma and urine pharmacokinetic (PK) assessments occurred during cycle 1 (C1) in the dose-escalation phase. Response was assessed using IWG modified criteria (lymphoma) or IWCLL criteria (CLL) between d22 and d29 (pre-dose) of C2 (both phases), then during C4, C6, and every 3 cycles (dose escalation), or every even numbered cycle until C12, then every 4 cycles (expansion). Cell-of-origin (COO) classification data were assessed locally. Ibrutinib resistance mutations were identified by high-sensitivity Sanger sequencing of DNA isolated from bone marrow or whole blood.

Results

At data cut-off (2 June 2017), 92 pts with lymphoma (DLBCL 69; iNHL 13; CLL 5; MCL 4; PTLD 1) received TAK-659 60-120 mg QD (84 pts at 100 mg). Median age was 65 yrs (range 23-85), 60 pts (65%) were male. Drug-related Gr ≥3 AEs occurred in 66 (72%) pts, including elevated amylase (24%), hypophosphatemia (20%), elevated blood creatine phosphokinase (14%), elevated lipase (14%), and neutropenia (14%). 26 (28%) pts had drug-related serious AEs (all grades included), most commonly pyrexia (8%), pneumonia (7%), and pneumonitis (3%). Laboratory grade changes over 3 cycles are shown in Figure 1. Changes in amylase, lipase, and LDH were not associated with symptoms; nearly all pts experienced a rise in serum LDH approximately 1 week into therapy. Of 31 lymphoma pts who died on study, 3 deaths were considered drug-related (respiratory failure, multiorgan failure, and disseminated varicella). Response and target lesion change among 65 response-evaluable pts are shown in Figure 2. Median treatment duration for DLBCL responders (27% of evaluable patients) was 327 days (range 53-1018) or 10.9 mos and median duration of response (DOR) was 277 days (range 1-905) or 9.2 mos. In DLBCL pts, responses were seen in both de novo (germinal center B-cell [GCB] n=3, non-GCB n=2) and transformed (GCB n=4, non-GCB n=1) disease and appeared to be independent of COO classification. COO classification was not available for 3 additional de novo DLBCL responders. In DLBCL pts, median PFS in responding (not reached) and non-responding pts (47 days [95% CI 31, 50]) is shown in Figure 3. Of the 5 CLL pts enrolled, none had BTK mutations. One of 2 CLL responders with prior ibrutinib and idelalisib exposure had PLC-gamma-2 mutations. Responses were also observed in 7 of 10 evaluable iNHL and 1 of 2 MCL pts. TAK-659 PK profile in the dose-escalation phase was approximately dose-proportional over 60-100 mg. TAK-659 was absorbed quickly (median T_max ≈ 2 h) with moderate variability (50% CV) in dose-normalized steady-state AUC. The overall mean accumulation was 2.1-fold in lymphoma pts after repeated QD dosing; geometric mean terminal half-life was ~34 h. Overall the mean proportion of TAK-659 excreted unmodified in the urine was 34%.

Conclusions

Our data suggest that TAK-659 is generally well tolerated in lymphoma pts, with evidence of single-agent activity across various lymphoma subtypes. Antitumor activity in DLBCL pts appeared durable (median DOR of >9 mos) and independent of COO subtype or disease history (de novo vs transformed). Antitumor activity was also observed in CLL, iNHL, and MCL pts. CLL pts who have failed prior ibrutinib and/or prior idelalisib therapy may remain responsive to TAK-659. The TAK-659 PK profile supports continuous oral QD dosing. Enrollment of pts with DLBCL, iNHL, CLL, MCL, and PTLD is ongoing, with planned accrual of ~152 pts in the expansion phase.

View largeDownload slide

View largeDownload slide

Close modal