Phase I Study of On 01910.Na (rigosertib), a Multikinase PI3K Inhibitor in Relapsed/Refractory B-Cell Malignancies

Chronic lymphocytic leukemia (CLL), mantle cell lymphoma (MCL), and related B-cell malignancies are incurable diseases that universally relapse after initial therapy. Resultant cytopenias in refractory patients are a common barrier to salvage therapy. Innovative targeted agents with favorable tolerability profiles that can overcome acquired mechanisms of resistance are urgently needed. ON 01910.Na (rigosertib) is a selective non-ATP competitive multikinase inhibitor that potently inhibits PI3 kinase and induces reactive oxygen species and NOXA-dependent apoptosis in vitro. Pre-clinical testing of rigosertib demonstrated selective cytotoxicity against CLL and MCL cells with minimal effects on normal B and T cells (Chapman et al 2012). Extensive clinical testing of rigosertib in patients with solid tumors or myelodysplastic syndromes (MDS) has indicated lack of myelosuppression and overall good tolerability (Raza et al ASH 2011 #3822). Here, we present the results from the phase I study assessing the safety and maximum tolerated dose (MTD) of intravenous rigosertib in patients with relapsed CLL, MCL, and related B-cell malignancies.

phase I dose-escalation clinical trial was conducted to evaluate the safety and efficacy of rigosertib in patients with CLL, MCL, MM, and HCL who were refractory or relapsed after ≥1 lines of therapy. Baseline cytopenias were permitted unless ANC < 500 or platelets were < 10K and unable to be supported with transfusion. Pts with GFR < 40ml/min, serum sodium < 134meq/L, and active ascites were excluded. Dose escalation followed a traditional 3+3 design and dosing cohorts were 1200mg/m2, 1500mg/m2 and 1800mg/m2 over 48 hours and 1800mg (flat dose), and 2100mg (flat dose) over 72 hours. Infusions were delivered via an ambulatory infusion pump and repeated in 14 day cycles for up to 4 cycles. Response was determined in patients who completed 4 cycles. Pts who demonstrated a biologic response without DLT were allowed to continue infusions until disease progression. Primary endpoint was toxicity after 2 cycles. Secondary endpoints included the toxicity with extended dosing and measures of biologic activity after 4 cycles.

Increasing doses of rigosertib were evaluated in 16 pts with relapsed CLL (10), MCL (2), MM (2), and HCL (2). All patients were evaluated for toxicity, while 10 patients completed 4 cycles of therapy and were evaluable for secondary endpoints. Median age was 61 yrs [range 52–65]. Drug-related adverse events (AEs) were reported in 15 pts (94%) and were almost exclusively grade ≤2. Most frequent drug-related AEs were fatigue 31%, musculoskeletal pain 31%, nausea 19%, constipation 19%, and diarrhea 12%. Grade 3/4 drug-related AEs included 2 cases of G4 neutropenia (both patients had neutropenia at baseline) and 1 case of syncope; there was 1 cardiac death in a patient with pre-existing heart disease that was classified as unrelated. No dose-limiting toxicities (DLTs) were observed. Analysis of blood samples collected for pharmacokinetics is planned. Response data in the 13 patients evaluable for response indicated that 7 had stable disease and 6 had disease progression. No clinical responses or evidence of biologic activity was observed.

Escalating doses of rigosertib were well tolerated in patients with relapsed/refractory B-cell malignancies with rare G3/G4 toxicities. Of note, most patients with baseline cytopenias tolerated the therapy well. The highest dose level studied in this study is one step up from the dose level of the ongoing pivotal trial of rigosertib in MDS. However, as a single agent no clinical responses were observed with rigosertib in B-cell malignancies. Further development of rigosertib in lymphoid malignancies will require either combination therapy or alternative dosing schedules.

Wilhelm:Onconova: Employment, Equity Ownership.