Abstract
Abstract 3678
Background:
CD37 is a tetraspanin protein expressed on the surface of normal and transformed B-cells across a wide range of maturational stages and demonstrates death signaling via SHP1. TRU-016 is a novel humanized anti-CD37 SMIP™ (mono-specific protein therapeutic) that has shown significantly greater direct killing of CLL cells than rituximab and greater Fc mediated cellular cytotoxicity of CLL cells than either alemtuzumab or rituximab in pre-clinical models. In preclinical in vitro and in vivo models of NHL significant activity of TRU-016 against multiple cell lines was observed. In a phase 1 study, TRU-016 showed activity in patients (pts) with CLL and NHL was observed. Given the single-agent clinical activity of TRU-016 and synergistic or additive effect of TRU-016 with multiple agents in preclinical models, this trial of TRU-016 with rituximab and bendamustine was conducted to establish the maximum tolerated dose, overall safety, and clinical activity of TRU-016 in pts with relapsed indolent NHL.
Methods:
Pts with relapsed or refractory indolent B-cell NHL with adequate organ function, ECOG ≤2, absolute neutrophil count ≥1000/μL, platelets'100,000/μL who were not refractory to bendamustine were eligible. After premedication with acetaminophen, diphenhydramine, and hydrocortisone pts received TRU-016 (10 or 20 mg/kg on Days 1 and 15) over 2–3 hrs combined with rituximab 375 mg/m2 (Day 2) and bendamustine 90 mg/m2 (Days 1 and 2) by IV infusion for up to six 28-day cycles. Safety was evaluated using CTCAE 4.03 and response was determined using the Revised Response Criteria for Malignant Lymphoma (Cheson 2007) after every 2 cycles.
Results:
12 pts, (9 with follicular lymphoma (FL) and 3 with small lymphocytic lymphoma (SLL)) were treated (6 each dose level). Pt characteristics: median age 57 yrs (range, 51–79), median prior regimens 3 (1–4), 67% ≥ stage III at diagnosis, 4 pts had bulky disease > 5cm. All pts had relapsed after prior rituximab (R) including 3 refractory to their most recent previous treatment. In addition, prior treatments included: CHOP-R (8), RICE (5), single agent R (5), transplant (2). FLIPI scores at study entry (FL pts) were: 8 intermediate, 1 low.
The most frequent (>2 pts) adverse events (AEs) were: neutropenia (8), fatigue, nausea, insomnia, and WBC decreased (7 each); anemia, diarrhea, headache, hypophosphatemia, thrombocytopenia, and vomiting (3 each). Grade 3/4 AEs that occurred in >1 pt were neutropenia (6), hypophosphatemia (3), WBC decreased (2). Serious AEs (SAE) included asymptomatic pulmonary thrombosis in 2 pts and febrile neutropenia, pneumonia, myelodysplastic syndrome (pt had prior transplant, CHOP, RICE), deep vein thrombosis, and retinal vein occlusion (last 2 events and pulmonary thrombosis occurred in same pt) in 1 pt each. There was no apparent dose relationship to the SAEs.
The best overall response was 10/12 (83%) with 4 CRs (32%). Four responding pts (1 CR, 3 PR) discontinued treatment prior to cycle 6 due to undergoing consolidation with transplant (2), development of myelodysplastic syndrome (1), and delayed neutrophil recovery (1). The 4 discontinuations occurred in the 20 mg cohort and limit the response evaluation of the 20 mg dose. However, after Cycle 2 Day 15, the overall response rate was 67% at 10 mg/kg and 100% at 20 mg/kg. Three of the 4 pts with bulky disease responded to the regimen
Conclusions:
TRU-016 in combination with rituximab and bendamustine was well-tolerated, induces responses in the majority of patients with relapsed indolent B-NHL. A MTD was not defined. Future combination studies of TRU-016 in NHL are warranted.
Disclosures:
Mato: Celgene, Milennium, Genentech, Seattle Genetics: Speakers Bureau. Stromatt:Emergent Product Development: Employment.
Hierarchical clustering optimizes the tradeoff between compositionality and expressivity of task structures in reinforcement learning
