A Phase I Study of PRO131921, a Novel Anti-CD20 Monoclonal Antibody in Patients with Relapsed/Refractory CD20⁺ Indolent NHL: Correlation Between Clinical Responses and AUC Pharmacokinetics.

Abstract 3742

Poster Board III-678

PRO131921 is a third-generation, humanized anti-CD20 monoclonal antibody engineered to have significantly increased antibody-dependent cytotoxicity and complement-dependent cytotoxicity compared to rituximab as shown in in vitro models. In preclinical in vivo lymphoma models, PRO131921 has superior anti-tumor efficacy compared to rituximab. In this Phase I study, PRO131921 was administered as a single agent to patients (pts) with CD20⁺, relapsed or refractory indolent non-Hodgkin's lymphoma (NHL) who had been treated with a prior rituximab-containing regimen. The aim of the study was to determine the safety and tolerability of PRO131921, the maximum tolerated dose (MTD), its pharmacokinetics (PK), and to establish a Phase II dose. Pts were treated with PRO131921 by intravenous infusion (premedication with acetaminophen and anti-histamine) weekly for 4 weeks on days 1, 8, 15 and 22. The dose of the first infusion was approximately 50% that of subsequent infusions. The dose was escalated based on safety in a 3+3 design. PK samples were obtained pre- and post-infusion on days 1, 8, 15, and 22, and once each on days 2, 23, 29, 50, and 78 (and at later time points for up to a year).

Twenty-four pts were treated with PRO131921 at doses from 25 mg/m² to 800 mg/m². Median age was 58 yrs (38-78). Histologies were follicular NHL (n=20), small lymphocytic lymphoma (n=3) or marginal zone NHL (n=1). Pts had received a median of 2 (range 1-6) prior regimens. PRO131921 was generally well-tolerated and no MTD was reached in the study. The most common adverse events were Grade 1 or 2 (CTCAE V3.0) chills, flushing, itching, fatigue, fever, nausea, dizziness, diarrhea, and hypotension, most of which were part of infusion-related reactions limited in general to the first infusion. These responded well to slowing or interruption of the infusion, and symptomatic treatment (including steroids). Grade 3 AEs (related and unrelated to study drug) included 3 episodes each of transient neutropenia and hypoxia, and single episodes of throat tightness, bronchospasm, syncope, fatigue, periarthritis, pneumonia, and deep venous thrombosis. There was 1 unrelated Grade 4 pulmonary embolism. Two pts did not receive all 4 doses of therapy due to DLTs. One dose limiting toxicity (DLT) was observed in the 200/400 mg/m² dose cohort due to a significant infusion reaction, and a second was observed at the 300/800 mg/m² dose cohort due to Grade 3 joint pain and fatigue after 2 infusions. Detailed PK studies of PRO131921 in all patients were broadly similar to rituximab with a dose-dependent increase in exposure, but with significant inter- and intra-patient variability. Analysis of PK data demonstrated a correlation between higher normalized drug exposure (normalized AUC) and both tumor shrinkage (p=0.049) and clinical response (p=0.034), consistent with the hypothesis that rapid drug clearance (e.g. by tumor in excess of drug) may result in decreased clinical efficacy. Best investigator-assessed responses to treatment in the 22 evaluable pts by day 78 or later were 6 PR, 13 SD, and 3 PD; 5/10 pts in the two highest dose cohorts responded.

In conclusion, PRO131921 has shown clinical activity in rituximab-relapsed and refractory indolent NHL pts. The observation that higher normalized AUC may be associated with improved clinical responses has potential implications in future trials of monoclonal antibody-based therapies, and emphasizes the importance of early PK studies to optimize antibody efficacy.