Abstract 1367: Relationship between exposure and safety/efficacy of loncastuximab tesirine (Lonca) in B-cell non-Hodgkin lymphoma (B-NHL)

Abstract

Abstract Lonca (ADCT-402) is an antibody (Ab) drug conjugate comprising a humanized anti-CD19 monoclonal Ab, conjugated with a pyrrolobenzodiazepine (PBD) dimer toxin. Lonca showed antitumor activity in a dose-finding Phase 1 study in relapsed/refractory (R/R) B-NHL (NCT02669017), and an open-label Phase 2 (NCT03589469) study in R/R diffuse large B-cell lymphoma (DLBCL), where patients received 150 µg/kg every 3 weeks (Q3W) for two doses, followed by 75 µg/kg Q3W thereafter until disease progression or unacceptable toxicity. Here, we evaluated the efficacy, safety, tolerability, and tolerable doses of Lonca PBD-conjugated Ab (cAb). A pooled Phase 1 and 2 population pharmacokinetics model of total Ab and cAb data was used to generate individual cAb exposure metrics (patients with DLBCL [n=284] for efficacy; all patients [N=328] for safety), including average cAb concentration (Cavg). SAS© v9.4 and R statistical software v4.0.1 were used for the exposure-response analysis. Efficacy endpoints included overall response rate (ORR), overall survival (OS), and duration of response (DoR); selected treatment-emergent adverse events (TEAEs) were used as safety endpoints. Univariate and multivariate logistic regression models were used to analyze ORR and safety endpoints. Time-to-event Kaplan-Meier and a Cox proportional hazards model were used to analyze OS and DoR, as applicable. Significant relationships between exposure and response (ORR and OS) were observed. ORR was 64.8% for patients in the highest quartile (Q4) of Cycle 1 Cavg compared with 23.9% for patients with drug exposure in Q1. The odds of response increased 1.2-fold for each 0.1 µg/kg increase in Cycle 1 Cavg (p<0.0001), and decreased 0.99-fold for 1 cm2 increase in baseline tumor sum of area and 0.48-fold for selected high-risk disease phenotype (double/triple hit/expressor, primary mediastinal or transformed DLBCL). Median OS was not reached in Q4, as hazard of death decreased 4.76% for 0.1 µg/kg increase in Cycle 1 Cavg (p=0.0472). Decreased baseline albumin, mild/moderate hepatic impairment, and presence of bulky tumor increased the hazard of death. No relationship was observed between exposure and DoR. Significant positive relationships were found between exposure and increased gamma-glutamyltransferase (GGT), skin and nail disorder, and liver function test abnormalities (but no evidence of hepatotoxicity). The odds of Grade ≥2 GGT occurrence increased with Cavg in Cycle 1 (p=0.000669). Higher Lonca exposure resulted in improved response. As expected, probability of Grade ≥2 TEAEs overall increased with exposure. The dosing regimen using initial dose of 150 µg/kg with dose reduction after two cycles resulted in rapid attainment and maintenance of high steady-state levels, and is anticipated to aid in response durability while minimizing toxicity. Citation Format: Brian Hess, Weiyun Ai, William Townsend, David Ungar, Sam Liao, Lori Liao, Xiaoyan Zhang, Joseph Boni. Relationship between exposure and safety/efficacy of loncastuximab tesirine (Lonca) in B-cell non-Hodgkin lymphoma (B-NHL) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1367.