Paper Details

ABSTRACT

Background 

Despite advances in targeted and cellular therapies, B-cell malignancies continue to relapse through immune escape and target-antigen loss. CD37 is expressed across mature B-cell malignancies independent for prior CD19 and CD20 directed therapy, with a limited normal-tissue expression profile, making it an attractive therapeutic target in heavily pretreated disease. HDP-102 is a first-in-human (FIH) CD37-targeted amanitin-based ADC designed to deliver an RNA polymerase II–inhibiting payload to CD37⁺ malignant cells, inducing apoptosis in both proliferating and quiescent tumor cells. Preclinical studies demonstrated CD37-specific binding, internalization, and dose-dependent antitumor activity in lymphoma and CLL models. Here, we report integrated preclinical data and FIH observations from the initial dose level of a Phase 1a/1b study of HDP-102 in patients with R/R B-cell malignancies.

Methods 

The preclinical program included in vitro pharmacology, efficacy studies in CD37⁺ lymphoma and leukemia mouse models, pharmacokinetic and toxicokinetic assessments in mice and cynomolgus monkeys, tissue cross-reactivity, repeat-dose toxicology, safety pharmacology, and allometric pharmacokinetic modeling to support FIH dose selection.

HDP-102-01 is a FIH, multicenter, multinational, open-label Phase 1a/1b trial in adults with R/R B-cell malignancies. Phase 1a was designed to determine the maximum tolerated dose and/or recommended dose for expansion, followed by Phase 1b dose expansion/optimization. HDP-102 was administered IV on Day 1 of each 21-day cycle, starting at 40 µg/kg, with a planned escalation through 65, 100, 133, 166, 200, and 250 µg/kg guided by an adaptive Bayesian logistic regression model with overdose control. The DLT window was Cycle 1, defined as the first 21 days after initial dosing. Safety, PK/PD, and dose-escalation decisions are reviewed by a Safety Review Committee. Safety endpoints included DLTs, TEAEs/SAEs, laboratory abnormalities, vital signs, ECG findings, physical examination findings, and dose modifications. Tumor response was assessed using standard criteria, including Lugano 2014 criteria for lymphoma and iwCLL criteria for CLL/SLL.

Results

Preclinical studies demonstrated high-affinity CD37 binding, efficient target-mediated internalization, robust in vitro cytotoxicity and potent antitumor activity in CD37⁺ xenograft models. HDP-102 achieved 70%–81% in vitro cytotoxicity in the CD37⁺ RAJI, MEC-2, and RAMOS, with EC50 values 9.1, 17.0 and 0.23 nM, respectively. PK and GLP toxicology studies identified a highest non-severely toxic dose (HNSTD) and no-observed-adverse-effect level (NOAEL) of 2.5 mg/kg in cynomolgus monkeys. Together with allometric scaling analyses, these data supported selection of the 40 µg/kg FIH starting dose, which provided an estimated nonclinical safety margin of 84- to 139-fold.

As of July 2026, 3 adult patients with heavily pretreated R/R B-cell malignancies were enrolled and treated at dose level 1 with HDP-102 40 µg/kg IV on Day 1 of a 21-day cycle. Patients had received multiple systemic therapy lines, including multi-agent chemotherapy and anti-CD20 antibody-based therapy. Two of 3 patients were refractory to their last prior therapy.

At dose level 1 (40 µg/kg), no DLTs were observed during the 21-day DLT window. After reviewing available safety data, the SRC recommended escalation to the next planned dose level.

Among 3 treated patients, best overall response included 1 partial response and 1 stable disease, which contributed a DCR of 67%. Preliminary antitumor activity observed at the starting dose was directionally consistent with the preclinical exposure-response model, supporting biological translation of CD37-targeted amatoxin delivery into clinical activity.

Conclusions

The initial clinical findings are consistent with preclinical data, demonstrating a manageable safety profile with no dose-limiting toxicities and antitumor activity at the lowest evaluated dose (40 µg/kg) encouraging continued dose escalation and clinical evaluation.