Pharmacokinetic and Pharmacodynamic Correlates from the Phase 1 Study of Camidanlumab Tesirine (Cami) in Patients with Relapsed or Refractory Hodgkin Lymphoma and Non-Hodgkin Lymphoma

Introduction: Cami is an antibody (Ab)-drug conjugate composed of the human monoclonal Ab, HuMax^®-TAC, directed against human CD25, and conjugated through a cleavable linker to SG3199, a pyrrolobenzodiazepine (PBD) dimer cytotoxin. Cami has shown notable single-agent antitumor activity and manageable toxicity in a phase 1 trial in patients (pts) with relapsed or refractory (R/R) Hodgkin lymphoma (HL) or non-Hodgkin lymphoma (NHL) (Collins et al, ICML Jun 18–22, 2019; https://onlinelibrary.wiley.com/doi/10.1002/hon.61_2629; Collins et al, ASH Dec 1–4, 2018; https://doi.org/10.1182/blood-2018-99-115986).

Methods: This open-label, single-arm, dose escalation and expansion trial (NCT04052997) enrolled pts aged ≥18 years with histologically confirmed R/R HL or R/R NHL. The primary objectives included characterizing the safety and tolerability of Cami. The secondary objectives addressed the pharmacokinetic (PK) profile of Cami, PBD-conjugated Ab (cAb, drug-to-Ab ratio [DAR] ≥1), total Ab (tAb, DAR≥0) and unconjugated warhead SG3199, and assessed the anti-drug Ab (ADA) response in serum. Also reported are exploratory PK/pharmacodynamic correlates to concentrations of soluble CD25 (sCD25), time course of peripheral blood B cells, NK cells, and T-cell subsets: T_reg(CD25+/CD127_low/FoxP3+[CD3+/CD4+]), T_eff (CD8+), and T_eff:T_reg ratio, using TruCount flow cytometry (Covance, Indianapolis, IN, USA), circulating cytokines, and immunohistochemistry (IHC) on archival or pre-treatment tumor biopsies for CD25 expression.

Results: Data from 133 pts with lymphoma (HL and NHL) were available for PK analysis following Cami doses of 3–300 µg/kg administered every 3 weeks. Mean exposures increased with dose. Apparent clearance for doses of 30–80 µg/kg was relatively constant (for cAb, 1.66 L/day [CV=64%] in the 45 µg/kg dose group during Cycle 1). Good linker stability of the immunoconjugate was demonstrated by close similarity of profiles between cAb and tAb. By Cycle 2, similar respective exposures with lower inter-patient variability were noted as relative to Cycle 1. Apparent half-life of cAb was 2.7 days (CV=40%), with no accumulation observed by the end of the 3-week cycle. Levels of SG3199 were generally too low for characterization. There were no instances of positive pre- or post-dose ADA response. For pts receiving the 30–60 µg/kg doses, preferential decreases over time for IL-6, IL-8, and IL-10 were noted for pts achieving clinical response, with decreases most evident by end of first cycle.

For total lymphocytes, CD19+ B cells, CD3+ and CD4+ T cells, CD8+ T_eff and CD16+/CD56+ NK cell subsets, similar modulations by time were observed at 30 and 45 µg/kg doses, comprising transient increases following dosing (Figure 1). Small absolute decreases in T_reg cell concentrations over time were seen following the 30 and 45 µg/kg doses, but with no distinction by best overall clinical response (complete response [CR] + partial response vs. non-responders). However, most pts had clear increases in T_eff:T_reg ratios over time, with greater effects in Cycle 2 relative to Cycle 1.

Mean sCD25 concentrations displayed apparent decrease from baseline over time following the 45 µg/kg dose. For pts with HL, higher baseline levels of sCD25 were associated with lower Cami exposure in both responders and non-responders. Particularly for pts exhibiting CR, clustering was apparent with sCD25 concentrations <~10,000 ng/mL at baseline (Figure 2). Data from IHC showed no relationship between CD25 histoscore and clinical response.

Conclusion: While T_reg modulation was seen with all populations, these data support further study, particularly of the Cami 45 µg/kg dose for pts with HL. Given that CD25 is the cognate target of Cami, these data suggest pts with HL achieving CR have higher drug exposure, resulting from lower baseline sCD25, and potential association with lower tumor burden. Although exposure at the 45 µg/kg dose showed moderate inter-patient variability, it was associated with noteworthy, cycle-related modulation in circulating T_reg and clear increases in T_eff:T_reg ratios, thought to favor clinical response.

Funding: Study funded by ADC Therapeutics SA.