Phase IIa Study of Single-Agent MOR208 in Patients with Relapsed or Refractory B-Cell Non-Hodgkin's Lymphoma

Introduction: There remains a high unmet medical need for new therapies for patients with relapsed or refractory (R-R) B-cell non-Hodgkin's lymphoma (NHL). CD19 is a B-lymphocyte, lineage-specific surface antigen that is highly expressed by most B-cell NHLs. CD19 expression is maintained on lymphoma cells which have CD20 expression downregulated following treatment with the CD20 antibody, rituximab. Consequently, MOR208 (XmAb5574; MOR00208), an Fc-engineered, humanized, monoclonal antibody that targets CD19, may have clinical utility as a new therapeutic approach to R-R NHL. A phase I study showed MOR208 to be safe and well-tolerated with encouraging single-agent activity in patients with chronic lymphocytic leukemia (CLL); an intravenous dose of 12 mg/kg was recommended for phase II studies.

Methods: This is a non-randomized, open-label, multicenter, two-stage, phase IIa study of MOR208 in adult patients with R-R NHL whose disease had progressed after at least one prior therapy containing the CD20 antibody, rituximab. In stage 1, 10 patients were to be enrolled into each of four NHL subtype-specific cohorts: diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), other indolent NHL (iNHL) and mantle cell lymphoma (MCL). Patients were to receive single-agent MOR208, 12 mg/kg intravenously, weekly, for 8 weeks (2 cycles). Those with at least stable disease by the 2007 International Response Criteria could continue MOR208 treatment for an additional 4 weeks (total of 12 weeks of therapy). Patients with a complete or partial response (CR or PR) after 12 weeks could then receive MOR208 as maintenance therapy, every 2 or 4 weeks depending on the investigator's decision, until progression. In stage 2, cohorts with ≥2 responses (CR or PR) were to be expanded by at least 20 additional patients. The primary endpoint was the overall response rate (ORR). Key secondary endpoints included duration of response, safety, immunogenicity of MOR208, pharmacokinetics and pharmacodynamics.

Results: The DLBCL and FL cohorts were expanded (to N=35 and N=34 patients, respectively), leading to a total enrollment of 92 patients: 56 (61%) were male; median age was 66.5 (range 35-90) years; 80 (87%) had stage III-IV disease; 41 (45%) had received ≥3 prior lines of therapy and 10 (11%) had received a prior stem-cell transplant. The investigator-assessed ORR across all NHL subtypes was 23% (21/92 patients; 16 not evaluable at cutoff) with clinical activity seen in the DLBCL (26% [9/35]; 2 CR, 7 PR); FL (26% [9/34]; 3 CR, 6 PR) and iNHL (27% [3/11]; 2 CR, 1 PR) cohorts (MCL, 0/12 responses). The iNHL cohort was not expanded as the response pattern in this subgroup was heterogeneous according to lymphoma subtype. The longest durations of response recorded to date are 15.4 months for FL and 14.2 months for DLBCL (both ongoing). Grade ≥3 non-hematologic and hematologic treatment-emergent adverse events (TEAEs) were recorded in 24 (26%) and 14 (15%) of 92 patients, respectively. The most commonly reported grade ≥3 hematologic TEAEs were neutropenia (7 [8%] of 92 patients, anemia (4 [4%]), and thrombocytopenia (4 [4%]); such TEAEs were seen most frequently in the DLBCL cohort (10 [29%] of 35 patients overall; neutropenia, 5 [14%], anemia, 4 [11%], thrombocytopenia, 2 [6%]). Dyspnea was the most commonly reported grade ≥3 non-hematologic TEAE (4 [4%] of 92 patients). Infusion-related reactions were seen in 9 (10%) of 92 patients; all were grade 1-2, except for one case of dyspnea, grade 4. There were no treatment-related deaths. Clinical activity in patients with R-R DLBCL appeared to be dependent on attaining a defined cumulative exposure (AUC_0-t) over 8 weeks of around 11,000 day*µg/mL; i.e., at the data cutoff date, all 8 patients with a PR after 2 cycles showed an exposure above this potential threshold level.

Conclusions: MOR208 demonstrated encouraging single-agent activity with CRs observed in patients with R-R DLBCL, FL, and iNHL. MOR208 was well tolerated without significant infusional toxicity. These data support further development of MOR208 in combination with other agents (including lenalidomide and bendamustine), and protocols for studies in patients with R-R DLBCL are now being developed.