Results of the Mantle Cell Lymphoma Subset from a Phase 2a Study of Copanlisib, a Novel PI3K Inhibitor, in Patients with Indolent and Aggressive Lymphoma

Introduction: Mantle cell lymphoma (MCL) is an aggressive disease that is incurable with conventional therapy and the outcome of which remains the poorest amongst B-cell lymphomas. Phosphoinositide-3 kinase (PI3K) pathway activation contributes to MCL pathogenesis, but early-phase studies of the PI3K-δ selective inhibitor idelalisib have reported lower responses in MCL compared with indolent non-Hodgkin lymphoma (NHL) subtypes (Kahl et al., Blood 123:3398-405, 2014; Gopal et al., NEJM 370:1008-18, 2014). In addition, although PI3K-δ is highly expressed in MCL, PI3K-α shows wide variation and expression increases with relapse (Iyengar et al., Blood 121;2274-84, 2013). Copanlisib is a novel pan-Class I PI3K inhibitor with potent preclinical inhibitory activity against both PI3K-α and PI3K-δ isoforms. Preliminary results from a phase 2a study of copanlisib in patients with relapsed/refractory NHL or chronic lymphocytic leukemia (CLL) have been reported (Dreyling et al., ASH 2013), with an expansion cohort in patients with aggressive lymphoma ongoing. We report here the final results of the MCL subset.

Methods: Patients with histologically confirmed indolent and aggressive NHL and relapsed after, or refractory to, ≥2 prior lines of treatment were eligible. Copanlisib was administered at the starting dose of 0.8 mg/kg as a 1 hour intravenous infusion on days 1, 8 and 15 of a 28-day cycle until disease progression or unacceptable toxicity. The primary endpoint was objective response rate (ORR) as assessed per independent radiologic review according to the response criteria for lymphoma (Cheson et al., JCO 17:1244, 1999). Secondary endpoints included progression-free survival (PFS) and duration of response (DOR), safety and tolerability.

Results: As of February 28, 2015, of the 81 patients enrolled, a total of 11 patients with MCL were treated. Median age was 70 years (range 60-85), M/F= 8/3. The median number of prior lines of treatment was 3 (range 3-9) with a median time of 1.4 months since last systemic anti-cancer therapy. All patients previously received rituximab and 8 patients (73%) were refractory to the last therapy. The median duration of treatment was 17 weeks (range 3-59), corresponding to a median number of 4 cycles (range 1-15), with a close adherence to planned dose (median 92%). The most common drug-related adverse events (AEs) of all grades were hyperglycemia (8 patients, 73%), hypertension (6 patients, 55%), neutropenia (5 patients, 46%) and fatigue (4 patients, 36%). Grade 3-4 AEs occurring in 2 or more patients included: neutropenia (2 patients with grade 3, 18%; 3 patients with grade 4, 27%), hypertension (3 patients with grade 3, 27%), hyperglycemia (2 patients with grade 3, 18%), and fatigue (2 patients with grade 3, 18%). One grade 5 AE of acute respiratory failure, was assessed as drug-related by the investigator. Dose reductions and delays possibly due to study drug-related adverse events (AE) were reported in 2 (18%) and 6 (55%) patients, respectively. Two patients were discontinued from study treatment due to adverse events (lung infection and non-melanoma skin cancer). All 11 patients were included into efficacy assessment, although one clinical PD was not confirmed by radiologic measurement. The ORR as determined by independent radiologic review was 64% (2 CRu and 5 PRs). The median duration of response was 150 days (95% CI: 56, 434); 33% of responders had a DOR of at least 270 days. PFS ranged from 7 to 547 days [median 112 days (95% CI: 42, 377)].

Conclusions: Copanlisib was active as a single-agent, with an ORR of 64%, and had a manageable safety profile in patients with heavily pretreated, advanced refractory/relapsed MCL. These results support the potential role of inhibiting both PI3k-α and PI3K-δ in relapsed MCL. Based on these results, a phase 2 trial of copanlisib in ibrutinib-pretreated patients with MCL is under way (NCT02455297).