Pooled Safety Analysis from Phase I-III Studies for Patients with Hematological Malignancies Treated with the PI3K Inhibitor Copanlisib

Abstract

Introduction: As of 2021, four phosphatidylinositol 3-kinase (PI3K) inhibitors had received accelerated approval by the FDA for treatment of patients (pts) with indolent follicular lymphoma (FL) having received ≥2 systemic therapies. Recently, during confirmatory studies, three of the PI3K inhibitors have been voluntarily withdrawn over concerns for toxicities impacting overall survival and this issue was the focus of an Oncologic Drugs Advisory Committee (ODAC) meeting in April 2022. Copanlisib is the only PI3K inhibitor currently available for treatment of FL and is distinguished by being administered intravenously on an intermittent schedule. We report here a comprehensive pooled analysis of copanlisib safety with long-term follow-up from 10 completed Phase I-III studies in pts with hematologic malignancies. Methods: Pts with hematologic malignancies treated with copanlisib monotherapy in completed open-label Phase I or II studies and a randomized, placebo-controlled Phase III in combination with rituximab were included. Copanlisib 60 mg (i.v.) was administered intermittently on days 1, 8, and 15 of a 28-day cycle until disease progression or unacceptable toxicity. AEs were reported using MedDRA version 24.1 and assessed by duration of treatment in 6-month intervals or time of onset (worst grade or new AEs occurring ≤180 days [d], 181-360 d, 361-540 d, 541-720 d, or ≥721 d). Results: A total of 684 pts received copanlisib and were evaluable for safety (data cut-off Feb 2022). Median age was 64 years (range 22-93). A total of 51% of pts had a history of FL, 14% MZL, and 10% DLBCL. More than half (54%) were treated with copanlisib for less than 6 months, whereas 27% were treated for >12 months and 13% >2 yrs. The mean overall duration of treatment was 10.9 months (range 0.2-80 months). 24% of pts had at least one dose reduction and 73% had at least one dose interruption or delay (mean duration 3 days). Copanlisib-related treatment-emergent AEs (TEAEs) were reported in 92% of pts; 55% of pts had TEAEs of worst grade [g] 3, and 24% had TEAEs of worst g4. The most common reasons for discontinuation were progression (41%), AE not associated with clinical disease progression (29%) and withdrawal by subject (11%). The most common (≥10%) copanlisib-related TEAEs (% all-grade/g3/g4) were hyperglycemia (58/37/6), hypertension (42/33/0), diarrhea (23/4/0), neutrophil count decreased (20/7/7), neutropenia (19/8/7), nausea (17/0.3/0), white blood cell count decreased (13/5/0.6), fatigue (13/1/0), and anemia (11/3/0). Hyperglycemia and hypertension, both of which were infusion related and transient, had the highest incidence in the first 180 days of treatment. For diarrhea, pts treated for longer duration were more likely to have experienced diarrhea, but g3 events were constant per 6-month onset of event interval (2.2/2.5/3.2/0.8/3.3%). For AEs of interest, the incidence of colitis was low (0.9%; 6 pts: 2 g1, 3 g3 and 1 g4) as was the overall incidence of pneumonitis (6/2/0.6; one g5) with all the g4 events occurring <180 days. The occurrence of late onset AEs of increasing severity has been an issue with orally administered PI3K inhibitors. The greater exposure of pts to copanlisib with longer treatment notwithstanding, we examined the incidence and severity of AEs in pts treated ≤6 months, 6-12 months and >12 months, by overall duration of treatment and by interval of onset (Table) to ascertain the timing and severity of AEs. For hyperglycemia and hypertension, the overall incidence rate was similar for pts treated ≤6 months, 6-12 months and >12 months but was mostly accounted for higher rates day 1-180. Diarrhea was higher in pts treated >12 months, but mostly due to low grade events. There was a trend for a higher incidence of neutropenia and/or neutrophil count decrease depending on treatment duration, but infections remained low. A total of 50 (7%) pts died within 35 days of discontinuing copanlisib, with 18 (2.6%) due to progressive disease (PD) and 9 (1.3%) AEs related to PD, and 20 (2.9%) AEs not associated with clinical progression. Copanlisib-related g5 (death) TEAEs were reported in 10 pts (1.5%), with 8 events in pts treated <6 months; infection-related AEs reported in 5 pts. Conclusions: Based on this expanded safety dataset with long-term treatment of copanlisib, there is no evidence of late-onset toxicities or worsening of severity of TEAEs. No new or unexpected safety issues were identified. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal