Long‐term safety and efficacy of the PI3K inhibitor copanlisib in patients with relapsed or refractory indolent lymphoma: 2‐year follow‐up of the CHRONOS‐1 study

Safety profiles of oral PI3K inhibitors have resulted in US FDA black box warnings regarding fatal/serious toxicities. The approved intravenous PI3K inhibitor copanlisib has low incidence of severe toxicities and no black box warnings, but chronic treatment effects were unknown. We provide an update on safety and efficacy of copanlisib with a minimum 2‐year follow‐up of the CHRONOS‐1 study. A total of 142 patients with histologically confirmed indolent B‐cell lymphoma who had relapsed after or were refractory to ≥2 prior treatments received intravenous copanlisib 60 mg on days 1, 8, and 15 (28‐day cycle). The primary efficacy endpoint was objective response rate (ORR) after ≥4 cycles (independent assessment). The predominant histology was follicular lymphoma (n = 104). The ORR was 60.6% (seven additional complete responses since primary analysis). Secondary endpoints of median duration of response, progression‐free survival, and overall survival were 14.1 months (median follow‐up, 16.1 months), 12.5 months (median follow‐up, 14.0 months), and 42.6 months (median follow‐up, 31.5 months), respectively. Median safety follow‐up was 6.7 months; 26% of patients received treatment for >1 year. Common treatment‐emergent adverse events (TEAEs) (all grade/grade 3/grade 4) were transient hyperglycemia (50.0%/33.1%/7.0%), diarrhea (35.2%/8.5%/0%), transient hypertension (29.6%/23.9%/0%), and neutropenia (28.9%/9.2%/14.8%). Serious AEs were largely unchanged, with no new cases of pneumonitis (4.2%), diarrhea (2.8%), or grade 5 events. Note, TEAEs showed no evidence for increased incidence or worsening following longer exposure in patients treated >1 year. Long‐term follow‐up of patients with relapsed/refractory indolent B‐cell lymphoma treated with intravenous copanlisib demonstrated durable, enhanced responses without evidence of worsening TEAEs, as reported for orally administered PI3K inhibitors.


Funding information
Bayer AG low incidence of severe toxicities and no black box warnings, but chronic treatment effects were unknown. We provide an update on safety and efficacy of copanlisib with a minimum 2-year follow-up of the CHRONOS-1 study. A total of 142 patients with histologically confirmed indolent B-cell lymphoma who had relapsed after or were refractory to ≥2 prior treatments received intravenous copanlisib 60 mg on days 1, 8, and 15 (28-day cycle). The primary efficacy endpoint was objective response rate (ORR) after ≥4 cycles (independent assessment). The predominant histology was follicular lymphoma (n = 104). The ORR was 60.6% (seven additional complete responses since primary analysis). Secondary endpoints of median duration of response, progression-free survival, and overall survival were 14.1 months (median follow-up, 16.1 months), 12.5 months (median follow-up, 14.0 months), and 42.6 months (median follow-up, 31.5 months), respectively.
Median safety follow-up was 6.7 months; 26% of patients received treatment for >1 year.

| INTRODUCTION
Indolent B-cell lymphomas are often incurable due to frequent relapse following an initial response to first-line therapy and eventual development of refractory disease, representing a significant challenge for treatment and a high unmet need. 1 Several phosphatidylinositol 3-kinase (PI3K) inhibitors are either approved or in development to target the aberrant PI3K signaling that underlies tumor progression in several types of cancer, including lymphoma. [2][3][4][5][6][7] Clinical benefit has been reported in patients with indolent lymphoma 2,5,6 ; however, data on the long-term effectiveness of PI3K inhibitors remain limited. In addition, several orally administered agents targeting the PI3K pathway have demonstrated adverse events (AEs), such as autoimmune dysfunction and opportunistic infections, 8 with reports of more severe late-onset toxicities such as diarrhea and ulcerative colitis. 9,10 Such PI3K-associated toxicities have caused the US Food and Drug Administration (FDA) to include warning statements as part of the prescribing information for orally administered PI3K inhibitors. 11,12 Hence, there is an unmet need for long-term treatment options that are both safe and effective in this patient population with advanced and difficult-to-treat lymphoma.
Copanlisib (Bayer AG, Berlin, Germany) is an intravenous panclass I PI3K inhibitor with predominant and potent activity against the PI3K-α and PI3K-δ isoforms. 13,14 Copanlisib is indicated in the US for the treatment of patients with relapsed follicular lymphoma (FL) who have received at least two prior systemic therapies. 15 Clinically significant activity was demonstrated in a population of patients with relapsed or refractory indolent B-cell lymphoma who had received at least two prior therapies in a large phase 2, multicenter, open-label study (CHRONOS-1; NCT01660451, Part B). The primary efficacy endpoint of that study was objective response rate (ORR); 84 of 142 patients achieved an objective response, giving an ORR of 59.2%. 2 At the time of the primary analysis, the minimum follow-up from the date of the last patient initiating treatment was only 4 months, with an overall median duration of safety follow-up of 24 weeks. There were low rates of severe hepatic transaminitis, diarrhea, pneumonitis, and colitis (0.7%) and infrequent opportunistic infections, fatal infections, or other fatal AEs. 2 Based largely on these results, copanlisib was approved by the US FDA without the so-called black box warning.
To gain a better understanding of the safety and efficacy associated with long-term use of an intravenous PI3K inhibitor, including patients treated for 1 year or more, we report here the safety and efficacy from a 2-year follow-up of patients treated with copanlisib in the CHRONOS-1 study.

| METHODS
Detailed methods of the CHRONOS-1 study have been reported previously. 2 Briefly, this was an open-label, single-arm, phase 2 study evaluating the efficacy and safety of single-agent copanlisib in patients with histologically confirmed indolent B-cell lymphoma who had relapsed after or were refractory to at least two prior lines of treatment (clinicaltrials.gov; NCT01660451, Part B). Eligibility criteria included patients aged ≥18 years with histologically confirmed indolent B-cell lymphoma, including FL grades 1 to 3a, marginal zone lymphoma (MZL), small lymphocytic lymphoma (SLL), and Waldenström macroglobulinemia/lymphoplasmacytoid lymphoma (WM/LPL), who had previously received rituximab and an alkylating agent or regimen. Patients were also required to have at least one bidimensionally measurable lesion 16 and an Eastern Cooperative Oncology Group performance status (ECOG PS) of ≥2.
Copanlisib was administered at a fixed dose of 60 mg via a 1-hour intravenous infusion on an intermittent schedule, on days 1, 8, and 15 of a 28-day cycle until progression or unacceptable toxicity.
The primary efficacy variable was ORR after 4 or more cycles of treatment, defined as the proportion of patients who had a best response rating of complete response or partial response according to central radiologic review. Secondary efficacy variables included duration of response (DoR), progression-free survival (PFS), overall survival (OS), best change in target lesions, and disease control rate.
Tumors were assessed by computed tomography scan or magnetic resonance imaging at screening and every 2 cycles during year 1, every 3 cycles during year 2, and every 6 cycles during year 3. Response was determined by independent radiologic review. 16

| Patients and treatment
The full analysis set comprised 142 patients who received treatment with copanlisib, of whom 141 had indolent lymphoma. Patient demographics and baseline characteristics are presented in Table 1

| Safety profile
The median duration of copanlisib treatment was 26 weeks (range . This corresponded to a median of 6.  Table 2). Newly occurring worst-grade events of any grade did not increase more than 5% vs the primary analysis set.
Drug-related TEAEs were reported in 127/142 patients (89.4%), one additional patient than reported at the primary analysis, and are presented in Supporting Information, Table S1.
Serious AEs (SAEs) were reported in 79 patients (55.6%), compared with 71 patients (50.0%) at the time of the primary analysis.
The most common all-grade SAEs occurring in at least three patients were pneumonia (16/142; 11.3%), pyrexia (9/142; 6.3%), hyperglycemia (7/142; 4.9%), and pneumonitis (6/  Since completion of the primary analysis, the WHO Drug Dictionary has been updated to include cisplatin as an alkylating agent; therefore, one additional patient who received cisplatin as the last regimen was included in the alkylating agents group who did not qualify at the time of the primary analysis.   Because of reports of delayed or late-onset AEs associated with continuously dosed oral PI3K inhibitors, in addition to the overall incidence of AEs as commonly reported (Table 2), we further evaluated the incidence and severity of AEs over the course of treatment. Specifically, we determined the incidence of new or worsening AEs within specific time intervals: patients treated up to 6 months (n = 142), between 6 and 12 months (n = 72), and >12 months (n = 37).
The assumption was that if there were cumulative toxicity or lateonset AEs, then AEs would occur at a higher incidence or higher grade following longer treatment. As shown in Figure 1A for AEs with overall incidence of ≥25% and in Supporting Information, Table S3 for all AEs >10%, the highest incidence of AEs was seen early in the <6-months interval, with reduction in incidence and grade in the   (Figure 2A).

| Efficacy
For the 41 patients (28.9%) who had stable disease as best response, the median duration of stable disease was 7.2 months (range 1.3-23.0).

| DISCUSSION
Long-term follow-up data from the large multicenter CHRONOS-1 study of copanlisib demonstrated continued safety and enhanced efficacy in patients with relapsed or refractory indolent B-cell lymphoma.
In the 2-year follow-up of the primary analysis, an overall ORR of 60.6% was achieved, compared with 59.2% at the time of the primary analysis, 2 and indicated some deepening of responses with time on treatment, principally partial response to complete response. Six patients who initially had a partial response at the primary analysis achieved complete response at the later cut-off, and one patient who initially had stable disease achieved a partial response at the later cutoff. Indeed, the median time to complete response (4.7 months) was more than twice the median time to first response (1.8 months). The ORR in patients with FL remained at 58.7%, while the ORR in patients with MZL increased to 78.3%. The proportion of patients with a primary response of progressive disease was low (2.1%). The ORR was consistent across the various subgroups, including patients aged <65 years or ≥65 years and patients who were or were not refractory to last treatment.
The median DoR overall was 14.1 months and was greater than 2 years for patients with complete responses. Median PFS was greater than 1 year. Median OS was demonstrated to be more than 3.5 years with copanlisib treatment.
These efficacy data for copanlisib suggest favorable response rates, generally consistent with or better than those reported for other PI3K inhibitors to date. A 20-month follow-up of the oral PI3K-δ inhibitor idelalisib reported an ORR of 56% in refractory indolent B-cell lymphoma 19 (ORR of 57% at the primary evaluation). 5 The median DoR (13.9 months) and median PFS (11.0 months) 19 were similar to those observed with copanlisib. The PI3K-δ and -γ inhibitor duvelisib achieved an ORR of 47.3% and median PFS of 9.5 months in patients with indolent refractory lymphoma. 6 The PI3K-δ inhibitor umbralisib demonstrated an ORR of 36.7% in patients with hematological malignancies, 20 while an ORR of 25% was reported in a phase 2 study of the oral pan-PI3K inhibitor buparlisib (median PFS of 9.8 months in patients with FL). 21 Overall, the most common TEAEs of any grade reported with long-term treatment with copanlisib were transient hyperglycemia (50.0%), diarrhea (35.2%), transient hypertension (29.6%), neutropenia (28.9%), and pyrexia (26.8%), remaining consistent with the primary analysis and with previous reports of copanlisib. 2,22,23 Events continued to be transient and manageable. Hyperglycemia has also been reported as a common TEAE with other PI3K inhibitors 21,24-26 and is believed to be an on-target effect of PI3K-α inhibition due to its welldescribed role in insulin signaling. 27 The incidence and severity of hyperglycemia with copanlisib were similar to those reported with the orally administered α-selective PI3K inhibitor alpelisib. 28 Although hyperglycemia with the latter agent is presumably more prolonged due to daily administration, hyperglycemia observed with copanlisib administered intravenously once weekly is transient and returns to baseline within 24 hours. 29 Interestingly, the incidences of hyperglycemia and hypertension appeared to decrease in later use compared with the earlier time points. Whether this reflects improved patient management or the development of tolerance cannot be determined at this time. In general, in later treatment intervals and for the patients treated for more than 1 year with copanlisib, the most commonly reported events showed no evidence for increased incidence or worsening severity following longer exposure, suggesting that worst-grade TEAEs tended to occur earlier in treatment. In addition, there was no evidence of new or unexpected TEAEs in patients exposed to copanlisib for longer duration. The safety profile that was observed from this extended analysis remains consistent with the known safety profile of copanlisib.
Of particular interest was whether there was an occurrence of lateonset worst-grade or severe toxicities with longer exposure to copanlisib. An additional four patients experienced grade 3 diarrhea (total incidence of grade 3 diarrhea was 8.5%). In the 37 patients treated with copanlisib for more than 1 year, four patients experienced grade 3 diarrhea with an onset after 12 months of treatment, although it is important to note that some patients in this subgroup had received treatment for more than 3 years. Overall, there were no cases of grade 4 diarrhea. Rates of individual severe toxicities also remained low, with all-grade SAEs of pneumonitis (4.2%), hyperglycemia (4.9%), and diarrhea (2.8%) unchanged at the extended cut-off date; there was one additional patient with all-grade SAE of lung infection. Colitis (one case overall) or other severe gastrointestinal toxicities or hepatic transaminase elevations remained low, suggesting a lack of late-onset toxicities of concern for patients receiving extended treatment with copanlisib. Higher rates of diarrhea and other severe gastrointestinal toxicities have been reported with other PI3K inhibitors that are approved for the treatment of B-cell lymphoma. This includes a phase 2 study of duvelisib with a similar median duration of treatment (6.7 months) 6 as well as a phase 2 study of idelalisib with a longer median duration of treatment (11.1 months). 19 In addition, higher rates of colitis, ulcerative colitis, and other severe gastrointestinal toxicities have been reported for idelalisib 9 and for duvelisib. 6 The low rate of severe gastrointestinal toxicities such as colitis or severe liver enzyme elevations reported with copanlisib may be attributable to lower exposure of the gastrointestinal tract to copanlisib due to the intermittent dose schedule and intravenous route of administration vs continuously administered oral agents such as idelalisib 9 or duvelisib. 30 In conclusion, long-term follow-up of copanlisib demonstrated sustained and enhanced efficacy over time, including additional durable complete responses. The AEs remained transient and manageable over time, including in those patients treated with copanlisib for more than 1 year. No new or unexpected TEAEs were reported in patients exposed to copanlisib for longer duration. In addition, the lack of late-onset toxicities and severe gastrointestinal toxicities suggests that the approved intravenous route of administration and the intermittent dose schedule for copanlisib benefit tolerability. Studies are currently investigating copanlisib in combination with standard of care treatments in relapsed indolent non-Hodgkin lymphoma (NCT02367040 and NCT02626455).

ACKNOWLEDGMENTS
This study was supported by research funding from Bayer AG. The authors wish to thank the patients and their families, co-investigators, and referring physicians who participated in this study. Laura Valenzo, PhD, of Complete HealthVizion, provided medical writing assistance with this manuscript, based on detailed discussion and feedback from all the authors; this assistance was funded by Bayer AG. collection, analysis, and interpretation. L.R., F.H., A.M., J.G.V., and B.H.C. participated in the literature review, study design and protocol development, and data analysis and interpretation. All authors participated in writing and/or critically reviewing the manuscript and approved this version for submission.