Duvelisib loses support for CLL/SLL indication following Advisory Committee (ODAC) meeting

Top concerns addressed for the continued approval for the highlighted long-term OS data which showed a higher rate of deaths with duvelisib vs ofatumumab (Kesimpta), a higher rate of fatal adverse effects (AEs) attributed to infection, and an OS detriment in the setting of progression-free survival (PFS) and overall response rate (ORR). Voters decided on the question, “Given the potential detriment in OS, duvelisib-associated toxicity, concerns with the selected dose, and the safety issues with the PI3K inhibitor class, is the benefit-risk profile of duvelisib favorable in patients with relapsed or refractory CLL or SLL after at least 2 prior lines of therapy?”

“All of us feel that this is a complex situation just by virtue of the design of the clinical trial,” Jorge A. Garcia, MD, FACP, chair of the ODAC meeting said prior to the vote. “Some of the themes of our discussion relate to our inability to use the clinical trial design to really determine the true detrimental outcome of patients receiving duvelisib and that related to the compounding effect of crossover…There were comments on the concerns of significant treatment-related AEs some of which could lead to death in the duvelisib arm. [But] there was the inability to know if they were related to true treatment effects or progression of disease while they were on treatment or on subsequent therapy.”

Supporting arguments for the continued approval noted that the safety profile was consistent with the data referenced for the approval. The rate of any-grade treatment-related AEs was 98.7% at the interim analysis vs 100% in the final analysis. Grade 3 or higher AEs were 87.3% vs 91.1% in the interim and final analyses, respectively. Additionally, the associated poor prognosis for patients for patients with resistance to both BLC2 and BTK inhibitors, can be curbed with PI3K inhibitors improving PFS outcomes. Susan O’Brien, MD, professor of medicine in the Division of Hematology/Oncology at the University of California at Irvine, highlighted that full drug approvals for agents for the treatment of patients with relapsed or refractory CLL have been based on PFS outcomes. She added that with since PI3K inhibitors have been available for many years, and idelalisib (Zydelig) has similar safety concerns and box warnings, clinicians have experience managing treatment-related AEs..

Initial Approval Duvelisib is the only agent approved specifically as a third-line monotherapy for relapsed or refractory CLL and is recommended in the National Comprehensive Cancer Network guidelines as of August 2022.

The application, submitted by Secura Bio Inc, was approved in 2018 based on data from the phase III DUO trial (NCT02004522), which evaluated duvelisib vs ofatumumab in 319 adults with CLL or SLL after at least 1 prior therapy. In data supporting the application, the median PFS in the intention-to-treat (ITT) population was 13.3 months (95% CI, 12.1-16.8) vs 9.9 months (95% CI, 9.2, 11.3) with duvelisib and ofatumumab, respectively (HR, 0.52; 95% CI, 0.39-0.69, P < .0001). The ORR in the duvelisib arm (n = 160) was 73.8% (95% CI, 66.9%-80.6%) vs 45.3% (95% CI, 37.5%-53.0%) in the ofatumumab arm (n = 159). At the time of approval, the median OS data were immature with a median of 24 months of follow-up, and the OS was not reached in either arm in the indicated population (HR, 0.82; 95% CI, 0.49-1.37).

Reports of substantial toxicity including fatal AEs led the FDA to issue a box warning to address the toxicities and a postmarketing requirement was issued to Secura Bio Inc requesting 5-year follow-up data for OS. The identified safety concerns included serious or fatal infections, diarrhea or colitis, rash, pneumonitis, hepatotoxicity, and neutropenia.

Updated Findings; Five-year OS follow-up data from the DUO trial were submitted in June 2021 with a data cutoff of January 22, 2021. During the hearing, Deepti Telaraja, MD, a medical officer in the Division of Hematologic Malignancies II at the FDA, presented the findings.“It is important to note that benefit/risk is continuously assessed as new information is available,” Telaraja said. “The 5-year OS data has prompted this assessment. I would also like to highlight some key considerations about OS as an end point…it is an objective measure of clinical benefit and is considered both an efficacy and safety end point. an evaluation of toxicity is embedded in an assessment of OS including the ability to assess short and long-term toxicity. further, the degree of statistical consideration [that is applied] when OS is used as a primary efficacy end point do not apply when OS is used as a safety end point. Finally, the FDA requires OS information in any trial using PFS as a primary end point in order to provide benefit/risk.”

A potential OS detriment was observed in the updated data. The median OS in the ITT population was 52.3 months (95% CI, 41.8-68.0) in the duvelisib arm vs 63.3 months (95% CI, 41.2–not estimable) in the ofatumumab arm (HR, 1.09; 95% CI, 0.79-1.51). This detriment extended to the indicated population with a median OS of 43.9 months (95% CI, 32.4-56.5) among patients who received duvelisib after 2 or more prior lines of therapy (n = 95) vs 46.8 (95% CI, 28.6-74.9) among those who received ofatumumab (n = 101; HR, 1.06; 95% CI, 0.71-1.58).The updated median PFS in the ITT duvelisib arm was 17.85 months (95% CI, 15.16-22.59) vs 9.47 months (95% CI, 9.14-11.14) in the ofatumumab arm (HR, 0.40; 95% CI, 0.27-0.59).Safety data also signaled a higher rate of death in duvelisib-treated population vs ofatumumab-treated population. In the ITT analysis 23 deaths were reported, 13 of which were among patients with 2 or more prior lines of therapy in the duvelisib cohort. In the ofatumumab population, 5 deaths occurred overall and 4 were in the indicated population. Infection was the primary cause of death in the duvelisib arm accounting for 14 events. One patient died of infection due to ofatumumab treatment.

Following this analysis, the FDA issued a safety alert in June 2022 intended to alert patients and health care providers of the potential risk of continued use of duvelisib.Of note, crossover upon disease progression was permitted in the DUO trial. Ninety patients (57%) crossed over from the ofatumumab arm to the duvelisib arm compared with 9 patients (6%) from duvelisib to ofatumumab upon disease progression. A sensitivity analysis was performed and demonstrated consistent potential OS detriment with the primary analysis. Further, most subgroups also reported an OS detriment with duvelisib, but for those that did not (ie, those who were refractory or had early relapse to purine treatment) would require more substantial analysis.

Troubles with P13K inhibitors: April 2022, ODCA voted unanimously (16-0) with a single abstention to recommend that future approvals of PI3K inhibitors for patients with hematologic cancers be supported by randomized clinical data. As a result of the hearing, indications were withdrawn for duvelisib, idelalisib, and umbralisib (Ukoniq), with each losing an indication in follicular lymphoma. Manufacturers also voluntarily withdrew indications for SLL and marginal zone lymphoma (MZL) for idelalisib and umbralisib, respectively. Further, new drug applications for copanlisib (Aliqopa) for follicular lymphoma and MZL and umbralisib for CLL were withdrawn.Cited problematic concerns during the meeting included concerning patterns of PFS benefit followed by OS detriment and the use of PFS as an end point for approval in the setting of significant toxicity.