Outcomes With Idelalisib Differ in the Clinical Trial Versus Real-World Setting

According to a real-world analysis published in JAMA Oncology, treatment outcomes with idelalisib differed between patients with follicular lymphoma (FL) and chronic lymphocytic leukemia (CLL) treated in the clinical setting or in a clinical trial, with Medicare beneficiaries faring worse than their clinical trial counterparts. The differences could be attributed to selective eligibility criteria and frequent dose reductions in clinical trials, the authors, led by Steven T. Bird, PhD, PharmD, MS, from the FDA’s Office of Pharmacovigilance and Epidemiology, noted.

“Idelalisib is approved as monotherapy in relapsed follicular lymphoma and with rituximab for relapsed chronic lymphocytic leukemia,” the authors explained. “Toxic effects can be severe and treatment-limiting, [and] outcomes in a real-world population are not yet characterized.”

With this analysis, investigators compared idelalisib treatment outcomes in the clinical setting with outcomes reported in clinical trial data of patients with FL and CLL.

Treatment duration, on-treatment mortality, overall mortality, and serious and fatal infections were compared between real-world patients and participants in two trials – a single-group, open-label phase II trial supporting idelalisib’s approval for relapsed or refractory FL and a randomized, double-blind phase III trial supporting approval of idelalisib plus rituximab for relapsed CLL.

The study included 115 clinical trial participants and 599 Medicare beneficiaries aged 65 years or older:

  • FL: 26 trial participants (mean age = 73 years) and 305 Medicare beneficiaries (mean age = 76 years)
  • CLL: 89 trial participants (mean age = 74 years) and 294 Medicare beneficiaries (mean age = 76 years)

“Medicare beneficiaries were older and sicker and had unfavorable imbalances in treatment duration, mortality, and fatal infections compared with clinical trial participants,” the authors reported.

The median treatment duration was shorter for real-world patients with CLL compared with clinical trial participants (173 days vs. 473 days; p<0.001); however, there was no significant difference in the FL cohort (114 days vs. 160 days; p=0.38). This could be related to the fact that there were higher rates of idelalisib dose reductions in the trial setting: 32.6% versus 18.0% for CLL (p=0.003) and 38.5% versus 16.1% for FL (p=0.02).

Compared with trial participants, Medicare beneficiaries in both disease cohorts had higher mortality rates (CLL: HR=1.40 [95% CI 0.93-2.11]; FL: HR=1.39 [95% CI 0.69-2.78]), as well as higher rates of fatal infections per 100 person-years (CLL: 18.4 vs. 9.8 [p=0.04]; FL: 27.6 vs. 18.6 [p=0.54). Among Medicare beneficiaries, an infection requiring hospitalization within 6 months of idelalisib treatment increased the risk of on-treatment fatal infections more than twofold. Ten of the serious on-treatment infections had Pneumocystis jirovecii pneumonia (PJP) listed as a contributory cause. Four of these infections were fatal; in the fatal cases, no PJP prophylaxis had been given (prophylaxis is recommended in the prescribing instructions).

These findings suggest that the data observed in clinical trials of patients treated with idelalisib may not be generalizable to clinical practice. “Efficacy and tolerability may be meaningfully different for optimally selected and closely monitored trial participants compared with unrestricted patients in the clinical setting,” the authors concluded.

“Compliance with [infection] prophylaxis and close monitoring of absolute neutrophil counts, with appropriate dose interruptions and reductions, may help reduce the rate of serious infections [for real-world patients],” the researchers concluded.

The authors report no relevant conflicts of interest.


Bird ST, Tian F, Flowers N, et al. Idelalisib for treatment of relapsed follicular lymphoma and chronic lymphocytic leukemia: a comparison of treatment outcomes in clinical trial participants vs Medicare beneficiaries. JAMA Oncol. 2019 December 19. [Epub ahead of print]