Comparison of Progression-Free Survival of Relapsed or Refractory Follicular Lymphoma Patients in a Single-Arm Trial of Tazemetostat with a Real-World External Control Arm

Introduction: Follicular lymphoma (FL) is a common type of non-Hodgkin's lymphoma. Currently, FL is not curable and often requires multiple lines of therapy. Tazemetostat is the first epigenetic-targeted agent approved for the treatment of FL patients. It received accelerated approval in the US for patients with relapsed or refractory FL based on the results of 2 single-arm cohorts from a multi-center Phase 2 trial (E7438-G000-101 [NCT01897571]).

Objective: This retrospective, observational, external control arm (ECA) study compared progression-free survival (PFS) of relapsed/refractory FL patients who received tazemetostat monotherapy in E7438-G000-101 with the real-world PFS (rwPFS) of patients included in the nationwide Flatiron Health electronic health record (EHR)-derived de-identified database who received the prescribing physician's treatment of choice.

Methods: ECA eligibility criteria were adapted from the inclusion and exclusion criteria for the E7438-G000-101 trial. All eligible third line or higher (3L+) treatments for ECA patients were considered for inclusion. For patients with multiple eligible lines of treatment, propensity score-based methods were used to select the line most comparable to those in the treatment arm. Real-world patients in the ECA were followed from the start date of their selected 3L+ treatment to the end of the patient record in the database. rwPFS in Flatiron Health is the time from the index date to death or a clinician-documented episode of worsening disease.

We adjusted for potential patient differences using overlap weighting of 11 baseline characteristics. Multiple imputations were used to handle missing values for baseline characteristics needed to apply the eligibility criteria and derive propensity scores. An impute-then-exclude approach was applied, where missing values were first imputed, and patients not meeting the eligibility criteria were excluded from the analysis. Patients not meeting the eligibility criteria were excluded from the analysis.

PFS and rwPFS were estimated using the Kaplan-Meier method. Data sets were compared using a hazard ratio estimated from Cox proportional hazards regression with and without overlap weights applied. The corresponding 95% confidence intervals (CIs) were calculated using robust standard errors. Sample sizes and event counts were reported as medians across the imputed data sets for each group.

Results: Of the 92 patients in the treatment arm, 55 progressed or died. In the ECA, 104 of the 181 patients progressed or died. The most common treatments in the ECA were CD20-monotherapy, bendamustine-containing regimens, and lenalidomide plus CD20. Before applying overlap weights, the median PFS was 12.2 months (95% CI: 11.0, 17.2), and rwPFS was 16.5 months (95% CI: 12.1, 25.7), respectively, and the hazard ratio was 1.31 (95% CI: 0.92, 1.87). After applying overlap weights, the median PFS was 14.8 months (95% CI: 11.2, not evaluable), and rwPFS was 12.3 months (95% CI: 7.5, 21.9), respectively, and the hazard ratio was 0.90 (95%: 0.54, 1.50).

Conclusion: PFS was numerically longer but not significantly different between tazemetostat monotherapy and real-world physician's choice of treatment. These findings contextualize the E7438-G000-101 trial results.

Salles:AbbVie: Consultancy, Research Funding; BeiGene: Consultancy; Merck: Consultancy; Kite/Gilead: Consultancy; Genentech/Roche: Consultancy, Research Funding; Janssen: Consultancy, Research Funding; Incyte: Consultancy; Ipsen: Consultancy, Research Funding; BMS/Celgene: Consultancy; Genmab: Consultancy, Research Funding; Molecular Partners: Consultancy; Nurix: Research Funding. Leonard:AbbVie, AstraZeneca, Astellas, Bayer, BeiGene,BMS, Calithera, Constellation, Eisai, Epizyme,GenMab, Grail, Incyte, Janssen, Karyopharm, Lilly,Merck, Mustang Bio, Pfi zer, Roche/Genentech,Seagen, Second Genome, Sutro: Consultancy. Morschhauser:Roche: Consultancy, Honoraria; Janssen: Honoraria; AbbVie: Consultancy, Honoraria; Takeda: Honoraria; Chugai: Honoraria; BMS: Consultancy; Novartis: Consultancy; Kite/Gilead: Consultancy. Crowley:Genesis Research Group: Current Employment; Ipsen: Consultancy, Research Funding. Sosinsky:Genesis Research Group: Current Employment; Ipsen: Consultancy, Research Funding. Lieb:Ipsen: Consultancy; Genesis Research Group: Current Employment. Zeldow:Ipsen: Consultancy, Research Funding; Genesis Research Group: Current Employment. Harton:Ipsen: Consultancy, Research Funding; Genesis Research Group: Current Employment. Wang:Ipsen: Consultancy, Research Funding; Genesis Research Group: Current Employment. Boyer:Ipsen: Current Employment, Current equity holder in private company. Doban:Ipsen: Current Employment, Current equity holder in private company. Bonnet:Ipsen: Current Employment, Current equity holder in private company. Perrot:Ipsen: Current Employment, Current equity holder in private company. Veazey:Ipsen: Current Employment, Current equity holder in private company.