Duvelisib in Patients with Relapsed/Refractory Peripheral T-Cell Lymphoma: Final Results from the Phase 2 PRIMO Trial

Introduction Peripheral T-cell lymphoma (PTCL) is a heterogeneous group of aggressive lymphomas, with poor prognosis and historically median overall survival (mOS) <6 months (mo) for relapsed or refractory (R/R) disease. In North America and Europe, the 3 most common subtypes are PTCL-NOS (not otherwise specified), angioimmunoblastic T-cell lymphoma (AITL), and anaplastic large-cell lymphoma (ALCL). Current single agent therapies for R/R PTCL have modest overall response rates (ORR) of <30% (with the exception of brentuximab vedotin for CD30+ ALCL). Duvelisib is an oral inhibitor of phosphatidylinositol 3-kinase (PI3K)-δ and PI3K-γ isoforms. PRIMO (NCT03372057) was an open-label, single-arm trial that evaluated the safety and efficacy of duvelisib monotherapy in R/R PTCL in patients from the US, EU, UK, and Japan. Methods Eligibility criteria included adults with pathologically confirmed R/R PTCL. Pneumocystis jirovecii prophylaxis (PJP) was required. After the dose optimization phase, optimal dose was determined as duvelisib 75 mg BID for 2 cycles (to maximize disease control), followed by 25 mg BID (to mitigate late toxicities), until progressive disease (PD) or unacceptable toxicity. The primary endpoint was ORR by Independent Review Committee (IRC) assessment (Lugano 2014 criteria). Here we report final results of the PRIMO dose expansion phase (EP). Results The PRIMO-EP included 123 patients with R/R PTCL (data cutoff: Feb 2, 2024). Baseline characteristics: median age: 65 (range 21-92) years, male sex: 54.5%, median prior lines of therapy: 2 (range 1-9), most common baseline histologies: PTCL-NOS (n=53), AITL (n=37), and ALCL (n=20). Common reasons for duvelisib discontinuation included: PD: 41%, adverse events (AEs): 22%, stem cell transplant (SCT): 10%, and death: 8%. The median follow-up for PFS was 6.24 mo. Efficacy outcomes: ORR: 48%, complete response rate (CRR): 33%, median progression-free survival (mPFS): 3.45 mo, median duration of response (DOR): 7.89 mo, and mOS: 12.35 mo. Outcomes stratified by baseline histology: ORR: 49.1% (PTCL-NOS), 62.2% (AITL), 15.0% (ALCL); mPFS: 3.45 mo (PTCL-NOS), 8.34 mo (AITL), 1.64 mo (ALCL); mOS: 11.04 mo (PTCL-NOS), 18.07 mo (AITL), 6.34 mo (ALCL); mDOR: 7.39 mo (PTCL-NOS), 11.70 mo (AITL), 1.97 mo (ALCL). Nineteen patients (15%) received SCT after PRIMO (11/12 with planned SCT at time of duvelisib treatment discontinuation plus 8 additional patients). There were 4 (3.3%) fatal treatment-related AEs (cryptococcosis, Epstein-Barr virus associated lymphoproliferative disorder, pneumonitis, and sepsis) and 6 (4.9%) fatal AEs other than PD which were unrelated or unlikely related to duvelisib. AEs of special interest (Grade ≥3) included: ALT/AST elevation: 24.4%, neutropenia: 17.9%, infections: 13.0%, cutaneous reactions: 10.6%, diarrhea: 9.8%, pneumonia: 4.1%, pneumonitis: 0.8%, and colitis: 0.8%. There were no cases of PJP. AEs leading to dose hold or dose reduction occurred in 44.7% and 9.8% of patients, respectively. The median relative dose intensity was 97.7%. Conclusions The final analysis of the PRIMO-EP demonstrated an ORR of 48% and a CRR of 33%, with 15% of patients continuing to SCT; these results compare favorably to currently available single-agent options. AEs were generally manageable with per-protocol dose modifications and were consistent with those observed previously. PRIMO-EP results confirm duvelisib to be a promising agent for this disease with a high unmet need, poor prognosis, and limited effective treatment options. Based on these results, the sponsor is initiating a randomized phase 3 study in the EU/UK to investigate duvelisib in R/R nodal T follicular helper cell lymphoma (TERZO™).