Dual-Targeted Therapy with Ruxolitinib Plus Duvelisib for T-Cell Lymphoma

Introduction: Both duvelisib, a phosphoinositive-3-kinase (PI3K)-γδ inhibitor, and ruxolitinib, a Janus kinase (JAK)-1/2 inhibitor, are effective agents for peripheral and cutaneous T cell lymphomas (PTCL and CTCL) that appear to target active, yet parallel, pathways in these diseases. In fact, PI3K signaling activation predicted for resistance to ruxolitinib while JAK/signal transducer and activator of transcription (STAT) signaling activation predicted for resistance to duvelisib-based therapy (Moskowitz, et al. Blood 2021; Horwitz, et al. Nature Medicine 2024). Building on these observations, we conducted a phase I study assessing dual-targeting of JAK and PI3K with ruxolitinib plus duvelisib for PTCL or CTCL.

Methods: This is an investigator-initiated multi-center phase I study for pts with relapsed or refractory PTCL or CTCL or untreated T-prolymphocytic leukemia (T-PLL). In part I, we aimed to identify the maximum tolerated dose (MTD) through a 3+3 design assessing 3 dose levels (ruxolitinib 20mg po BID plus duvelisib 25mg, 50mg, or 75mg PO BID). In part II, we assessed efficacy in 2 expansion cohorts for pts with lymphoma characterized by presence of genetic or immunohistochemical JAK/STAT pathway activation (cohort A) or absence of JAK/STAT activation (cohort B). Prophylaxis against herpes zoster and Pneumocystis pneumonia (PCP) was required.

Results: 49 pts enrolled, including 8 in part I and 41 in part II. Median age was 65 (28-86); 47% were female; 80% white, 12% black, 4% Asian, and 4% unknown. Median number of prior therapies was 2 (range 0-12). Histologies included 14 (29%) T-follicular helper (TFH) lymphomas, 13 (27%) peripheral T-cell lymphoma, not otherwise specified (PTCL, NOS), 7 (14%) mycosis fungoides (MF), 5 (10%) T-PLL, 3 (6%) ALK-negative anaplastic large cell lymphoma (ALCL), 3 (6%) T-cell large granular lymphocytic leukemia (T-LGL), and 1 pt each with ALK-positive ALCL, adult T-cell lymphoma/leukemia, and monomorphic epitheliotropic intestinal T cell lymphoma.

In part I, 8 pts enrolled on dose level 1 and 6 were evaluable for dose-limiting toxicity (DLT, 2 pts not evaluable for DLT due to early disease progression). One DLT (grade 4 thrombocytopenia) occurred, leading to dose level 1 (ruxolitinib 20mg BID plus duvelisib 25mg BID) being accepted as the MTD.

In part II, 23 pts enrolled onto cohort A, 14 pts onto cohort B, and 4 pts are not yet assigned (JAK/STAT data pending). All 8 pts enrolled on part I had lymphomas associated with JAK/STAT activation and thus were included with cohort A for efficacy analysis.

Treatment-related grade 3 drug-related adverse events included neutropenia (24% G3, 14% G4), anemia (16% G3), thrombocytopenia (6% G3, 6% G4), lung infection (4% G3), hypertension (4% G3), hypertriglyceridemia (4% G3), transaminitis (4% G3), sepsis (2% G3, 2% G5), urinary tract infection (2% G3), diarrhea (2% G3), weight gain (2% G3), leukopenia (2% G3), and mucositis (2% G3).

All enrolled patients were response evaluable and the overall response rate (ORR) was 41%, complete response (CR) rate 24%, and partial response (PR) rate 16%. Median duration of response (DOR) was not reached. ORR and CR rates in cohort A were 52% and 29% compared to 14% and 14% in cohort B (p=0.023). Highest activity was observed for TFH lymphomas (ORR 79%, CR 64%) and T-PLL (ORR 60%, CR 0%). ORR and CR rates among other histologies were 67% and 33% for T-LGL, 23% and 15% for PTCL, NOS, and 14% and 0% for MF.

Conclusion: Ruxolitinib 20mg BID plus duvelisib 25mg BID was determined to be the MTD. The toxicity profile of this regimen was manageable with low rates of typical PI3Ki-related toxicities such as hepatitis, rash, or colitis, which were possibly diminished by ruxolitinib. The combination is particularly active in TFH lymphomas and T-PLL as well as cases with evidence of JAK/STAT activation. Further expansion for TFH lymphomas and T-PLL is planned.