A Phase 2 trial of AI-BEL (AZT/Interferon-α plus Belinostat) for HTLV-1 related adult T-cell leukemia-lymphoma: Final safety and efficacy Results

Abstract Background: HTLV-1 related adult T-cell leukemia-lymphoma (ATL) is generally fatal and incurable with modern drugs alone. Patients with acute type ATL have the worst outcome with median survival of < 8 months in the largest retrospective studies. At the University of Miami and affiliated hospitals we encounter amongst the highest number of ATL cases in the U.S.A. due to our proximity to HTLV-1 endemic regions (the Caribbean and South America). Our pre-clinical work using patient derived ATL models demonstrated that belinostat (BEL), a pan-HDAC inhibitor, reactivates HTLV-1 (Tax protein) expression in ATL cells in a dose-dependent manner, an effect that was augmented by azidothymidine (AZT, or zidovudine). Belinostat (and other HDAC inhibitors) blocked HTLV-1 b-ZIP factor (HBZ) protein expression resulting in apoptosis of ATL cells. Thus, we hypothesized that reactivating HTLV-1 in host cells of ATL patients using BEL treated with AZT + interferon-α (AI) would provoke anti-HTLV-1/ATL immune responses. We conducted a phase 2 trial of AI-BEL for patients with ATL (Clinicaltrials.gov NCT02737046). Methods: Eligibility criteria included adult patients with newly diagnosed, pre-treated, or relapsed ATL with adequate organ function, ECOG PS ≤ 3, and active disease in peripheral blood without progressive disease after AI or steroids, or after standard or pre-phase chemotherapy ≥ 2 weeks prior. The regimen consisted of up to eight 21-day cycles of starting doses of intravenous belinostat 1,000 mg/m2 (Days 1-5), oral AZT 300 mg three times daily, and interferon-α 5-10 million units daily (n=2) or pegylated interferon-α forms (n=12) 90-180 mcg weekly followed by maintenance AZT and/or pegylated interferon-α until disease progression or drug intolerance. Dose adjustment guidelines for all drugs were based on grade 4 hematologic events. The primary objectives were 1) to evaluate safety, and 2) to determine the regimen's ability to induce a complete molecular response (CMR) by clearing clonal ATL disease from the blood compartment. Secondary objectives included determining overall response and survival rates and evaluating molecular and immune responses in vivo. Results: 15 eligible patients (ages 31-75) with acute type ATL were enrolled and treated between December 2016 and June 2024. Patients completed 1-8 cycles of BEL, which was dose-reduced or discontinued after grade 4 thrombocytopenia (33%) or neutropenia (53%) under protocol guidelines (n=7), or after progressive disease (n=4). Febrile neutropenia and prolonged agranulocytosis were seen in 2 separate subjects (6.7% each). Grade 1 or 2 nausea/vomiting occurred in 75% of subjects. Among 14 of 15 (93.3%) patients evaluable for clinical response, 10 (71%) had a clinical benefit: with overall response rate of 57% (complete response, n=3, 21%; partial response n=5, 36%) and stable disease (n=2, 14%). Additionally, 5/15 (33%) achieved CMR in blood compartment (peripheral blood and bone marrow), which occurred spontaneously in 3 patients after treatment discontinuation. Two patients were early censored for progression-free survival (PFS) due to switching to new treatments. 12-month PFS was 27% (95%CI: 7.3%, 53%) with median PFS of 4.8 months (95%CI: 0.8, 13 months). Estimated 12-month overall survival (OS) was 80% (95%CI: 50%, 93%) with median OS 15.5 months (95%CI: 7.1 months, not estimable). One subject continues to be in complete remission and CMR at 51 months while receiving maintenance pegylated interferon-α. Immunophenotypic studies by investigational flow cytometry studies revealed the expansion of highly cytolytic CD8+ CTLs in patients responding to treatment. Conclusions: AI-BEL showed good efficacy in patients with aggressive ATL resulting in CMR in blood compartment, which occured after treatment discontinuation in multiple subjects. The overall results compare favorably to those observed in historical trials for ATL that included previously treated or relapsed patients. Judicious use of HDAC inhibitors in combination with AZT/interferon-α based treatment can provoke sustained immunologic effects resulting in deep molecular responses. AI-BEL was relatively safe to administer inducing expected hematologic adverse events warranting dose and treatment schedule modifications in future studies. Our data support future clinical testing combining BEL and AZT/interferon-α based treatment as upfront therapy for ATL.