Multi-Center Phase II Study of Oral Azacitidine (CC-486) Plus CHOP As Initial Treatment for Peripheral T-Cell Lymphoma (PTCL)

INTRODUCTION:

Nodal PTCL with T-follicular helper phenotype (PTCL-TFH), which includes angioimmunoblastic T-cell lymphoma (AITL), is characterized by recurrent mutations affecting epigenetic regulators such as TET2, DNMT3A, IDH2 and RHOA. The association of aberrant DNA methylation with lymphomagenesis provides rationale for clinical application of hypomethylating agents. Azacitidine, an epigenetic modifier which inhibits DNA methyltransferase, has shown clinical activity as a single agent and in combination in R/R PTCL. We report the findings of the first study of oral azacitidine (CC-486) plus CHOP as initial treatment for PTCL (ClinicalTrials.gov - NCT03542266).

METHODS:

This phase 2 study prioritized enrollment of PTCL-TFH. Subjects received CHOP on day 1 of each cycle for 6 cycles. Priming with oral azacitidine (CC-486) at 300 mg daily was administered for 7 days prior to cycle 1 of CHOP, and for 14 days before CHOP cycles 2-6. Supportive care included mandatory G-CSF. The primary endpoint is CR per 2014 IWG criteria. Secondary endpoints include ORR, safety and survival. Correlative biomarker studies are planned to assess genomic mutations by next-generation-sequencing (NGS), in addition to methylation and transcription profiles. Using a Simon two-stage design comparing an CR of ≥60% with treatment to an unacceptable CR of ≤35% (alpha=10%, power=80%), 9 or more CR out of 17 enrolled patients were required to declare the treatment worthy of further study.

RESULTS:

From 6/2018 to 3/2020, 21 subjects with previously untreated PTCL were enrolled at 4 centers, and the study met its accrual. At study entry, 17 patients (81%) had PTCL-TFH (16 AITL and 1 TFH), 3 with PTCL-NOS, 1 with ATLL, including 5 (24%) with CD30+ disease. The median age was 66 years (range 22-77), and the M:F ratio was 1.6:1. Nineteen (90%) had stage III/IV disease, 10 (48%) had elevated LDH, 7 (33%) had bone marrow involvement, and 9 (43%) had IPI 3-5. Treatment was generally well tolerated with expected side effects. Grade 3-4 hematologic toxicities included neutropenia (71.4%), thrombocytopenia (9.5%) and anemia (14.3%), with febrile neutropenia uncommon (14.3%). Grade 3-4 non-hematologic toxicities included fatigue (14.3%), hyponatremia (14.3%), diarrhea (4.8%), vomiting (4.8%), rash (4.8%), and elevated ALT (4.8%). One incidence each of influenza A, COVID-19 pneumonia, C.diff and strongyloides hyperinfection were observed and treated. There was no study treatment-related mortality to date. As of July 2020 at a median follow-up of 7 months (range 4-25 months), one subject withdrew consent after cycle 1 (patient preference), and 20 subjects had at least one response assessment, including 15 completed treatment, 2 progressed during treatment, and 3 nearing completion of therapy. At interim assessment after cycle 3 (n=20), the ORR was 85% with CR at 55% (90%CI of 34.7%-74.1%). To date, the preliminary end-of-treatment (EOT, n=17) CR was 76.5% (90%CI of 53.9%-91.5%) for all evaluable patients and was 86.7% for 15 PTCL-TFH, exceeding primary endpoint threshold. CR did not correlate with CD30 expression. The estimated 1-yr PFS for all patients was 56.8% (95%CI of 26.3%-87.3%), with 1-yrs PFS for PTCL-TFH at 61.1% (95%CI of 29.5%-92.7%), and the estimated 1-yr OS for all patients was 74.4% (95%CI of 48.8%-100.0%), with 1-yr OS for PTCL-TFH at 88.9% (95%CI of 68.4%-100.0%). Mutational status by NGS was determined in 15 patients to date. The frequencies of the TET2, RHOA, DNMT3A, and IDH2 mutations were 73%, 40%, 13% and 13%, respectively. TET2 mutations were significantly associated with CR (p=0.014), favorable PFS (p-0.012) and OS (p=0.042). In contrast, DNMT3A mutations were associated with adverse OS (p=0.028).

CONCLUSIONS:

This study provides the first demonstration that addition of hypomethylating agent oral azacitidine (CC486) to CHOP as initial therapy is feasible, safe, and induces high CR rate in PTCL-TFH subtype, with expected side effects. Although preliminary, the EOT CR to date exceeds the threshold of meeting study primary endpoint. Final efficacy data as well as response according to subtype and mutational profiling will be updated at ASH. This active combination will be further evaluated in the upcoming ALLIANCE/Intergroup randomized study A051902, comparing oral azacitidine-CHO(E)P with duvelisib-CHO(E)P against CHO(E)P in CD30 negative PTCL.