Targeting the methylome in T-LGL leukemia: Pre-clinical analysis and Results of a Phase I trial with oral 5-azacytidine

Abstract

Abstract Epigenetic dysregulation, particularly changes in DNA methylation, are major contributors to the onset and progression of Large Granular Lymphocytic Leukemia (LGLL). We previously identified hypermethylation of tumor suppressor genes as a factor in LGLL pathogenesis; however, the full landscape of epigenetic dysregulation and causality versus correlation to transcriptomic dysregulation remains unknown. In this study, we performed reduced representation bisulfite sequencing to profile the methylome, alongside RNA sequencing, on flow cytometry sorted CD3⁺CD8⁺CD5⁻/dim peripheral blood cells isolated from treatment-naïve LGLL patients and healthy donors. The analysis of differentially methylated regions (DMR) revealed increased hypermethylation in patients relative to healthy donors, spanning exons, introns, promoters, and transcription start sites. While promoter methylation usually suppresses gene expression, some genes showed increased transcription despite being hypermethylated. A notable example is IL-15, whose expression was upregulated following promoter methylation, due to silencing of a repressor of IL-15. To investigate the role of methylation in this atypical gene activation, we treated MOTN1 cells (human LGLL cell line) with the demethylating agent 5-aza-2′-deoxycytidine (5-aza). This intervention reversed the methylation status, resulting in reduced expression of IL-15, thereby confirming that its transcriptional upregulation was driven by promoter methylation (p<0.001). Further supporting the role of IL-15 in disease pathogenesis, transplanting MOTN1 cells revealed a markedly increased LGLL burden and reduced overall survival in NSG-Tg(Hu-IL15) transgenic mice compared to NSG control mice (n=5 each, p<0.05). To assess whether treatment 5-aza and reversal of the associated methylation signature influenced LGLL cell viability and proliferation, MOTN1 cells labeled with CellTrace dye were incubated with 5-aza for 48 hours in an in vitro assay. Cell viability was measured using Zombie Aqua staining, while proliferation via dilution of the CellTrace dye. The results demonstrated a dose-dependent increase in LGLL cell death and a corresponding reduction in LGLL proliferation. Parallel experiments conducted on primary patient samples (N=5) revealed similar alterations in viability and proliferative capacity following 5-aza treatment in vitro. In vivo, NSG huIL-15 mice engrafted with LGLL cells and treated with 5-aza (3 mg/kg) showed significant LGLL reduction via IVIS imaging and improved overall survival (n=5 each, p<0.001), supporting the potential of epigenetic therapy in LGLL therapy. Given the strong preclinical impact of 5-aza on T-LGLL cells, we initiated a multi-center Phase I/II trial using oral 5-aza (CC-486) in relapsed/refractory T-LGLL patients (NCT05141682). Here we report results from the Phase 1 dose-escalation trial, where patients received 3 escalating doses of oral 5-aza (100, 200, 300 mg) on days 1-14 of a 28 day cycle. Dose-limiting toxicities (DLT) were evaluated during the first two cycles to establish the recommended phase II dosage (RP2D) using a Baysian Optimal Interval Design (BOIN). Secondary endpoints included overall partial and complete response rates per ECOG5998 criteria. Nine patients were enrolled and evaluated for DLT and response. The median age was 60 (range: 32-80), and 7 (78%) were female. 3 (33%) patients were treated for transfusion/symptomatic anemia, 6 (67%) patients were treated for neutropenia (ANC<500 or ANC <1500 with frequent infections) and the median number of prior therapies was 4 (range: 2-5). Three patients each received 100, 200, or 300mg doses. No dose-limiting toxicities were observed and the RP2D dose was determined to be 300mg. Per ECOG 5998, one patient per dose level attained a partial response (ORR 33%), with two patients attaining resolution of transfusion-dependent anemia, and one patient achieving resolution of long-standing severe neutropenia. All three patients remain on therapy. LGLL counts showed a marked decline with 5-aza in patients, underscoring the role of hypermethylation in T-LGLL. In conclusion, methylation is a therapeutic target in LGLL. In vitro, in vivo, and Phase I trial data show that epigenomic targeting with 5-aza yields clinical responses across doses without DLTs. The Phase II component of the oral 5-aza clinical trial is currently ongoing, reinforcing the critical role of the methylome in T-LGLL pathogenesis.