Background: Lysine-specific demethylase-1 (LSD1) is an activity critical for the self-renewal of malignant myeloid cells and maturation of megakaryocytes, cells central to the pathogenesis of MF. Bomedemstat is an orally active LSD1 inhibitor that in mouse models ameliorated the hallmarks of MPNs and improved survival (Kleppe et al. 2015; Jutzi et al. 2018).
Aims: IMG-7289-CTP-102 is an ongoing, global, open-label Phase 2 study evaluating bomedemstat dosed QD in MF patients (NCT03136185). Key eligibility criteria include patients intolerant, refractory, resistant, or inadequately controlled by approved therapy, and platelet count ≥100 x 109/L. Key objectives are safety and reduction of spleen volume (SVR) by MRI/CT and total symptoms scores (TSS) using the MPN-SAF instrument.
Methods: Serial bone marrow (BM) biopsies and imaging studies are read centrally. 261 genes are serially sequenced to quantify changes in the allelic frequencies of mutations (MAF) and identify new mutations. The starting dose is 0.6 mg/kg/d titrating, as needed, to a platelet count of 50-75x109/L.
Results: At 89 patients, the study is now fully enrolled: 46% primary MF, 33% post-essential thrombocythaemia-MF, 21% post-polycythaemia vera-MF. Median age is 68 (35-88) with 52% males. Prior treatment with ruxolitinib was reported in 83% (74/89); 46% had also received at least 1 additional treatment. 30% of patients (27/89) had received ≥1 RBC transfusion prior to dosing. By IPSS, 53% were high-risk, 40% int.-2, and 7% int.-1. At screening (N=103), sequencing to a mean depth of >1000 bp, JAK2 was mutated in 69%, CALR in 21%, MPL in 6%; 59% had ≥2 mutations of which 71% were high-molecular risk mutations in ASXL1, IDH1/2, EZH2, U2AF1, TP53 and/or SRSF2.
At data cutoff (3 Feb 2022), the median duration of treatment is 28 weeks (2-131). Of patients for whom TSS data is available at 24 weeks in those with baseline values ≥20, 72% (18/25) had a reduction in TSS; 24% (6/25) reported a ≥50% reduction. Of patients evaluable for SVR (N=50), 64% had a reduction in spleen volume from baseline; in 28%, SVR was ≥20%. Of evaluable patients (N=41), 90% had stable (∆ <±1.0 g/dL) or improved (≥1.0 g/dL) hemoglobin. Of patients with BM fibrosis scoring post-baseline (N=52), 31% improved by 1 grade and 50% were stable. CCL5 and S100A8/A9 cytokines were elevated at baseline in 50% (16/32) and 78%, respectively; at Day 84, 81% and 68%, respectively, showed reductions of at least 10% and of those, 100% (CCL5) and 47% (S100A8/A9) normalized. In follow-up sequencing at around Week 24, of 60 mutant alleles in 32 patients, the mean MAF fell by 39% in 48%. JAK2 MAFs fell by 31% [SD 5%] in 46% (N=24); ASXL1 MAFs fell 40% in 71%. Clones with JAK2 and/or ASXL1 mutations were most affected. No new mutations have been identified and no patient has transformed to AML.
The most common non-hematologic AEs reported by patients was dysgeusia in 36% (32/90) and diarrhoea in 34% (31/90). All dysgeusia events were grade 1/2 and 1 led to treatment discontinuation. Of 14 related SAEs, 4 were Grade 2, 9 Grade 3 and 1 Grade 4 (thrombocytopenia). Twenty-nine patients remain on bomedemstat. Early terminations due to AEs occurred in 18 (20%) patients (9 related to bomedemstat), and 13 discontinued for other reasons. There have been no safety signals, DLTs, or deaths related to drug.
Summary/Conclusion: In patients with advanced MF, bomedemstat alone had an acceptable tolerability profile, relieved symptoms, reduced spleen volume and mutation burden while improving fibrosis and anemia without safety signals.