Modulation of Biomarkers By BET Inhibitor, BMS-986158, Including JAK2 Variant Allele Frequency (VAF), Bone Marrow (BM) Fibrosis, and Reversal of Abnormal Cytokine Production in Intermediate- or High-Risk Myelofibrosis (MF)

Background

MF, driven by JAK/STAT pathway activation, results in an aberrant stem cell niche, leading to abnormal cytokine production, an inflammatory BM microenvironment, splenomegaly, and an array of other symptoms. Bromodomain and extra terminal (BET) proteins play a role in cancer cell proliferation, survival, and oncogenic progression. In combination with Janus kinase inhibitors (JAKi), ruxolitinib (RUX) or fedratinib (FED), BET inhibitors (BETi) have been shown to reduce inflammatory signals and disease burden in preclinical models of MF. BMS-986158 is an orally bioavailable, potent, and selective small-molecule BETi (Hilton et al. ESMO 2018. Abstr 1429). BMS-986158 in combination with RUX or FED is being evaluated in the phase 1b/2 CA011-023 (NCT04817007) study in patients (pts) with first-line (1L) and second-line (2L) MF, respectively. Clinical observations, including spleen volume reduction (SVR), were previously presented (Ayala et al. EHA 2023. Presentation S213). In this analysis, exploratory biomarkers were assessed for potential early disease-modifying activities of BMS-986158. These included: measurement of longitudinal JAK2 VAF, which may be indicative of mutation clonal burden; evaluation of the BM microenvironment, including fibrosis, megakaryocyte clusters, and other morphological features; and circulatory cytokines and other factors associated with MF (Hasselbalch. Cytokine Growth Factor Rev. 2013).

Methods

In dose escalation, RUX-naïve pts (1L MF) received BMS-986158 2.0, 3.0, or 3.75 mg QD 5d on/2d off over 28-day cycles and RUX 15 mg BID; pts relapsed, refractory or intolerant to prior RUX treatment (2L MF) received BMS-986158 0.5, 0.75, or 1.25 mg QD 5d on/2d off over 28‑day cycles and FED 400 mg QD. Longitudinal assessment of JAK2 VAF was performed on multiple sample types, including isolated circulatory CD34+ cells, CD66b+ cells, and whole blood, using the MLL next-generation sequencing targeted exosome gene panel. Variants and indels were called using Pisces and Pindel, followed by annotation to COSMIC, ClinVar, gnomAd, and dbNSFP databases. BM assessments were performed on BM biopsies at baseline and post-treatment at the end of cycles 3 (optional), 6, and 12; fibrosis scores were determined by independent pathologist assessment (prognostic scoring system DIPSS PLUS-WHO 2017) and digital analysis of megakaryocyte clustering, reticulin pattern, and overall cellularity. The modulation of MF-associated circulatory cytokines and growth factors (79 soluble factors [SF] in total) was evaluated by multiple Rules-Based Medicine (RBM) custom MAP-HMP panels; aggregated data were analyzed.

Results

JAK2 VAF reduction was observed post-treatment with BMS-986158+RUX in 6/7 (86%) 1L MF pts with available data (Figure A). The maximum VAF reduction from baseline was 52%, where > 25% VAF reduction looks to be associated with improved spleen response (6/6 pts [100%] achieved SVR35). In 2L MF pts treated with BMS-986158+FED, JAK2 VAF reduction was observed in 1/4 (25%) pts. Similar degree of changes in JAK2 VAF were observed between CD34+ isolated cells and whole blood samples. In addition, BM fibrosis was either stabilized (3/9; 33%) or had ≥ 1 grade reduction (5/9; 56%) with BMS-986158+RUX at 24 weeks compared with baseline in 1L MF. Digital analysis further suggests potential normalized cellularity and decreased megakaryocyte clusters and reticulin levels in fibrosis-reduced samples. BM fibrosis reduction was also observed with BMS-986158+FED in 2L MF (3/8 [38%] stabilized and 2/8 [25%] reduced fibrosis). Lastly, sustained reduction of inflammatory cytokines downstream of NF-κB pathway was observed, including TNFα, IL-1ra, and VEGF (Figure B). In addition, increased levels of metabolic adipokine/hormone SF (eg, leptin, erythropoietin) and reduction of SF specific to BETi+JAKi combination (eg, DKK1, CD27, TIMP3; Mascarenhas et al. J Clin Oncol. 2023) were observed as expected.

Conclusions

Preliminary data suggest combination treatment with BMS-986158 and JAKi in MF may modulate JAK2 VAF, BM microenvironment, circulatory cytokines, and other SF, providing evidence of early disease modifying potential of these drug combinations.