Abstract Background: The Myelofibrosis and Essential Thrombocythemia Observational Study (MOST; NCT02953704) is a prospective observational study of pts with low-risk MF defined by DIPSS criteria. A previous analysis of data from MOST showed that 61% of pts with low-risk MF progressed to higher-risk disease over a median 54 months of follow-up (Yacoub A. Blood 2024). This analysis investigated the correlation between molecular gene signatures at enrollment and disease progression. Methods: MOST included pts with a physician-reported MF diagnosis (primary MF, post-polycythemia vera, or post-essential thrombocythemia) and low-risk status by DIPSS (except age 65 y). Optional biospecimens were collected at usual care visits. Of 232 pts with MF enrolled, 158 (68%) met these criteria by central review and had available biospecimens. MF progression was defined previously (Yacoub A. Blood 2024). RNA sequencing (RNAseq) was performed on samples from 140 pts with available RNA specimens (MF cohort; 85 pts with and 55 pts without progression), and 98 samples from age-matched healthy pts (control cohort, Discovery Life Sciences, AL, USA), using Illumina NovaSeq™ X Plus (100 paired-end PE base reads with 50 M PE reads per sample). RNAseq count data were analyzed by differential gene expression (DGE) analysis via limma; pathway enrichment analyses utilized gene set variation analysis (GSVA) and gene set enrichment analysis (GSEA). Results: The MF cohort (n=140) consisted of 81 pts with a driver mutation in JAK2, 35 in CALR, 8 in MPL; 4 pts had 2 driver mutations; 12 were triple-negative. DGE analysis of pts in MF vs control cohorts identified 560 downregulated and 1524 upregulated genes (fold change 1.5; adjusted P0.05). CREB3L1, previously correlated with increased risk of MF progression, was among the most upregulated genes in the MF cohort. Pathway enrichment analysis identified 10 differentially expressed pathways (adjusted P0.01), highlighting heme metabolism, cell cycle checkpoint (G2M), DNA replication (E2F targets), and epithelial mesenchymal transition as differentially expressed in the MF cohort. To identify MF progression biomarkers, the MF cohort was split by progression status (with vs without). No differences in driver mutation status were seen between groups, but JAK2V617F variant allele frequency (VAF) at enrollment was predictive of progression (with vs without progression, mean 0.48 vs 0.37; P=0.026). Comparing each subgroup with the control cohort showed more genes were differentially expressed in pts with (n=2836) than without (n=1347) MF progression. These gene expression and pathway enrichment (GSEA) patterns reflected those seen for total MF vs the control cohort; however, the magnitude of change trended higher in pts with progression. For example, fold change vs control cohort in expression of genes involved in tumor microenvironment remodeling (eg, MMP8 and FGFR3)were greater in pts with vs without progression. To investigate progression patterns and potential differences due to treatment effects, pts in the MF cohort were split by treatment history. Focusing on the subset of JAK inhibitor-treatment naive patients, a subject score per pathway of interest was derived using GSVA which identified statistically significant (P0.05) differences in cell cycle checkpoint (G2M) and DNA replication (E2F targets) between with and without MF progression subgroups.Conclusions: This analysis of molecular markers of MF progression identified JAK2V617F VAF as predictive of disease progression, with significantly higher VAF at enrollment seen in pts with vs without progression. DGE analysis identified multiple other biomarkers of interest, including CREB3L1, previously associated with progression in myeloproliferative neoplasms and other cancers, and involved in cellular response to ER stress and DNA damage. DGE analysis also identified several pathways differentially expressed in pts with MF vs controls, with trends consistent in pts with MF with and without progression. These differences vs controls were greater in pts with vs without MF progression, suggesting progression involves worsening of already dysregulated oncogenic pathways. Genes associated with these pathways may represent biomarkers to identify patients at high risk for MF progression. Additional analyses are ongoing to validate biomarkers of disease progression and identify targets for disease monitoring.
Ritchie et al. (Mon,) studied this question.