Abstract Background In polycythemia vera (PV), reduction of the JAK2 V617F allele burden is increasingly recognized as a marker of disease modification. Molecular response (MR) is commonly assessed using the 2009 European LeukemiaNet (ELN) criteria, which define response based on relative reduction from baseline. However, the clinical relevance of these thresholds remains uncertain. Recent studies have suggested that absolute allele burden—particularly 2%, defined as deep molecular response (DMR)—may better reflect therapeutic effect. In this study, we report long-term outcomes from a prospective cohort of PV patients treated with ropeginterferon alfa-2b. Molecular responses at week 132 were evaluated using both ELN-defined criteria and absolute allele burden thresholds. RNA sequencing was performed to explore transcriptomic features associated with molecular response depth. Methods We analyzed data from a phase 2, multicenter, prospective clinical trial in Korea evaluating the efficacy of ropeginterferon alfa-2b in patients with PV. The trial began in October 2021, with a 1-year core treatment phase completed in November 2023. Ropeginterferon was administered subcutaneously every two weeks, starting at 250 μg, escalating to 350 μg at week 3 and 500 μg at week 5, followed by maintenance dosing. After the core phase, patients entered long-term follow-up with visits every 12 weeks. At the 132-week visit (Visit 13), peripheral blood samples were collected for molecular and transcriptomic analyses. JAK2 V617F allele burden was assessed by quantitative PCR. For RNA sequencing, strand-specific libraries were prepared from total RNA and sequenced using the MGI platform. After quality control and adapter trimming, sequencing reads were aligned to the human reference genome using HISAT2. Gene expression levels were quantified with StringTie, and differentially expressed genes (DEGs) were identified. Functional enrichment analyses of DEGs were performed using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway databases. Results A total of 75 patients completed the week 132 visit and were included in the analysis. Based on ELN 2009 criteria, 64 patients (88.9%) achieved MR and 9 were non-responders. Two patients with a baseline JAK2 V617F allele burden 10% were excluded from MR analysis. At week 132, 31 patients (41.3%) achieved DMR (allele burden 2%) and 45 patients (60.0%) had a burden 10%. Transcriptomic analysis comparing DMR (n=27) and non-DMR (n=43) groups identified 650 DEGs. Hierarchical clustering revealed distinct gene expression profiles. GO analysis showed significant upregulation of immune-related processes in the DMR group, including response to stimulus, signal transduction, cell migration, and cytokine receptor activity. Enriched cellular components included secretory vesicles, granules, and the plasma membrane, reflecting immune effector function. Molecular function analysis highlighted pattern recognition receptor activity and signaling receptor binding. KEGG pathway analysis revealed enrichment of cytokine–cytokine receptor interaction, PI3K-Akt signaling, complement and coagulation cascades, and hematopoietic cell lineage, indicating immune and hematopoietic remodeling in DMR patients. These findings suggest that deep molecular responders exhibit a distinct transcriptional signature involving immune activation and receptor-mediated signaling. Further details of the RNA sequencing analysis will be presented at the ASH Annual Meeting 2025. Conclusions In patients with PV receiving long-term ropeginterferon alfa-2b, molecular response defined by ELN 2009 criteria may not fully reflect the depth of treatment effect. There is a growing need to refine response definitions to better capture biologic remission and disease modification. DMR, defined by an absolute JAK2 V617F allele burden 2%, may serve as a more robust surrogate marker. Transcriptomic profiling supports DMR as a biologically distinct state, highlighting the potential role of immune-related molecular signatures in future response assessments.
Yoon et al. (Mon,) studied this question.