Mutational architecture defines disease behaviour in an Indian cohort of Myeloproliferative Neoplasms: Foundations for a Genomics-first classification.

AIMS: Data on BCR::ABL1-negative MPN genomics in India are limited. We aimed to define mutational architecture and evaluate co-mutation burden for risk stratification.

METHODS

We analysed 179 patients (median age 57 years; 36.3% ≤50 years). Targeted NGS covering relevant myeloid genes was performed in 104 patients (58%). Prognostic factors and latent molecular subgroups were evaluated using Cox proportional hazards and DBSCAN clustering.

RESULTS

Single-hit genotypes predominated (65.9%), but MF/pre-PMF showed ≥ 2 mutations in 42% and ≥ 3 in 18%. Double-hit status (≥1 driver + ≥1 DTA/MDS-HR) occurred in 26% and conferred a 4.87-fold mortality risk (p = 0.0132). TET2 independently predicted inferior OS (HR 11.83; p < 0.005). DBSCAN identified a poor risk JAK2 +CoMut subgroup (n = 29; 18.3% of NGS cohort) enriched for TET2/ASXL1 with median OS 3.3 years versus NR for the remaining cohort. Pre-PMF and MF cases had higher mutation burden and co-mutations (≥1 DTA/MDS-HR: 62% vs 15% in PV/ET). Survival differences were independent of WHO diagnostic labels, with molecular architecture alone consistently predicting outcomes across all clusters

CONCLUSIONS

In this Indian MPN cohort, molecular architecture, not WHO label, dictated prognosis. The reproducible impact of high-risk co-mutations demonstrates both the feasibility and need for a genomics-first, tiered classification to guide risk stratification and therapy.