Molecular International Prognostic Scoring System (IPSS-M) in Myelodysplastic Syndromes Arising from Aplastic Anemia and Paroxysmal Nocturnal Hemoglobinuria

Clonal evolution to myelodysplastic syndromes (MDS) or acute myeloid leukemia represents the most serious late complication of patients with aplastic anemia (AA) and paroxysmal nocturnal hemoglobinuria (PNH). The majority of secondary MDS (sMDS) cases present high-risk features with chromosome 7 abnormalities and mutations in ASXL1, RUNX1, SETBP1, and RAS pathway genes as the most frequent molecular lesions. Other cases tend to show more favorable outcomes with distinct genomic architecture mainly driven by generally considered low-risk cytogenetic alterations (e.g., del(13q), del(20q)). Such heterogeneity of genomic makeup, the rarity of the disease, and the scarcity of information make the prognostication of sMDS challenging for routine clinical practice. In recognition of the growing role of somatic mutations, the Molecular International Prognostic Scoring System (IPSS-M) has been conceived to improve the precision of prognostication of prior schemes by adding objective molecular markers. Whereas validated on de novo and secondary/therapy-related cases, its applicability specifically on post-AA/PNH MDS has not yet been explored. Here, we applied IPSS-M to 69 sMDS with complete clinico-genomic information from a large multicenter dataset of 1008 AA/PNH patients, using a comparator cohort of primary MDS (n=1281).

The median age at MDS progression was 63 years (IQR, 34-70), and most patients were male (M:F ratio 1.65). Despite low bone marrow blast percentages (median, 2; 0-7), 71% of patients showed high-risk features according to IPSS-R (scores >3.5), chiefly due to the enrichment of poor risk cytogenetics, most frequently monosomy 7. Overall, 48% of patients received disease-modifying treatments (DMT) with either HMA (28%), HSCT (60%) or both (12%). Leukemia progression was registered in 14% of cases after a median of 5.4 months (1-14) from sMDS onset. With a median follow-up of 55.6 months (95%CI, 42.5-75.8), the 60-month overall (OS) and leukemia free-survival (LFS) achieved 44% (32-60) and 42% (31-58), respectively.

The IPSS-R distribution was as follows: Very Low (6%), Low (23%), Intermediate (30%), High (23%) and Very High (18%). By applying IPSS-M, patients were redistributed in Very Low (5%), Low (19%), Moderate Low (14%), Moderate High (14%), High (29%), and Very High (19%) risk categories. When comparing five-to-five (merging together the Moderate Low/High IPSS-M categories), 55% of cases were re-stratified, of which 63% up- and 37% down-staged (Fig.1A). Indeed, 77% of sMDS patients harbored at least 1 mutation with 43% presenting >2, thereby substantiating the differences observed between the two prognostication systems. Although according to IPSS-M the majority (62%) of sMDS cases still clustered in higher-risk categories, a non-negligible fraction (38%) was assigned to lower-risk groups. To further explore whether IPSS-M could be used to predict OS/LFS also in sMDS cases, we analyzed survival outcomes by taking advantage of a cohort of de novo cases. However, approximately half of sMDS received DMT, potentially generating a bias when using survival as a surrogate for outcomes comparisons, even in cases falling into equivalent risk categories. Therefore, we applied a propensity score matching by using age, gender and DMT as relevant covariates to reduce any possible confounder. As reported for others MDS subtypes in the original IPSS-M study, OS/LFS outcomes of sMDS paralleled those observed in the de novo group within the various IPSS-M risk categories (Fig.1B).

We applied for the first time IPSS-M in a real-world cohort of sMDS, showing its potential for refining the assignment of patients into risk categories prognostic of disease outcomes. The actuarial survival of sMDS cases was not distinct from de novo patients matched for basic clinical features in equivalent IPSS-M risk groups. The complexity of the genomic landscape typical of AA/PNH clonal evolution makes sMDS a prototypic model whereby the incorporation of molecular information is of paramount importance to ensure adequate clinical management.