Molecular Classification System May Advance Personalized Care in CMML

A large international study (NCT04889729) published in the Journal of Clinical Oncology has introduced a comprehensive molecular framework that integrates genomic classification, individualized prognostic scoring, treatment response prediction, and transplant timing optimization into the clinical decision-making process for patients with chronic myelomonocytic leukemia (CMML).^1,2 The findings represent a substantive advance in applying molecular data to clinical decision-making in a disease that has historically offered limited precision medicine tools.

CMML is a rare myeloid neoplasm categorized among myelodysplastic syndrome/myeloproliferative overlap neoplasms, predominantly affecting individuals older than 70 years. The disease is characterized by peripheral blood monocytosis, clinical heterogeneity, and an inherent risk of transformation to acute myeloid leukemia (AML). Allogeneic hematopoietic stem cell transplantation (HSCT) remains the only potentially curative treatment; however, most patients are ineligible due to age or comorbidities.

The observational study enrolled a large retrospective cohort of over 3000 patients with CMML to serve as a training set and a prospective cohort of over 500 patients to validate the framework. Using integrated molecular and clinical variables, investigators developed and validated a multilayered decision-support system encompassing various components: a molecular disease classification, an individualized prognostic scoring system (iCPSS), and a transplant timing decision tool.

“Overall, these findings emphasize the value of molecularly informed strategies in therapeutic decision making for CMML, define the priority of genomic alterations to be included in routine screening as clinically informative, and offer a practical framework for personalized transplantation policies aimed at maximizing both survival and quality of life,” Luca Lanino, MD, and colleagues wrote in the publication.¹

Molecular Classification Defines 9 Genomic Classes

The investigators first identified 9 genomically defined disease classes with distinct clinical features and outcomes, collectively capturing 69% of the retrospective cohort (2076 of approximately 3000 patients). Two classes characterized by isolated splicing factor mutations, SRSF2 (n = 392; 13%) and ZRSR2 (n = 83; 3%), frequently co-mutated with TET2 (82%) and were associated with predominantly myelodysplastic features, low marrow blast counts (<5%), and favorable outcomes (median overall survival [OS], 5.4 and 8.2 years, respectively).

By contrast, 2 additional classes combined SRSF2 (n = 556; 18%) or ZRSR2 (n = 74; 2%) with adverse-risk mutations, including ASXL1, EZH2, and RUNX1, and were associated with proliferative disease features, higher blast counts (≥10%), and markedly inferior survival (median OS, 3.3 and 2.7 years, respectively; P <.001). The remaining classes reflected further genomic complexity with varying clinical correlates.

This classification expands upon prior risk stratification systems, including the Chronic Myelomonocytic Leukemia-Specific Prognostic Scoring System incorporating molecular data (CPSS-Mol) and the Mayo Molecular Model, by offering more granular entity definitions that align genomic features with clinical phenotype and prognosis.

Development and Clinical Utility of the iCPSS

A key translational component of the framework is the iCPSS, which integrates individual-level genomic and clinical variables to generate patient-specific survival estimates. In an accompanying editorial note, JCO associate editor Charles Craddock, MD, described the iCPSS as "an important advance in disease prognostication in CMML," noting its potential to more accurately identify patients who may benefit from HSCT and to inform transplant timing decisions in an era of expanding transplant options.

An evaluation of the iCPSS in the retrospective cohort found that the score was able to significantly distinguish post-HSCT OS and relapse risk (both P <.001) into low- and high-risk groups, and was independently associated with OS and relapse-free survival (HR, 1.34 and 1.54, respectively; P <.001).

The iCPSS was developed to address the difficult clinical question of optimal HSCT timing in CMML, a determination complicated by the advanced age and comorbid burden of most patients, as well as the absence of prospective randomized data. The decision-support tool models individualized life expectancy under different transplant timing strategies, enabling risk-adapted recommendations that incorporate molecular risk, disease dynamics, and patient fitness. Additionally, the ability of the iCPSS to be computed on a web calculator makes it a suitable tool for integration into clinical workflows.

The study arrives at a juncture when the CMML treatment landscape is beginning to diversify, with several investigational approaches under evaluation. The molecular framework is positioned to complement these emerging therapeutic strategies by enabling more precise patient selection, informing treatment sequencing, and facilitating enriched clinical trial design.

The integration of artificial intelligence-based frameworks such as MOSAIC, a multimodal classification tool for rare hematologic cancers developed by the same research group, further reflects the potential of computational tools to operationalize complex genomic data at the point of clinical decision-making.

Validation in prospective cohorts and translation into accessible clinical workflows will be necessary before widespread implementation, according to the authors. Nonetheless, the framework described in this study represents a meaningful step toward molecularly informed, individualized management of CMML.

REFERENCES

1. Lanino L, Hunter AM, Gagelmann N, et al. Molecular-based ecosystem to improve personalized medicine in chronic myelomonocytic leukemia. J Clin Oncol. Published online March 27, 2026. doi:10.1200/jco-25-02116

2. GENOMED4ALL: Improving MDS classification and prognosis by AI. ClinicalTrials.gov. Updated September 9, 2022. Accessed April 6, 2026. https://clinicaltrials.gov/study/NCT04889729