Unveiling relative lymphopenia and elevated monocyte‐to‐lymphocyte count as novel independent adverse prognostic factors in chronic myelomonocytic leukemia (CMML)—Proposal of the objective prognostic index for CMML (OPIC)

Abstract Chronic myelomonocytic leukemia (CMML) shows marked prognostic heterogeneity. Although leukocytosis is a recognized adverse prognostic factor, the contribution of its individual components remains insufficiently defined. In a cohort of 240 patients classified according to International Consensus Classification (ICC) and World Health Organization (WHO) 2022 criteria—including 23% with oligomonocytic CMML—we evaluated the prognostic impact of neutrophil and monocyte percentage, along with surrogate markers of their relative increase, including relative lymphopenia ( 1). Both relative lymphopenia and MLR > 1 emerged as independent adverse prognostic factors, correlating with adverse mutations ( TP53 , RAS pathway) and high‐risk clinical features. Notably, MLR > 1 identified a subset of patients with dysplastic CMML with molecular and clinical profiles resembling proliferative CMML (MP‐CMML). These variables retained their prognostic impact after adjustment for established prognostic models (CMML‐specific prognostic scoring system [CPSS], CPSS with the addition of the variable platelet < 100 × 10⁹/L [CPSS‐P], and Mayo prognostic model), and their addition improved predictive performance. Based on these and other objective variables readily available from a routine complete blood count (CBC), we developed the objective prognostic index for CMML (OPIC), which integrates hemoglobin < 11 g/dL, platelets < 100 × 10⁹/L, MP‐CMML, and MLR > 1. OPIC stratified patients into four risk categories with distinct survival outcomes (median overall survival [OS]: 104, 66.6, 34.3, and 18.3 months), demonstrating strong discriminatory power. Variable selection was performed using stepwise and elastic net regression, and random survival forests. Model performance metrics, including the C‐index, time‐dependent area under the receiver operating characteristic curve, and the Brier Score, were internally validated using bootstrapping‐based resampling methods and externally validated in a cohort of 250 patients. OPIC provides a robust, accessible tool for CMML risk stratification, supporting its integration into routine clinical workflows and early therapeutic decision‐making.