Estimating the burden of undiagnosed CMML: Prevalence by phenotype and cytopenia stratification in NHANES

Abstract Background: Chronic myelomonocytic leukemia (CMML) is a rare clonal myeloid neoplasm defined by persistent monocytosis, cytopenias, and overlapping dysplastic and proliferative features. Given its often indolent course, especially in its early or lower-risk forms, CMML may be substantially underrecognized. Historical estimates suggest a prevalence of fewer than 1 in 100,000 U.S. adults but are limited by methodological challenges identifying CMML cases. To approximate the potential burden of unrecognized CMML, we conducted a cross-sectional analysis using nationally representative survey data. Methods: We analyzed data from the National Health and Nutrition Examination Survey (NHANES), a population-based survey administered by the CDC that includes clinical interviews, laboratory data, and physical assessments. NHANES cycles from 1999–2020 were included. Participants were classified as having a possible CMML phenotype if they had monocytosis (absolute monocyte count ≥1.0 × 10⁹/L and monocytes ≥10% of WBCs) along with at least one cytopenia: anemia (hemoglobin <13 g/dL in men or <12 g/dL in women), thrombocytopenia (platelets <150 × 10⁹/L), or neutropenia (ANC <1.8 × 10⁹/L). Those with self-reported malignancy or recent illness were excluded to reduce the likelihood of reactive findings. These parameters were selected to approximate CMML-like biology in the absence of diagnostic marrow or molecular data. Cytopenias were further categorized as mild or severe using predefined thresholds: mild anemia (hemoglobin <13 g/dL in males; <12 g/dL in females), severe anemia (hemoglobin <11.5 g/dL in males; <10.5 g/dL in females); mild thrombocytopenia (platelet count 100–150 × 10⁹/L), severe thrombocytopenia (<100 × 10⁹/L); mild neutropenia (ANC 1.0–1.8 × 10⁹/L), severe neutropenia (<1.0 × 10⁹/L). Red cell distribution width (RDW), mean corpuscular volume (MCV), and participant age were also extracted for exploratory analysis of potential biologic heterogeneity. Results: We estimated the prevalence of possible undiagnosed CMML at 181.3 per 100,000 U.S. adults. The identified cohort had a mean age of 59.3 years (weighted), 46.4% were male, and 37.0% were non-Hispanic White. Mild anemia was the most common cytopenia (50.2%), followed by mild thrombocytopenia (24.5%) and severe anemia (22.8%). Severe neutropenia and thrombocytopenia were uncommon (9.1% and 2.9%, respectively), and no individuals had mild neutropenia alone. To evaluate heterogeneity, we stratified individuals into two groups based on cytopenia severity: “CMML-mild,” defined by the presence of only mild cytopenias, and “CMML-severe,” defined by at least one severe cytopenia. The estimated prevalence of CMML-mild was 121.7 per 100,000 U.S. adults, while CMML-severe was less common, at 59.6 per 100,000. We compared RDW, MCV, and age between groups. RDW trended higher in the CMML-severe group (16.1% vs. 14.3%, p=0.087), while MCV (85.8 vs. 88.0 fL, p=0.48) and age (58.2 vs. 59.9 years, p=0.80) did not significantly differ between groups. Conclusion: This analysis identifies a sizable subset of U.S. adults with blood count patterns suggestive of CMML, many of whom have only modest abnormalities. Our results reinforce the notion that milder phenotypes may represent clinically relevant disease, reflected in hematologic features suggestive of erythropoietic stress (elevated RDW), potential marrow dysplasia (MCV), and clonal risk (age). Our findings suggest a potential gap in current diagnostic practices: individuals with persistent monocytosis and at least one accompanying cytopenia, which reflect key components of CMML, yet may not meet conventional thresholds that trigger hematologic evaluation. In aggregate, these results support a more vigilant approach to unexplained monocytosis, particularly when coupled with even a mild cytopenia. Key limitations include the use of a single time point per participant and lack of confirmatory testing (bone marrow, molecular, cytogenetic). Nonetheless, this work offers a novel, population-based perspective that builds upon and refines prior approaches, supporting further epidemiologic study into underrecognized or early CMML phenotypes.