Establishment and validation of a novel risk model for estimating time to first treatment in 120 patients with chronic myelomonocytic leukaemia

Abstract

Chronic myelomonocytic leukaemia is a rare disease and data on the treatment are often extrapolated from myelodysplastic syndrome studies. Although several scores exist for the prognosis of overall survival in chronic myelomonocytic leukaemia, so far there is no designated score for the prediction of the time to first treatment. We tested clinical parameters and cytogenetic information for their ability to predict the time to first treatment in our single center cohort of 55 unselected consecutive chronic myelomonocytic leukaemia patients. In multivariate analysis we identified elevated lactate dehydrogenase (≥223 U/l), higher bone marrow blast percentage (≥7.5%) and thrombocytopenia (<55 G/l) at initial diagnosis as the most relevant parameters for the time to first treatment. Using these three parameters we developed a risk score that efficiently estimates the time to treatment initiation with azacitidine or hydroxyurea (p < 0.001; log-rank). In the high-risk group (≥2 risk factors) 85% of patients required treatment within 1 year, whereas this was the case in 48% in the intermediate-risk (1 risk factor) and in 0% in the low-risk group (0 risk factors). Our risk model was validated in an external test cohort of 65 patients and may serve as a simplified and easily applicable tool for identifying patients who may not require early treatment initiation.