Platelet transcriptome yields progressive markers in chronic myeloproliferative neoplasms and identifies putative targets of therapy

Predicting disease progression remains a particularly challenging endeavor in chronic degenerative disorders and cancer, thus limiting early detection, risk stratification, and preventive interventions. Here, profiling the spectrum of chronic myeloproliferative neoplasms (MPNs) as a model, we identify the blood platelet transcriptome as a proxy for highly sensitive progression biomarkers that also enables prediction of advanced disease via machine learning algorithms. Using RNA sequencing (RNA-seq), we derive disease-relevant gene expression in purified platelets from 120 peripheral blood samples constituting two time-separated cohorts of patients diagnosed with one of three MPN subtypes at sample acquisition - essential thrombocythemia, ET (n=24), polycythemia vera, PV (n=33), and primary or post ET/PV secondary myelofibrosis, MF (n=42), and healthy donors (n=21). The MPN platelet transcriptome reveals an incremental molecular reprogramming that is independent of patient driver mutation status or therapy and discriminates each clinical phenotype. Leveraging this dataset that shows a characteristic progressive expression gradient across MPN, we develop a machine learning model (Lasso-penalized regression) and predict advanced subtype MF at high accuracy and under two conditions of external validation: i) temporal: our two Stanford cohorts, AUC-ROC of 0.96; and ii) geographical: independently published data of an additional n=25 MF and n=46 healthy donors, AUC-ROC of 0.97). Lasso-derived signatures offer a robust core set of View larger version (42K): org.highwire.dtl.DTLVardef@72dc16org.highwire.dtl.DTLVardef@cf5096org.highwire.dtl.DTLVardef@b3d460org.highwire.dtl.DTLVardef@3c0ddc_HPS_FORMAT_FIGEXP M_FIG C_FIG