Plasma cell-free RNA profiling distinguishes cancer from pre-malignant conditions

Cell-free RNA (cfRNA) in plasma reflects gene expression profiles of both localized sites of cancer and at the systemic host response. Here we explore the diagnostic potential of cell-free transcriptomes by mRNA sequencing. We sequenced total cell-free plasma RNA from 90 plasma samples from two independent sample cohorts representing two cancer types, two pre-cancerous conditions and non-cancer donors. We identified distinct gene sets and built classification models that could distinguish cancer patients with specific cancer types from premalignant conditions and non-cancer individuals with high accuracy. Determination of multiple myeloma from its pre-cancerous monoclonal gammopathy of undetermined significance (MGUS) yielded an accuracy of 90% (17/19). Detection of primary liver cancer from its premalignant condition cirrhosis yielded an accuracy of 100% (12/12). This work lays the foundation for developing low cost assays by measuring mRNA transcript levels in plasma using a small panel of genes for detection that can distinguish the presence of cancer from pre-cancerous conditions and non-cancer individuals. We sequenced total cell-free plasma RNA from 90 plasma samples from two independent sample cohorts representing two cancer types, two pre-cancerous conditions and healthy donors: 10 multiple myeloma and 8 liver cancer; 13 pre-malignant conditions including 9 monoclonal gammopathy of undetermined significance and 4 Cirrhosis; and 20 aged and gender matched healthy donor in our pilot cohort.