Decoding the RNA cargo of plasma-derived extracellular vesicles in metastatic castration-resistant prostate cancer

Liquid biopsies offer a minimally invasive and readily accessible complement to tissue biopsies for monitoring cancer by capturing multiple molecular components circulating in body fluids, such as circulating tumor cells (CTCs), cell-free DNA (cfDNA), cell-free RNA (cfRNA) and extracellular vesicles (EVs). EVs are nanoscale structures surrounded by a lipid bilayer. They carry a rich molecular cargo that includes nucleic acids, proteins, and lipids. The EV RNA cargo encompasses various types of RNAs, both coding and non-coding RNAs and the presence of tissue-specific transcripts offers insights into EV cellular origin, potentially informing about the molecular features of the disease and the ongoing cellular processes. The current study performs a comprehensive analysis of plasma-derived EVs from a prospectively collected observational trial cohort of metastatic castration-resistant prostate cancer (mCRPC) patients, treated with Enzalutamide in a chemo-naïve setting. Specifically, it incorporates EV physical characterization, EV-RNA and matched cfDNA genomics. The use of total RNA sequencing (RNA-seq) enabled the detection of both coding and non-coding RNAs, regardless their integrity status, providing a complete view of the RNA landscape of plasma EVs, significantly adding to the recent poly-A-enriched sequencing based study in the same setting. In addition, by capturing a broad spectrum of circulating tumor DNA (ctDNA) levels, as opposed to high ctDNA patients only, it offers a more inclusive perspective on prognostic indicators. Integrating EV transcriptomics, ctDNA estimation, physical measurements of EVs, nucleic acid quantifications, blood counts and standard of care markers, this study underscores the value of a holistic profiling strategy. We show that for patients with undetectable ctDNA, EV-RNA provides additional insights. Short responders exhibit larger EVs, suggesting a link between EV characteristics and tumor biology. A prostatic tissue-associated signal is detectable in patients compared to healthy donors (HDs), underscoring the capability of EV RNA to capture tumor-derived information even in cases with undetectable ctDNA. Deconvolution analysis of RNA levelsrevealed significant contributions from both the immune and tumor-related components. Specifically, it identified distinct immune cell populations, including monocytes, macrophages M2, and T cells CD4+ memory activated, each exhibiting significant associations with patient prognosis. Notably, monocytes relative contribution detected at baseline (BL) emerged as independent prognostic factors with a protective effect. This association was further observed in an mCRPC tissue cohort, highlighting the importance of the tumor microenvironment, including immune cell profiles, in shaping disease outcomes. Altogether these findings emphasize the potential of EV RNA in enhancing prognostic assessments in mCRPC, with EV RNA deconvolution analysis revealing a tumor-related and an immune component, as well as specific gene signatures, associated with patient prognosis.