Prostate-derived circulating microRNAs add prognostic value to prostate cancer risk calculators

Abstract

Prostate cancer is the second leading cause of malignancy-related deaths among American men. Active surveillance is a safe option for many men with less aggressive disease, yet definitively determining low-risk cancer is challenging with biopsy alone. Herein, we sought to identify prostate-derived microRNAs in patient sera and serum extracellular vesicles, and determine if those microRNAs improve upon the current clinical risk calculators for prostate cancer prognosis before and after biopsy. Prostate-derived intracellular and extracellular vesicle-contained microRNAs were identified by small RNA sequencing of prostate cancer patient explants and primary cells. Abundant microRNAs were included in a custom microRNA PCR panel that was queried in whole serum and serum extracellular vesicles from a diverse cohort of men diagnosed with prostate cancer. The levels of these circulating microRNAs significantly differed between indolent and aggressive disease and improved the area under the curve for pretreatment nomograms of prostate cancer disease risk. The microRNAs within the extracellular vesicles were the most informative and improved the AUC to 0.739 compared to the existing nomogram alone, which has an AUC of 0.561. The microRNAs in the whole serum improved it to AUC 0.675. In summary, quantifying microRNAs circulating in extracellular vesicles is a clinically feasible assay that may provide additional information for assessing prostate cancer risk stratification.

FIGURE 1

Characterization of prostate‐derived extracellular vesicles and microRNA profiles of patient‐derived prostate samples. (a) Flow chart of PCa patient‐derived samples for small RNA NGS profiling (4 tissue slice explants (TSC), 4 epithelial cell cultures (PrE), 4 stromal cell cultures (PrS) and 4 EV samples from each sample type (TSC EV, PrE EV and PrS EV). (b) Nanoparticle particle tracking analysis of two PrE EV samples, two PrS EV samples and one TSC EV sample. (c) CD63 ELISA of PrE‐, PrS‐ and TSC‐derived EVs (N = 3 PrE, 3 PrS, 2 TSC; Error bars show SEM). (d) Transmission electron microscopy (TEM) images of EVs derived from PrE, PrS and TSC (size bars = 100 nm). (e and f) PCA plots and unbiased hierarchical clustering of the miR profiles of (e) Patient‐derived prostate epithelial cells, stromal cells and tissue slice explants and (f) EVs isolated from patient‐derived prostate epithelial cells, stromal cells and tissue slice explants (N = 4 per sample type). Heatmaps show differentially expressed miRs with an adjusted p‐value ≤0.01 using the Benjamini–Hochberg method. (g) Venn diagram representation of TMM‐normalized miRs with >500 counts per million (cpm) within each sample type. (h) RT‐qPCR validation of cell/EV‐specific miRs as determined by NGS miR profiles (N = 2 replicates per sample, Error bars = SD, ND = not detected).

FIGURE 2

Characterization of patient‐derived pretreatment serum extracellular vesicles and their microRNA profiles. (a) Nanoparticle tracking analysis of N = 2 patient‐derived serum EV samples. (b) Transmission electron microscopy image of patient‐derived serum EVs (size bar = 100 nm). (c) Western blot of EV‐specific markers (CD63 and CD81) and intracellular protein control (GRP78) of two patient‐derived serum EV samples and one patient‐derived stromal cell control. (d) CD63 ELISA of patient‐derived serum EV samples stored at 4, −20, and −80°C for 1 and 7 days (N = 3 samples per day, error bars = SD). (e) PCA plot of the miR profiles of patient‐derived pretreatment serum EVs (N = 20 patients). Data are grouped by pathological Gleason Grade group (Pathological GG). (f) Venn diagram representation of TMM‐normalized miRs with >500 counts per million (cpm) within EVs isolated from patient‐derived serum, epithelial cells, stromal cells and tissue slice explants. (g) RT‐qPCR validation of serum EV miRs as determined by NGS miR profiles (N = 2 replicates per sample, Error bars = SD, ND = not detected).

FIGURE 3

Patient cohort and study design. (a) Study design flow chart for determining serum and serum EV miRs that are predictive of prostate cancer outcomes. (b) MiRs in the panel and how they were selected, either from our prior study, the new NGS data or a literature search. For the miR counts chart, green indicates >5000 counts, yellow >100 counts, and red <100 counts.

FIGURE 4

Circulating serum and serum EV microRNAs significantly improve AUCs for prostate cancer risk. (a) Violin plot of serum miRs and (b) serum EV miRs that are predictive of adverse pathology (AP). The lines on the violin plots represent the mean and first and third quartiles. Significant miRs had a p‐value <0.05. (c) Significant serum and serum EV miRs were included with the CAPRA score in random forest models to predict AP. The area under the curve (AUC) is reported for CAPRA alone (black), CAPRA + serum miRs (red) and CAPRA + serum EV miRs (blue). (d) Violin plot of serum miRs and (e) serum EV miRs that are predictive of low‐grade and high‐grade PCa. Patients stratified in biopsy Gleason grade group 1 were considered low‐grade; patients with biopsy Gleason grade group 2 or higher were considered high‐grade. The lines on the violin plots represent the mean and first and third quartiles. Significant miRs had a p‐value <0.05. (f) Significant serum and serum EV miRs were included with the PBCG risk in random forest models to predict low‐grade and high‐grade PCa. The area under the curve (AUC) is reported for PBCG alone (black), PBCG + serum miRs (red) and PBCG + serum EV miRs (blue).