Development and Analytical Validation of a 6-Plex Reverse Transcription Droplet Digital PCR Assay for the Absolute Quantification of Prostate Cancer Biomarkers in Circulating Tumor Cells of Patients with Metastatic Castration-Resistant Prostate Cancer

Background Gene expression in circulating tumor cells (CTCs) can be used as a predictive liquid biopsy test in metastatic castration-resistant prostate cancer (mCRPC). We developed a novel 6-plex reverse transcription droplet digital PCR (RT-ddPCR) assay for the absolute quantification of 4 prostate cancer biomarkers, a reference gene, and a synthetic DNA external control (DNA-EC) in CTCs isolated from mCRPC patients. Methods A novel 6-plex RT-ddPCR assay was developed for the simultaneous absolute quantification of AR-FL, AR-V7, PSA, and PSMA, HPRT (used as a reference gene), and a synthetic DNA-EC that was included for quality control. The assay was optimized and analytically validated using DNA synthetic standards for each transcript as positive controls. Epithelial cellular adhesion molecule (EpCAM)-positive CTC fractions isolated from 90 mCRPC patients and 11 healthy male donors were analyzed, and results were directly compared with reverse transcription quantitative PCR (RT-qPCR) for all markers in all samples. Results Linear dynamic range, limit of detection, limit of quantification, intra- and interassay precision, and analytical specificity were determined for each marker. Application of the assay in EpCAM-positive CTC showed positivity for AR-FL (71/90; 78.9%), AR-V7 (28/90; 31.1%), PSA (41/90; 45.6%), PSMA (38/90; 42.2%), and HPRT (90/90; 100%); DNA-EC concentration was constant across all samples. Direct comparison with RT-qPCR for the same markers in the same samples revealed RT-ddPCR to have superior diagnostic sensitivity. Conclusions Our 6-plex RT-ddPCR assay was highly sensitive, specific, and reproducible, and enabled simultaneous and absolute quantification of 5 gene transcripts in minute amounts of CTC-derived cDNA. Application of this assay in clinical samples gave diagnostic sensitivity and specificity comparable to, or better than, RT-qPCR.