Introduction:

Prostate cancer (PC) is the most common cancer in men in the U.S., the current standard clinical test for PC is the quantification of Prostate Specific Antigen (PSA) level in blood which has a sensitivity of 21% with specificity of 91%. This has led to a decline in PC screening, new cases and deaths related to PC, thus earlier and more accurate screening is warranted. Seminal fluid is a novel sample type that comes into direct contact with prostate epithelium; collecting cellular debris improving biomarker source. A diverse panel of DNA methylation markers (EVX1, CAV1, FGF1), Protein (AMACR) and extracellular MicroParticle (MP) surface proteins (PSMA, STEAP1, GHSR) is quantified from this sample source improving prostate cancer detection over blood PSA.

Methods:

Seminal fluid samples were collected in a multi-center biorepository clinical trial of men aged 50 or older. Fourteen (14) sites participated across the United States.  A single sample was collected from each study participant and were assigned into cohorts defined as: Cohort 1– PSA value (< 2.0 ng/ml); Cohort 2– Negative prostate biopsy results for PC; Cohort 3– Biopsy diagnosed low risk prostate cancer (Gleason Score 6 to 7 (3+4)); and Cohort 4– Biopsy diagnosed high risk prostate cancer (Gleason Score 7 (4+3) to 10). The semen samples were processed, and AMACR protein level was quantified using ELISA assay. Microparticle quantity and surface marker (PSMA, STEAP1, and GHSR) profiles were characterized using flow cytometry. DNA methylation profile was characterized by methyl specific PCR assays (EVX1, CAV1, FGF1). Data normalization, univariate and multivariate LDA analysis were conducted to determine the sensitivity and specificity of PC diagnosis.

Results:

Results indicate that the complete biomarker panel analysis had higher sensitivity and specificity compared to PSA alone or any individual biomarker alone. Many samples had technical processing issues, such as low volume requiring dilution, hyper viscosity or difficulties forming a tight pellet and defined supernatant phase for biomarker extraction. Furthermore, the study design defined Cohort 1 as individuals with a low PSA < 2.0 ng/ml inferring this group as control healthy individuals, however PSA may not be elevated in all cases of prostate cancer, and this cohort was not confirmed with biopsy. Therefore, a subgroup of subjects without any processing issues and with biopsy confirmed true positives and true negatives were analyzed. Here, when the PSA cutoff was set at 9.6ng/ml and specificity at 88%, sensitivity increased to 79% when analyzing all markers compared to 18% for PSA alone, a significant increase in PC diagnosis accuracy.

Conclusion:

This preliminary study demonstrates that a panel of biomarkers including DNA methylation, protein, MP, and PSA greatly improved prostate cancer detection sensitivity over PSA alone or any single marker alone. However, the improved sensitivity was manifested in samples without any processing issues and biopsy confirmed samples. Thus, future improvements in both the sample processing procedures in parallel with further optimization of biomarker detection assays are in progress. These improvements could further increase both sensitivity and specificity using seminal fluid as a biomarker for prostate cancer detection.

Funding: Biotech startup