Objectives: Prior investigations evaluating the presence of human papillomavirus (HPV) in prostatic tissue by polymerase chain reaction (PCR) technology have yielded detection rates of 0% to 100%. Contamination by viral DNA from prostatic urethral colonization or less than optimal laboratory conditions have been suggested to explain this discrepancy. In addition, the various investigations have differed in the specific oligonucleotide primers utilized for amplification and, therefore, have searched for different segments of the viral genome. The objective of this study is to address these differences.

Methods: Forty-one archival radical prostatectomy specimens were evaluated, identifying areas of normal and abnormal histology. Meticulous technique was used during tissue acquisition, histologic confirmation, DNA isolation, PCR amplification, polyacrylamide gel electrophoresis, and staining. Primers for a 126- and 99-base pair (bp) fragment of the E6 portion of HPV 16 as well as a consensus primer for the L1 portion of the papillomavirus genome were utilized.

Results: Of the normal prostatic tissues, 13.5% (5/37) contained the 126-bp E6 viral DNA as did 33.3% (7/21) of benign prostatic hyperplasia (BPH) samples, 25% (5/20) of dysplasia, 18.2% (2/11) of Gleason grades 1 and 2 adenocarcinoma, 25.9% (7/27) of Gleason grade 3 adenocarcinoma, and 6.7% (1/15) of Gleason grade 4 adenocarcinoma. Sections from the urethras of the prostatectomy specimens contained viral DNA in 31.7% (13/41). Viral detection was variable among different specimens in the same patient. With amplification for the 99-bp fragment of HPV 16, 1 of 37 normal (2.7%), 2 of 21 BPH (9.5%), 1 of 20 dysplasia (5.0%), and 2 of 53 cancer (3.8%) specimens revealed HPV DNA. In none of the specimens was DNA amplified using primers for a 450-bp fragment of the L1 portion of HPV.

Conclusions: Previously published discrepancies in HPV detection may be solely due to the differences in primer sets utilized.