Abstract
We evaluated p16INK4a as a reliable option to detect human papilloma virus (HPV) DNA in penile tumor specimens. Formalin-fixed paraffin embedded samples of 26 patients with penile cancer and another 18 cases with non-tumorigenic lesions were stained by three different widely used commercially available chromogenic in-situ hybridization assays high-risk HPV CISH Y1443 (Genpoint, DAKO), pan HPV CISH Y1404 (Genpoint, DAKO), INFORM HPV III (Ventana, Tucson, Arizona) and p16INK4a immunohistochemistry, then compared to the known gold standard polymerase chain reaction detecting HPV 16, 18, 31, and 33. Immunoreactivity for p16INK4a was evaluated by using a 4-tiered (0, 1, 2, and 3) pattern based system. 19 cases were positive for p16INK4a, 13 of which showed a continuous transepithelial staining (pattern 3). Pan HPV ISH showed positivity in 9 cases, high-risk HPV ISH in 7 cases and INFORM HPVIII ISH in 7 cases. p16INK4a IHC pattern 3 versus pattern 0, 1 and 2 exhibited a specificity and positive predictive value of 100%, with a sensitivity and negative predictive value of 72% and 62%, respectively, which was much better than all HPV in-situ hybridization methods referred to polymerase chain reaction. p16INK4a seems to be a superior marker for the detection of HPV-associated penile squamous cell carcinoma compared to CISH tests, but is not recommend for the detection of non-tumorigenic lesions, where PCR should be used for the initial assessment.