ABSTRACT
BACKGROUND: Colorectal cancer (CRC) is one of the most common cancers. As in other cancer locations, the involvement of human papillomaviruses (HPV) has been suggested but remains highly debated with wide differences among reported prevalence of HPV infection in CRCs. AIM: To determine the actual prevalence of high risk HPV16 and 18 in a large case-control study. METHODS: CRC specimens were used for analysis of both tumor and distant healthy tissue. As a non-malignant control group, samples from sigmoid diverticulosis resections were studied. Detection of HPV16 and HPV18 DNA was performed using a real time polymerase chain reaction (qPCR). Ten percent of tumor samples were also randomly subjected to a complete HPV genotyping using the INNO-LiPA technique. RESULTS: 467 samples were analyzed: 217 tumor samples from 210 CRCs, 210 distant healthy tissue samples, and 40 sigmoid samples. HPV18 DNA was never amplified and HPV16 was amplified only three times in tumor tissues with viral loads under or at the limit of quantification. New extraction from the same tumor blocks for these samples revealed no HPV with qPCR and INNO-Lipa assays. CONCLUSION: With adequate procedures and reliable techniques, no HPV was detected in the largest case-control study so far, bringing more evidence on the absence of involvement of HPV in CRCs.
