Abstract
Using multiple PCR primer sets, we tried to optimize the detection of human papillomavirus (HPV) in DNA samples isolated from 361 frozen biopsy specimens from patients with invasive cervical carcinomas. The HPVs detected were placed into three distinct groups, including group I/Inex at Telelab (Skien, Norway) and group Ineg and group II at the Norwegian Radium Hospital (Oslo, Norway). The consensus primer sets were Oli-1b-oli-2i, My09-My11, Gp5-Gp6, and Gp(5+)-Gp6+ from the HPV L1 gene and CpI-CpIIG from the E1 gene. Using these consensus primers together with the type-specific primers from E6-E7, we found that 355 patients (98%) were HPV positive. Type-specific primers for HPV types 11, 16, 18, 31, 33, and 35 detected more HPV-infected patients than the most sensitive consensus primer set, while the three consensus primer sets My, Gp/Gp+, and Cp together detected more HPV-positive patients than the type-specific primers. Testing of sensitivity of the PCR with SiHa cells serially diluted in lymphocytes (HPV-negative cells) indicated a detection limit of 6,300 HPV type 16 DNA copies with consensus primers (My, Gp+, and Cp) and 126 original HPV type 16 DNA copies with type-specific primers. Comparison of the amplification results for consensus L1 primers and type-specific E6-E7 primers indicated the presence of L1 deletions in 23 of 56 samples. The conclusion is that in PCR detection systems, multiple consensus primers and type-specific primers should be used in order to detect all patients harboring HPV.