We established a new assay to detect the E6–E7 DNA of mucosal human papillomaviruses (HPV) by a PCR-based method using four pairs of degenerate LCR and E7 primers (LCR-E7 PCR). This assay amplifies the full length of E6 and the N-terminal part of E7. HPV typing was performed using restriction-fragment-length polymorphism (RFLP), and by analyzing the sequences of cloned PCR products. We compared this assay with the first generation hybrid captured assay (HCA-I) and the MY09/11-PCR method. LCR-E7 PCR was able to detect more than 34 mucosal HPV types and theoretically should detect two additional types. LCR-157 PCR and HCA-I detected HPV DNA in 70% (69/99) and 55% (54/99) of low-grade cervical intraepithelial lesions (LSIL), 89% (105/118) and 76% (90/118) of high-grade cervical intraepithelial lesions (HSIL), and 90% (56/62) and 79% (49/62) of invasive squamous cell carcinomas (SCC), respectively. LCR-E7 PCR was more sensitive than the HCA-1 test. Discordant results between the LCR-E7 and MY 11/09-PCR tests were observed in one of 185 (0.5%) normal samples, seven of 85 (8.2%) LSIL samples, seven of 82 (8.5%) HSIL samples, and four of 72 (5.6%) SCC samples. The discordant results were mostly observed in samples with a low-copy number of the HPV genome or with multiple HPV infection. The sensitivity of LCR-E7 PCR was equivalent to that of MY 11/09 ECR, and false positives were less frequent in LCR-E7 PCR. LCR-E7 PCR may be useful for determining the biological activity of detected HPV types, since this method amplifies the entire E6 gene.
As part of an ongoing cohort study in the Hokuriku region of Japan, cervical cell samples from histologically confirmed normal (n5114) or abnormal (n5286) women were examined for the presence of HPVDNAusing a second-generation hybrid capture assay (HCA-II) and LCR-E7 PCR. HCA-II detected low-risk (HPV-6, -11, -42, 43 and -44) and high-risk (HPV-16, -18, -31, -33, -35, -39, -45, -51, -52, -56, -58, -59 and -68) HPV types, while LCR-E7 PCR detected an additional 7 HPV types and some uncharacterized types. In screening of high-grade squamous intraepithelial lesions (HSILs) and invasive cervical cancer, the sensitivities of HCA-II and LCR-E7 PCR testing the high-risk HPV types were 83% and 81%,respectively, while the specificity of both assays was 93%. The sensitivity of LCR-E7 PCR increased to 87%, which was significantly higher than that in HCA-II, when testing both highrisk and other HPV types. Sixty-eight inconsistent results (17% of total tested) fromHCA-II and LCR-E7 PCR were due to (i) low copy number of HPV genome (false-negative for HCA-II, 5.3% and for LCR-E7 PCR, 1.3%), (ii) infection with HPV types undetectable by HCA-II (4.8%), (iii) multiple HPV infections (5%) or (iv) unknown reasons (0.8%). LCR-E7 PCR revealed that infections with HPV-16, -18, -31, -33, -35, -51, -52, -56, -58 or -67 was a high risk for cancer since these types predominated in HSIL and invasive cervical cancer. Samples showing high relative light units (>20) with a high-risk probe in HCA-II also gave positive results in LCR-E7 PCR and were generally associated with abnormal cervical lesions. Thus, we propose that both HCA-II and LCR-E7 PCR are valuable screening tests for premalignant and malignant cervical lesions.