Allison P. Cullen

The integration of human papillomavirus (HPV) DNA into the human genome has been generally accepted as a characteristic of malignant lesions. To gain a better understanding of this phenomenon, genomic DNA from 181 cervical biopsy specimens was isolated and analyzed for HPV type and physical state of the HPV genome. These specimens represented the full spectrum of cervical disease, from condyloma to invasive carcinoma. Discrimination between integrated and episomal HPV DNA was accomplished by the detection of HPV-human DNA junction fragments on Southern blots. In most cases in which ambiguous Southern blot results were obtained, the specimens were reanalyzed by two-dimensional gel electrophoresis. Of the 100 biopsy specimens of cervical intraepithelial neoplasia analyzed, only 3 showed integrated HPV DNA, in contrast to 56 (81%) of 69 cervical carcinomas (P less than 0.001) showing integrated HPV DNA. Of the 40 carcinomas containing HPV 16 DNA, 29 (72%) had integrated HPV DNA, of which 8 (20%) also had episomal HPV DNA. In 11 (27%) cancers, only episomal HPV 16 DNA was detected. All 23 HPV 18-containing carcinomas had integrated HPV DNA, and 1 also had episomal HPV 18 DNA. The difference between HPV types 16 and 18 with respect to frequency of integration was statistically significant (P less than 0.01). The results of this study indicate that detectable integration of HPV DNA, regardless of type, occurs infrequently in cervical intraepithelial neoplasia. The absence of HPV 16 DNA integration in some carcinomas implies that integration is not always required for malignant progression. In contrast, the consistent integration of HPV 18 DNA in all cervical cancers examined may be related to its greater transforming efficiency in vitro and its reported clinical association with more aggressive cervical cancers.

The Digene Hybrid Capture II (HCII CT/GC) test is a combination test designed to detect Chlamydia trachomatis and Neisseria gonorrhoeae in a single specimen. It is a nucleic acid hybridization test which uses signal amplification to increase sensitivity. We compared its performance to that of culture on cervical specimens from 1,370 women. Direct fluorescent-antibody assay was used to resolve discrepant results for C. trachomatis. Samples were collected with a proprietary cervical brush or with endocervical swabs. The HCII CT/GC test proved to be sensitive and specific in detecting these organisms. Compared to N. gonorrhoeae culture, it had a sensitivity of 93% (87/94) and a specificity of 98.5% (1,244/1,263). Compared to C. trachomatis culture, the sensitivity was 97.7% (129/132) and specificity was 98.2% (1,216/1,238). Testing of some specimens with discrepant results by PCR suggested that the test would actually prove to be even more specific if it were compared to a nucleic acid amplification test (NAAT). The sensitivity of C. trachomatis culture was somewhat less, at 88.6% (117/132). The endocervical brush appeared to be better than Dacron swabs for collecting specimens. The HCII CT/GC test offers an attractive format that allows simultaneous detection of C. trachomatis and N. gonorrhoeae with a single specimen. An initial positive result is followed by repeat tests with probes to identify chlamydiae or gonococci. This test is more sensitive than C. trachomatis culture and is at least as sensitive as culture for gonococci. It deserves further evaluation and comparison with NAATs and may well offer an attractive alternative for diagnosis and screening of these infections.

Cox, J. Thomas; Lorincz, Attila T.; Schiffman, Mark H.; Sherman, Mark E.; Cullen, Allison; Kurman, Robert J. Author Information

The prevalence and associated cytologic manifestations of cervical infection with human papillomavirus (HPV) types 6, 11, 16, 18, 31, 33, 35, 42, 43, 44, 45, 51, 52, and 56 were studied among 500 consecutive women attending the Harborview Medical Center Sexually Transmitted Diseases (STD) Clinic in Seattle, WA. Using radiolabeled-probes without prior amplification of DNA, HPV DNA was detected in cervical specimens from 120 (24%) of the women and was found to be more prevalent than Chlamydia trachomatis (13%), Neisseria gonorrhoeae (12%), or mucopurulent cervicitis (20%). High-risk HPV types 16 or 18 were present alone in 5% of the women; intermediate-risk types 31, 33, 35, 45, 51, 52, or 56 in 3%; and low-risk types 6, 11, 42, 43, and 44 in 5%. In an additional 8% HPV DNA was detected but could be characterized only as being type 6, 11, 16, 18, 31, 33, or 35. Each grouping of HPV types was equally associated with squamous intraepithelial lesions (SILs) of the cervix. In the absence of SIL and koilocytosis, the cytologic changes associated with HPV infection included frequent binucleation and variation in nuclear size and chromatin distribution. Parakeratosis and hyperkeratosis without nuclear atypia were not associated with HPV DNA. The natural history and clinical significance of these HPV-associated lesions remain to be defined.