Anna Marie Beckmann

A cohort study of the risk of cervical intraepithelial neoplasia grade 2 or 3 in relation to papillomavirus infection.

Abstract Background. Human papillomavirus (HPV) has been associated with cervical intraepithelial neoplasia, but the temporal relation between the infection and the neoplasia remains unclear, as does the relative importance of the specific type of HPV, other sexually transmitted diseases, and other risk factors. Methods. We studied prospectively a cohort of 241 women who presented for evaluation of sexually transmitted disease and had negative cervical cytologic tests. The women were followed every four months with cytologic and colposcopic examinations of the uterine cervix and tests for HPV DNA and other sexually transmitted diseases. Results. Cervical intraepithelial neoplasia grade 2 or 3 was confirmed by biopsy in 28 women. On the basis of survival analysis, the cumulative incidence of cervical intraepithelial neoplasia at two years was 28 percent among women with a positive test for HPV and 3 percent among those without detectable HPV DNA. The risk was highest among those with HPV type 16 or 18 infe...

Human papilloma viruses and p53 mutations in normal pre-malignant and malignant oral epithelia.

HPV infections have been previously observed in oral cancers, and inactivation of the p53 gene has been shown to be one of the most common genetic alterations in human tumors. We examined 179 oral specimens from 70 individuals with histologic findings of either normal mucosa (n = 6) or oral disease that ranged from mild dysplasia to invasive squamous‐cell carcinoma (n = 64) to determine the occurrence of both HPV infection and p53 mutations and their relationship with several clinical factors. HPV infection was detected by PCR amplification of viral DNA, and the presence of p53 mutations was assayed using the single‐strand conformation polymorphism (SSCP)‐PCR technique. HPV infection was found in 31% of individuals with oral disease and was not seen in healthy individuals. Mutations in exons 5, 6, 7 or 8 of the p53 gene were detected in 37.5% of patients with oral lesions and in a biopsy from 1 healthy individual who was a heavy smoker. Approximately one‐third of lesions classified as pre‐malignant (dysplasia and carcinoma in situ) and 42% of invasive carcinomas contained p53 mutations. The majority of these mutations were G:T transversions located within exons 7 and 8. Tumor tissues from 6 patients with oral lesions were found both to be HPV‐16‐positive and to contain p53 mutations; of these, 4 were poorly differentiated carcinomas that were diagnosed as late‐stage disease. In this study, p53 mutations were detected in the early stages of cancer development.

Detection of human papillomavirus capsid antigens in various squamous epithelial lesions using antibodies directed against the L1 and L2 open reading frames

HPV6 and HPV16 infect the squamous epithelium of the genital tract and are thought to be involved in the pathogenesis of benign and malignant lesions. HPV6 is primarily found in benign condylomas whereas HPV16 is present in dysplasias and in invasive squamous cell carcinomas. To examine the expression of the major and minor capsid proteins in these lesions polyclonal antisera directed against bacterially derived fusion proteins harboring different restriction fragments of the L1 and L2 ORFs of HPV6b and HPV16 were generated. L1 ORF-specific antisera were not type-specific and detected the major capsid antigen in lesions infected with related HPV types. Anti-L2 ORF antisera could distinguish among HPV1, HPV6, and HPV16 when the fusion protein used as the immunogen did not harbor the amino-terminus of the L2 ORF. The anti-L1 ORF antisera were employed to detect the major capsid protein in various lesions by immunohistochemical staining. Lesions harboring HPV16 were positive in a high percentage of cervical intraepithelial neoplasia I-II (87%), and less frequently in carcinomas in situ (29%) or invasive carcinomas (17%). In all cases capsid antigen expression was restricted to cells showing some differentiation at the surface or periphery of the lesion.

Detection of viral DNA and RNA by in situ hybridization.

Using cloned restriction endonuclease fragments of Herpes simplex virus (HSV), human papillomavirus (HPV), and cytomegalovirus (CMV) DNA as probes, viral DNA and RNA sequences have been detected in human tissues. The probes were labeled either with a radioactive isotope, for subsequent detection by autoradiography, or with biotin. This latter technique has been successfully used to visualize HPV DNA in tissues that have been fixed in formalin and embedded in paraffin, and is therefore of value in retrospective studies of histological specimens. HPV DNA was detected under non-stringent conditions (Tm = -42 degrees C) with heterologous probes in plantar and common warts, laryngeal papillomas, and anogenital condylomas. The specific type of HPV was established using stringent hybridization conditions (Tm = - 17 degrees C). Results from these and from malignant tissues show the distribution and localization of HSV and HPV RNA and DNA sequences in malignancies of squamous cell origin in the anogenital region. Both HSV and HPV DNA sequences have occasionally been detected in the same tumor, providing a further impetus to test the hypothesis that an initiator-promoter relationship might involve these common human viruses in the development of some tumors.

Human papillomavirus infection in women with multicentric squamous cell neoplasia

Tissues from 32 women with multicentric squamous cell neoplasia of the anogenital region (72 anatomically distinct lesions at the cervix, vagina, vulva, perineum, or anus) were tested for the presence of human papillomavirus with the polymerase chain reaction or in situ hybridization. All the women had invasive carcinomas or grade 3 intraepithelial neoplasia lesions at a minimum of one site and one or two squamous cell lesions at another site(s). Human papillomavirus was detected in all of the multicentric lesions in 87.5% (28/32) of the women and in at least one lesion in 12.5% (4/32). In the 28 women with detectable human papillomavirus at all sites, 61% (17/28) had the same virus type(s) at all sites (types 6, 16, 6 and 16, 33) and 25% (7/28) had 6 or 16 at one site and both viruses at the other site(s). Four women (15%) had different virus patterns in the separate lesions.

Human papillomavirus type 16 DNA in esophageal carcinomas from Alaska natives

The possible etiological role of human papillomavirus (HPV) in esophageal carcinogenesis was evaluated in Alaska Natives in whom the incidence of esophageal cancer is 1.3 and 3.8 times higher than in US Caucasian men and women, respectively. Fixed paraffin‐embedded esophageal tissues from 32 cases of squamous‐cell carcinoma (SCC) and 3 cases of adenocarcinoma (AC) diagnosed between 1957 and 1988 were analyzed by polymerase chain reaction (PCR) and in situ hybridization for HPV DNA sequences. Detection of the human β‐globin gene by PCR was used as a control for sufficiency of DNA and its potential for amplification in the tissue samples. Twenty‐five of the tumor tissues were considered adequate for PCR analyses; HPV DNA was detected in 10 of 22 SCCs and was not found in 3 ACs. Seven of the 10 HPV‐positive tissues contained sequences from the E6 gene of HPV type 16. Kollocytosis, an epithelial change consistent with HPV infection, was found in 80% of the esophageal squamous‐cell tumors with HPV DNA and in 75% of those without HPV DNA. The detection of amplifiable cellular DNA was related to recentness of diagnosis; however, the detection of HPV DNA within amplifiable specimens was not related to recentness of diagnosis. A 413‐bp sequence from the LI open reading frame of HPV 16 from esophageal tissue of 2 patients was identical to sequences previously identified in cervical cells from other Alaska Natives. Our results provide molecular evidence of HPV infection, especially type 16, in archival esophageal cancer tissues from 45% of those patients whose specimens contain adequate DNA for PCR analysis. Int. J. Cancer 71:218–222, 1997.

Genital-type human papillomavirus infection is not associated with surface epithelial ovarian carcinoma

Tumor tissues from 29 women with borderline or malignant epithelial ovarian tumors were examined for the presence of human papillomavirus (HPV) DNA by the polymerase chain reaction (PCR). The PCR analysis used a set of consensus primers that are complementary to highly conserved sequences in the genital HPVs (M. M. Manos, Y. Ting, D. K. Wright, A. J. Lewis, T. R. Broker, and S. M. Wolinsky, Cancer Cells 7, 209-214, 1989). Amplification products were detected by Southern hybridization with consensus oligonucleotide probes. A total of 70 paraffin-embedded tissue sections from ovarian carcinomas were tested and we did not detect genital-type HPV DNA sequences in any of these specimens. However, all of the tissue specimens were considered adequate for PCR analysis because a human cellular gene (beta-globin) was successfully amplified in each tissue specimen. In addition, HPV 16 DNA was found in a concurrent invasive squamous-cell carcinoma of the cervix from one ovarian cancer patient, indicating that the PCR was able to detect HPV in the lower genital tract of this individual. We conclude that there is no association between infection with the most common genital HPVs and borderline and malignant epithelial ovarian tumors.

Human papillomavirus type 16 in multifocal neoplasia of the female genital tract

SummaryWe examined lesions from six women with multifocal neoplasia of the lower genital tract (vulva and cervix or vulva and vagina) for the presence of human papillomavirus (HPV) DNA sequences by DNA-DNA in situ hybridization. Cervical tissue specimens from four of the five women with carcinoma in situ (CIS) of the cervix were found to contain HPV type 16 DNA sequences. Vulvar lesions from three of these women also contained HPV-16 DNA and were histologically consistent with CIS. In each instance, hybridization with HPV-16 DNA probes was seen only in areas of the epithelium that contained evidence of surface maturation or koilocytotic atypia (KA). The HPV DNA was not visualized in regions of these lesions in which the full thickness of the epithelium was occupied by poorly differentiated cells. Virus capsid antigens were only detected in a very few cells in two of the four cervical lesions that contained HPV-16 DNA. HPV type 16, which has consistently been associated with the development of cervical cancer, is further implicated as an agent in the pathogenesis of genital cancers by demonstration of the virus genome in neoplasia at nonadjacent sites.

Subclinical manifestations of vulvar human papillomavirus infection

We studied the presence of human papillomavirus (HPV) DNA by in situ hybridization in 74 colposcopically defined subclinical vulvar lesions obtained from 51 patients (mean age 25 years, range 18-42 years) referred for atypical Papanicolaou smears to the Colposcopy Clinic. Serial sections of formalin-fixed, paraffin-embedded biopsy tissue were stained by routine hematoxylin and eosin stain or were separately hybridized with biotin-labeled probes of HPV types 6, 11, 16, and 18 at high stringency and with mixed HPV probes at low stringency. Histologic evidence of HPV infection (koilocytosis or dysplasia) was found in eight lesions. Overall, 10 lesions were positive by in situ hybridization with HPV DNA. Although certain colposcopic features were more strongly associated with HPV DNA than others were, we were unable to demonstrate any consistent correlation between HPV DNA positivity or specific HPV DNA types and colposcopic categories. There was little correlation between HPV DNA positivity and the histopathologic incidence of HPV infection and dysplasia. Three of the 10 hybridization-positive lesions were severely dysplastic as compared with 0 of 64 nonhybridizing lesions. In spite of this association between HPV DNA and dysplasia, HPV DNA hybridization added no additional clinically relevant information to the histopathologic findings. We emphasize the significance of careful colposcopic examination of the vulva with directed biopsies.

A Cohort Study of the Risk of Cervical Intraepithelial Neoplasia Grade 2 or 3 in Relation to Papillomavirus Infection

BACKGROUND Human papillomavirus (HPV) has been associated with cervical intraepithelial neoplasia, but the temporal relation between the infection and the neoplasia remains unclear, as does the relative importance of the specific type of HPV, other sexually transmitted diseases, and other risk factors. METHODS We studied prospectively a cohort of 241 women who presented for evaluation of sexually transmitted disease and had negative cervical cytologic tests. The women were followed every four months with cytologic and colposcopic examinations of the uterine cervix and tests for HPV DNA and other sexually transmitted diseases. RESULTS Cervical intraepithelial neoplasia grade 2 or 3 was confirmed by biopsy in 28 women. On the basis of survival analysis, the cumulative incidence of cervical intraepithelial neoplasia at two years was 28 percent among women with a positive test for HPV and 3 percent among those without detectable HPV DNA: The risk was highest among those with HPV type 16 or 18 infection (adjusted relative risk as compared with that in women without HPV infection, 11; 95 percent confidence interval, 4.6 to 26; attributable risk, 52 percent). All 24 cases of cervical intraepithelial neoplasia grade 2 or 3 among HPV-positive women were detected within 24 months after the first positive test for HPV. After adjustment for the presence of HPV infection, the development of cervical intraepithelial neoplasia was also associated with younger age at first intercourse, the presence of serum antibodies to Chlamydia trachomatis, the presence of serum antibodies to cytomegalovirus, and cervical infection with Neisseria gonorrhoeae. CONCLUSIONS Cervical intraepithelial neoplasia is a common and apparently early manifestation of cervical infection by HPV, particularly types 16 and 18.