Herpes simplex virus type 2 infection and cervical cancer: a prospective study of 12 years of follow-up in Finland

Association between human herpesvirus infection and cervical carcinoma: a systematic review and meta-analysis

BACKGROUND

Cervical cancer (CC) is one of the most common gynecologic tumors among women around the world. Although the etiological role of human papillomavirus (HPV) in CC is well established, other factors in CC carcinogenesis remains unclear. Here, we performed a systematic review and meta-analysis to explore the association between infections of human herpesvirus (HHVs) and CC risk.

METHODS

Embase and PubMed databases were utilized to search the relevant studies. The revised JBI Critical Appraisal Tool was used to assess the quality of the included studies. Prevalence and odds ratios (ORs) with 95% confidence intervals (CI) were calculated to evaluate the association between viral infection and CC or precancerous cervical lesions (PCL).

RESULTS

Totally 67 eligible studies involving 7 different HHVs were included in meta-analysis. We found an increased risk of CC or PCL that was associated with the overall infection of HHVs (CC, OR = 2.74, 95% CI 2.13-3.53; PCL, OR = 1.95, 95% CI 1.58-2.41). Subgroup analysis showed a trend towards positive correlations between herpes simplex virus type 2 (HSV-2) infection and CC (OR = 3.01, 95% CI 2.24 to 4.04) or PCL (OR = 2.14, 95% CI 1.55 to 2.96), and the same is true between Epstein-Barr virus (EBV) infection and CC (OR = 4.89, 95% CI 2.18 to 10.96) or PCL (OR = 3.55, 95% CI 2.52 to 5.00). However, for HSV-1 and cytomegalovirus (HCMV), there was no association between viral infection and CC or PCL. By contrast, the roles of HHV-6, HHV-7, and Kaposi sarcoma-associated herpesvirus (KSHV) in cervical lesions were unclear due to the limited number of studies.

CONCLUSIONS

This study provided evidence that HHVs infection as a whole increase the risk of CC incidence. In addition, some types of HHVs such as EBV and HSV-2 may serve as potential targets in the development of new interventions or therapeutic strategies for cervical lesions.

Herpes simplex virus type 2 and the risk of cervical cancer: a meta-analysis of observational studies

OBJECTIVES

To assess whether herpes simplex virus type 2 (HSV-2) infection has an effect on the risk of cervical cancer.

METHODS

A systematic literature search of PubMed, Embase, Web of Science, and Scopus from their inception through July 2013 was conducted and reference lists of retrieved articles were reviewed. Information on the characteristics of the included studies, risk estimates, and control for possible confounding factors was extracted independently by two investigators. A random effects model was used to calculate the pooled risk estimates.

MAIN RESULTS

Sixteen articles with 20 studies (14 case control and 6 longitudinal) involving 3,337 patients with cervical cancer were included. Compared with individuals who did not experience HSV-2 infection, the pooled ORs of cervical cancer for individuals with HSV-2 infection were 1.37 (95 % CI 1.12-1.69) for traditional case-control studies and 1.04 (95 % CI 0.82-1.31) for prospective or retrospective nested case-control studies.

CONCLUSIONS

Given that a longitudinal study gives a higher level of evidence than a traditional case-control study, the existing observational epidemiological evidence do not support a harmful effect of HSV-2 infection on cervical cancer.

Vaccines against human papillomaviruses--a major breakthrough in cancer prevention

Carcinoma of the cervix (CaCer) is the second most frequent malignancy in women on a global scale. Epidemiological studies carried out at the beginning of the second half of the 20th century showed that CaCer was of infectious nature and that its agent was transmitted by sexual intercourse. For some 15 years, herpes simplex virus type 2 (HSV2), the genital herpes virus, was suspected to be the etiological agent. This hypothesis was disproved just in the time when the first convincing evidence that the agents of the disease were human papillomaviruses (HPVs) was produced. Copious new findings obtained during the 1980's and 1990's unequivocally confirmed that HPVs were the causative agents. The most dangerous among the over 100 HPV types are types 16 and 18, which together account for over 70% of CaCer cases and very likely also for most of the other malignancies of the anogenital region and the oropharynx. Extensive research of the HPV biology and immunology enabled the development of vaccines based on the s.c. virus-like particles (VLP) prepared by genetic engineering. At present, there is one HPV vaccine on the market; it contains, besides types 16 and 18, also types 6 and 11, the causative agents of certain benign tumours of the genital area and of the larynx. A new vaccine, comprising types 16 and 18 only, the product of another firm, is to appear on the market soon. Both vaccines have already been tested in extensive clinical trials. They are nearly 100% effective, only very weakly reactogenic and they undoubtedly belong among the most perfect vaccines ever produced. The darker side of the anti-HPV vaccines is their high price, the fact that the highest benefits they bring will only become evident in 20 or 30 years, and that they do not afford protection against all oncogenic HPVs. It is therefore imperative that organized cytological screening be continued: it is destined to remain the main instrument of CaCer prevention for several decades. With all probability also other types of vaccine are under development, viz. VLP-based vaccines, whose range of applicability will be wider than that of the present preventive vaccines, as well as vaccines that will, hopefully, be able to inhibit already progressing infection or will be utilizable in CaCer immunotherapy.

Interactions between human papillomavirus and other sexually transmitted agents in the etiology of cervical cancer

The interplay between human papillomavirus, notably type 16, and HIV in cervical carcinogenesis leads to persistent infection and cervical neoplasia by destruction of the afferent arm (Langerhans cells) of the host immune system. The joint effect takes place at the early stages of squamous intraepithelial lesions and has severe consequences if left untreated. The recent increase of cervical cancer mortality in young women in developed countries may well be a result of the HIV epidemic. Infection with Chlamydia trachomatis is associated with cervical squamous cell carcinoma but not with cervical adenocarcinoma, and the association remains after adjusting for human papillomavirus 16. Joint effects of C. trachomatis and the human papillomaviruses have not been studied at the population level but indirect evidence from epidemiological studies suggests that the interaction might be different (synergistic versus antagonistic) at different stages (cervical intraepithelial neoplasia versus invasive cervical cancer) of cervical carcino-genesis. Concomitant exposure to human papillomaviruses 6 or 11 and human papillomavirus type 16 has not been shown to result in excess risk of cervical squamous cell carcinoma. This antagonistic joint effect was also discovered between human papillomavirus types 18 and 16, as well as 33 and 16. Herpes simplex virus type 2 antibodies are associated with a modest risk of cervical cancer, which is not surprising since the presence of herpes simplex virus antibodies reflects risk-taking sexual behaviour. However, no excess risk remains after adjustment for human papillomavirus type 16, and no interaction between these two viruses has been found in epidemiological studies. Evidence of interaction between human papillomavirus type 16 and the other members of the herpesvirus family is still at an experimental level and difficult to judge. Little progress has been made in the most promising experimental association between the oncogenic human papillomaviruses and adeno-associated viruses. In addition to the well established interaction between human papillomaviruses and HIV, intriguing interactions are emerging between the human papillomaviruses and C. trachomatis, as well as between the different human papillomavirus types.

Cigarette smoking and cervical cancer: Part I: a meta-analysis

Cancer of the cervix is the third most common cancer among women worldwide and its etiology is not clearly understood. Human papillomavirus can be found in approximately 95% of cervical cancers, but it does not appear to be necessary or sufficient to induce malignancy. In 1977, Winkelstein suggested that cigarette smoking was a causative factor in the development of cervical cancer. We report a meta-analysis of cigarette smoking and cervical disease and conclude that the data support a role for cigarette smoking as a risk factor for cervical cancer. We propose a multifactorial hypothesis involving a virus-tar interaction as the etiology of cervical cancer.

Causality in medicine: the case of tumours and viruses

Clarification of the aetiology of chronic human diseases such as atherosclerosis or cancer is one of the dominant topics in contemporary medical research. It is believed that identification of the causal factors will enable more efficient prevention and diagnosis of these diseases and, in some instances, also permit more effective therapy. The task is difficult because of the multistep and multifactorial origin of these diseases. A special case in contemporary aetiological studies is definition of the role of viruses in the pathogenesis of human cancer. Virus-associated cancer develops only in a small minority of infected subjects, which implies that, if the virus does play a role in the pathogenesis of the malignancy, other factors must also be involved. In this paper the author attempts to review the present methodological approaches to aetiological studies of chronic diseases, discusses the role of criteria for identifying causal relationships and proposes guidelines that might help to determine the role of viruses in human cancer.

Prospective study on cervical neoplasia IV. Presence of HPV antibodies

Sera collected in the course of a prospective study carried out in Prague in 1975-1983 were assayed for the presence of human papillomavirus (HPV) antibodies. Women with cervical neoplasia proven by biopsy at enrollment possessed antibodies to peptides derived from E2, E4 and E7 proteins of HPV16 and to virus-like particles (VLPs) of HPV16, -18 and -33 significantly more frequently than matched controls. Women without cervical neoplasia at enrollment who developed the disease in the course of the study differed from matched controls by a higher prevalence of antibodies against VLPs of HPV16 and -18 but not against early antigens of HPV16. In 19 of the latter subjects, paired serum specimens were tested, the first samples having been taken at enrollment and the second at diagnosis. Development of the disease was associated with seroconversion from negativity to positivity to at least one HPV antigen in 11 (57.9%) women.

Cervical cancer: is herpes simplex virus type II a cofactor?

In many ways, cervical cancer behaves as a sexually transmitted disease. The major risk factors are multiple sexual partners and early onset of sexual activity. Although high-risk types of human papillomaviruses (HPV) play an important role in the development of nearly all cases of cervical cancer, other sexually transmitted infectious agents may be cofactors. Herpes simplex virus type 2 (HSV-2) is transmitted primarily by sexual contact and therefore has been implicated as a risk factor. Several independent studies suggest that HSV-2 infections correlate with a higher than normal incidence of cervical cancer. In contrast, other epidemiological studies have concluded that infection with HSV-2 is not a major risk factor. Two separate transforming domains have been identified within the HSV-2 genome, but continued viral gene expression apparently is not necessary for neoplastic transformation. HSV infections lead to unscheduled cellular DNA synthesis, chromosomal amplifications, and mutations. These observations suggest that HSV-2 is not a typical DNA tumor virus. It is hypothesized that persistent or abortive infections induce permanent genetic alterations that interfere with differentiation of cervical epithelium and subsequently induce abnormal proliferation. Thus, HSV-2 may be a cofactor in some but not all cases of cervical cancer.

Cervical cancer and herpes simplex virus type 2: case-control studies in Spain and Colombia, with special reference to immunoglobulin-G sub-classes

Two case-control studies, including 449 histologically confirmed cases of cervical intra-epithelial neoplasia (CIN) III and 425 controls, and 2 studies on invasive cervical cancer, involving 316 histologically confirmed cases and 330 population controls, were conducted in Colombia and Spain to assess the role of herpes simplex virus type 2 (HSV-2) in cervical neoplasia. Antibodies to this virus were also measured in the sera of 931 husbands of cases and controls. A serological assay using type-specific antigens, glycoprotein C for type I (gC-I) and glycoprotein G for type 2 (gG-2) was employed. Immunoglobulin-G (IgG) sub-classes, IgG1 and IgG3, were measured in women positive for HSV-2 antibodies. No increase in risk of CIN III or invasive cancer was found in women whose sera or whose husbands' sera were positive to HSV-2. However, compared with women negative to HSV-2, the risk of CIN III progressively increased with increasing levels of IgG1. The trend was statistically significant in Colombia. There was also a statistically significant increasing trend in risk of invasive cancer with levels of IgG1 in Spain. The levels of IgG3 and its ratio to IgG1, which may indicate recurrent infections, were not associated with the risk of either type of cancer. When the association with IgG1 was analyzed by human papillomavirus (HPV) DNA status, as determined by polymerase chain reaction, the trend was clearer in women whose HPV status was not determined or in those with negative HPV DNA. These results suggest that the role of HSV-2 is merely marginal and do not support the hypothesis that recurrent HSV-2 infections are of importance for cervical neoplasia.

Increased risk of malignant lymphoma indicated by elevated Epstein-Barr virus antibodies--a prospective study

We estimated Epstein-Barr virus (EBV) antibody-associated relative risks (RR) of malignant lymphoma/leukemia within a cohort of 39,000 healthy Finnish adults followed up for 12 years. Antibody analyses to EBV capsid antigen (VCA), early antigen (EA), and nuclear antigens (EBNA, EBNA1, and EBNA2) were based on concomitantly evaluated ELISA techniques. No increased risk was associated with mere EBV seropositivity. However, elevated EBV EA and EBNA antibody levels were associated with a statistically significant excess risk of malignant lymphoma/leukemia (RREA = 3.4, 95 percent confidence interval [CI] = 1.0-11.0; RREBNA = 4.5, CI = 1.2-16.9). These elevated antibody responses may be due either to destruction of neoplastic EBV positive B-cells and/or to activation of latent EBV infection early in the lymphomagenesis.