Women who have human papillomavirus (HPV) infection of the cervix have a greater risk of invasive cervical cancer if they also have genital herpes, according to a pooled analysis of case-control studies.1 Women with invasive cervical cancer were much more likely than women without cervical cancer to have HPV-infected cervical cells, but they were also nearly twice as likely to have antibodies to herpes simplex virus type 2 (HSV-2). Among all women who had HPV-infected cervical cells, women who also had antibodies to HSV-2 had more than twice the risk of squamous cell carcinoma and more than three times the risk of adenocarcinoma or adenosquamous cell carcinoma relative to women who did not have these antibodies. Neither past sexual behavior nor chlamydial infection modified these associations.

Data were obtained from seven studies conducted in Thailand, the Philippines, Morocco, Peru, Brazil, Colombia and Spain. The analysis included 1,263 women with invasive cervical cancer (1,158 with squamous cell carcinoma and 105 with adenocarcinoma or adenosquamous cell carcinoma) and 1,117 women without cervical cancer who were the same age. Exfoliated cervical cells were tested by a polymerase chain reaction assay to determine if they contained HPV DNA and, if so, the HPV type. Serum samples were tested for the presence of type-specific antibodies to HSV-2 and HSV-1, and for antibodies to Chlamydia trachomatis. Personal interviews covered social, demographic, reproductive and other characteristics. Unconditional logistic regression was used to generate summary odds ratios.

On average, women with invasive cervical cancer were 48-49 years old, and women without cancer were 47 years old. Almost all of the women with cervical cancer were HPV-positive (91-95%), compared with 15% of the women without cervical cancer. Women with cervical cancer were significantly more likely than women without cancer to test positive for HSV-2 (44% in both cancer subgroups vs. 26%).

Among women without cervical cancer, several markers of sexual behavior were significantly associated with the odds of testing positive for HSV-2. Compared with married women, both cohabiting women and noncohabiting unmarried women had elevated odds of infection (2.2 and 1.6, respectively). The odds were nearly three times as high among women who had had three or more lifetime sex partners as among those who had had one or none (2.9). The odds were more than twice as high for women who had antibodies to C. trachomatis as for women who did not (2.2), and were 60% higher among women who had used the pill for five or more years than among never-users (1.6). However, the odds of testing positive for HSV-2 were not elevated among women who were infected with HPV.

A multivariate analysis was performed among HPV-positive women, taking into account age, study center, HPV type, history of Pap smears, pill use, number of full-term pregnancies and presence of antibodies to C. trachomatis. HPV-infected women who were also positive for HSV-2 had 2.2 times the odds of squamous cell carcinoma found among those women who tested negative for HSV-2, and 3.4 times the odds of adenocarcinoma or adenosquamous cell carcinoma. Compared with HSV-2-positive women who had low-risk types of HPV, those who had high-risk HPV other than type 16 had 2.6-4.2 the odds of invasive cervical cancer, and those who were positive for type 16 had 4.0-6.7 times the odds.

In analyses taking into account a woman's number of lifetime sexual partners and her age at first intercourse, HPV-positive women who were also infected with HSV-2 still had nearly twice the odds of squamous cell carcinoma as did those who tested negative for HSV-2 (1.9). This risk was not significantly modified by their age, pill use, marital status, number of full-term pregnancies or presence of C. trachomatis antibodies. In contrast, HPV-positive women who tested positive for HSV-1 were not at a higher risk of squamous cell carcinoma relative than those who were negative for HSV-1.

"[G]enital HSV-2 infection may act in conjunction with HPV infection to modestly increase the risk of invasive cervical cancer," the investigators comment. They add that past sexual behavior and presence of chlamydial infection do not modify this association, supporting a direct link between genital herpes and cancer risk in HPV-positive women.

The investigators suggest several mechanisms that may explain genital herpes's role as a cofactor in HPV-induced cervical cancer. Herpes lesions may allow HPV easier access to deeper cell layers of the cervix; alternatively, the inflammation caused by these lesions may interfere with an immune response to HPV or may damage the DNA in HPV-infected cells. The herpes virus may also stimulate HPV to replicate or to integrate its DNA into the DNA of cervical cells. The investigators conclude that "Future studies are needed to elucidate at which step in the pathogenesis of HPV-induced cervical carcinogenesis HSV-2 infection may be relevant."--S. London


1. Smith JS et al., Herpes simplex virus-2 as a human papillomavirus cofactor in the etiology of invasive cervical cancer, Journal of the National Cancer Institute, 2002, 94(21):1604-1613.