Ao. Olsen et al., HERPES-SIMPLEX VIRUS AND HUMAN-PAPILLOMAVIRUS IN A POPULATION-BASED CASE-CONTROL STUDY OF CERVICAL INTRAEPITHELIAL NEOPLASIA GRADE II-III, APMIS. Acta pathologica, microbiologica et immunologica Scandinavica, 106(3), 1998, pp. 417-424
In order to evaluate the association between seropositivity for herpes
simplex virus (type 1 and type 2) and cervical intraepithelial neopla
cia (CIN), we analysed data from a population-based case-control study
of CIN grade II-III which included Norwegian women aged 20 to 44 year
s, 94 cases and 228 controls. Our objectives were to determine if HSV-
1 and/or HSV-2 seropositivity were independent risk factors for CIN, t
aking human papillomavirus exposure into account, and to elucidate the
combined effect of HPV positivity and seropositivity for HSV. In logi
stic regression analyses, the association between HSV-2 or HSV-I serop
ositivity and CIN II-III was not explained by HPV (adjusted OR 3.0; 95
% CI 1.3-7.2 and adjusted OR 3.3; 95% CI 1.3-8.4, respectively). In an
alyses restricted to HPV-16 DNA-positive individuals, seropositivity f
or HSV-2 increased the risk of CIN (OR 11.1; 95% CI 1.2-105.7), wherea
s HSV-1 seropositivity was not significantly associated with CIN. In w
omen positive for other HPV types. only HSV-1 seropositivity was assoc
iated with CIN (OR 8.5; 95% CI 1.3-55.8). In analyses of the HPV-16-se
ropositive individuals, neither HSV-1 nor HSV-2 seropositivity was ass
ociated with CIN. Compared to the reference group of jointly unexposed
subjects, the HPV16 DNA-positive women who were anti-HSV-2 negative h
ad an increased risk of CIN (OR 29; 95% CI 12-67), whereas the risk in
women who were both HPV-16 DNA-positive and HSV-2 was OR=247 (95% CI
31-1996). The estimate of interaction was strong, but did not reach si
gnificance, and our findings may suggest that the combined effect of t
he two viruses is of aetiological importance in cervical carcinogenesi
s. Furthermore, the results indicate that HPV DNA positivity is not su
fficient to explain the sexual behaviour-associated risk of cervical n
eoplasia and that further studies on the role of genital HSV (type 1 a
s well as type 2) and other STDs are warranted.