Clearly HPV infection is one of the most important risk factors for anogenital SIL and cancer, and HPV infection is likely to be necessary for development of almost all of these lesions. Sexual risk factors such as number of sexual partners, age at first intercourse, and parity have long been associated with cervical cancer, but these likely reflect risk for acquisition of HPV infection. However, although the age-related prevalence of cervical
LSIL parallels that of infection, only a small proportion of women who acquire HPV infection develop clinically detectable LSIL. An even smaller proportion of these women develop HSIL or invasive cervical cancer. This reflects the current understanding that HPV infection may be necessary but insufficient for development of CSIL and cervical cancer. The paradigm is that HPV infection contributes to carcinogenesis indirectly by stimulating cellular proliferation and rendering the epithelial cells susceptible to genetic damage through induction of chromosomal instability (19,20). Consistent with this hypothesis, chromosomal mutations as shown by studies of loss of heterozygosity may be found in cervical cancer (21-23). Other events that lead to genetic damage in the cells may play a more direct role in carcinogenesis.
Cigarette smoking has been shown to be associated with CSIL, and this was presumed to be mediated by exposure of tobacco-related carcinogens to HPV-infected epithelium and through attenuation of cervical epithelium Langerhans cell number and function. In more recent studies, however, in which the data were adjusted for HPV infection, cigarette smoking was not an independent risk factor (24,25). The association between cervical cancer and oral contraceptive use is not clear, with some but not all studies showing an association. Dietary factors have been proposed and some studies suggest an association between folate deficiency and CSIL (26,27). However, dietary supplements of folic acid appear to have little effect on the natural history of the disease (28,29). In a study of women with CSIL compared to controls without CSIL, after adjusting for HPV infection and demographic factors there was an inverse correlation between plasma a-tocopherol and ascorbic acid levels and risk of CSIL (30), a finding confirmed by a more recent study (31). Further studies are needed to clarify these findings.
In vitro, several studies show that retinoic acid may stimulate cellular differentiation and inhibit proliferation of HPV-infected keratinocytes (32-34). In clinical studies, however, CSIL was not associated with plasma retinol or P-carotene levels (30); and in an earlier study, the association between increased risk of CSIL and retinol levels did not reach statistical significance (35). Overall, the relationship between dietary factors and the incidence or natural history of CSIL remains unclear. In general, in vitro findings have not been matched by clear in vivo results, reflecting either a limited role for these factors in vivo or difficulties in precisely quantifying these factors in clinical studies.
Another risk factor of clear importance is that of immunodeficiency. The epidemio-logic data described previously show that the prevalence of cervical HPV infection and CSIL decline with age. The presumed mechanism of this decline is a cell-mediated immune (CMI) response. Examples of loss of immune competence and its association with increased risk of anogenital and skin malignancies included women undergoing renal transplant who were iatrogenically immunosuppressed to prevent graft rejection (36,37), as well as individuals with a rare immune disorder known as epidermodysplasia verruciformis.
More recently, HIV has emerged as the most common cause of immune suppression. There is a large body of literature that documents an increased risk of detection of HPV DNA in HIV-positive women, increased risk of CSIL, a slightly increased risk of cervical cancer, and greater difficulty treating the lesions. In a study from New York, the prevalence of HPV infection of any type was significantly greater in the 344 HIV-positive (60%) than the 325 HIV-negative (36%) women (38). In addition, multiple types of
HPV were present in 51% of the HIV-positive women compared to 26% of the HIVnegative women. These findings were confirmed in a more recent, larger study conducted at six Women's Interagency HIV study (WIHS) sites around the United States, in which HPV infection was significantly more common among HIV-positive women than HIV-negative matched controls (39). The strongest risk factors for HPV infection among HIV-positive women were indicators of more advanced HIV-related disease such as higher HIV viral load and lower CD4 levels. However, other factors that are commonly found in studies of HIV-negative women were important in HIV-positive women as well, including racial/ethnic background, current smoking, and age. Persistence of HPV infection may also be affected by HIV infection. In one study, persistent HPV infection with "high-risk" types was found in 20% of the HIV-positive women and 3% of the HIV-negative women (40). The likelihood of persistent infection among the HIV-positive women was related to the level of immunosuppression. Women with CD4 counts < 200/mm3 were more than twice as likely to have persistence as women with CD4 counts > 500/mm3.
Cervical cytology is more often abnormal among HIV-positive women than in HIV-negative women in earlier studies (41-43). In a study reported from San Francisco General Hospital (SFGH), CSIL was detected in 9 (20%) of 44 HIV-positive women compared to 2 (4%) of 52 HIV-negative women (44). In the New York cohort, 20% of the HIV-positive women had CSIL compared to 4% of the HIV-negative women (45). In a large recent study from the WIHS, cervical cytology was abnormal in 38% of HIV-infected women and 16% of HIV-negative women (46). Among the risk factors for abnormal cytology in multivariate analysis in this study were HIV infection, lower CD4 cell count, higher HIV RNA level and HPV positivity, a prior history of abnormal cytology, and greater number of male sex partners within 6 mo of enrollment.
Many questions remain about the anti-HPV CMI response, including the nature of the effector cells, the HPV antigens to which they are directed, and whether the immune response leads to complete clearance of HPV infection or suppression of HPV to a level that is below the level of detection of current HPV DNA detection tests. The possibility of suppression, rather than clearance, is important because it implies that HPV could be reactivated at some point in the future by unknown factors, some of which may include loss of the CMI response that suppressed HPV.
Several lines of evidence point to the role of the CMI response in controlling cervical disease, but results have been somewhat conflicting. Earlier studies have shown that women with cervical lesions have decreased cellular proliferative responses to HPV type 16 E6 and E7 proteins compared to women without lesions (47,48). In contrast both healthy women and women with CSIL produce T-cell responses to the HPV 16 L1 protein (49-51), and responses to HPV 16 E7 peptides were linked to viral persistence and disease progression (52). A study of cytotoxic T-cell response showed that fewer women with cervical lesions had responses to E6 and E7 than women without disease (53). These results suggest the possibility that expression of E7 may induce T-cell tolerance, consistent with results in mouse models (54-56). Again, conflicting results have been obtained in other studies (57). Although the role of E7-specific CTL response in clearance of tumors in animal models has been demonstrated (58,59), the role of these responses in vivo in prospective studies has not yet been determined. Host factors such as HLA haplotype have been shown to play a role as well, but these remain poorly defined (60,61).
Was this article helpful?