In this study of cervical cancer risk factors in HIV-positive Nigerian women, we found a 6% prevalence of cervical pre-cancer and cancer. This finding is inconsistent with previous reports from other Sub-Saharan African countries where a higher prevalence of pre-cancer and cancer has been reported [31–38]. Studies conducted in Rwanda, Kenya, South Africa, Uganda and Zambia reported prevalence of cervical pre-cancer and cancer among HIV-positive women of 24.3%, 26.7%, 66.3%, 73.0% and 76%, respectively [31–35].
In order to address these inconsistencies, knowing that the age-standardized incidence of cervical cancer in Nigeria is 34.5 per 100,000 people/year and this has not changed significantly in the last few decades , that 20% of cervical abnormalities if untreated progress to carcinoma, and assuming that the average duration from a cervical abnormality to cancer is about 15 years given that the average duration of cervical cancer from the time of infection with the HPV to cervical cancer has been estimated at 30 years , it is possible to estimate the expected prevalence of cervical abnormalities. The relationship between the incidence, I, and the point prevalence of disease, p, can be used to calculate the expected prevalence of cervical abnormalities as follows:
is the average duration of the disease . Applying this formula and based on the assumptions made above, an expected prevalence of 2.5% is obtained. Further assumptions based on the sensitivity of VIA/VILI will increase this prevalence estimate. It is also easy to see how, with small departures from these assumptions, the prevalence of cervical abnormality observed in the study could have been easily obtained. For example, if only 10% of untreated cervical abnormalities in this setting progress to frank cancer, the expected prevalence of cervical abnormalities would be 4.9%. It is noted that the incidence of cervical cancer in East Africa is about 50% higher than in West Africa . For prevalence to be as high as that reported in some of the other studies in Sub-Saharan African, such as 27% as reported in a recent study among HIV-infected women in Kenya, the cervical cancer incidence rate would have to be about 493/100,000 people/year; approximately 10-fold greater than what has been observed in African population-based registries. Thus, it is probable that the cervical pre-cancer and cancer prevalence obtained in these other studies were over-estimated and the results of the present study are likely to be much closer to the true value.
The higher prevalence reported by some of these other studies may also be partly explained by differences in the sexual practices of the women studied . Having multiple sexual partners increases the risk of acquiring HPV, and in turn, the development of cervical pre-cancer and cancer. In Cote D’Ivoire, 56% of the women had ≥5 lifetime sexual partners and in South Africa the median number of sexual partners was 4, while in the present study 96% of women reported ≤ 2 sexual partners [2, 33, 41]. It is, however, unlikely that these differences are sufficient to account for such high rates of prevalence as those reported in some of the previous studies. The limitations of self-reported sexual practices data are noted, but it is not believed that this would differ systematically among Sub-Saharan countries .
This study is similar to other studies where participants are recruited through HIV and ART clinics [5, 32, 33, 43]. The median age of women in this study was similar to that of women in other studies [32, 33, 44]. In general, women in the current study had higher CD4 counts than women in other studies, as all HIV-infected women in this study had initiated ART treatment [32, 33, 44].
The prevalence and incidence of cervical cancer is only marginally affected by HIV/AIDS . Since the introduction of anti-retroviral therapy, the incidence of cervical cancer has not decreased substantially, unlike that of other AIDS-defining cancers such as Kaposi’s sarcoma and non-Hodgkin’s lymphoma. This suggests that immunosuppression caused by HIV infection is only marginally associated, at best, with the progression to cervical cancer [9, 14–18]. The lack of a decrease in the incidence of cervical cancer in this era of effective ART is supported by data from HIV/AIDS cohorts in developed countries [18, 45, 46]. This finding may be partially explained by the lack of impact of immunosuppression on carcinogenetic HPV, a necessary but not sufficient cause of cervical pre-cancer and cancer [44, 47].
The majority of the women in this study were aged between 30 and 39 (55%), and the median age observed for an invasive cancer diagnosis or positive screening result was 32 years, which is about 10 years younger than the median age of 47 years for cervical cancer diagnosis in the general population . This result is similar to that of other studies from Africa, which showed HIV-positive women presenting with cervical pre-cancer and cancer at an earlier age. In Kenya and South Africa, HIV-positive women with invasive cervical cancer were on average about 10 years younger compared with women who were HIV-negative [49, 50]. One hypothesis is that unsafe sexual behavior puts women at risk of both HPV and HIV, and that this may explain the younger age of presentation of HIV-infected women . However, given the results from HIV cohorts in developed countries that suggest a limited role of HIV-associated immunosuppression in cervical carcinogenesis [18, 45, 46], the differences in age between HIV-positive and HIV-negative people may be a reflection of the lower life expectancy of persons living with HIV/AIDS in Africa.
The results of this study suggest that HIV-infected women with higher CD4 cell counts were at a reduced risk for invasive cancer diagnosis or a positive VIA result. This result agrees with findings from other studies. For example, in Rwanda, an inverse association was found between the CD4 count and cervical pre-cancer among HIV-infected women . Harris et al. found that HIV-negative and HIV-positive women with a CD4 cell count of greater than 500 cells per mm3 had low incidence of cervical pre-cancer compared with those with lower CD4 counts . These findings may reflect competing risks in HIV-positive patients with low CD4 counts, and the need to focus more on immune reconstitution in that population. It also appears that once pre-cancerous lesions develop, disease progression is not affected by the CD4 count, consistent with the marginal role previously reported for immunosuppression in disease progression .
We also found the risk for cervical pre-cancer and cancer to be associated with age (≥40 years) which supports the natural history of cervical carcinogenesis from hrHPV infection through cervical dysplasia, pre-cancer and cancer; having five or more abortions and the presence of vaginal wall abnormalities suggesting a generalized increased risk of sexually transmitted diseases in women who have cervical pre-cancer and cancer. The non-linearity in the relationship between age and screening status may be due to the natural history of cervical carcinogenesis where infection needs to persist before engendering malignant change .
As this was a cross-sectional study it was not possible to explore cause-and-effect relationships. We did not perform HPV DNA testing, cytology or biopsies to confirm the result of visual inspections; as a result, visually inapparent pre-cancerous lesions could have been missed. Also, since this study was conducted among HIV-infected women on ART, the results may not be generalizable to uninfected women. Nevertheless, we believe this study demonstrates the risk factors of positive cervical cancer screening test in this population.