Cervical carcinoma is the second most prevalent cancer in the world, and the most common female cancer in sub-Saharan Africa. Infection with high-risk human papillomavirus (hr-HPV) is the main risk factor for the development of cervical carcinoma. HIV-positive women are reported to be almost twice as likely to be concurrently infected with HPV than HIV-negative women. Recently introduced HPV vaccines against genotypes 16 and 18 are designed to prevent about 70% of cervical cancers. The basis for selecting these genotypes was their prevalence in Europe and North America [
There are limited data on hr-HPV genotypes among HIV-positive women in Africa, and little is known about their relationship with cervical cytology in these populations. We carried out a study to determine the prevalence of HPV genotypes among HIV-positive women in Africa with normal and abnormal cervical cytology scores.
Cervical samples were collected from women attending the HIV treatment clinic of Sekou-Toure regional hospital and surrounding hospitals in Mwanza, Tanzania between February and March 2010, as part of a randomized controlled trial. Eligible subjects were nonpregnant, HIV-positive females, over 18 years of age who had used antiretroviral therapy (ART) for at least 6 months. Subjects were excluded if pregnant, breastfeeding, intolerant to lactose or fermented milk. In total 168 Tanzanian women were included in the study, of whom 143 women had cervical swabs adequate for HPV testing at baseline. The Medical Research Coordinating Committee of the National Institute for Medical Research, Tanzania, approved the study design and protocol. Participants were informed of the purpose of the study and gave their signed or thumb-printed informed consent before participation (ClinicalTrials.gov Identifier: NCT01258556) [
Cervical samples collected in 2008 from HIV-positive women attending the Ndlovu Medical Centre in Elandsdoorn in Moutse District, South Africa, were also included in this analysis. In total, 51 women testing positive for HIV were included. Information was provided to the study participants and oral consent was obtained by the doctors for a cervical sample to be examined. Pregnant women and those aged ≤16 years were excluded. None of the HIV-positive women were receiving ART at the time of sample collection.
Cervical samples from all participants were collected by a clinician using a standard sampling brush, and placed into a vial with coagulant fixative Boonfix, which consists of ethyl alcohol, low molecular weight PEG, and acetic acid. This substance has no detrimental effects on DNA preservation or DNA isolation [
A ThinPrep Cytology slide was made of the Boonfixed samples, and stained with the Papanicolaou method. The Cytology ThinPrep slides were classified according to the Bethesda classification system as WNL (within normal limits), ASCUS or LSIL (atypical squamous cells of undetermined significance and low-grade squamous intraepithelial lesions), or HSIL (high-grade squamous intraepithelial lesions). Of the cases with an abnormal cytology score (ASCUS, LSIL, and HSIL), the residue of the cervical sample was used to prepare histologic paraffin slides [
The purification of DNA from the Boonfix solution was performed in the BioRobot M48, Qiagen (Germany), which uses the MagAttract DNA Mini M48 Kit. This magnetic-particle technology provides high-quality DNA. Samples were analyzed using a sensitive SPF 10 PCR-reverse hybridization line probe assay (Innogenetics, Belgium). This system is based on PCR of part of the L1 region of the HPV genome, which is amplified using SPF10 primers [
Type-specific HPV prevalence was compared between women with HSIL, LSIL, and WNL. The chi-square test was used to test for significant associations between country and HPV type. Analyses were conducted using SPSS and Mathematica.
One hundred three (72%) Tanzanian women and 35 (69%) South African women had a normal cytology score (WNL) (Table
Cervical cytology results, overall and by country.
Tanzanian women ( |
South African women ( |
Total ( |
|
---|---|---|---|
Negative Cytology (WNL) | 103 (72.0%) | 35 (68.6%) | 138 (71.1%) |
ASCUS-LSIL | 31 (21.7%) | 8 (15.7%) | 39 (20.1%) |
HSIL | 9 (6.3%) | 8 (15.7%) | 17 (8.8%) |
(a) LSIL, with polyploid (enlarged) nuclei and koilocytosis. Tested positive for hr-HPV 52 and hr-HPV 56. (b) HSIL, dyskeratotic cells. Tested positive for lr-HPV 44, hr-HPV 52, and undetermined HPV types.
Overall, 109 (56%) women tested positive for hr-HPV (Table
Cytology score versus HPV testing.
Negative cytology (WNL) | ASCUS-LSIL | HSIL | Total | |
---|---|---|---|---|
Hr-HPV positive | 58 (42.0%) | 35 (89.7%) | 16 (94.1%) | 109 (56.2%) |
Lr-HPV positive | 68 (49.3%) | 27 (69.2%) | 11 (64.6%) | 106 (54.6%) |
HPV negative | 43 (31.2%) | 0 | 0 | 43 (22.2%) |
| ||||
Total | 138 (71.1%) | 39 (20.1%) | 17 (8.8%) | 194 |
(a) Age distribution versus hr-HPV genotypes. (b) Age distribution versus lr-HPV genotypes.
Overall, the most prevalent hr-HPV subtypes were hr-HPV 16 (26%) and hr-HPV 52 (30%), and the prevalence of hr-HPV genotypes varied by country. Among Tanzanian women testing positive for any hr-HPV, the most common genotype was hr-HPV 52 (35%), followed by hr-HPV 16 (23%) and hr-HPV 66 (20%). In contrast, among South African women testing positive for any hr-HPV, the most common genotype was hr-HPV 16 (34%), followed by hr-HPV 35 (24%), hr-HPV 18 (21%), hr-HPV 33 (21%), and hr-HPV 66 (21%). Genotypes 18 and 35 were significantly more common in South Africa than Tanzania; the prevalence of genotype 18 was 21% in South Africa compared to 4% in Tanzania (
(a) Percentage weighted prevalence with country specific hr-HPV prevalence is shown. Significant differences between Tanzania and South Africa are marked with an asterisk if
All women with abnormal cytology (ASCUS, LSIL, or HSIL) had detectable HPV (Table
Cytology score versus hr-HPV.
Negative cytology (WNL) (Total: 138) | ASCUS-LSIL (Total: 39) | HSIL (Total: 17) | |
---|---|---|---|
Hr-HPV 16 | 8 (5.8%) | 15 (38.5%) | 5 (29.4%) |
Hr-HPV 18 | 6 (4.3%) | 1 (2.6%) | 2 (11.8%) |
Hr-HPV 31 | 5 (3.6%) | 2 (2.9%) | 0 |
Hr-HPV 33 | 11 (8.0%) | 3 (7.7%) | 1 (5.9%) |
Hr-HPV 35 | 6 (4.3%) | 3 (7.7%) | 4 (23.5%) |
Hr-HPV 39 | 3 (2.2%) | 1 (2.6%) | 2 (11.8%) |
Hr-HPV 45 | 1 (0.7%) | 0 | 0 |
Hr-HPV 51 | 10 (7.2%) | 3 (7.7%) | 2 (11.8%) |
Hr-HPV 52 | 17 (12.3%) | 9 (23.1%) | 7 (41.2%) |
Hr-HPV 56 | 3 (2.2%) | 3 (7.7%) | 0 |
Hr-HPV 58 | 3 (2.2%) | 8 (20.5%) | 0 |
Hr-HPV 59 | 0 | 1 (2.6%) | 0 |
Hr-HPV 66 | 7 (5.1%) | 13 (33.3%) | 2 (11.8%) |
Hr-HPV 68 | 2 (1.4%) | 2 (2.9%) | 2 (11.8%) |
This study found that hr-HPV genotypes not included in the currently licensed HPV vaccines are highly prevalent in HIV-positive women from Tanzania and South Africa. As expected, among women with hr-HPV the prevalence of hr-HPV 16 was high (26%).
Despite the high rate of abnormal cytology among women in Tanzania, hr-HPV 18 was uncommon in women with abnormal cytology (4%). In contrast, hr-HPV 18 was significantly more common among South African women with abnormal cytology (21%). We also found significant regional differences in the prevalence of hr-HPV 35; the prevalence of hr-HPV 35 was 8% among Tanzanian women compared to 24% for South African women. In rural Mozambique, Castellsagué et al
The hr-HPV 52 prevalence found in this study (35% of Tanzanian women and 17% of South African women with any HPV) is vastly different from that in women in The Netherlands (5.5%) [
The prevalence of hr-HPV 45 was low in our cohort (Figure
The strengths of our study include using a standardized HPV type-specific PCR protocol and cytology diagnosis across the two sites. There were differences between the women studied that may have contributed to the different prevalence estimates of hr-HPV types. Tanzanian women were all using ART. In contrast, the women in South Africa were recently diagnosed with HIV and not yet eligible for ART. In HIV positive women there is less HPV clearance due to a lack of an appropriate immune response. The impact of ART in HPV infection and cervical cancer still needs to be defined [
For an HPV vaccine to effectively reduce incidence of cervical cancer, it will need to protect against the major hr-HPV infection types progressing to cervical cancer. Because infection types may vary across settings and ethnicities, the impact a vaccine has in preventing cancer in particular settings needs to be considered. In this study, hr-HPV16 was found in 29% of women diagnosed with HSIL; however, hr-HPV 18 was only found in 11%.
Certainly the high prevalence of hr-HPV found here and reported by others, [
In summary, the present study showed some differences in the prevalence of HPV genotypes among HIV-positive women in two African countries, and an overall high HPV prevalence. The differences found in hr-HPV genotypes warrant the need to consider different monitoring programmes for cervical preneoplasia, especially in HIV-positive women.
Leiden Cytology and Pathology Laboratory (LCPL) is an independent laboratory whose main occupation is to analyze samples of screening programs for cervical carcinoma in The Netherlands and develop better screening methods. No conflict of interests were present, either at LCPL, Erasmus University, Lawson Health Research Institute, Elandsdoorn Health Clinic, DDL, or Bontekoe Research.
J. Dols was supported by Lawson Health Research Institute funding and by LCPL. We would like to thank all people included for making this trial possible, particularly dr. J. Changalucha and dr. N. Butamanya. And mama Mlawa, mama Karoko and mama Judith, three local women who worked with us in Sekou-Toure regional hospital.