A Study of Genital Human Papillomavirus (HPV) and Evaluation of HPV Testing for Cervical Cancer Screening in Women from the Eastern Cape Province, South Africa

Abstract

Introduction: Human papillomavirus (HPV) infection is an important public health problem facing black African women. Persistent infection with high-risk (HR) HPV types is the key factor for the development of cervical cancer. Coinfection of HPV with other sexually transmitted pathogens contributes to the progression of cervical cancer. Preventative measures including screening for and treating pre-cancerous cervical lesions as well as HPV vaccination have been implemented in parts of South Africa. However, in the rural Eastern Cape Province there is limited information on the prevalence of HPV and the HPV types associated with cervical lesions. Two cohorts were chosen to study HPV in the Eastern Cape (South Africa), a community clinic, and a referral hospital for treatment of cervical lesions. This study aimed at determining the prevalence of HPV, risk factors of HPV, coinfection of HPV with sexually transmitted pathogens and evaluate the performance of a number of HPV tests for HPV detection and cervical cancer screening. The objectives of the study were: • To investigate the prevalence of HR-HPV and factors associated with HR-HPV infection among women from rural Eastern Cape, South Africa. • To investigate the distribution of HPV genotypes among women with cervical intraepithelial lesions according to HIV status from Eastern Cape Province, South Africa. • To investigate HR-HPV prevalence and compare agreement between cliniciancollected and self-collected genital specimens as well as two different HPV tests on clinician-collected samples. • To investigate the prevalence of sexually transmitted pathogens and co-infection of with HR-HPV infection among women from rural Eastern Cape Province, South Africa. Methods: A total of 741 participants were recruited from the Mbekweni Community Clinic (N=417) and the Nelson Mandela Hospital Referral Clinic (N=324) located in the OR Tambo municipality of the Eastern Cape Province. Clinician-collected cervical scrapes from women attending the Community Clinic were screened for HR-HPV prevalence and HR-HPV viral load using Hybrid Capture 2 (HC-2, Qiagen Inc., Gaithersburg, MD; USA); Cervical clinician-collected and vaginal self-collected specimens of women with or without abnormal cytology from both study cohorts were also screened for HR-HPV infection using hpVIR real-time PCR. HPV typing of clinician-collected cervical specimens from women with cervical intraepithelial neoplasia grades 2 and 3 (CIN 2 / 3) was done using Direct Flow Chip HPV kit (Master Diagnostica, Spain). Cervical specimens from the Community Clinic (N=205) were also tested for sexually transmitted infections (STIs) namely Chlamydia trachomatis: CT, Haemophilus ducreyi, Herpes Simplex Virus type 2, Neisseria gonorrhoeae: NG, Treponema pallidum, and Trichomonas vaginalis: TV) and pathobionts (Ureaplasma spp: (UP), Mycoplasma genitalium: MG, and Mycoplasma hominis: MH) using the STD Direct Flow Chip kit (Master Diagnostica, Spain). The univariate and multivariate analysis was used to determine the correlation between HPV infection and potential behavioural risk factors using STATA 14.2 (Stata Corp, College Station, Texas). A chi squared test was used determine the difference in estimated HR-HPV prevalence between self-collected and clinician-collected samples. STIs prevalence and association with behavioural risk factors were analysed using GraphPad Prism v6.01 (GraphPad Software, Inc., San Diego, CA). Results: Of the 417 women from the community clinic, HR-HPV prevalence was significantly higher in HIV-positive women compared to HIV-negative women (40.6%, 63/155 vs 21.4%, 56/262, p< 0.0001). Among women referred to Nelson Mandela Hospital with cervical intraepithelial lesions, HPV prevalence was observed to be significantly higher in HIV-positive than HIV-positive women (98.0% vs 89.1%, p=0.012). Similarly, HIV-positive women (65.3%, 96/147) had higher multiple HPV infections than HIV-negative women (47.8%, 22/46; p=0.034). HPV35 (23.9%), HPV58 (23.9%), HPV45 (19.6%), and HPV16 (17.3%) were the most frequently detected HPV types in CIN2, while HPV35 (22.5%), HPV16 (21.8%), HPV33 (15.6%), HPV58 (14.3%) were commonly detected in women with CIN3 regardless of HIV status. HR-HPV prevalence in clinician-collected samples was equivalent to self-collected samples from both study sites, the community clinic (26.4% vs 27.9%, p=0.601) and the referral clinic (83.6% vs 79.9%, p=0.222). HR-HPV positivity between self-collected and clinician-collected samples showed an agreement of 86.9% for community clinic (k=0.669) and 91.4% for referral clinic (k=0.711). The distribution of HR-HPV genotypes was similar between self-collected and clinician-collected samples from both study sites. The agreement of HR-HPV genotypes between self-collected and clinician ranged from moderate to almost perfect (0.571-0.888). A majority of women reported a high positive response of acceptance for self-collection (community-based clinic: 77.2% and referral clinic: 83.0%). HR-HPV detection agreement between hpVIR real-time PCR and HC-2 was almost perfect (87.7%, k=0.754). The prevalence of the six traditional STIs (CT, TV, NG, HSV-2, TP, and Haemophilus ducreyi) was high (22.9%, 47/205). TV was the most frequently detected STI (15.6%, 32/205). UP (70.2%, 144/205) and MH (36.6%, 47/205) were the most frequently identified pathobionts. Multiple infections/coinfections with more than two STIs/pathobionts was found in 52.7% (108/205) of women with UP/MH (26.9%) and UP/HPV (21.3%) the frequently identified coinfections. HR-HPV infection was significantly associated with HIV infection (p=0.017) and HSV-2 (p=0.026). Conclusion: This study shows that HIV infection and sexual behaviour increased the risk of HPV infection among women from the community clinic. HIV-positive women had significantly higher HPV viral load and multiple HPV type infections compared with HIV-negative women with or without cervical lesions. Since HIV positive women are at higher risk of HPV infection they need to continue to be screened more regularly for cervical lesions and treated when appropriate. In addition, the high prevalence of HPV in the community of HIV negative women indicates that a robust cervical screening programme is needed to implement the cervical screening policy of South Africa. Thus, the women get the allocated three cervical smears in a life time. Distribution of HPV types was similar among women with CIN2 &amp; 3 with HPV35 being the most frequently detected HPV type regardless of HIV status. This highlights the importance for the inclusion of HPV-35 in the next generation of HPV prophylactic vaccines. The findings of this study add to the limited information on genital HPV infection in women from this province. Moreover, our data now acts as a baseline/reference data for future investigations. The data will also contribute to discussions on HPV testing as the primary screening strategy for cervical cancer and HPV vaccination in South Africa. The hpVIR real-time PCR test between self-collected and clinician-collected specimens showed comparable agreement for the detection of HPV infection. The type-specific concordance between self-collected and clinician-collected showed moderate to an excellent agreement, indicating that self-collection can be utilised as the alternative screening tool for cervical cancer. The participants showed a high positive response for the self-collection method, indicating that introducing this method can positively impact the cervical cancer screening program. However, hpVIR real-time PCR is an in-house test which is not practical to introduce on a large scale in South Africa. Therefore, future research should be done to determine what other HPV tests could be done on these types of specimens. Presently, syndromic management is used to treat STI at clinics in South Africa. The high prevalence of sexually transmitted pathogens necessitates the need to enhance the current screening methods for these pathogens.