HPV self-sampling as primary screening test in sub-Saharan Africa: implication for a triaging strategy

Our objective was (i) to assess if a self-collected test for human papillomavirus (HPV) may serve as a primary cervical cancer screening method in a low-resource setting, (ii) to evaluate its implication in a screen and treat approach and (iii) to identify the most eligible age group in a screening program. Women were recruited through a cervical cancer screening campaign conducted in Cameroon. Written and oral instructions were given to participants by a health-care professional to carry out an unsupervised self-collected HPV-test (Self-HPV), followed by a physician-collected cervical sample for HPV testing (Physician-HPV) and cytology. Differences in performance between Self-HPV versus Physician-HPV and their ability to detect abnormal cytology results (ASC-US+) were evaluated. Descriptive analyses were used to examine the correlation between HPV positivity and cervical abnormalities by age. A sample of 789 women was prospectively enrolled. HPV prevalence was 14.6% and 12.7% for Self-HPV and Physician-HPV, respectively (Cohen's kappa = 0.74). HPV positivity by cytological diagnosis for ASC-US+ was similar with the two tests. positive predictive value of the Self-HPV for ASC-US+ was 20.4; odds ratio and number needed to treat were 6.5 (3.2-13.4) and 6 (4.2-10.9), respectively. We observed a trend of increasing cytological abnormalities in 30-49 year-old women and a concomitant trend of decreasing HPV prevalence supporting that this age group might be the most eligible group for screening. In conclusion, Self-HPV can be used as a primary screening test but needs to be followed by a triaging test that would identify the subset of women affected by clinically significant precancer or cancer.