A comparison of human papillomavirus testing of clinician-collected and self-collected samples during follow-up after screen-and-treat.

Screen-and-treat cervical cancer prevention programs based on high-risk human papillomavirus (HPV) testing and cryotherapy have been shown to be effective in resource-limited settings. However, because cryotherapy is not 100% effective, follow-up is needed after treatment to detect post-treatment failures. We compared the test performances of high-risk HPV testing (Hybrid Capture 2) using self-collected and clinician-collected samples as well as cervical cytology for identifying cervical intraepithelial neoplasia grades 2 or 3 or invasive cervical cancer (CIN2+) among women who did (n=812) and did not (n=1858) undergo cryotherapy in a South African screen-and-treat trial. At 6 months after enrolment (and after cryotherapy, if performed), women were tested using all three screening methods and then underwent colposcopy/biopsy. A predefined subset of women (n=1,455) had extended follow-up with colposcopy/biopsy at 12 months. A total of 33 and 91 cases of CIN2+ were detected among treated and untreated women, respectively. The sensitivity of HPV testing using clinician-collected samples and cervical cytology did not differ by treatment status. HPV testing of clinician-collected samples detected the most cases of CIN2+ among both treated (85%) and untreated (91%) women (p=0.31). Cytology (at a cutoff of atypical squamous cells of undetermined significance or greater) detected 76% of cases among both treated and untreated women. However, the sensitivity of HPV testing using self-collected samples was significantly lower among treated versus untreated women (55% vs. 78%, p=0.01). HPV testing using self-collected vaginal specimens may be useful in primary screening but performs poorly for detecting post-treatment failures.