Evaluation of cytology versus human papillomavirus-based cervical cancer screening algorithms in Bhutan

Abstract

To evaluate the performance of existing versus alternative cervical cancer screening protocols in Bhutan, cervical exfoliated cells were collected for cytology and high-risk human papillomavirus (HR-HPV) testing among 1,048 women aged 30-69 years. Conventional smears were prepared and read locally. HR-HPV was tested by GP5+/6+ polymerase chain reaction, followed by genotyping and human DNA methylation analysis among HR-HPV-positives, in Europe. Test positivity was 7.5% for ASCUS or worse (ASCUS+) cytology and 14.0% for HR-HPV. All women with ASCUS+ and/or HR-HPV positivity (n=192) were recalled for colposcopy, among whom a total of 29 cases of histologically confirmed cervical intraepithelial neoplasia grade 2 or worse (CIN2+) were identified. An additional 7 CIN2+ cases were imputed among women without colposcopy. Corrected sensitivities for CIN2+ and CIN3+ were 61% and 74% for ASCUS+, 86% and 96% for HR-HPV, and 47% and 70% for ASCUS+ triage of HR-HPV. Specificity varied from 88% for HR-HPV up to 98% for ASCUS+ triage of HR-HPV, similarly for CIN2+ and CIN3+. Among HR-HPV-positive women with biopsies, methylation analysis offered similar discrimination of CIN2/3 and cervical cancer as ASCUS+, and better than HPV16/18 genotyping alone, but sample sizes were limited. In conclusion, the performance of cytology in Bhutan is in the mid-range of that reported in other screening settings. HR-HPV testing has the potential to improve detection of CIN2+, albeit with a higher referral rate for colposcopy. Cytological triage of HR-HPV-positives (performed in the absence of knowledge of HR-HPV status) reduced referral but missed more than one third of CIN2+.