Evaluating the performance of anal cytology and high-risk HPV genotyping for detecting anal HSIL in a clinic-based sample of people living with and without HIV in Puerto Rico.

Abstract

Given the disproportionately elevated anal cancer risk in high-risk populations, it is important to assess the performance of commonly used anal cancer screening tools to improve the effectiveness of detection and treatment methods. This study evaluates 1) the concordance between anal cytology and histology results and 2) the performance of cytology and high-risk human papillomavirus (HR-HPV) genotyping as screening tools for detecting histologically confirmed anal high-grade squamous intraepithelial lesions (HSIL). Data from the Anal Neoplasia Clinic in Puerto Rico (2014-2021; n = 466) was used. The clinical performance of anal cytology and HR-HPV genotyping to detect HSIL was compared to the gold standard: high-resolution anoscopy-guided biopsy. Sensitivity, specificity, positive predictive value, negative predictive value, and κ coefficients were calculated. A total of 66.95% of the patients were men, 74.0% were people living with HIV, 76.2% had anal HR-HPV infection, and 40.34% had histologically confirmed anal HSIL. The weighted κ statistic between the tests (cytology and histology) was 0.25 (p < .001). The sensitivity and specificity of cytology alone to detect anal HSIL were 84.3% (95% confidence interval [CI], 78.3%-89.1%) and 36.0% (95% CI, 30.3%-42.0%), respectively. Anal HR-HPV genotyping had higher sensitivity (92.2%; 95% CI, 87.4%-95.6%) and similar specificity (34.8%; 95% CI, 29.2%-40.7%) compared to cytology. The two tests combined (positive results following cytology or HR-HPV test) improved sensitivity to detect anal HSIL (97.9%; 95% CI, 94.8%-99.4%), but specificity was compromised (19.2%; 95% CI, 14.7%-24.4%). Although HR-HPV genotyping improved the detection of anal HSIL, HR-HPV testing had lower specificity than anal cytology alone.