Automated evaluation of p16/Ki-67 dual stain cytology as a biomarker for detection of anal precancer in MSM living with HIV

Abstract Background Human papillomavirus (HPV)-related biomarkers such as p16/Ki-67 “dual stain” (DS) cytology have shown promising clinical performance for anal cancer screening. Here, we assessed the performance of automated evaluation of DS cytology to detect anal precancer in men who have sex with men (MSM) living with human immunodeficiency virus (HIV). Methods We conducted a cross-sectional analysis of 320 MSM with HIV undergoing anal cancer screening and high-resolution anoscopy (HRA) between 2009-2010. We evaluated the performance of automated DS based on a deep-learning classifier compared to manual DS evaluation to detect anal intraepithelial neoplasia (AIN) grades 2 and 3 (AIN2+) and AIN3. We evaluated different DS-positive cell thresholds quantified by the automated approach and modeled its performance compared to other screening strategies in a hypothetical population of MSM with HIV. Results Automated DS had significantly higher specificity (50.9% vs. 42.2%, p=0.0004) and similar sensitivity (93.2% vs. 92.1%) for detection of AIN2+ compared to manual DS cytology. HPV testing with automated DS triage was significantly more specific (56.5% vs. 50.9%, p=0.0003) with the same sensitivity (93.2%) compared to automated DS alone. In a modeled analysis assuming a 20% AIN2+ prevalence, automated DS detected the most precancers compared to manual DS and anal cytology (186,184, and 162, respectively) and had the lowest HRA referral per AIN2+ detected (3.1,3.5, and 3.3, respectively). Conclusions Compared to manual evaluation, automated DS cytology detects the same number of precancers with lower HRA referral.