DNA high-risk HPV, mRNA HPV and P16 tests for diagnosis of anal cancer and precursor lesions: a systematic review and meta-analysis

BACKGROUND: Anal cancer prevention has two critical points: the incidence rate is several fold higher for some groups, such as people living with human immunodeficiency virus (HIV) and men who have sex with men (MSM), and there is not a well-defined guideline for its screening. This systematic review evaluates the accuracy of DNA HRHPV (high-risk human papillomavirus), mRNA HPV, DNA HPV16 isolated and p16 staining biomarkers in anal canal smears for identifying anal intraepithelial neoplasia (AIN) 2 or 3, summarised as anal high-grade squamous intraepithelial lesions (aHSIL), and cancer. METHODS: We searched the MEDLINE, Cochrane Library and Embase electronic databases as well as Grey literature to identify eligible papers published up to 31st July 2022. This systematic review and meta-analysis included observational studies comparing biomarker tests to histopathology after HRA (High-resolution Anoscopy) as a reference standard. We (ACM, TF) analysed studies in which patients of both sexes were screened for anal cancer using DNA HRHPV, mRNA HPV, DNA HPV16 and/or p16 biomarkers. The analysis was performed in pairs, for instance AIN2 or worse (AIN2+) vs. AIN1, HPV infection and normal (AIN1-). PROSPERO CRD42015024201. FINDINGS: We included 21 studies with 7445 patients. DNA HR HPV showed a higher sensitivity 92.4% (95% CI 84.2–96.5), specificity 41.7% (95% CI 33.9–44.9) and AUC 0.67, followed by the mRNA HPV test, with a sensitivity 77.3% (95% CI 73.2%–80.9%), specificity 61.9% (95% CI 56.6–66.9) and AUC 0.78. DNA HPV16 showed higher specificity 71.7% (95% CI 55.3–83.8), followed by p16 test, 64.1% (95% CI 51.0–75.4); Sensitivity of DNA HPV16 was 53.3% (95% CI 35.4–70.3) and AUC 0.69, while p16 had a sensitivity of 68.8% (95% CI 47.9–84.1) and AUC 0.74. Subgroup analysis of MSM with HIV, with 13 studies and 5123 patients, showed similar accuracy, with a bit higher sensitivities and lower specificities. Considering the measure of the total between-study variability, mRNA HPV tests showed the smallest area of the 95% prediction ellipse, 6.0%, influenced by the low logit sensitivity, 0.011. All other groups of tests exceed 50% prediction ellipse area, which represent a high heterogeneity. INTERPRETATION: Our findings suggested that DNA HR HPV can be a useful tool for screening for aHSIL and anal cancer if followed by biomarker with a higher specificity. As an isolated test, mRNA HPV had better performance. FUNDING: There was no funding source for this study.