Rationale. The risk of anal cancer is markedly increased (40–130 fold) in HIV-infected subjects, especially among men who have sex with men (MSM) and in women with history of anogenital HPV disease. The current screening strategy is based on the detection of high degree squamous intraepithelial lesions (HSIL), a cancer precursor, using anal cytology. While this approach is highly sensitive, the specificity is poor, leading to an excess number of invasive procedures, explaining the poor implementation of this screening strategy. Emerging evidence supports that epithelial-adherent bacteria amplify HPV-associated cancers and contribute to impairment of mucosal immunity since the early stage of HIV-infection.
Hypothesis. The diagnostic value of the anal cytology for the diagnosis of HSIL can be improved by measuring bacterial biomarkers derived from anal cytologies.
Aims. Primary: to identify in HIV-infected MSM a set of anal-associated bacterial biomarkers to improve the accuracy of anal cytology for the diagnosis of biopsy-proven HSIL. Bacterial biomarkers may include bacterial species, proteins and metabolites. Secondary: to externally validate the diagnostic accuracy of the microbiota-based screening of HSIL in an external cohort of HIV-infected MSM and women. Third: to generate a model that allows linking metabolically active bacteria to proteins being actively expressed and subsequently to fluxes of metabolites being produced in the context of anal cancer in HIV patients and to identify potential targets for therapeutic interventions.
Methods. We will include 250 HIV-infected MSM undergoing screening for anal cancer by anal cytology and high-resolution anoscopy. We will extract the DNA/RNA from the anal bacteria and sequence the 16S ribosomal genes to infer the composition of the dead, dormant, inactive and active bacteria. Mass spectrometry will be used to examine proteins and metabolites produced by the active bacteria. Direct relationships between species abundance, protein synthesis and metabolic alterations will be then studied using integrative models and multivariate logistic regression to identify predictors of HSIL. We will internally validate the model using bootstrapping and perform external validation in two independent cohorts of 100 HIV-infected MSM and in 50 HIV-infected women.
Expected results and potential impact. We expect to improve the diagnosis of HSIL, to reduce the incidence of anal cancer, to reduce the harms and costs associated with the screening of anal cancer, and to gain insight into the potential mechanisms driven by microbiota fuelling anal cancer. To achieve this, a consortium of 2 clinical sites and 3 academic partners has been established.