Abstract
Recent studies suggest that the treatment response and survival from head and neck tumours can be stratified according to biomarker status, particularly human papillomavirus (HPV) status and p16 expression, but the evidence for predictive biomarkers in anal squamous cell carcinoma (ASCC) remains limited. The aim of this study was to determine which biomarkers were associated with locoregional recurrence (LRR), overall survival and disease-free survival (DFS) in ASCC. A systematic search was undertaken of the MEDLINE, Embase, Cochrane Library, CINAHL and Web of Science databases using validated terms for ASCC, biomarkers and prognosis. Biomarkers were included in the meta-analysis if they were reported by at least four studies and provided sufficient data to permit the calculation of survival effect estimates. HPV status, p16, p53 and epidermal growth factor receptor (EGFR) met the inclusion criteria for meta-analysis and were reported by 17 retrospective cohort studies describing 1635 patients. When compared with HPV-negative tumours, HPV-positive tumours were associated with reduced LRR (pooled hazard ratio = 0.27 [95% confidence interval 0.16–0.48]; P < 0.001), improved overall survival (hazard ratio =0.26 [0.12–0.59]; P = 0.001) and DFS (hazard ratio = 0.33 [0.16–0.70]; P = 0.003). Likewise, p16-positive tumours were associated with reduced LRR (hazard ratio = 0.26 [0.13–0.52]; P < 0.001), improved overall survival (hazard ratio = 0.44 [0.24–0.81]; P = 0.009) and DFS (hazard ratio = 0.44 [0.23–0.83]; P = 0.012) when compared with p16-negative tumours. HPV-positive/p16-positive tumours had improved overall survival when compared with HPV-negative/p16-negative tumours (hazard ratio = 0.27 [0.15–0.48], P < 0.001), but not HPV-negative/p16-positive tumours (hazard ratio = 0.64 [0.21–1.90]; P = 0.421). p53 mutation was associated with worse DFS (hazard ratio = 1.63 [1.33–2.01]; P = 0.003). There was no association between EGFR status and any survival outcome. HPV status, p16 and p53 expression are of prognostic utility in ASCC. Future studies should prospectively validate these findings with a view to conducting subsequent randomised controlled trials where patients are stratified according to biomarker status and randomised to different treatment regimens.