Meta-analysis of prognostic factors of overall survival in patients undergoing oesophagectomy for oesophageal cancer

ABSTRACT Introduction Currently, the American Joint Commission on Cancer (AJCC) staging system is used for prognostication for oesophageal cancer. However, several prognostically important factors have been reported but not incorporated. This meta-analysis aimed to characterize the impact of preoperative, operative, and oncological factors on the prognosis of patients undergoing curative resection for oesophageal cancer. Methods This systematic review was performed according to PRISMA guidelines and eligible studies were identified through a search of PubMed, Scopus, and Cochrane CENTRAL databases up to 31 December 2018. A meta-analysis was conducted with the use of random-effects modeling to determine pooled univariable hazard ratios (HRs). The study was prospectively registered with the PROSPERO database (Registration: CRD42018157966). Results One-hundred and seventy-one articles including 73,629 patients were assessed quantitatively. Of the 122 factors associated with survival, 39 were significant on pooled analysis. Of these. the strongly associated prognostic factors were ‘pathological’ T stage (HR: 2.07, CI95%: 1.77–2.43, P < 0.001), ‘pathological’ N stage (HR: 2.24, CI95%: 1.95–2.59, P < 0.001), perineural invasion (HR: 1.54, CI95%: 1.36–1.74, P < 0.001), circumferential resection margin (HR: 2.17, CI95%: 1.82–2.59, P < 0.001), poor tumor grade (HR: 1.53, CI95%: 1.34–1.74, P < 0.001), and high neutrophil:lymphocyte ratio (HR: 1.47, CI95%: 1.30–1.66, P < 0.001). Conclusion Several tumor biological variables not included in the AJCC 8th edition classification can impact on overall survival. Incorporation and validation of these factors into prognostic models and next edition of the AJCC system will enable personalized approach to prognostication and treatment.