Factors affecting survival in patients with colorectal cancer: an umbrella review

BACKGROUND: This study aimed to evaluate the strength and validity of the evidence for reported associations between multiple factors and colorectal cancer (CRC) survival. METHODS: We searched PubMed, Embase, Cochrane Database of Systematic Reviews from inception to 20 June 2024 for systematic reviews and meta-analyses of observational studies on factors and CRC prognosis. The quality of the methodology was evaluated using A Measurement Tool to Assess systematic Reviews 2 (AMSTAR-2). RESULTS: 90 meta-analyses investigated 191 associations (123 main evidence and 68 supplementary evidence) with CRC overall survival (OS). Factors were categorized into: demographic characteristic, perioperative condition, pathology of neoplasia, clinical laboratory indicators, comorbidities, behavior patterns and others. Totally, 6 associations were recommended as strong evidence (class I). While the presence of chemotherapy-induced neutropenia (CIN) (hazard ratio HR = 0.40, 95% confidence interval CI 0.32-0.51) denoted a better OS outcome, the existence of log odds of positive lymph nodes (LODDS) (HR = 3.49, 95% CI 2.88-4.23), liver cirrhosis (HR = 3.10, 95% CI 2.68-3.59) and the increment of metastatic lymph node ratio (mLNR) (HR = 1.62, 95% CI 1.39-1.88), pretreatment neutrophils-lymphocytes ratio (NLR) (HR = 2.17, 95% CI 1.83-2.58) and pretreatment lymph node ratio (LNR) (HR = 3.13, 95% CI 2.36-4.15) were considered as poor predictors of OS. Notably, 2 strong evidence was also found in supplementary evidence including extranodal tumor deposits (ENTDs) (HR = 1.72, 95% CI 1.46-2.04) and absolute monocyte count (AMC) (HR = 1.83, 95% CI 1.47-2.29). CONCLUSION: Beyond high-risk pathological signs of lymph node status and ENTDs, optimizable and cost-effective laboratory indicators, such as CIN, NLR, LNR and AMC, may effectively predict the prognosis of CRC. Furthermore, our study provides valuable insights for stratified CRC patient management, reinforcing the development of evidence-based prognostic survival strategies and recommendations. TRIAL REGISTRATION: This protocol was registered with PROSPERO under CRD42024560003.