Neutrophil to lymphocyte ratio and cancer prognosis: an umbrella review of systematic reviews and meta-analyses of observational studies

Abstract

Background: Although neutrophils have been linked to the progression of cancer, uncertainty exists around their association with cancer outcomes, depending on the site, outcome and treatments considered. We aimed to evaluate the strength and validity of evidence on the association between either the neutrophil to lymphocyte ratio (NLR) or tumour-associated neutrophils (TAN) and cancer prognosis.

Methods: We searched MEDLINE, Embase and Cochrane Database of Systematic Reviews from inception to 29 May 2020 for systematic reviews and meta-analyses of observational studies on neutrophil counts (here NLR or TAN) and specific cancer outcomes related to disease progression or survival. The available evidence was graded as strong, highly suggestive, suggestive, weak or uncertain through the application of pre-set GRADE criteria.

Results: A total of 204 meta-analyses from 86 studies investigating the association between either NLR or TAN and cancer outcomes met the criteria for inclusion. All but one meta-analyses found a hazard ratio (HR) which increased risk (HR > 1). We did not find sufficient meta-analyses to evaluate TAN and cancer outcomes (N = 9). When assessed for magnitude of effect, significance and bias related to heterogeneity and small study effects, 18 (9%) associations between NLR and outcomes in composite cancer endpoints (combined analysis), cancers treated with immunotherapy and some site specific cancers (urinary, nasopharyngeal, gastric, breast, endometrial, soft tissue sarcoma and hepatocellular cancers) were supported by strong evidence.

Conclusion: In total, 60 (29%) meta-analyses presented strong or highly suggestive evidence. Although the NLR and TAN hold clinical promise in their association with poor cancer prognosis, further research is required to provide robust evidence, assess causality and test clinical utility.

Trial registration: PROSPERO CRD42017069131 .

Keywords: Cancer; Neutrophil to lymphocyte ratio; Neutrophils; Prognosis; Tumour-associated neutrophils; Umbrella review.

Fig. 1

Flowchart of systematic review and meta-analysis selection

Fig. 2

Assessment of consistency in meta-analyses. a Log (HR) of largest study versus log (HR) of random effects estimates for each meta-analysis. The Y-axis labelled “log (HR) Largest Study” represents the log of the HR of the largest component study. The X-axis labelled “log (HR) Random Effects” represents the log of the HR of the random effects estimate calculated in each meta-analysis. b Random effects estimates versus inverse variance. The Y-axis labelled “Random Effects Estimates Hazard Ratio” represents the HR of random effects estimate for each meta-analysis. The X-axis labelled “Inverse Variance” represents the inverse of the variance for each meta-analysis. c, d Box plots of random effects HR estimates for each meta-analysis by cancer site and outcome. The Y-axis labelled “HR” details the effect size for each meta-analysis describing an association between NLR or TAN and cancer prognosis for each site grouping. The X-axis labelled “Site” in c represents each site group meta-analyses have been sorted into. The composite endpoints subgroup is defined as a grouping of cancer diagnosis unrelated to site, stage or treatment. The X-axis labelled “Outcome” in d represents the prognostic outcome assessed in each meta-analysis. The outlier of HR = 14 for NLR and OS in rectal cancer has been excluded from these figures