Inexpensive Systemic Inflammatory Biomarkers in Ovarian Cancer: An Umbrella Systematic Review of 17 Prognostic Meta-Analyses

The association of several inflammation-based biomarkers (lymphocyte-to-monocyte, neutrophil-to-lymphocyte, and platelet-to-lymphocyte ratios) with the survival of epithelial ovarian cancer (EOC) patients has been extensively investigated in several systematic reviews and meta-analyses (MAs) of observational studies. The aim of this umbrella systematic review is to appraise all available results in published MAs that explored the association between these biomarkers and EOC outcomes. An umbrella systematic review of the current evidence for systemic inflammatory biomarkers in the peripheral blood of EOC patients was performed by searching several databases including Pubmed/Medline and Web of Science. The quality of the MAs was appraised using the AMSTAR-2 tool as well as other qualitative criteria. The evidence was graded from convincing (class I) to weak (class IV). Our umbrella review appraised 17 MAs of retrospective studies (range: 7-16) with a number of enrolled patients ranging from 1636 to 4910 patients in each meta-analysis. All these MAs demonstrated that pretreatment high NLR and PLR, as well as low LMR, were independent predictors of poor overall survival and progression-free survival in EOC. Nearly all published MAs were conducted by Chinese researchers (16/17) and were redundant in their character. Another issue in these MAs is the absence of prior PROSPERO database registration as well as the earlier exclusion of the grey literature. On the other hand, PRISMA and MOOSE-based reporting guidelines were used in 9 out of the 17 MAs. A good number of MAs have transparently provided funding acknowledgment. The AMSTAR 2.0-based assessment showed low-quality in 11 out of the 17 reviewed MAs. This negative rating was largely due to the absence of critical domains. Finally, all evaluated MAs were rated as Class III or IV (suggestive and weak respectively). Despite the power of MAs in increasing sampling and precision, the quality of the current non-randomized evidence on this topic is still weak. Study registration: PROSPERO: CRD42020201493.