Meta-analysis Of Platelet Lymphocyte Ratio As A Prognostic Factor For Non-small Cell Lung Cancer

Objective:The present study showed platelet lymphocyte ratio?platelet to lymphocyte ratio,PLR?can be used as a prognostic factor to predict certain cancers,Such as colon cancer and hepatocellular carcinoma,etc.However,the prognostic value of PLR in patients with non-small-cell lung cancer?NSCLC?remains controversial.Therefore,We evaluated the prognostic value of PLR in patients with NSCLC through Meta-analysis.Methods:Literature search for PubMed,EMBASE,Web of Science,Medline,Cochrane Library,China National Knowledge Internet?CNKI?,China Biomedical Medicine disc?CBMdisc?,VIP,Wanfang Database using computer electronic system to study the association between PLR and overall survival?OS?and disease-free survival?DFS?,inclusion of criteria and quality assessment studies,the search time is from the establishment of the database to November 2018.OS refers to the time from the start of treatment in random groups to the time of death due to any cause or the time of the last follow-up before the loss of follow-up.DFS refers to the time from the start of treatment in random groups to the time of recurrence or death due to disease progression.Each eligible study data was extracted and all included analyses used Revman5.3 and Stata12.0 statistical software.Hazard risk?HR?and 95%confidence interval?95%CI?in each study were combined to evaluate the prognostic value of PLR in NSCLC patients.Heterogeneity between different studies was analyzed using Cochran's Q statistic test.Heterogeneity was acceptable as long as I² less than or equal to 50%,P>0.10.When no significant heterogeneity existed among studies,the fxed-effects model was used to generate the HRs and 95%CIs,Otherwise,the random-effects model was applied.Subgroup analysis and Meta regression were performed to discuss the source of heterogeneity.A combined HR>1 indicated a worse survival,and it was considered statistically signifcant if the 95%CI did not overlap with the null value of one.Sensitivity analysis was also required to assess the stability of the results,Begg method and Egger method were used for quantitative analysis of publication bias.All statistical tests were two sided,and the signifcance level was set at 5%.Results:We included a total of 15 research literatures involving 5524 patients for meta-analysis.According to the results of the meta-analysis:The OS of the higher PLR group was significantly lower than that of the lower PLR group(HR:1.69,95%CI:1.45-1.97,P<0.00001,I²=46.2%,P _{heterogeneity}=0.026);the DFS of the higher PLR group was significantly lower than that of the lower PLR group(HR:1.41,95%CI:1.14-1.74,P=0.001,I²=0%,P _{heterogeneity}=0.462).Through the analysis of I² value and P _{heterogeneity} value,there was no obvious heterogeneity in the literatures included in DFS-related NSCLC patients,but there was moderate heterogeneity in the literatures included in OS-related NSCLC patients.Subsequently,subgroup analysis,Meta regression analysis and sensitivity analysis were conducted to further explore the source of heterogeneity.Subgroup analyses by ethnicity indicated that high PLR predicted poor prognosis for patients both in Caucasian populations(HR:1.75,95%CI:1.29-2.38,P=0.0003,I²=25.1%,P _{heterogeneity}=0.254)and in east Asian populations(HR:1.68,95%CI:1.39-2.01,P<0.00001,I²=55.3%,P _{heterogeneity}=0.017).When grouping according to PLR cutoff value>162 and PLR cutoff value less than or equal to 162,the results show that the combined HRs were 1.78(95%CI:1.44-2.21,P<0.00001,I²=36%,P _{heterogeneity}=0.153)and 1.61(95%CI:1.27-2.03,P<0.0001,I²=55.7%,P _{heterogeneity}=0.027),When grouped by sample size,>202 patients(HR:1.58,95%CI:1.33-1.88,P<0.00001,I²=41.9%,P _{heterogeneity}=0.099)and less than or equal to 202 patients?HR:1.85,95%CI:1.39-2.48,P<0.0001,I²=47.5%,P heterogeneity=0.076?.When the patients were grouped according to treatment methods,4470 cases of surgical treatment(HR:1.77,95%CI:1.43-2.18,P<0.00001,I²=51.4%,P _{heterogeneity}=0.044)and 909cases of non-surgical treatment(HR:1.63,95%CI:1.20-2.23,P=0.002,I²=56.2%,P _{heterogeneity}=0.044)were obtained.To sum up,OS in the higher PLR group was significantly lower than that in the lower PLR group?P<0.05?after grouping by ethnicity,sample size,PLR cutoff value and treatment.Meta regression analysis concluded that ethnicity,sample size,PLR cutoff value and treatment may not be the main sources of inter-study heterogeneity.Sensitivity analysis showed that the combined HRs in this Meta analysis had good stability.Finally,the Egger method and Begg method were used for quantitative analysis of publication bias and it was concluded that there was no obvious publication bias.Conclusion:Through the analysis of this paper,increased PLR is associated with poor prognosis in NSCLC,the OS of the higher PLR group was significantly lower than that of the lower PLR group(HR:1.69,95%CI:1.45-1.97,P<0.00001,I²=46.2%,P _{heterogeneity}=0.026),the DFS of the higher PLR group was significantly lower than that of the lower PLR group(HR:1.41,95%CI:1.14-1.74,P=0.001,I²=0%,P _{heterogeneity}=0.462).So PLR may be an important biological predictive marker for NSCLC patients,however,its clinical application still needs to be verified through more research in the future.