Clinical Benefit From The Traditional Chinese Medicine Combined With Chemotherapy For Non-small Cell Lung Cancer: A Meta-Analysis

Objective:1)To evaluate the clinical benefits from the Traditional Chinese Medicine (TCM) combined with chemotherapy for NSCLC.2) To probe into problems about the clinical researches of TCM combined with chemotherapy for NSCLC, to provide reference for regulatting TCM clinical research.Methods:Chinese and English databases:Chinese Medical Current Contents (CMCC), China National Knowledge Infrastructure (CNKI), Chinese Sci & Tech Journals (VIP), PubMed and the Cochrane Collaboration Controlled Trials Center registration database, were searched for RCTs, involving TCM combined with chemotherapy for NSCLC versus chemotherapy alone. Only trials with confirmed randomisation procedures and laparoscopic diagnosis of endometriosis, published during 1999-2009 and the Cochrane Collaboration randomized controlled trials what has not yet published the results were included. Literatures were selected with a strict criteria. Evaluated the quality of all the literatures in the standards of Jadad scale and found the problems of the clinical research. Data were analysed using standard Cochrane software Revman 4.2 and Stata v10.0 for a second test of the publication bias.Results:1)Forty-two trials involving 3524 participants met the inclusion criteria. Most of the trials (73.81%)were of poor quality with a considerable risk of bias, but eight(19.05%)of them had a good quality.Three trials (7.14%) used an incorrect random method.2) There was a significant improvement in clinical symptoms, when TCM combined with chemotherapy for NSCLC compared to chemotherapy alone (Z=8.11,P<0.00001).All the literatures had a good sensitivity and homogeneity (P=0.64), the publication bias were well controlled (P=0.083).In sub-group analysis, Chinese Patent medicine(PCm) group had a greater difference than others (RR 1.53,95% CI 1.18-1.99, P=0.001)VS (RR 1.68,95% CI 1.48-1.91,P<0.000),and came to a uncertain result,because of literature deficiencies. Other sub-groups had a good sensitivity and came to an authentic result.3)On account of publication bias, there was no evidence of a significant improvement in rates of 1-year survival, when TCM combined with chemotherapy for NSCLC compared to chemotherapy alone (Z=6.37,P<0.00001).All the literatures had a good sensitivity. Contains the negative result of the literatures, so they were heterogeneity (P=0.03)and publication bias. (P=0.013).In sub-group analysis, CPm group had a same Efficacy to chemotherapy alone group. Other sub-groups had a good sensitivity and came to a authentic result. 4) Because of publication bias, there was no evidence of a significant improvement in rate of objective response, when TCM combined with chemotherapy for NSCLC compared to chemotherapy alone (Z=5.85,P<0.00001).All the literatures had a good sensitivity and homogeneity (P=1.00).Contains the negative result of the literatures, so they were heterogeneity (P=0.03)and publication bias (P=0.013).In sub-group analysis, CPm group had a had a greater difference than others(RR 1.68,95%CI 1.10-2.56, P=0.02) VS (RR 1.40, 95% CI 1.25-1.56,P<0.000),and came to a uncertain result,because of literature deficiencies. Other sub-groups had a good sensitivity and came to an authentic result.5) There was a significant improvement in increased Karnofsky Performance Status (KPS),when TCM combined with chemotherapy for NSCLC compared to chemotherapy alone (Z=8.82,P<0.00001).All the literatures had a good sensitivity and homogeneity (P=0.62), the publication bias were well controlled (P=0.347). In sub-group analysis, CPm group had a greater difference than others (RR 1.98,95%CI 1.34-2.92,P<0.000) VS (RR 2.25,95% CI 1.88-2.69,P<0.000),and came to a uncertain result,because of literature deficiencies. Other sub-groups had a good sensitivity and came to an authentic result.6) There were forty-two literatures selected in this research. Among the four blind trials, one of them was double-blind study, and others were single-blind. Eight researchs used properly hidemethod, one trial unspecified its hidemethod, and others never used or used unproperly hidemethod. All data of the nine literatures, refered to drop out and lost to follow-up, were not intention-to-treat analysised. Only one literature among the one hundand and twenty-eight random experiments was intention-to-treat analysis and per-protocol analysis. According to the ethical requirement of the clinical study, the subjects of six literatures signed informed consent. Twenty-one literatures researched on syndrome classification criterion, but in inconsistency. Six of them never referred to the baseline between treatment and controlled groups.Only one study referred to the security and adverse event when took TCM.Conclusions:1)It came to an authentic result, that a significant improvement in clinical symptoms and increased KPS when TCM combined with chemotherapy for NSCLC compared to chemotherapy alone. On account of publication bias, there was no evidence of a significant improvement in rate of 1-year survival and objective response, and it's authenticity must be further confirmed.2) During the past decade, random allocation and blinding was not be appreciated in the clinical.researches about TCM combined with chemotherapy for NSCLC.Beyond that, most of the researchs used an inconsistency criterion about syndrome classification and therapeutic evaluation. To get a higher quality result, ethical requirement,the compliance,drop out and follow-up of the clinical study should be pay more attention.