| Objective:Meta-analysis was used to evaluate the clinical efficacy of palliative treatment of advanced non-small cell lung cancer with Chinese medicine,providing a reliable basis for clinical practice.Methods:Electrionic and manual searches were made both in the Chinese and English databases.Chinese database included: China National Knowledge Infrastructure(CNKI),Wanfang database,VIP database from the building to 2016.English database included: EMBASE(OVID),All EBM Reviews(OvidSP),MEDLINE(OVID),searching words: Herbal Medicine / Plants,Medicinal / medicine,oriental traditional,East Asian Traditional / Chinese herbal / lung neoplasms / Lung cancer or pulmonary cancer,pulmonary carcinoma / randomized controlled clinical trial.Inclusion criteria(1)Type of study: prospective randomized controlled trials(RCTs)published at home and abroad.Where the references in the paper refer to "random" are included.Including journals,conferences and dissertations,which are limited to Chinese and English.(2)subjects:non-small cell lung cancer patients with stage Ⅲ and Ⅳ confirmed by pathology / cytology.Age,sex,race and nationality are not limited.(3)interventions: the treatment group simply use traditional Chinese medicine decoction or Western symptomatic treatment plus traditional Chinese medicine decoction;control group only observed or Western symptomatic treatment.(4)final indicators: the main indicators(1)physical strength of the score(2)tumor stability rate(3)the efficiency of Chinese medicine symptom score improvement.Secondary indicators(1)survival rate:1 year survival rate,2-year survival rate.(2)weight stability rate.(3)changes in the score of TCM symptoms.(4)EORTC QLQ-C3 O,QLQ-LC13,FACT-L scale changes.(5)immune function.Exclusion criteria(1)non-randomized controlled clinical trials;(2)subjects included small cell lung cancer;(3)subjects included I and stage II NSCLC;(4)chinese medicine includes Chinese patent drug,Chinese medicine injection,acupuncture,moxibustion,acupoint application,acupoint injection and other Chinese medicine treatment methods;(5)the control group was given other anti-tumor treatment such as surgery,chemotherapy,immunotherapy,target therapy,radiation therapy,radiofrequency ablation,particle implantation,hyperthermia etc.(6)outcome indicators are not available(7)no corresponding outcome data.Meta-analysis was performed using the Revman 5.3 software provided by the Cochrane Collaboration Network.standardized mean difference(MD)was used as the effect of Measurement data,and relative risk(RR)was used as the effect of enumeration data.Each effect was expressed as 95% confidence interval(CI)The heterogeneity was assessed by using the chi-square test P and I2.If the heterogeneity test P > 0.1,suggesting that there was no significant heterogeneity among the studies,the fixed effect model was used for meta-analysis.If the heterogeneity test P ≦ 0.1,then used the random effect model for meta-analysis,and did the analysis of heterogeneity sources.The results of the meta-analysis are shown in the forest map.Sensitivity analysis was used to evaluate the stability of the analytical results.The effects of different statistical models on the results of the analysis were compared,and also it was performed using a method that was excluded from the study.Funnel map was used to determine if there was a publication bias.If the scatter plot on the funnel map showed a symmetrical distribution,it could be considered without publication bias;if the funnel map was asymmetric or incomplete,there was a greater likelihood of publication bias.And the TSA analysis of the main results determined whether the meta-analysis met the requirements of the sample size to ensure the reliability of the comprehensive analysis conclusion.Finally,the results of the analysis were discussed and summarized.Results:5191 articles were obtained from Chinese databases:2307 articles from CNKI,2138 articles from Wanfang data,746 articles from VIP data.441 articles from English databases: 340 articles from EMBASE,156 articles from MEDLINE,and130 articles from ALL EBM REVIEWS.Finally,18 domestic studies was included in the evaluation after reviewing the title,abstract,keyword and full text.18 RCTs were observed in 1070 patients with advanced non-small cell lung cancer.including 565 cases in the treatment groups and 565 cases in the control groups.According to Bias Risk Assessment Tool in the Cochrane 5.0.1 System Evaluation Manual,bias risk assessment was conducted on the quality of the18 studies,(1)the generation of random sequences.Among all the 18 studies,patients were reported randomized.11 articles described the correct randomization method which were evaluated as low risk.7 articles only reported that the patients were randomized,which were evaluated as unclear risk.(2)Allocation concealment.1 study reported using sealed envelope,which was evaluated as low risk.17 articles were evaluated as unclear risk.(3)Blinding.2 articles refered to the single-blind method,but did not report the implementation way.The other 16 articles did not mention whether they conducted blinding,which were are identified as unclear risk.(4)Blinding evaluation evaluation of the outcome,all were evaluated as unclear risk.(5)Loss of bias.All were rated as low risk.(6)Selective data.All were evaluated as unclear risk.(7)Other potential bias.All were evaluated as unclear risk.In summary,Bias Risk Assessment of all the 18 studies were unclear.Meta-analysis was performed on the 17 items: The improvement rate of objective efficacy of the tumor,the improvement rate of the KPS score,the KPS score,the improvement of TCM symptoms,the stability of body weight,the1-year survival rate,the 2-year survival rate,the rate of increase of NK cells,the score of cough,anhelation,bloody sputum,insomnia,fever,fatigue,chest pain,pain,and physical function score.11 aticles reported the stability of the objective efficacy of the tumor,including 665 patients.The result of meta-analysis showed that the tumor stability of chinese medicine group was better than that of the control group,RR=1.67,95%CI=(1.34,2.08),Z=4.54,P<0.00001.The result of the TSA showed that the cumulative Z-value curve passed through the traditional boundary line and also passed through the TSA boundary value curve,indicating that no more tests required and a positive conclusion can be drawn.10 articles reported improvement rates of KPS scores,including 592 patients.Meta-analysis showed that the improvement rate of KPS score of chinese medicine groups was better than that of the control groups RR = 2.15,95% CI =(1.72,269),Z = 6.68,P <0.00001.The results of the TSA showed that the cumulative Z-value curve passed through the traditional boundary line and also passed through the TSA boundary value curve,indicating that no more tests required and a positive conclusion can be drawn.5 articles reported KPS score,including 288 patients.The results of meta-analysis showed that the improvement of KPS scores of chinese medicine groups was better than that of the control group,MD=8.55,95%CI=(8.20,8.90),Z=47.87,P<0.000013 articles reported the improvement of traditional Chinese medicine symptoms,including 152 patients.Meta analysis showed that the effective rate of improvement of TCM symptoms in Chinese medicine treatment groups was higher than that in the control groups.RR= 4.50,95% CI =(2.45,8.25),Z = 4.86,P <0.000014 articles reported stability rate of body weight,including 262 patients.Meta analysis showed that the stability rate of body weight in chinese medicine groups was better than that in the control groups,RR = 1.41,95% CI =(1.18,1.68),Z = 3.84,P <0.01.2 articles reported the increase rate of NK cells in the traditional Chinese medicine group was better than that in the control group,including156 patients.RR = 2.33,95% CI =(1.48,3.66),Z = 3.68,P = 0.0002<0.014 articles reported 1-year survival rate,including 251 patients.Meta-analysis showed that the 1-year survival rate of the Chinese medicine groups was not better than that of the control groups,but the heterogeneity was large.The results of the sensitivity analysis showed that the stability and reliability were poor.After removal of the article of Liu Ling shuan,it showed the homogeneous.Meta-analysis showed the 1-year survival rate of the Chinese medicine groups was better than that of the control groups,.RR=2.35,95%CI=(1.37,4.05),Z=3.1,P=0.002<0.01.2 articles reported 2-year survival rate,including 124 patients.Meta-analysis showed that there was no significant difference in the 2-year survival rate between Chinese medicine groups and control groups.RR = 2.00,95%CI=(0.38,10.52),Z=0.82,P=0.41>0.052 articles reported changes in fever,including 124 patients.Meta analysis showed that the improvement of fever in chinese medicine groups was not better than that in the control group,MD =-0.11,95% CI =(-0.28,-0.05),Z = 1.33,P= 0.18>0.05.3 articles reported the change of fatigue score,including 164 patients.Meta analysis showed that the improvement of fatigue score in the Chinese medicine groups was better than that in the control groups,MD =-0.52,95% CI=(-0.77,-0.27),Z = 4.02,P <0.0001.3 articles reported the change of cough scores,including 164 patients.Meta analysis showed that the improvement of cough scores in chinese medicine groups was better than that of the control groups,MD =-0.38,95% CI =(-0.62,-0.13),Z = 3.05,P = 0.002<0.01.3 articles reported the change of anhelation scores,including 164 patients.Meta analysis showed that the improvement of anhelation in chinese medicine groups was better than that of the control groups,MD =-0.47,95% CI =(-0.73,-0.20),Z = 3.44,P = 0.0006<0.01.2 articles reported the change of blood sputum scores,including 124 patients.The results of meta-analysis showed that there was no difference between the two groups in the improvement of blood sputum scores.MD =-0.10,95% CI =(-0.58,0.37),Z = 0.42,P = 0.67>0.05.3 articles reported the change of chest pain scores,including 164 patients.Meta analysis showed that there was no difference between the two groups.MD=-0.13,95% CI =(-0.38,0.12),Z = 1.02,P = 0.31>0.052 articles reported the change of pain scores,including 121 patients.Meta analysis showed that was no difference between the two groups.MD =-0.96,95%CI =(-1.93,0.01),Z = 1.95,P = 0.05,2 articles reported the change of insomnia scores,including 124 patients.Meta analysis showed was no difference between the two groups.MD =-0.19,95%CI =(-0.54,0.16),Z = 1.05,P = 0.30>0.052 articles reported the change of physical function scores,including 124 patients.Meta analysis showed was no difference between the two groups.MD =0.70,95% CI =(0.02,1.43),Z = 1.90,P = 0.06>0.05Conclusion:Meta analysis and TSA analysis of 18 articles can prove that Chinese medicine decoction is effective in the palliative treatment of advanced non-small cell lung cancer.In general,Chinese medicine decoction treatment can stabilize the development of the tumor,improve the symptoms,improve the KPS score,improve the patients living quality.It can reduce some symptomsincluding cough,fatigue and anhelation.Besides,it helped to keep weight stable and also improve the 1-year survival rate.The effects might be related to improve the patient’s immune system.While the improvement of fever,bloody sputum,pain,insomnia,physical function score and 2-year survival rate were not statistically significant.The quality of the articles included in this study was not high,and with the small number of cases,the results were not sure.More researches are needed for further analysis.At the same time,due to the complexity of traditional Chinese medicine,it is difficult to carry out high-quality interpretation of RCT research.On the basis of this effective result,practical RCT researches can carry out to provide reliable evidences for clinical application in the future. |
