Überlebenszeitanalyse bei nicht-kleinzelligem Lungenkarzinom - Meta-Analysen klinischer Studien zur Chemotherapie -

Lung cancer still continues to be a malignant tumor with comparatively poor prognosis 140 years after it was first documented by Langhans in Marburg [49]. Despite progress in therapy and diagnosis, the 5-year survival rate lies currently between 13 and 19% [154]. In most cases the patients are treated with chemotherapy. Besides research of new therapeutic concepts, optimization of existing concepts, especially of chemotherapy, is very important. Varying quality is common, even in randomized controlled trials. This dissertation deals with this particular problematic aspect and targets various related issues. A systematic literature analysis of four application areas of chemotherapy to non-small cell lung cancer (NSCLC) establishes a survey of the actual data. Each study was assessed for its methodic quality by using the Level of Evidence model of the Oxford Centre for Evidence-Based Medicine. The main goal was to establish four meta-analyses for the purpose of determining the total effect on survival rate in a comparison of studies of varying quality: 1. neoadjuvant chemotherapy, stage I to IIIA 2. adjuvant chemotherapy, stage I to IIIA 3. combination of gemcitabine and platinum compared to other combinations, stage IIIB and higher 4. combination of paclitaxel und platinum compared to other combinations, stage IIIB and higher Systematic research on studies dating from January 1980 until June 2011 was performed in the Medline and Cochrane Library. Title and abstract were examined on the basis of predefined criteria for inclusion. Ineligible studies were excluded. The quality of methodology was evaluated. The relevant data of each study were extracted by two scientists who worked independently of each other. If both assessments differed, the researchers came to a consensus with the help of a third party. Meta-analysis were calculated by ReviewManager. The Hazard Ratio (HR) was used to measure the benchmark of overall survival. Only those studies which passed the evaluation criteria as being level 1b and 2b were included. In addition to subgroup analyses of stages, therapy combinations and current studies since 2004, studies of various quality were compared. Eight studies (involving 1, 849 patients) were included in the meta-analysis of neoadjuvant chemotherapy. Among these eight studies, four are level 1. Overall an HR of 0.85 (95%-CI 0.75 - 0.96) was calculated. Level 1 studies were measured at a HR of 0.87 (95%-CI 0.74 - 1.03). All studies taken together showed a significant survival advantage for neoadjuvant chemotherapy. Level 1b studies did not confirm this result. The meta-analysis of adjuvant chemotherapy included 30 studies (involving 10, 403 patients). The pooled HR was 0.88 (95%-CI 0.83 - 0.93). Similar values were calculated in analyses of 13 studies (6, 659 patients) in level 1 (HR 0.88 [95%- CI 0.82 - 0.95]). Significant survival benefit was calculated in subgroup analysis of stage I, II, III, each viewed in light of all studies. There was no significant survival benefit in analysis of stage IB and of high quality studies in stage I. These results confirm the significant survival benefit of adjuvant chemotherapy save in stages IB and I. Because of insufficient data, an analysis of stage IA was not practicable. The combination of gemcitabine with platinum, in comparison with other treatment combinations, yielded a slight significant survival rate (HR 1.05 [95%-CI 0.99 - 1.11]). This meta-analysis covered 34 studies involving 12, 468 patients. The analysis of level 1 studies did not indicate a relevant difference between the experimental and control group. The analysis of the subgroups of vinca alkaloidbased, paclitaxel-based, platinum-based combinations or newer studies since 2004 showed no significant survival rates compared to the rate for control group. The comparison of the paclitaxel with platinum to other combinations of chemotherapy, both in overall analysis and subgroup analysis, indicated no survival differential. This meta-analysis of 22 studies (7, 571 patients) yielded a HR of 1.02 (95%-CI 0.97 - 1.07). The HR result from the analysis of twelve level 1b studies (3, 844 patients) was 1.01 (95%-CI 0.93 - 1.09). CONCLUSION: In majority, there was no relevant difference shown in the result of comparing studies in various quality. The results of meta-analyses of adjuvant and neoadjuvant chemotherapy in different stages of NSCLC confirmed existing results. Most of the analysis of treatment of the inoperable NSCLC in combination with gemcitabine and platinum or paclitaxel and platinum presented no cause to a significant survival benefit, compared to other combinations.