Evaluation of the extent of lymphadenectomy in a randomized trial of Western- versus Japanese-type surgery in gastric cancer.

PURPOSE: In the context of a prospective, randomized trial of gastric cancer treatment, comparing Western surgical resection with limited lymphadenectomy (R1) versus Japanese surgical resection with extended lymphadenectomy (R2), we analyzed adherence to the specified surgical-pathologic guidelines. PATIENTS AND METHODS: Following evaluation of 389 patients, we quantified noncompliance (ie, performance of less dissection than specified) and contamination (ie, performance of more extensive dissection than specified). Of 389 patients, pathologic data permitted identification of 237 eligible patients treated with curative intent. RESULTS: Noncompliance occurred in 84% of R1 and R2 cases, with magnitude significantly (P < .001) higher in R2 cases versus R1 cases. Contamination occurred in 48% of R1 cases and 52% of R2 cases, with the magnitude of contamination moderate and equally distributed between the two groups. The contamination in R1 resections and the noncompliance in R2 resections lead to a partial homogenization of the groups, undermining the likelihood of detecting any potential therapeutic advantage to R2 dissection. CONCLUSION: The observed tendency to perform R1 resections combined with insufficient retrieval of lymph nodes underlines the need for increased surgical-pathologic standardization in this trial. Potential remedies are discussed. Proper conduct of clinical trials requires reliable means of standardizing performance of the surgical-pathologic team, an elusive but important goal.

Adjuvant therapy after curative resection for gastric cancer: meta-analysis of randomized trials.

PURPOSE: An overview is presented of reports published since 1980, in which postoperative adjuvant chemotherapy is compared with surgery alone for patients with gastric cancer. A MEDLINE literature review yielded 123 reports, 14 of which were relevant randomized trials; data from 11 of these trials were (or became) available for analysis of crude mortality odds. These 11 trials included 2,096 patients. METHODS: Odds ratios were calculated by comparing the adjuvant treatment arm with the observation-only arm. Those odds ratios that could be considered homogeneous yielded an estimated common odds ratio of 0.88 (95% confidence interval [CI], 0.78 to 1.08), which was slightly, but far from significantly, in support of adjuvant treatment. RESULTS: The results confirm the common opinion that the adjuvant chemotherapy regimens prescribed in these trials, although effective in phase II studies, do not improve survival. Furthermore they indicate that postoperative chemotherapy in general offers no additional survival benefit for patients with curatively resected gastric cancer. CONCLUSION: In conclusion, at present, postoperative chemotherapy cannot be considered as standard adjuvant treatment. New trials of adjuvant therapy for gastric cancer must include a no-treatment control arm.