Meta-analysis to Examine Effect of Amifostine on Survival in Radiotherapy-Treated Patients

VILLEJUIF, FRANCE- After more than 25 randomized trials, the effect of amifostine(Drug information on amifostine) (Ethyol) on the survival of patients treated with radiotherapy is still unknown. This is because most of the trials were neither designed nor powered to detect small survival differences and had followup periods that were either short (less than 2 years) or very short (a few months), according to Jean Bourhis, MD, PhD, of the departments of biostatistics and radiotherapy, Institut Gustave-Roussy, Villejuif, France. To assess whether this drug is associated with a survival benefit in radiotherapytreated patients, he and his col- leagues are conducting a metaanalysis of amifostine trials. Nonmetastatic Ca Only
Selection criteria include properly randomized trials performed between 1990 and June 2002, in which patients were randomized to receive radiotherapy or chemoradiotherapy plus amifostine, or to radiotherapy or chemoradiotherapy without amifostine. Patients had nonmetastatic head and neck, lung, or pelvic carcinomas. Updated data are being collected for all randomized patients from published and unpublished trials, and extensive checking and validation will be done to ensure the integrity of randomization and foloni low-up. Dr. Bourhis noted "pseudorandomization" might happen in some clinical trials. Tumor Control an Endpoint
The primary endpoint of the meta-analysis is the effect of amifostine on survival. The secondary endpoint is effect on tumor control. The meta-analysis will include more than 2,000 patients from 8 trials of non-small-cell lung cancer, 13 of head and neck cancer, and 7 of pelvic cancer. The investigators expect to report results in about 1 year. Dr. Bourhis's decision to include unpublished trials evoked some questions during the discussion period. "We are trying to avoid publication bias in favor of positive trials," he said. "We have to be careful not to avoid trials that have not been published because the results were not exactly as expected."