Efficient designs for testing new agents and regimens.

Current clinical chemotherapy trial procedures in non-small cell lung cancer (NSCLC) require large numbers of patients and last up to 10 years before providing results that are often disappointing. Different designs are available that offer the advantages of smaller numbers of patients, shorter duration, and earlier identification of ineffective or highly active agents and regimens. An alternative method is proposed, combining phase II screening, randomized phase II studies, and multicenter, comparative, randomized phase III trials. Issues under discussion included the combined study in phase II of different disease stages and the need for recruitment to larger trials.