Randomized international phase III trial of ERCC1 and RRM1 expression-based chemotherapy versus gemcitabine/carboplatin in advanced non-small-cell lung cancer.

We assessed whether chemotherapy selection based on in situ ERCC1 and RRM1 protein levels would improve survival in patients with advanced non-small-cell lung cancer (NSCLC). Eligible patients were randomly assigned 2:1 to the trial's experimental arm, which consisted of gemcitabine/carboplatin if RRM1 and ERCC1 were low, docetaxel/carboplatin if RRM1 was high and ERCC1 was low, gemcitabine/docetaxel if RRM1 was low and ERCC1 was high, and docetaxel/vinorelbine if both were high. In the control arm, patients received gemcitabine/carboplatin. The trial was powered for a 32% improvement in 6-month progression-free survival (PFS). Of 331 patients registered, 275 were eligible. The median number of cycles given was four in both arms. A tumor rebiopsy specifically for expression analysis was required in 17% of patients. The median time from informed consent to expression analysis was 11 days. We found no statistically significant differences between the experimental arm and the control arm in PFS (6.1 months v 6.9 months) or overall survival (11.0 months v 11.3 months). A subset analysis revealed that patients with low levels for both proteins who received the same treatment in both treatment arms had a statistically better PFS (P = .02) in the control arm (8.1 months) compared with the experimental arm (5.0 months). This demonstrates that protein expression analysis for therapeutic decision making is feasible in newly diagnosed patients with advanced-stage NSCLC. A tumor rebiopsy is safe, required in 17%, and acceptable to 89% (47 of 53) of patients.