A landmark point analysis with cytotoxic agents for advanced NSCLC

As a result of recent publications, we hypothesized that period of 8 weeks after initiation of treatment is a useful landmark point for cytotoxic agents for advanced non-small cell lung cancer (NSCLC). To test this hypothesis, we conducted landmark analyses with clinical trials employing cytotoxic agents. Our goal was to assess the proper design of clinical trials with cytotoxic agents for NSCLC for maximizing patients' benefit.

We conducted landmark analyses of a phase II study of pemetrexed in locally advanced or metastatic NSCLC and a phase III study of Four-Arm Cooperative Study for advanced NSCLC. A total of 806 patients who received chemotherapy (pemetrexed, cisplatin and irinotecan, paclitaxel and carboplatin, cisplatin and gemcitabine, cisplatin and vinorelbine) were included in this assessment.

Tumor-shrinkage rate at 8 weeks was significantly associated with longer survival in the study with pemetrexed (p = 0.043), whereas tumor-shrinkage rate at 4 weeks did not correlated with survival (p = 0.139). Similarly, using the Four-Arm Cooperative Study data, the optimal landmark point was 8 weeks (p = 0.002), not 4 weeks (p = 0.190).

The landmark point for NSCLC was 8 weeks with all cytotoxic agents in our analysis when the therapy was given as a frontline or subsequent therapy. Our result suggests the concept of a disease-specific landmark point, which may lead to a change of phase II/III clinical study design to evaluate cytotoxic agents and clinical investigators, and their sponsors may consider an early look to assess the efficacy of cytotoxic agents for NSCLC.