Daniel Shao-Weng Tan1, António Araújo2, Jie Zhang3, James Signorovitch4, Zheng-Yi Zhou5, Xiaopeng Cai5, Geoffrey Liu6. 1. Division of Medical Oncology, National Cancer Centre Singapore, Republic of Singapore. Electronic address: daniel.tan.s.w@nccs.com.sg. 2. Central Hospital of Porto, University of Porto, Porto, Portugal. 3. Novartis Pharmaceuticals Corporation, East Hanover, New Jersey. 4. Analysis Group, Inc., Boston, Massachusetts. 5. Analysis Group, Inc., New York, New York. 6. Princess Margaret Cancer Centre, Toronto, Ontario, Canada.
Abstract
INTRODUCTION: Crizotinib and ceritinib have been developed to treat advanced or metastatic NSCLC by inhibiting anaplastic lymphoma receptor tyrosine kinase gene (ALK). No randomized trial has compared these treatments head-to-head. We compared efficacy outcomes between patients receiving ceritinib and an external control group receiving crizotinib, both as initial ALK-targeted therapies for previously treated advanced or metastatic ALK-positive NSCLC. METHODS: Individual patient data for the ceritinib-treated patients were drawn from two single-arm trials (ASCEND-1 and ASCEND-3); published summary data for the crizotinib-treated patients were extracted from three trials (PROFILE 1001, PROFILE 1005, and PROFILE 1007). To adjust for cross-trial differences, average baseline characteristics were matched using propensity score weighting. Overall survival (OS), progression-free survival (PFS), and overall response rate were then compared between treatment groups. RESULTS: Before matching, the ceritinib-treated patients (n = 189) were significantly different from the crizotinib-treated patients (n = 557) in the distribution of race and number of prior regimens. After matching, all available baseline characteristics were balanced. Compared with crizotinib, ceritinib was associated with longer OS (hazard ratio = 0.59, 95% confidence interval: 0.46-0.75) and longer PFS (median 13.8 versus 8.3 months, hazard ratio = 0.52, 95% confidence interval: 0.44-0.62) in Cox proportional hazards models. The 12-month OS was 82.6% with ceritinib and 66.0% with crizotinib in a Kaplan-Meier analysis (log-rank p < 0.001). There was no significant difference in overall response rate between ceritinib and crizotinib. CONCLUSIONS: In an adjusted comparison across separate clinical trials, ceritinib was associated with prolonged OS and PFS compared with crizotinib when used as initial ALK-targeted therapy for previously treated ALK-positive NSCLC.
RCT Entities:
INTRODUCTION:Crizotinib and ceritinib have been developed to treat advanced or metastatic NSCLC by inhibiting anaplastic lymphoma receptor tyrosine kinase gene (ALK). No randomized trial has compared these treatments head-to-head. We compared efficacy outcomes between patients receiving ceritinib and an external control group receiving crizotinib, both as initial ALK-targeted therapies for previously treated advanced or metastatic ALK-positive NSCLC. METHODS: Individual patient data for the ceritinib-treated patients were drawn from two single-arm trials (ASCEND-1 and ASCEND-3); published summary data for the crizotinib-treated patients were extracted from three trials (PROFILE 1001, PROFILE 1005, and PROFILE 1007). To adjust for cross-trial differences, average baseline characteristics were matched using propensity score weighting. Overall survival (OS), progression-free survival (PFS), and overall response rate were then compared between treatment groups. RESULTS: Before matching, the ceritinib-treated patients (n = 189) were significantly different from the crizotinib-treated patients (n = 557) in the distribution of race and number of prior regimens. After matching, all available baseline characteristics were balanced. Compared with crizotinib, ceritinib was associated with longer OS (hazard ratio = 0.59, 95% confidence interval: 0.46-0.75) and longer PFS (median 13.8 versus 8.3 months, hazard ratio = 0.52, 95% confidence interval: 0.44-0.62) in Cox proportional hazards models. The 12-month OS was 82.6% with ceritinib and 66.0% with crizotinib in a Kaplan-Meier analysis (log-rank p < 0.001). There was no significant difference in overall response rate between ceritinib and crizotinib. CONCLUSIONS: In an adjusted comparison across separate clinical trials, ceritinib was associated with prolonged OS and PFS compared with crizotinib when used as initial ALK-targeted therapy for previously treated ALK-positive NSCLC.
Authors: Darren S Thomas; Aaron Y Lee; Philipp L Müller; Roy Schwartz; Abraham Olvera-Barrios; Alasdair N Warwick; Praven J Patel; Tjebo F C Heeren; Catherine Egan; Paul Taylor; Adnan Tufail Journal: Clin Transl Sci Date: 2021-03-02 Impact factor: 4.689