FS benefit of Zykadia was consistent across various subgroups including age, gender, race, smoking class, ECOG performance status, and presence of brain metastases or prior response to crizotinib. The PFS benefit was further supported by local investigator assessment, and analysis of overall response rate (ORR) and disease control rate (DCR).
OS data was immature with 48 (41.7%) events in the ceritinib arm and 50 (43.1%) events in the chemotherapy arm, corresponding to approximately 50% of the required events for the final OS analysis. In addition, 81 patients (69.8%) in the chemotherapy arm received subsequent Zykadia as first antineoplastic therapy after study treatment discontinuation.
Efficacy data from Study A2303 are summarised in Table 4, and the Kaplan-Meier curves for PFS and OS are shown in Figure 3 and 4, respectively.
Table 4 ASCEND-5 (Study A2303) – Efficacy results in patients with previously treated ALK-positive metastatic/advanced NSCLC
Ceritinib
(N=115)
Chemotherapy
(N=116)
Duration of follow-up
Median (months) (min – max)
16.5
(2.8 – 30.9)
Progression-free survival (based on BIRC)
Number of events, n (%)
83 (72.2%)
89 (76.7%)
Median, months (95% CI)
5.4 (4.1, 6.9)
1.6 (1.4, 2.8)
HR (95% CI)a
0.49 (0.36, 0.67)
p-valueb
<0.001
Overall survivalc
Number of events, n (%)
48 (41.7%)
50 (43.1%)
Median, months (95% CI)
18.1 (13.4, 23.9)
20.1 (11.9, 25.1)
HR (95% CI)a
1.00 (0.67,1.49)
p-valueb
0.496
Tumour responses (based on BIRC)
Objective response rate (95% CI)
39.1% (30.2, 48.7)
6.9% (3.0, 13.1)
Duration of response
Number of responders
45
8
Median, monthsd (95% CI)
6.9 (5.4, 8.9)
8.3 (3.5, NE)
Event-free probability estimate at 9 monthsd (95% CI)
31.5% (16.7%, 47.3%)
45.7% (6.9%, 79.5%)
HR=hazard ratio; CI=confidence interval; BIRC=Blinded Independent Review Committee; NE=not estimable
a Based on the stratified Cox proportional hazards analysis.
b Based on the stratified log-rank test.
c OS analysis was not adjusted for the potentially confounding effects of cross over.
d Estimated using the Kaplan-Meier method.
Figure 3 ASCEND-5 (Study A2303) – Kaplan-Meier plot of progression-free survival as assessed by BIRC
Figure 4 ASCEND-5 (Study A2303) – Kaplan-Meier plot of overall survival by treatment arm
Patient reported outcome questionnaires were collected using the EORTC QLQ C30/LC13, LCSS and EQ-5D-5L. 75% or more of patients in the ceritinib and chemotherapy arms completed the LCSS questionnaires at most of the time points during the course of the study. Significant improvements were reported for the majority of lung cancer specific symptoms for Zykadia compared to chemotherapy (four out of six LCSS and 10 out of 12 QLQ-LC13 symptom scores). Ceritinib significantly prolonged time to deterioration for the lung cancer specific symptoms of interest of cough, pain and dyspnoea (composite endpoint LCSS: HR=0.40; 95% CI: 0.25, 0.65, median Time to Deterioration [TTD] 18.0 months [95% CI: 13.4, NE] in the ceritinib arm versus 4.4 months [95% CI: 1.6, 8.6] in the chemotherapy arm; LC13: HR=0.34; 95% CI: 0.22, 0. 52, median TTD 11.1 months [95% CI: 7.1, 14.2] in the ceritinib arm versus 2.1 months [95% CI: 1.0, 5.6] in the chemotherapy arm). The EQ-5D questionnaire showed a significant overall health status improvement for Zykadia in comparison to the |
