This randomized, open-label, phase III study enrolled 221 patients with pathological Stage I-IIIA and completely resected high-grade neuroendocrine carcinoma (HGNEC) of the lung. Patients were assigned to receive either irinotecan plus cisplatin (IP) or etoposide plus cisplatin (EP). The primary outcome was relapse-free survival (RFS) with a 5-year follow-up.
At 5 years, RFS was 65.7% in the EP arm and 65.2% in the IP arm. The hazard ratio for RFS was 1.026 (95% CI, 0.670-1.569), indicating no significant difference. Overall survival at 5 years was 73.5% with EP and 72.4% with IP, with a hazard ratio of 1.175 (95% CI, 0.742-1.861), also not significant.
Safety and tolerability data were not reported. A key limitation was the concordance proportion of pathological diagnoses between institutions and central reviews, which was 75.6% (95% CI, 69.4%-81.1%).
The study suggests no clear advantage for IP over EP in this setting. Practice relevance was not reported, and the findings should be considered in the context of the reported limitations.
View Original Abstract ↓
INTRODUCTION: We aimed to evaluate the long-term follow-up overall survival (OS) of irinotecan plus cisplatin (IP) versus etoposide plus cisplatin (EP) as postoperative adjuvant chemotherapy in patients with pathological Stage I-IIIA high-grade neuroendocrine carcinoma (HGNEC) of the lung.
METHODS: The JCOG1205/1206 randomized, open-label, phase III study compared the efficacy of IP and EP as adjuvant chemotherapy. Patients with pathological Stage I-IIIA and completely resected HGNEC of the lung were randomized to receive either the IP or EP arm. The primary endpoint was relapse-free survival (RFS), and the secondary endpoint included overall survival (OS). The analyses were performed using data from 5 years after the last patient enrollment. In addition, a central pathological review was planned for this study.
RESULTS: Between April 2013 and October 2018, 221 patients were enrolled (EP arm, 111 patients; IP arm, 110 patients). Updated respective 3- and 5-year RFS rates were 68.5% and 65.7% in the EP arm versus 71.8% and 65.2% in the IP arm, with a hazard ratio (HR) of 1.026 (95% confidence interval [CI], 0.670-1.569). OS at 3 and 5 years was 85.6% and 73.5%, respectively, in the EP arm versus 83.6% and 72.4%, respectively, in the IP arm (HR, 1.175; 95% CI, 0.742-1.861). The concordance proportion of pathological diagnoses between each institution and central pathological reviews was 75.6% (95% CI, 69.4%-81.1%).
CONCLUSIONS: Our results showed no significant difference in both updated RFS and OS between the two arms for patients with completely resected HGNEC.