Prognostic factors for overall survival in patients with advanced digestive neuroendocrine carcinoma treated with first-line cisplatin-based chemotherapy: A post-hoc analysis of JCOG1213.

Abstract

594 Background: Advanced digestive neuroendocrine carcinoma (ADNEC) is a rare and aggressive malignancy without evidence of prognostic factors assessed in prospective studies. We aimed to evaluate prognostic factors in patients with ADNEC receiving first-line cisplatin-based chemotherapy. Methods: This is a post-hoc analysis of JCOG1213, which is a phase 3 randomized trial showing equivalent overall survival (OS) for cisplatin plus etoposide versus cisplatin plus irinotecan as first-line chemotherapy for patients with ADNEC. The primary endpoint of JCOG1213 was OS. For this post-hoc analysis of prognostic factors, patients who had ineligible histology based on the central pathological review or who lacked the necessary clinical data for analysis were excluded. A Cox proportional hazard model was used to assess the effect of independent variables on OS. Priori variables selected on previous reports (performance status (PS), Ki-67 index, and serum LDH level), as well as variables that remained at a level of p-value <0.20 by backward stepwise selection, were incorporated in the final model of OS. Results: Among 170 patients enrolled in JCOG1213, a total of 129 patients with ADNEC were included in this analysis. In multivariable analysis, serum LDH level (> upper normal limit vs. normal) was identified as a significant prognostic factor for OS (HR = 1.721, 95% CI: 1.144–2.589, p = 0.009). Other clinicopathological factors were not found to be significant, including PS (1 vs. 0; HR = 0.784, 95% CI: 0.507–1.214, p = 0.276), Ki-67 (≥ 55% vs. < 55%; HR = 1.475, 95% CI: 0.588–3.698, p = 0.407), liver metastasis (present vs. absent; HR = 1.434, 95% CI: 0.968–2.124, p = 0.072), sex (female vs. male; HR = 1.387, 95% CI: 0.937–2.053, p = 0.102), and serum ALP level (> upper normal limit vs. normal; HR = 0.948, 95% CI: 0.639–1.406, p = 0.789). OS of patients with elevated serum LDH levels was shorter than those with normal serum LDH levels (median: 9.5 months vs. 15.6 months, p = 0.0019). No statistically significant differences were observed between the two groups regarding objective response rate (58.8% vs. 54.1%, p = 0.594) and progression-free survival (median: 4.5 months vs. 5.9 months, p = 0.141). A numerically lower proportion of patients with elevated serum LDH levels received second-line chemotherapy than those with normal serum LDH levels (76.5% vs. 88.3%). Conclusions: Serum LDH level may serve as a prognostic factor in the daily practice of patients with ADNEC and as a risk stratification factor in future randomized clinical trials. Clinical trial information: UMIN000014795 .