Metastatic patterns and outcomes among early-onset versus average-onset gastroenteropancreatic neuroendocrine tumors (GEP-NETs): Insights from a multicenter database analysis.
Abstract
621 Background: The incidence of early-onset GEP-NETs (EO, <50 years) is rising, with data suggesting better outcomes compared to average-onset (AO, ≥50 years) GEP-NETs. While comorbidities and tumor biology may contribute to these differences, the role of metastatic burden and distribution of metastasis remains unclear. We sought to characterize and compare the clinical profiles of early and average onset GEP-NETs. Methods: We conducted a multicenter retrospective analysis using the TriNetX database, a federated de-identified EMR network. Adults (18–90 years) with GEP-NETs and at least one metastatic site between January 1, 2010, and August 5, 2025, were included. Metastatic sites were identified using ICD codes: lung (C78), lymph nodes (C77), brain (C79.3), retroperitoneum/peritoneum (C78.6), liver (C78.7), and bone (C79.5). EO-GEPNETs were defined as age 18–50 at diagnosis, AO-GEPNETs as ≥51. Propensity score matching (1:1 greedy nearest neighbor, caliper 0.1) was performed for sex, treatments (5-FU, octreotide, lanreotide, everolimus, cabozantinib, sunitinib, temozolomide, lutathera), surgeries (small/large bowel resection, liver resection/lobectomy), and primary site (small bowel, pancreatic, unspecified). Survival analysis was done using kaplan-meier survival estimate. The index event was metastatic disease. Baseline demographics, clinical features, and treatments were collected and compared using t-test statistics. Results: Among 5,438 patients, 961 (17.7%) had EO-GEPNETs. Compared with AO-GEPNETs, EO-GEPNETs had a higher proportion of females (54.5% vs 52.2%, p=0.05), more African Americans (17.2% vs 13.2%, p=0.002), more pancreatic primaries (5.3% vs 3.2%, p=0.002), and fewer small bowel primaries (38% vs 47.8%, p<0.0001). After matching, each cohort included 959 patients. Median overall survival (OS) in AO-GEPNETs was 88 months with 327 deaths (34.1%), whereas EO-GEPNETs did not reach median OS, with 245 deaths (25.5%), (HR 1.50, 95% CI 1.27–1.78; p<0.0001). Metastatic patterns showed modest differences: liver metastases were more frequent in EO-GEPNETs (53.6% vs 48.1%; RR 0.90, 95% CI 0.82–0.98, p=0.02), while bone (16.8% vs 13.8%, p=0.06), lymph node (25.3% vs 25.0%, p=0.88), and lung metastases (9.2% vs 10.7%, p=0.25) were comparable. Peritoneal metastases were slightly higher in EO-GEPNETs (20.3% vs 16.8%; RR 0.83, 95% CI 0.68–0.99, p=0.04). Conclusions: EO-GEPNETs display distinct demographics, greater pancreatic involvement, and modestly higher liver and peritoneal metastases, yet are associated with better overall survival than AO-GEPNETs. Differences in metastatic patterns alone may not account for these disparities, underscoring the need to elucidate biological underpinnings of age-related survival differences.
Article Details
Journal Info
Journal of Clinical Oncology
Lippincott Williams & Wilkins
Authors (7)
Deevyashali Parekh
2SUNY Upstate University, Department of Internal Medicine, Syracuse, United States
Suriya Baskar
1The Brooklyn Hospital Center, Brooklyn, United States
Timothy J. Brown
Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX
Nicholas James Hornstein
Northwell Health Cancer Center, New York, NY
Sarbajit Mukherjee
Department of Medicine, Roswell Park Comprehensive Cancer Center , Buffalo, NY,
Udhayvir Singh Grewal
Winship Cancer Institute of Emory University, Atlanta, GA
Deepak Vadehra
Department of Medicine, Roswell Park Comprehensive Cancer Center , Buffalo, NY,