Oxaliplatin- and temozolomide-based therapy in metastatic GEP-NETs: a monocentric comparison.

Background: Regimens based on oxaliplatin (OXA) and temozolomide (TEM) have demonstrated efficacy and are widely used in metastatic gastroenteropancreatic neuroendocrine tumors (GEP- NETs), a rare and heterogenous group of malignancies. However, their comparative effectiveness in randomized clinical trials and in real-world settings remains unexplored. We design the present study to provide real-world evidence on the performance of OXA- and TEM-based treatments sequencing in metastatic GEP-NETs. Methods: We conducted a retrospective study at the Veneto Institute of Oncology. Eligible patients (pts) had confirmed diagnosis of metastatic GEP-NETs, received at least one OXA- or TEM-based regimen between January 2008 and October 2025, had detailed clinicopathological data and adequate follow up. Data collected included demographic and pathological characteristics, treatments, and overall survival. The primary endpoint was overall survival (OS) estimated using the Kaplan–Meier method. The log-rank test and Cox proportional hazards model were used for univariate and multivariate analyses. To adjust for baseline characteristic imbalances, inverse probability weighting (IPW) with propensity score matching was applied. Results: Among 160 pts with well-differentiated neuroendocrine neoplasm treated with OXA or TEM, 81 (51%) had metastatic GEP-NETs. Of these, 60 pts (74%) received a TEM-based regimen first (TEM- PRE), while 21 pts (26%) started with an OXA-based regimen (OXA-PRE). Baseline characteristics were largely comparable between groups, with a median age of 62 years (55–71) in the OXA-PRE group and 61 years (54–69) in the TEM-PRE group (p = 0.4). Extra-pancreatic NETs were more frequent overall (62% in OXA-PRE vs 52% in TEM-PRE; p = 0.2) and carcinoid syndrome was present in 32% of patients. Notably, the OXA-PRE group had significantly higher Ki-67 proliferation index and tumor grade (p = 0.046 and p = 0.02, respectively). Overall, 24% of patients received both treatments, with chemotherapy most commonly administered in the second-line setting. Median PFS was similar among the two groups (TEM 11.31 vs 10.38, p=0.58). Median OS (mOS) was numerically longer in the TEM-PRE group compared with the OXA-PRE one (51.11 vs 37.98 mo; log-rank p=0.26) without reaching statistical significance. After IPW adjustment, 50 pts were evaluable: mOS in the TEM-PRE group was longer than in the OXA-PRE one (55.35 vs 37.98 mo), although the difference was not statistically significant (p=0.29). Therapy in the OXA-PRE group was associated with a significantly higher risk of death compared with TEM (HR 2.67, 95% CI 1.05–6.79, p=0.04). Among TEM-PRE pts, the presence of liver metastases (HR 1.68, p=0.45), FDG positivity (HR 1.93, p=0.45) and synchronous disease (HR 1.91, p=0.47) were associated with a non-significantly increased mortality risk, whereas pancreatic primary (HR 0.88, p=0.86) and Ki67 index ≥5% (HR 0.55, p=0.31) appeared protective, though not significantly. Conclusions: This exploratory analysis confirms in a monocentric cohort, the role of temozolomide-based regimens in patients with metastatic GEP-NETs, suggesting a potential advantage of a temozolomide-first approach vs oxaliplatin-first, warranting confirmation in larger and possibly biomarkers-oriented cohorts.