Résumé
664 Background: The phase 3 NETTER-2 trial recently demonstrated a statistically significant improvement in progression-free survival (PFS) with [ 177 Lu]Lu-DOTA-TATE ( 177 Lu-DOTATATE) plus long-acting octreotide (Oct LAR) vs. high-dose Oct LAR in patients with G2/G3 gastroenteropancreatic neuroendocrine tumors in the 1L setting, across tumor grades and origins (including pNETs). However, no head-to-head comparison of 177 Lu-DOTATATE to other treatment options was conducted. Sunitinib is a currently recommended 1L treatment for patients with advanced G2 pNETs. The Raymond 2018 study is the only study that assessed sunitinib in advanced pNETs and provided sub-group data for the 1L setting (80% of the patients had G2 tumors). We aimed to determine the relative efficacy (PFS) of 177 Lu-DOTATATE + Oct LAR vs. sunitinib as 1L treatment for advanced G2 pNETs using MAIC. Methods: An unanchored MAIC (due to absence of a common comparator) of PFS in patients with advanced G2 pNETs treated in 1L with 177 Lu-DOTATATE + Oct LAR vs. sunitinib was conducted using individual patient data from NETTER-2 and aggregated data published in the Raymond 2018 study. Baseline characteristics and PFS data for the sunitinib arm were available for the treatment-naïve subgroup and were considered for the analysis. Baseline characteristics considered in the MAIC included age, sex, race, prior somatostatin analogue use, and number as well as location of involved disease sites. Multiple scenarios with various combinations of covariates were considered with multivariate Cox-regression of NETTER-2 used to identify covariates predictive of PFS. Finally, the combination of covariates with a reasonable effective sample size (ESS) was selected. Results: The final ESS was estimated to be 22 for 177 Lu-DOTATATE + Oct LAR, representing 43% of the original sample size (n=50), whereas sample size for sunitinib arm was 61 patients. After matching and weighting, patient characteristics were well balanced. MAIC analysis showed a statistically significant PFS benefit with 1L 177 Lu-DOTATATE + Oct LAR vs. sunitinib in G2 pNET (PFS hazard ratio: 0.23 [95% CI: 0.12–0.45]). Unadjusted comparison showed similar results (PFS hazard ratio: 0.34 [95% CI: 0.19–0.61]). Results were consistent across various scenarios assessed. Conclusions: Using MAIC to adjust for cross trial differences, PFS comparison favored 177 Lu-DOTATATE + Oct LAR over sunitinib as 1L treatment of patients with G2 pNET.