Background <p>The prognosis of gastroenteropancreatic neuroendocrine tumors (GEP-NETs) following metastasis is often poor. The efficacy of ¹⁷⁷Lu-DOTATATE therapy and the subgroups that benefit from it remain unclear. Our objective is to characterize the composition of the tumor immune microenvironment in GEP-NETs and to identify predictive biomarkers associated with response and PFS following ¹⁷⁷Lu-DOTATATE.</p> Methods <p>Multiplex immunofluorescence (mIF) staining of tumor sections was used to characterize the cellular density and spatial organization of the microenvironment of 50 NET patients. The relationship between baseline immune microenvironment and ¹⁷⁷Lu-DOTATATE efficacy or prognosis in 20 ¹⁷⁷Lu-DOTATATE-treated patients was explored.</p> Results <p>Patients with GEP-NET exhibited an immunosuppressive microenvironment. The overall response rate (ORR) to ¹⁷⁷Lu-DOTATATE therapy was 60%. Patients with a good response (partial response or complete response) had a lower density of regulatory T cells (Tregs) and CD8⁺TIM-3⁺ cells at baseline, with greater nearest neighbor distances between Tregs and CD8⁺ T cells, Tregs and CD11c⁺ dendritic cells (DCs), as well as Tregs and CD163⁺ M2 macrophages. Conversely, patients with a poor response (stable disease or progressive disease) had greater distances between CD8⁺ T cells and DCs, and more clustering pattern between CK⁺ tumor cells and CD8⁺TIM-3⁺ cells. Furthermore, a logistic regression-based predictive model for ¹⁷⁷Lu-DOTATATE efficacy was established, which maintained good discrimination after internal validation. Finally, the study revealed that Counts-(CD163)(CD11c) positively correlated with progression-free survival (PFS) post-¹⁷⁷Lu-DOTATATE.</p> Conclusions <p>The immune microenvironment parameters of GEP-NET patients are closely associated with ¹⁷⁷Lu-DOTATATE therapy efficacy. Our findings provide a potential tool for predicting ¹⁷⁷Lu-DOTATATE efficacy and patient selection, offering direction for precision clinical strategies and potential combination therapies in GEP-NET.</p>