Pancreatic neuroendocrine tumors (pNETs) form a group of rare tumors, the incidence of which is increasing (0.27 to 1 case per 100,000 individuals from 2000 to 2016), and they have the particularity of being able to produce hormones. Despite the fact that these tumors frequently develop metastases, their relatively slow growth results in a prolonged survival rate (50% at 5 years). The NEMETABO study (led by Dr. Arnaud Jannin, CHU Lille) gathers multi-omics data from two cohorts of patients with pNET: a first cohort with clinical and transcriptomic data (n=63) and then a second cohort with clinical and metabolomic data (n=76), 8 patients of which are common between these two cohorts. Using the data from these two cohorts, we performed analyses integrating transcriptomic and metabolomic data using the RGCCA algorithm. These analyses were initially performed in an unsupervised manner, without any prior assumptions about potential groups, and then in a supervised manner based on certain factors of interest.