This is a retrospective, monocentric study involving 50 patients with pancreatic neuroendocrine neoplasia resected between January 2008 and June 2020 at Paoli Calmettes Institute.

The primary objective of the study is to evaluate the grade concordance rate, based on Ki67 obtained on the pre-operative micro-biopsy and the surgical specimen.

Based on the histology slides obtained in the course of the treatment, several Ki67 recounts will be performed on pre-operative tumor micro-biopsies and on tumors resected after surgery:

• a manual count (on photo printed in the hotspot area according to World Health Organization (WHO) 2017 recommendations, by an expert pathologist and a junior pathologist.
• Automated counting using specific software based on artificial intelligence (Qpath software).

On the other hand, clinical, surgical and anatomopathological data will be collected in order to follow the patient evolution.

Inclusion Criteria:

• Patients with resection of pancreatic neuroendocrine neoplasia by surgery at Paoli Calmettes Institute between January 2008 and June 2020
• Patients who had a pre-surgical evaluation of Ki67 on micro-biopsy equipment obtained by echo-endoscopy.

Exclusion Criteria:

• Patient with a mixed tumor of the pancreas.
• Patient with a pancreas tumor composed of several nodules of similar size without a main mass and for which it is not possible to determine the biopsied nodule during echo-endoscopy.
• Initial micro-biopsy in a private pathology laboratory
• Initial subechoendoscopic puncture cytology only (Ki67 not possible)
• No initial micro-biopsy and surgery

Patients who had a pre-surgical evaluation of Ki67 on micro-biopsy equipment obtained by echo-endoscopy and a resection of pancreatic neuroendocrine neoplasia by surgery at Paoli Calmettes institute between January 2008 and June 2020