Chromogranin a and pancreatic polypeptide are not suitable for the screening of pancreatic neuroendocrine tumors in MEN1 - a long-term follow-up study.

Objective: In patients with multiple endocrine neoplasia type 1 (MEN1) followed up at ENETS centers of Excellence, chromogranin A (CgA) and pancreatic polypeptide (PP) are widely used screening tools for pancreatic neuroendocrine tumors (panNETs). Previous studies have demonstrated conflicting results regarding their performance in MEN1. This retrospective study aims to bring clarity to the question by investigating a well-characterized MEN1 cohort. We studied the impact of long-term biomarker follow-up on the clinical management of panNETs in MEN1.

Methods: We calculated the sensitivity and specificity and performed ROC analysis of CgA and PP for diagnosing any panNET, ≥20 mm panNET, and metastatic panNET in comparison to imaging reference standard in 58 MEN1 patients. All patients had undergone somatostatin receptor PET/CT and conventional imaging. Longitudinal impact of 10-year annual biomarker measurements on real-life clinical management was analyzed from patient records.

Results: Sensitivity of CgA (n = 48) and PP (n = 47) for diagnosing any panNET, ≥20 mm panNET, and metastatic panNET was 35%, 30%, and 60 and 23%, 33%, and 0%, respectively. For CgA, the AUC for diagnosing any panNET, ≥20 mm panNET, and metastatic panNET was 0.30 (95% CI 0.09-0.51), 0.49 (95% CI 0.29-0.68), and 0.69 (95% CI 0.42-0.95), respectively. For PP, the AUC for detection of metastatic panNET was 0.28 (95% CI 0.11-0.46). The annual biomarker measurements during 514 patient-years of follow-up did not affect the clinical management of panNETs.

Conclusions: CgA and PP are not helpful in diagnosing panNETs in MEN1. It is time to revise the surveillance protocols in practice.