Active surveillance in metastatic pancreatic neuroendocrine tumors: A 20-year single-institutional experience

BACKGROUND Pancreatic neuroendocrine tumors (PanNETs) are heterogeneous and indolent; systemic therapy is not essential for every patient with metastatic PanNET. The National Comprehensive Cancer Network guidelines state that delaying treatment is an option for PanNET with distant metastasis, if the patient has stable disease. However, specific factors that influence surveillance were not mentioned. In addition, data regarding the period of active surveillance in patients with metastatic PanNET are lacking. AIM To specifically determine factors influencing active surveillance in patients with liver metastatic nonfunctioning PanNETs (NF-PanNETs). METHODS Seventy-six patients with liver metastatic NF-PanNETs who received active surveillance from a high-volume institution were enrolled. Time to disease progression (TTP) and time to initiation of systemic therapy were determined. RESULTS Thirty-one (40.8%) patients had recurrent liver disease after R0 resection; 45 (59.2%) were diagnosed with liver metastasis. The median follow-up period was 42 mo and 90.7% patients were observed to have disease progression. The median TTP (mTTP) was 10 mo. Multivariate analysis showed that the largest axis of the liver metastasis > 5 mm (P = 0.04), non-resection of the primary tumor (P = 0.024), and T3-4 stage (P = 0.028) were associated with a shorter TTP. The mTTP in patients with no risk factors was 24 mo, which was significantly longer than that in patients with one (10 mo) or more (6 mo) risk factors (P < 0.001). A nomogram with three risk factors showed reasonable calibration, with a C-index of 0.603 (95% confidence interval: 0.47-0.74). CONCLUSION Active surveillance may only be safe for metastatic NF-PanNET patients with favorable risk factors, and other patients progressed rapidly without treatment. Further studies with a larger sample size and a control group are needed.