Challenges in evaluating high-grade pancreato-hepato-biliary neuroendocrine tumors

Abstract

We read with great interest and commend Tasneem et al for their valuable study on pancreato-hepatobiliary neuroendocrine tumors diagnosed via endoscopic ultrasound-guided biopsy. While the study offers important insights, we raise concerns regarding lesion classification, the lack of correlation between tumor size and aggressiveness, and the limited predictive value of individual clinical factors. Notably, many extra-pancreatic lesions may represent metastases rather than primary tumors. We advocate for a more comprehensive risk stratification approach and suggest incorporating novel molecular markers and analysis of the tumor microenvironment. Larger prospective studies are essential to enhance understanding and management of these rare and heterogeneous neoplasms.

Key Words: Pancreato-hepatobiliary neuroendocrine tumors; Endoscopic ultrasound; Malignant transformation; Pancreatic tumors; Screening

Core Tip: The pancreato-hepatobiliary neuroendocrine tumors are rare, heterogeneous neoplasms. Therefore, individual tumor prognosis prediction is essential. Although various clinicopathological factors influence the aggressiveness of tumors, no single factor can accurately predict prognosis. Hence, efforts towards the development of novel markers, such as identifying gene mutations and the tumor microenvironment in biopsy specimens, would have brought new research insights.

TO THE EDITOR

We would like to commend the efforts made by Tasneem et al[1] to study pancreato-hepatobiliary (PHB) neuroendocrine tumors (NETs) diagnosed through endoscopic ultrasound-guided biopsy and to identify the risk factors associated with high-grade PHB-NETs. We read the study by Tasneem et al[1] with great interest. PHB-NETs are among the rarest neoplasms arising from the pancreas, liver, and bile ducts. Several clinicopathological factors, including age, tumor stage, tumor grade, and tumor functionality, influence the aggressiveness of the tumor[2,3]. However, to date, there has been no single clinicopathological marker that can accurately predict the aggressiveness of PHB-NETs. Regarding this study, we would like to make a few pertinent remarks.

Firstly, lesions were divided into pancreatic and extra-pancreatic NETs. Extrapancreatic NET location is one of the predictors for high-grade NET. However, almost two-thirds of the extrapancreatic lesions were located in the porta hepatis and peri-gastric region, which were more likely to be metastatic lesions rather than the primary site of disease and may have been missed on cross-sectional imaging. Therefore, classifying the NET based on the location of the primary lesion rather than metastatic lesions would have been more rational. Secondly, it was surprising to find that there was no correlation between the size of the PHB-NET and the aggressiveness of the tumor. A clear relationship has been established between the size of the tumor and the risk of lymph node metastasis in previous studies, as listed in Table 1[4-8]. Thirdly, single clinical predictors have limited clinical utility and inadequate sensitivity and predictive efficacy in NETs. Therefore, it would have been more intriguing to combine multiple predictors and develop a scoring system or stratification scale to classify them into low, intermediate, and high-risk groups. Lastly, efforts towards the development of novel markers, such as identifying gene mutations and the tumor microenvironment in biopsy specimens, would have yielded new research insights. To conclude, PHB-NETs are relatively rare neoplasms, and individual clinicopathological factors have limited clinical utility. Therefore, extensive multicenter studies are needed to investigate the behaviour of these neoplasms. The development of nomograms and the performance of multi-analyte assays will help generate more robust risk factors for tumor aggressiveness.

Table 1 Association between tumor size and survival outcomes in published studies.

Ref.	Tumor size	Hazard ratio (95%CI)
Krogh et al[4], 2022	≥ 20 mm	3.1 (1.3-7.3)
Liu et al[5], 2019	> 40 mm	1.28 (1.116-1.474)
La Rosa et al[6], 2009	> 30 mm	3.45 (1.46-8.15)
Merath et al[7], 2018	≥ 30 mm	1.67 (1.11-2.51)
Ruzzenente et al[8], 2017	≥ 30 mm	1.01 (1.00-1.01)

CI: Confidence interval.