Use of quantitative endoscopic ultrasound elastography for diagnosis of pancreatic neuroendocrine tumors

Reliability of grading preoperative pancreatic neuroendocrine tumors on EUS specimens: a systematic review with meta-analysis of aggregate and individual data

BACKGROUND AND AIMS

Therapy and prognosis of pancreatic neuroendocrine tumors (PanNETs) are strictly related to the Ki-67 index, which defines tumor grading. The criterion standard for the assessment of grading of PanNETs is EUS-guided FNA (EUS-FBAFNA) or EUS-guided fine-needle biopsy sampling (EUS-FNB). Because data on diagnostic accuracy of EUS-FNA and EUS-FNB are heterogeneous, we aimed to analyze the variability in concordance between EUS grading and surgical grading.

METHODS

The MEDLINE, SCOPUS, and EMBASE databases were searched until November 2021 to identify studies reporting the concordance rate between EUS grading and surgical grading. The study was conducted in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Pooled events were calculated using a random-effects model and expressed in terms of pooled prevalence rates. A multivariate meta-regression was performed to find possible sources of heterogeneity. Where available, individual data were analyzed.

RESULTS

Twenty-six studies with 864 patients undergone EUS-FNA or EUS-FNB and surgical resection for PanNETs were included. The pooled estimate rate for the overall concordance of EUS grading and surgical grading was 80.3% (95% confidence interval, 75.6-85.1). Undergrading (EUS grading < surgical grading) was significantly more frequent with respect to overgrading (14.7% vs 3.5%, P < .001). Individual data analysis showed that among nonconcordant patients, the median Ki-67 difference was 3% (interquartile range, 2-6.15). The type of World Health Organization classification adopted and the median lesion diameter were significantly associated with heterogeneity at meta-regression.

CONCLUSIONS

EUS is an accurate technique in defining grading in patients with PanNETs, but a margin of error still exists, which should be the focus of future studies to minimize the risk of over- and/or undertreatment.

Role of Endoscopic Ultrasound Elastography Strain Histograms in the Evaluation of Patients with Pancreatic Masses

BACKGROUND

The Endoscopic Ultrasound (EUS) quantitative elastography strain ratio (SR) and strain histogram (SH) methods for noninvasive pancreatic masses differentiation have been recently developed. The aim of this research was to investigate the accuracy of the diagnostic differentiation methods for patients with pancreatic masses, based on the EUS SR and SH.

METHODS

This is a prospective study involving 100 cases with pancreatic masses. Patients were classified into 2 groups: group that was diagnosed with pancreatic malignancy with positive histopathology by biopsy obtained by fine-needle aspiration or postoperative pathology (72 patients) and the group diagnosed with pancreatitis with negative pathology and follow-up for at least 1 year (28 patients).

RESULTS

Based on the ROC curve, the cut-off point for Mode 1 was set at 97. Values under it showed the presence of malignant pancreatic masses. Mode 1 achieved a sensitivity of 89% and a specificity of 43% with an overall accuracy of 76%. The predictive positive value was 70%, and the predictive negative value was 60%. The cut-off point for SR was set at 3.04, and the values were equal or above the suggested pancreatic malignancy. The SR achieved a sensitivity of 95.83%, a specificity of 61%, with an overall accuracy of 86%. The predictive positive and negative values were 86.2% and 85%, respectively.

CONCLUSION

Mode 1 SH showed good sensitivity in the identification of pancreatic malignant tumors but were disappointingly of low specificity. Higher sensitivity, specificity, and overall accuracy were obtained by using the SR.

Automatic Image Selection Model Based on Machine Learning for Endobronchial Ultrasound Strain Elastography Videos

Background

Endoscopic ultrasound (EBUS) strain elastography can diagnose intrathoracic benign and malignant lymph nodes (LNs) by reflecting the relative stiffness of tissues. Due to strong subjectivity, it is difficult to give full play to the diagnostic efficiency of strain elastography. This study aims to use machine learning to automatically select high-quality and stable representative images from EBUS strain elastography videos.

Methods

LNs with qualified strain elastography videos from June 2019 to November 2019 were enrolled in the training and validation sets randomly at a quantity ratio of 3:1 to train an automatic image selection model using machine learning algorithm. The strain elastography videos in December 2019 were used as the test set, from which three representative images were selected for each LN by the model. Meanwhile, three experts and three trainees selected one representative image severally for each LN on the test set. Qualitative grading score and four quantitative methods were used to evaluate images above to assess the performance of the automatic image selection model.

Results

A total of 415 LNs were included in the training and validation sets and 91 LNs in the test set. Result of the qualitative grading score showed that there was no statistical difference between the three images selected by the machine learning model. Coefficient of variation (CV) values of the four quantitative methods in the machine learning group were all lower than the corresponding CV values in the expert and trainee groups, which demonstrated great stability of the machine learning model. Diagnostic performance analysis on the four quantitative methods showed that the diagnostic accuracies were range from 70.33% to 73.63% in the trainee group, 78.02% to 83.52% in the machine learning group, and 80.22% to 82.42% in the expert group. Moreover, there were no statistical differences in corresponding mean values of the four quantitative methods between the machine learning and expert groups (p >0.05).

Conclusion

The automatic image selection model established in this study can help select stable and high-quality representative images from EBUS strain elastography videos, which has great potential in the diagnosis of intrathoracic LNs.

Diagnostic value of endobronchial ultrasound image features: A specialized review

Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) technology is important in the diagnosis of intrathoracic benign and malignant lymph nodes (LNs). With the development of EBUS imaging technology, its role in noninvasive diagnosis, as a supplement to pathology diagnosis, has been given increasing attention in recent years. Many studies have explored qualitative and quantitative methods for the three EBUS modes, as well as a variety of multimodal analysis methods, to find the optimal method for the noninvasive diagnosis using EBUS for LNs. Here, we review and comment on the research methods and predictive diagnostic value, discuss the existing problems, and look ahead to the future application of EBUS imaging.

Diagnostic imaging of gastrointestinal neuroendocrine neoplasms with a focus on ultrasound

The diagnosis of gastrointestinal neuroendocrine neoplasms represents a significant diagnostic challenge since these tumours have a various, often non-specific clinical presentation. Currently, more than half of gastroenteropancreatic neuroendocrine neoplasms are detected incidentally, usually during surgery, diagnostic imaging studies or endoscopic procedures performed for other indications. Sometimes the first symptom of the disease is the presence of metastatic lesions in the liver. A neuroendocrine tumour is diagnosed based on the clinical presentation, assessment of specific and non-specific biochemical markers, imaging studies and histopathological examination. Focal lesions, both primary and metastatic may be small and often have an atypical location. Diagnostic imaging of neuroendocrine tumours is of fundamental importance for determining the location of the primary lesion, staging of the disease, selection of treatment and monitoring of its effects. In addition, diagnostic imaging make it possible not only to detect tumours, but also to perform therapeutic procedures based on the result. Transabdominal ultrasound is one of the first diagnostic imaging method for neuroendocrine neoplasms. New ultrasound techniques such as ultrasound elastography, contrast-enhanced ultrasound, endoscopic ultrasound, intraductal and intraoperative ultrasound improve the efficacy of ultrasound examination. Endoscopic ultrasound is a fundamental diagnostic tool for the detection of neuroendocrine tumours of the pancreas and the distal part of the colon. Due to the large variety of neuroendocrine tumours and differences in tumour biology, clinical stage and expression of somatostatin receptors, no single imaging method is sufficient; therefore, in order to determine the right diagnosis and select the best treatment, it is recommended that a combined morphological and functional assessment be used.

Real time endoscopic ultrasound elastography and strain ratio in the diagnosis of solid pancreatic lesions

AIM

To evaluate the accuracy of the elastography score combined to the strain ratio in the diagnosis of solid pancreatic lesions (SPL).

METHODS

A total of 172 patients with SPL identified by endoscopic ultrasound were enrolled in the study to evaluate the efficacy of elastography and strain ratio in differentiating malignant from benign lesions. The semi quantitative score of elastography was represented by the strain ratio method. Two areas were selected, area (A) representing the region of interest and area (B) representing the normal area. Area (B) was then divided by area (A). Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated by comparing diagnoses made by elastography, strain ratio and final diagnoses.

RESULTS

SPL were shown to be benign in 49 patients and malignant in 123 patients. Elastography alone had a sensitivity of 99%, a specificity of 63%, and an accuracy of 88%, a PPV of 87% and an NPV of 96%. The best cut-off level of strain ratio to obtain the maximal area under the curve was 7.8 with a sensitivity of 92%, specificity of 77%, PPV of 91%, NPV of 80% and an accuracy of 88%. Another estimated cut off strain ratio level of 3.8 had a higher sensitivity of 99% and NPV of 96%, but with less specificity, PPV and accuracy 53%, 84% and 86%, respectively. Adding both elastography to strain ratio resulted in a sensitivity of 98%, specificity of 77%, PPV of 91%, NPV of 95% and accuracy of 92% for the diagnosis of SPL.

CONCLUSION

Combining elastography to strain ratio increases the accuracy of the differentiation of benign from malignant SPL.