Establishment of a multi-parameters MRI model for predicting small lymph nodes metastases (<10 mm) in patients with resected pancreatic ductal adenocarcinoma

Preoperative carbohydrate antigen 19.9 level predicts lymph node metastasis in resectable adenocarcinoma of the head of the pancreas: a further plea for biological resectability criteria

INTRODUCTION

Lymph-nodal involvement (N+) represents an adverse prognostic factor after pancreatoduodenectomy (PD) for pancreatic adenocarcinoma (PDAC). Preoperative diagnostic and staging modalities lack sensitivity for identifying N+. This study aimed to investigate preoperative carbohydrate antigen 19.9 (CA 19.9) in predicting the N+ stage in resectable-PDAC (R-PDAC).

METHODS

Patients included in a multi-institutional retrospective database of PDs performed for R-PDAC from January 2000 to June 2021 were analysed. A preoperative laboratory value of CA 19.9 greater than 37 U/l was used in univariate and multivariate logistic regression analysis to determine a possible association with N+. Additionally, different cut-offs of CA 19.9 related to the preoperative clinical T (cT) stage was assessed to evaluate the risk of N+.

RESULTS

A total of 2034 PDs from thirteen centres were included in the study. CA 19.9 greater than 37 U/l was significantly associated with higher N+ at univariate and multivariate analysis ( P <0.001). CA 19.9 levels greater than 37 U/l were associated with N+ in 75.9%, 81.3%, and 85.7% of patients, respectively, in cT1, cT2, and cT3 tumours and with higher cut-off values for all cT stages.

CONCLUSION

Lymph-nodal involvement is strongly related to preoperative CA 19.9 levels. Specially in patients staged as cT3 the CA 19.9 could represent a valid and easy tool to suspect nodal involvement. Due to these findings, R-PDAC patients with elevated CA 19.9 values should be considered in a more biologically advanced stage.

Histogram array and convolutional neural network of DWI for differentiating pancreatic ductal adenocarcinomas from solid pseudopapillary neoplasms and neuroendocrine neoplasms

PURPOSE

This study aimed to investigate the diagnostic performance of the histogram array and convolutional neural network (CNN) based on diffusion-weighted imaging (DWI) with multiple b-values under magnetic resonance imaging (MRI) to distinguish pancreatic ductal adenocarcinomas (PDACs) from solid pseudopapillary neoplasms (SPNs) and pancreatic neuroendocrine neoplasms (PNENs).

METHODS

This retrospective study consisted of patients diagnosed with PDACs (n = 132), PNENs (n = 45) and SPNs (n = 54). All patients underwent 3.0-T MRI including DWI with 10 b values. The regions of interest (ROIs) of pancreatic tumor were manually drawn using ITK-SNAP software, which included entire tumor at DWI (b = 1500 s/m²). The histogram array was obtained through the ROIs from multiple b-value data. PyTorch (version 1.11) was used to construct a CNN classifier to categorize the histogram array into PDACs, PNENs or SPNs.

RESULTS

The area under the curves (AUCs) of the histogram array and the CNN model for differentiating PDACs from PNENs and SPNs were 0.896, 0.846, and 0.839 in the training, validation and testing cohorts, respectively. The accuracy, sensitivity and specificity were 90.22%, 96.23%, and 82.05% in the training cohort, 84.78%, 96.15%, and 70.0% in the validation cohort, and 81.72%, 90.57%, and 70.0% in the testing cohort. The performance of CNN with AUC of 0.865 for this differentiation was significantly higher than that of f with AUC = 0.755 (P = 0.0057) and α with AUC = 0.776 (P = 0.0278) in all patients.

CONCLUSION

The histogram array and CNN based on DWI data with multiple b-values using MRI provided an accurate diagnostic performance to differentiate PDACs from PNENs and SPNs.

Prediction of hepatic lymph node metastases based on magnetic resonance imaging before and after preoperative chemotherapy in patients with colorectal liver metastases underwent surgical resection

BACKGROUND

Patients with colorectal liver metastases (CRLM) combined with hepatic lymph node (HLN) metastases have a poor prognosis. In this study, we developed and validated a model using clinical and magnetic resonance imaging (MRI) parameters to predict HLN status before surgery.

METHODS

A total of 104 CRLM patients undergoing hepatic lymphonodectomy with pathologically confirmed HLN status after preoperative chemotherapy were enrolled in this study. The patients were further divided into a training group (n = 52) and a validation group (n = 52). The apparent diffusion coefficient (ADC) values, including ADC_mean and ADC_min of the largest HLN before and after treatment, were measured. rADC was calculated referring to the target liver metastases, spleen, and psoas major muscle (rADC_-LM, rADC_-SP, rADC_-m). In addition, ADC change rate (Δ% ADC) was quantitatively calculated. A multivariate logistic regression model for predicting HLN status in CRLM patients was constructed using the training group and further tested in the validation group.

RESULTS

In the training cohort, post-ADC_mean (P = 0.018) and the short diameter of the largest lymph node after treatment (P = 0.001) were independent predictors for metastatic HLN in CRLM patients. The model's AUC was 0.859 (95% CI, 0.757-0.961) and 0.767 (95% CI 0.634-0.900) in the training and validation cohorts, respectively. Patients with metastatic HLN showed significantly worse overall survival (p = 0.035) and recurrence-free survival (p = 0.015) than patients with negative HLN.

CONCLUSIONS

The developed model using MRI parameters could accurately predict HLN metastases in CRLM patients and could be used to preoperatively assess the HLN status and facilitate surgical treatment decisions in patients with CRLM.