18F-FDG PET/CT Radiomics for Preoperative Prediction of Lymph Node Metastases and Nodal Staging in Gastric Cancer

Abstract

Objectives: The accurate assessment of lymph node metastases (LNMs) and the preoperative nodal (N) stage are critical for the precise treatment of patients with gastric cancer (GC). The diagnostic performance, however, of current imaging procedures used for this assessment is sub-optimal. Our aim was to investigate the value of preoperative 18F-FDG PET/CT radiomic features to predict LNMs and the N stage.

Methods: We retrospectively collected clinical and 18F-FDG PET/CT imaging data of 185 patients with GC who underwent total or partial radical gastrectomy. Patients were allocated to training and validation sets using the stratified method at a fixed ratio (8:2). There were 2,100 radiomic features extracted from the 18F-FDG PET/CT scans. After selecting radiomic features by the random forest, relevancy-based, and sequential forward selection methods, the BalancedBagging ensemble classifier was established for the preoperative prediction of LNMs, and the OneVsRest classifier for the N stage. The performance of the models was primarily evaluated by the AUC and accuracy, and validated by the independent validation methods. Analysis of the feature importance and the correlation were also conducted. We also compared the predictive performance of our radiomic models to that with the contrast-enhanced CT (CECT) and 18F-FDG PET/CT.

Results: There were 185 patients-127 men, 58 women, with the median age of 62, and an age range of 22-86 years. One CT feature and one PET feature were selected to predict LNMs and achieved the best performance (AUC: 82.2%, accuracy: 85.2%). This radiomic model also detected some LNMs that were missed in CECT (19.6%) and 18F-FDG PET/CT (35.7%). For predicting the N stage, four CT features and one PET feature were selected (AUC: 73.7%, accuracy: 62.3%). Of note, a proportion of patients in the validation set whose LNMs were incorrectly staged by CECT (57.4%) and 18F-FDG PET/CT (55%) were diagnosed correctly by our radiomic model.

Conclusion: We developed and validated two machine learning models based on the preoperative 18F-FDG PET/CT images that have a predictive value for LNMs and the N stage in GC. These predictive models show a promise to offer a potentially useful adjunct to current staging approaches for patients with GC.

Keywords: N stage; PET/CT; gastric cancer; lymph nodes metastases; radiomics.

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2021 Liu, Li, Xin, Sun, Feng, Fulham, Wang and Song.

Figures

Figure 1

Radiomic flowchart for the prediction of LNMs (task A) and the N stage (task B).

Figure 2

Methodology and the results of feature selection: (A) feature selection pipeline, and (B) number of selected features during the selection procedure.

Figure 3

The performance of predicting LNMs and the N stage. (A) The AUC curve for predicting LNMs. (B) The AUC curve for predicting the N stage. (C) Accuracy of the prediction of LNMs. (D) Accuracy of the prediction of the N stage.

Figure 4

Normalized feature importance. (A–C) Feature importance in predicting LNMs for all validation patients and patients with/without metastases. (D–I) Feature importance in predicting the N stage for all validation patients and patients with five N stages (N0, N1, N2, N3a, and N3b).

Figure 5

Case studies for seven patients with GC. Top Panels: (A) patient with no lymph nodes metastases. (B) patient with lymph nodes metastases. The image at the bottom of (A, B) contains the feature value of the patients and the corresponding LIME interpretation. The top left and top right sections in panel (A, B) demonstrated the 3D model constructed based on the input CT and PET images from different viewpoints, while the red section represented the tumor of the patients. Our predictive model correctly identified the status for both patients in panel (A, B). (C) Bottom Panel - Five patients with different stages N0, N1, N2, N3a, and N3b from left to right. Our machine learning model predicted the N stage of the five patients accurately. 18F-FDG PET/CT, however, did not detect LNMs in all five patients; and CECT also did not assess the N stages correctly.

Figure 6

Pearson Correlations between the selected PET/CT features and the pathological features. (A) Correlation analysis for predicting LNMs. (B) Correlation analysis for predicting the N stage. Pairwise correlations with p < 0.05 are shown in the figure.