Quantitative Prediction of Microsatellite Instability in Colorectal Cancer With Preoperative PET/CT-Based Radiomics.

Jiaru Li,Junyan Xu,Xiuying Wang,Yichao Hao,Shaoli Song,Bowen Xin,Lisheng Wang,Ziyi Yang

doi:10.3389/fonc.2021.702055

Abstract

ObjectivesMicrosatellite instability (MSI) status is an important hallmark for prognosis prediction and treatment recommendation of colorectal cancer (CRC). To address issues due to the invasiveness of clinical preoperative evaluation of microsatellite status, we investigated the value of preoperative 18F-FDG PET/CT radiomics with machine learning for predicting the microsatellite status of colorectal cancer patients.MethodsA total of 173 patients that underwent 18F-FDG PET/CT scans before operations were retrospectively analyzed in this study. The microsatellite status for each patient was identified as microsatellite instability-high (MSI-H) or microsatellite stable (MSS), according to the test for mismatch repair gene proteins with immunohistochemical staining methods. There were 2,492 radiomic features in total extracted from 18F-FDG PET/CT imaging. Then, radiomic features were selected through multivariate random forest selection and univariate relevancy tests after handling the imbalanced dataset through the random under-sampling method. Based on the selected features, we constructed a BalancedBagging model based on Adaboost classifiers to identify the MSI status in patients with CRC. The model performance was evaluated by the area under the curve (AUC), sensitivity, specificity, and accuracy on the validation dataset.ResultsThe ensemble model was constructed based on two radiomic features and achieved an 82.8% AUC for predicting the MSI status of colorectal cancer patients. The sensitivity, specificity, and accuracy were 83.3, 76.3, and 76.8%, respectively. The significant correlation of the selected two radiomic features with multiple effective clinical features was identified (p < 0.05).Conclusion 18F-FDG PET/CT radiomics analysis with the machine learning model provided a quantitative, efficient, and non-invasive mechanism for identifying the microsatellite status of colorectal cancer patients, which optimized the treatment decision support.

Highlights

Colorectal cancer (CRC) is one of the most common malignancies in China and ranks the fifth in terms of cancer mortality [1]
There were four features selected during the multivariant feature selection process, including one clinical feature, one feature obtained from the positron emission tomography (PET) images that applied the wavelet filter, and two features obtained from wavelet decomposed computed tomography (CT) images
Since the original dataset was imbalanced, conducting the crossvalidation method on the training group would cause a significant increase in the imbalanced ratio of the dataset

Summary

Introduction

Colorectal cancer (CRC) is one of the most common malignancies in China and ranks the fifth in terms of cancer mortality [1]. Microsatellite instability (MSI) is an essential molecular hallmark of hereditary non-polyposis colorectal cancer (HNPCC) and Lynch syndrome (LS). It occurs in 15% of sporadic colorectal cancers and is often associated with the deficiency of the mismatch repair (MMR) system caused by the failure of one of the primary MMR genes, including MSH2, MLH1, MSH6, or PMS2 [2]. CRCs with high MSI (MSI-H) are usually located in the right colon, more common in stage II, and relatively infrequent among metastatic tumors [3]. The patients with MSI-H CRCs had a poor prognosis for stage IV CRCs, which constituted about 2– 4% of all metastatic CRCs

Objectives

Methods

Results

Conclusion