Researchers used a radiomics approach to predict brain metastases (BM) in patients with non-small cell lung cancer (NSCLC), according to a study presented at the 2023 American Society of Clinical Oncology Annual Meeting.
NSCLC makes up the largest portion of BM from solid cancer, with 40% of NSCLC patients developing brain tumors. Currently, there are no viable prediction modalities to discern risk in this population, especially in the early-stage setting when MRI is not performed.
In an effort to identify high-risk patients for BM that can benefit from MRI surveillance lead investigator Xiancheng Wu and colleagues assessed 162 lung adenocarcinoma (LUAD) patients, of which 66 had BM, and 96 did not have BM. They used the Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression feature selection method to choose the most relevant features for analysis, and constructed models using the machine learning method XGBoost classifier. They noted that training and testing sets with random splitting were used for cross validation, and then reported the accuracy, sensitivity, specificity, and area under the curve (AUC) for each model.
According to the results of the analysis, the XGBoost was able to predict BM with a robust 79% accuracy, 83% sensitivity, 72% specificity, and 79% AUC (p= 0.01) in the overall population. Moreover, the model distinguished those with metachronous vs synchronous BM with 84% accuracy, 83% sensitivity, 86% specificity, and 83% AUC (p = 0.04). Importantly, the investigators noted that the model was predictive in early-stage patients with 92% accuracy, 96% sensitivity, 83% specificity, and 95% AUC (p= 0.0005). “Moreover, our model predicted for high vs. low overall survival, and was BM-specific as it was not predictive of other sites of metastases,” the researchers said of the results.
“Utilizing a radiomics approach, we were able to predict BM from primary lung CT features including in stage I and II disease, predict synchronous vs metachronous BM, and distinguish distinct molecular LUAD subtypes,” the researchers concluded. They added that are currently “validating our BM prediction model in a large independent cohort and developing models to classify targetable LUAD-BM molecular alterations utilizing brain MRI scans. These studies will identify patients that require MRI surveillance in the early-stage setting and more intensive surveillance in the late-stage” setting for BM.”
