Researchers from China developed a radiomic algorithm that could identify gastrointestinal stromal tumors (GISTs) with KIT exon 11 mutation, as published in Diagnostic and Interventional Radiology.

“The combined radiomics algorithm performed better than the other predictive models, which confirmed it as a potential predictor to supplement conventional approaches for selective genetic analysis and to support the clinical decision about imatinib therapy in the precision medicine of GISTs,” they said.

The area under the curve (AUC) for the combined radiomics algorithm was 0.836 (95% CI, 0.640-0.951) in the validation set. The other models explored included a clinical model, which was based on significant preoperative and postoperative clinical features, and radiomics nomograms (preoperative and postoperative), which were built using independent predictive factors.


Continue Reading

Read more about GIST diagnosis

The clinical model had the lowest AUC (0.606, 95% CI, 0.397-0.790) in the validation set. The AUCs for the preoperative and postoperative radiomics nomograms were 0.715 (95% CI, 0.506-0.873) and 0.679 (95% CI, 0.468-0.847), respectively.

The authors developed 4 algorithms using contrast-enhanced computed tomography images from the training set: arterial phase, venous phase, delayed phase, and tri-phase combined radiomics algorithms.

New Deep Learning Framework Could Improve Classification of GIST

To train and validate the method, the study enrolled 106 patients with GIST from 2 clinical centers, who were distributed between a training set (n=80) and a validation set (n=26). All the patients were diagnosed with GISTs.

The exon 11 mutation accounts for most of the KIT mutations in GISTs. Also, it is the most common imatinib mesylate-target mutation. Therefore, patients with KIT exon 11 mutation usually respond well to imatinib mesylate treatment.

Reference

Liu B, Liu H, Zhang L, et al. Value of contrast-enhanced CT based radiomic machine learning algorithm in differentiating gastrointestinal stromal tumors with KIT exon 11 mutation: a two-center study. Diagnostic Interv Radiol. 2022;28(1):29-38. doi:10.5152/dir.2021.21600