Virtual biopsy using MRI radiomics for prediction of BRAF status in melanoma brain metastasis

April 29, 2020

Brain metastases are common in patients with advanced melanoma and constitute a major cause of morbidity and mortality. Between 40% and 60% of melanomas harbor BRAF mutations. Selective BRAF inhibitor therapy has yielded improvement in clinical outcome; however, genetic discordance between the primary lesion and the metastatic tumor has been shown to occur. Currently, the only way to characterize the genetic landscape of a brain metastasis is by tissue sampling, which carries risks and potential complications. The aim of this study was to investigate the use of radiomics analysis for non-invasive identifcation of BRAF mutation in patients with melanoma brain metastases, based on conventional magnetic resonance imaging (MRI) data. We applied a machine-learning method, based on MRI radiomics features for noninvasive characterization of the BRAF status of brain metastases from melanoma (BMM) and applied it to BMM patients from two tertiary neuro-oncological centers. All patients underwent surgical resection for BMM, and their BRAF mutation status was determined as part of their oncological work-up. Their routine preoperative MRI study was used for radiomicsbased analysis in which 195 features were extracted and classifed according to their BRAF status via a support vector machine. The BRAF status of 53 study patients, with 54 brain metastases (25 positive, 29 negative for BRAF mutation) was predicted with mean accuracy = 0.79±0.13, mean precision = 0.77±0.14, mean sensitivity = 0.72±0.20, mean specifcity = 0.83±0.11 and with a 0.78 area under the receiver operating characteristic curve for positive BRAF mutation prediction. Radiomics-based noninvasive genetic characterization is feasible and should be further verifed using large prospective cohorts.


Shofty, B., Artzi, M., Shtrozberg, S., Fanizzi, C., Dimeco, F., Haim, O., … Grossman, R. (2020). Virtual biopsy using MRI radiomics for prediction of BRAF status in melanoma brain metastasis. Scientific Reports10(1). doi: 10.1038/s41598-020-63821-y

« Back to News