Evaluation of EGFR mutation status by radiographic characteristics of MRI in NSCLC patients with brain metastasis

Di-san junyi daxue xuebao

HE Rong¹  ZHANG Zhimin¹  HE Xuan¹ 

[1] Cancer Center, Daping Hospital, Army Medical University (Third Military Medical University), Chongqing, 400042, China ;

关键词: non-small cell lung cancer;    epidermal growth factor receptor;    gene mutation;    brain metastasis;   

DOI  :  10.16016/j.1000-5404.201811007

来源: DOAJ

【 摘 要 】

Objective To investigate the correlation between brain metastatic MRI features and EGFR mutation status in the patient with non-small cell lung cancer (NSCLC) and explore the value of radiographic characteristics of MRI for evaluation of EGFR mutation. Methods Clinical data of 116 patients diagnosed with brain metastases from NSCLC admitted in our hospital during January 2011 and January 2017 were collected. Their EGFR mutation status in pulmonary primary tumors and metastatic tissues was detected by amplification block mutation system (ARMS). The features of MR images were analyzed for the correlation of EGFR mutation status with number, size, location of the metastatic brain tumors, strengthen approaches, size of peritumoral brain edema of the largest lesion and peritumoral brain edema index. Results Among the 116 patients with brain metastasis of NSCLC, 50 patients had EGFR mutation, at a mutation rate of 43.1%. The patients younger than 60 years had significantly higher ratio of EGFR 19Del mutation (25.4% vs 10.2%, P=0.040) and lower ratio of L858R mutation (16.4% vs 34.7%, P=0.023) when compared with those older than 60 years. Of the 201 brain metastatic lesions measured, the diameter of peripheral edema were significantly larger in the patients with EGFR wild type than those harboring EGFR 19Del mutation (median: 2.87 vs 1.54, P=0.008), and in those with L858R mutation than those with 19Del mutation (median: 3.31 vs 1.54, P=0.034). However, the index of peripheral edema in the patients with EGFR wild type was obviously higher than those harboring EGFR 19Del mutation (median: 0.979 vs 0.413, P=0.012). The predictive model for EGFR 19 deletion mutation was developed based on age, diameter and index of peripheral edema of metastatic tumor, and the capacity of this model was 0.733 (95%CI=0.625~0.842) by ROC analysis. Conclusion Clinical features and MRI radiographic characteristics of brain metastatic lesions of NSCLC can be used to predict EGFR mutation status. Our study provides theoretical evidence for selecting subpopulation benefiting from EGFR-TKI from the patients who can't detect the mutation of EGFR medically.

【 授权许可】

Unknown