Increasing discrepancy of MR imaging and CSF study in patients with leptomeningeal seeding from lung adenocarcinoma after targeted therapy using a tyrosine kinase inhibitor.

PURPOSE: To evaluate the correlation between contrast-enhanced (CE) MRI and cerebrospinal fluid (CSF) cytology for the evaluation of leptomeningeal metastasis (LM) on MRI after targeted therapy with tyrosine kinase inhibitors. METHODS: We retrospectively reviewed the data of nonsmall cell lung cancer patients registered with NCT03257124 from May 2017 to December 2018, with progressive disease despite targeted therapy. Twenty-nine patients whose MRI scans exhibited LM at the time of registration were enrolled. During the targeted therapy with osimertinib, MRI scans, and subsequent CSF examinations were performed in every 2 months. In total, 113 MRI scans and CSF cytology data after treatment were collected. For each CE MRI scan, LM positivity was evaluated on 3D T1-weighted image (T1WI) and 2D FLAIR. The correlation between MRI and CSF cytology results and the diagnostic performance of MRI with CSF cytology as a reference standard were evaluated. RESULTS: After treatment, MRI revealed positivity for LM in 81 and negativity in 32. CSF results were positive in 69 examinations and negative in 44. The diagnostic accuracy of CE 3D T1WI and 2D FLAIR was 0.52 and 0.46, respectively. After targeted therapy, discrepancy in the CSF and MRI results tended to increase over time. The proportions of concordant MRI and CSF cytology results after targeted therapy were 66%, 58%, 62%, and 47% at the first, second, third, and fourth follow-up, respectively. CONCLUSION: The discrepancy of MRI in evaluation of LM and CSF cytology increases over time after targeted therapy with osimertinib. LM positivity on MRI could be a surrogate imaging marker in the pre- and immediate posttargeted-treatment with Osimertinib but not after sessions of osimertinib.

A Deep Learning-Based Automatic Collateral Assessment in Patients with Acute Ischemic Stroke.

This study aimed to develop a supervised deep learning (DL) model for grading collateral status from dynamic susceptibility contrast magnetic resonance perfusion (DSC-MRP) images from patients with large vessel occlusion (LVO) acute ischemic stroke (AIS) and compare its performance against experts' manual grading. Among consecutive LVO-AIS at three medical center sites, DSC-MRP data were processed to generate collateral flow maps consisting of arterial, capillary, and venous phases. With the use of expert readings as a reference, a DL model was developed to analyze collateral status with output classified into good and poor grades. The resulting model was externally validated in a later-collected population from one medical center site. The model was trained on 255 patients and externally validated on 72 patients. In the all-site internal validation population, DL grading of good collateral probability yielded a c statistic of 0.91; in the external validation population, the c statistic was 0.85. In the external validation population, there was moderate agreement between the experts' grades and DL grades (kappa = 0.53, 95% CI = 0.32-0.73, p < 0.0001). Day 7 infarct growth volume was higher in DL-graded poor collateral group than good collateral group patients (median volume [26 mL vs. 6 mL], p = 0.01) in patients with successful reperfusion (modified treatment in cerebral infarction (mTICI) = 2b-3). In all patients with a 90-day modified Rankin Scale (mRS) score, there was a shift to more favorable outcomes in the good collateral group, with a common odds ratio of 2.99 (95% CI = 1.89-4.76, p < 0.0001). The DL-based collateral grading was in good agreement with expert manual grading in both development and validation populations. After exclusion of patients with large infarct volume, early reperfusion is more likely to benefit patients with the poor collateral flow, and the DL method has the potential to aid the assessment of collateral status.