Colin P. Derdeyn

BACKGROUND: Final infarct volume (FIV) on 24-hour magnetic resonance imaging is a well-established imaging biomarker linked to functional recovery after ischemic stroke, yet its prognostic value in intracranial atherosclerosis–related large vessel occlusion remains poorly explored. The impact of adjunct intracranial stenting on both infarct size and progression also remains unclear in this population. This study aimed to examine the association between FIV and clinical outcome, evaluate the effect of adjunct stenting on FIV and infarct progression, and assess the relationship between infarct progression and functional independence. METHODS: We conducted a post hoc secondary analysis of the RESCUE-ICAS registry (Registry of Emergent Large Vessel Occlusion due to Intracranial Stenosis); only patients with anterior circulation large vessel occlusion with magnetic resonance imaging after thrombectomy were included. FIV was measured on diffusion-weighted magnetic resonance imaging performed 24 to 36 hours postthrombectomy. Infarct progression was defined as the difference between baseline computed tomography perfusion infarct volume (cerebral blood flow P =0.007). Adjunct stenting was associated with a significant relative reduction in FIV, with inverse probability of treatment weighting–adjusted analyses demonstrating a 37.8% lower FIV compared with no stenting ( β =−0.47 [95% CI, −0.93 to −0.02]; P =0.043). Among 108 patients with baseline computed tomography perfusion data, infarct progression was strongly associated with outcome, with each additional 10 mL associated with ≈25% lower odds of achieving modified Rankin Scale score 0 to 2 at 90 days (adjusted odds ratio, 0.74 [95% CI, 0.60–0.90]; P =0.004), although infarct progression did not differ significantly between stented and nonstented groups (Δ −9.53 mL [95% CI −37.9 to 18.8]; P =0.506). In inverse probability of treatment weighting–weighted models adjusted for successful recanalization, stenting was not independently associated with reduced FIV ( β =−0.41; P =0.066), but remained independently associated with higher odds of 90-day functional independence (odds ratio, 2.29 [95% CI, 1.24–4.27]; P =0.008). CONCLUSIONS: Among intracranial atherosclerosis–related large vessel occlusion patients, 24- to 36-hour FIV is a strong predictor of functional outcome. Adjunct stenting is associated with smaller FIV. Lower infarct progression was also associated with favorable outcome. These findings highlight FIV as a reliable imaging biomarker and potential surrogate end point in future trials.

The MRI model was run on 34 patients with tumefactive demyelination. (A) Distribution of predicted MRI score for the 34 patients. (B) Distribution of predicted MRI score by age at diagnosis. The blue line denotes a loess fit and the grey shaded area denotes the 95% confidence interval. (C) Distribution of predicted MRI score by gender (F=female, M=Male). (D) Distribution of predicted MRI score by MRI manufacturer. (E) Distribution of predicted MRI score by MRI field strength. (F) Distribution of predicted MRI score by T1Gd acquisition type. T2 acquisition type is not shown because 33 of the 34 patients were sequenced using 2D.

Model performance on the 256 GBM and 73 CNS-DLBCL independent test cases from stage 3 from the model developed using the ensemble approach (A) overall, (B) stratified by sex, and (C) stratified by age. Results for the model developed using the loss approach (D) overall, (E) stratified by sex, and (F) stratified by age. AUCs (95% CI) are provided.

Glioblastoma (GBM), isocitrate dehydrogenase wild-type (IDHwt) and central nervous system diffuse large B-cell lymphoma (CNS-DLBCL) are aggressive brain tumors with overlapping MRI features, yet distinct treatment approaches. Noninvasive tools are needed to aid in differential diagnosis. Deep learning on T1 postcontrast and T2-weighted MRI sequences were used to differentiate GBM and CNS-DLBCL. A three-stage temporal study design was utilized. Model development was performed on 146 patients with CNS-DLBCL and 146 age-matched, sex-matched, and MRI year–matched patients with GBM diagnosed at Mayo Clinic between 1998 and 2019. Models were tested on independent temporal test cohorts. Initial testing included 240 independent GBM diagnosed at Mayo Clinic between 1998 and 2019. The prospective test cohort included 37 patients with CNS-DLBCL and 256 patients with GBM diagnosed at Mayo Clinic after January 1, 2020, and 36 patients with CNS-DLBCL diagnosed at an external institution. Of the patients diagnosed at Mayo Clinic, 47% had MRIs generated from non-Mayo institutions. Two different model approaches were compared: (i) ensemble approach using area under the receiver operating characteristic curve (AUC) and cross-validation for model selection and (ii) loss approach minimizing cross-entropy loss and cross-validation to evaluate prediction performance. The AUCs on the prospective test cohort were 0.84 [95% confidence interval (CI), 0.78–0.90] and 0.83 (95% CI, 0.77–0.88) for the ensemble and loss approaches, respectively. Stability of ensemble prediction improved with the increasing number of models. Stratified AUC analysis demonstrated consistent performance across sex and age. We utilized a robust temporal study design and applied 2 different analytic approaches to develop a classification model. The findings confirm the feasibility of using MRI-based deep learning models to differentiate GBM from CNS-DLBCL.Significance:GBM, IDHwt and CNS-DLBCL are aggressive brain tumors with overlapping MRI features, yet distinct treatment approaches. Noninvasive tools are needed to aid in differential diagnosis. We developed MRI-based deep learning models to differentiate GBM, IDHwt from CNS-DLBCL using a rigorous three-stage temporal design that included prospective validation. The model AUC on a prospective cohort was 0.84.

Cartoon comparing AUC and cross entropy loss on five GBM patients and five CNS-DLBCL cases. Grey denotes GBM and black denotes CNS-DLBCL. All models have the same AUC (AUC=100%); however, Model 4 has the smallest cross entropy loss.

Violin plots showing the model scores for the 256 GBM and 73 CNS-DLBCL cases in stage 3, obtained from the models developed using the (A) ensemble approach and (B) loss approach.

Evaluating the effect on model performance (AUC) when performing ensemble across an increasingly larger number of models. Five independent sets of 5-fold cross-validation were evaluated. One 5-fold cross-validation set denotes performing ensemble across the resulting k = 5 models. Two 5-fold cross-validation sets denote ensemble across 10 models, three sets denote ensemble across 15 models, four sets denote ensemble across 20 models, and five sets denote ensemble across 25 models.

Sensitivity and specificity of the loss model applied to the stage 3 cohort. All possible model score thresholds from the loss model are shown. The yellow highlighted row denotes the score threshold required to obtain 90% sensitivity. The orange highlighted row denotes the score threshold required to obtain 90% specificity.

Description of the three-stage study design.