Torbey, Michel

Features of new bleeding on conventional imaging in cerebral cavernous malformations (CCMs) often disappear after several weeks, yet the risk of rebleeding persists long thereafter. Increases in mean lesional quantitative susceptibility mapping (QSM) ≥ 6% on MRI during 1 year of prospective surveillance have been associated with new symptomatic hemorrhage (SH) during that period. The authors hypothesized that QSM at a single time point reflects features of hemorrhage in the prior year or potential bleeding in the subsequent year. Twenty-eight features were extracted from 265 QSM acquisitions in 120 patients enrolled in a prospective trial readiness project, and machine learning methods examined associations with SH and biomarker bleed (QSM increase ≥ 6%) in prior and subsequent years. QSM features including sum variance, variance, and correlation had lower average values in lesions with SH in the prior year (p < 0.05, false discovery rate corrected). A support-vector machine classifier recurrently selected sum average, mean lesional QSM, sphericity, and margin sharpness features to distinguish biomarker bleeds in the prior year (area under the curve = 0.61, 95% CI 0.52–0.70; p = 0.02). No QSM features were associated with a subsequent bleed. These results provide proof of concept that machine learning may derive features of QSM reflecting prior hemorrhagic activity, meriting further investigation. Clinical trial registration no.: NCT03652181 ( ClinicalTrials.gov )

Introduction and Hypothesis: New bleeding in cerebral cavernous malformations (CCM) heralds increased risk of future hemorrhage for several years, yet conventional imaging only detects new bleeding that occurred in the prior weeks. A biomarker of hemorrhage could help identifying high risk lesions. An increase of mean lesional quantitative susceptibility mapping (QSM) ≥6% on MRI has been adjudicated as reflecting new bleeding in CCM during longitudinal follow-up. However, mean lesional QSM from a single acquisition could not diagnose or prognosticate a bleed. We hypothesize that machine learning (ML) may identify diagnostic and prognostic features of bleeding within QSM maps at a single point in time. Material and Methods: Two hundred and sixty-five QSM maps of CCM lesions were acquired in 120 patients enrolled in National Institute of Health (NIH) multisite trial readiness project (NCT03652181). Each map was classified (Yes/No) in association with symptomatic hemorrhage (SH) and/or biomarker event with QSM increase ≥6% in the prior (diagnostic association) and subsequent (prognostic association) year. Twenty-eight features were extracted including 14 texture, 5 first-order statistical, as well as 3 size, shape, and morphological. Five-fold cross-validation was conducted on a support-vector machine (SVM) with linear stepwise kernel for both diagnostic and prognostic associations. Performance of individual features and composite classifiers was evaluated using student t-test ( p <0.05) with Bonferroni-correction and receiver operating characteristic (ROC) analysis, area under the curve (AUC). Results: Lesions with SH in the prior year had lower average values for sum variance (AUC=0.79, p<0.001), variance (AUC=0.79, p<0.001), and correlation (AUC=0.71, p=0.004) as compared to lesions without SH. The SVM classification method tasked with distinguishing lesions with mean QSM increase ≥6% in the prior year included recurrent selection of sum average, mean, sphericity and margin sharpness, yielding an AUC of 0.61 (p=0.02). In the prognostic cohort, no individual feature nor any SVM classification model was able to distinguish cases with a subsequent clinical bleed or biomarker event. Conclusion: This proof-of-concept suggests that ML may assist in deriving features on QSM maps acquired at a single timepoint, which reflect prior hemorrhagic activity in CCM. Further investigation is planned in larger cohorts and in conjunction with clinical trials of bleeding in CCM.

BACKGROUND: Quantitative susceptibility mapping (QSM) and dynamic contrast-enhanced quantitative perfusion (DCEQP) magnetic resonance imaging sequences assessing iron deposition and vascular permeability were previously correlated with new hemorrhage in cerebral cavernous malformations. We assessed their prospective changes in a multisite trial-readiness project. METHODS: Patients with cavernous malformation and symptomatic hemorrhage (SH) in the prior year, without prior or planned lesion resection or irradiation were enrolled. Mean QSM and DCEQP of the SH lesion were acquired at baseline and at 1- and 2-year follow-ups. Sensitivity and specificity of biomarker changes were analyzed in relation to predefined criteria for recurrent SH or asymptomatic change. Sample size calculations for hypothesized therapeutic effects were conducted. RESULTS: We logged 143 QSM and 130 DCEQP paired annual assessments. Annual QSM change was greater in cases with SH than in cases without SH ( P =0.019). Annual QSM increase by ≥6% occurred in 7 of 7 cases (100%) with recurrent SH and in 7 of 10 cases (70%) with asymptomatic change during the same epoch and 3.82× more frequently than clinical events. DCEQP change had lower sensitivity for SH and asymptomatic change than QSM change and greater variance. A trial with the smallest sample size would detect a 30% difference in QSM annual change during 2 years of follow-up in 34 or 42 subjects (1 and 2 tailed, respectively); power, 0.8, α=0.05. CONCLUSIONS: Assessment of QSM change is feasible and sensitive to recurrent bleeding in cavernous malformations. Evaluation of an intervention on QSM percent change may be used as a time-averaged difference between 2 arms using a repeated measures analysis. DCEQP change is associated with lesser sensitivity and higher variability than QSM. These results are the basis of an application for certification by the US Food and Drug Administration of QSM as a biomarker of drug effect on bleeding in cavernous malformations. REGISTRATION: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT03652181.

BACKGROUND: Cerebral cavernous malformation with symptomatic hemorrhage (SH) are targets for novel therapies. A multisite trial-readiness project ( https://www.clinicaltrials.gov ; Unique identifier: NCT03652181) aimed to identify clinical, imaging, and functional changes in these patients. METHODS: We enrolled adult cerebral cavernous malformation patients from 5 high-volume centers with SH within the prior year and no planned surgery. In addition to clinical and imaging review, we assessed baseline, 1- and 2-year National Institutes of Health Stroke Scale, modified Rankin Scale, European Quality of Life 5D-3 L, and patient-reported outcome-measurement information system, Version 2.0. SH and asymptomatic change rates were adjudicated. Changes in functional scores were assessed as a marker for hemorrhage. RESULTS: One hundred twenty-three, 102, and 69 patients completed baseline, 1- and 2-year clinical assessments, respectively. There were 21 SH during 178.3 patient years of follow-up (11.8% per patient year). At baseline, 62.6% and 95.1% of patients had a modified Rankin Scale score of 1 and National Institutes of Health Stroke Scale score of 0 to 4, respectively, which improved to 75.4% ( P =0.03) and 100% ( P =0.06) at 2 years. At baseline, 74.8% had at least one abnormal patient-reported outcome-measurement information system, Version 2.0 domain compared with 61.2% at 2 years ( P =0.004). The most common abnormal European Quality of Life 5D-3 L domains were pain (48.7%), anxiety (41.5%), and participation in usual activities (41.4%). Patients with prospective SH were more likely than those without SH to display functional decline in sleep, fatigue, and social function patient-reported outcome-measurement information system, Version 2.0 domains at 2 years. Other score changes did not differ significantly between groups at 2 years. The sensitivity of scores as an SH marker remained poor at the time interval assessed. CONCLUSIONS: We report SH rate, functional, and patient-reported outcomes in trial-eligible cerebral cavernous malformation with SH patients. Functional outcomes and patient-reported outcomes generally improved over 2 years. No score change was highly sensitive or specific for SH and could not be used as a primary end point in a trial.

Background: Cerebral cavernous angiomas with symptomatic hemorrhage (CASH) carry highest rebleed rates and are targeted for novel interventions. A multisite clinical trial readiness project (clinicaltrials.gov NCT03652181) aims to identify clinical and functional changes over time in trial eligible CASH patients. Methods: Adult patients from 5 high volume centers with a CASH in the prior year and no planned surgical intervention were enrolled. In addition to demographic, clinical, and imaging review, patients underwent baseline, 1 year and 2 year NIH-stroke scale (NIHSS), modified Rankin scale (mRS), Euro-QOL 5D-3L (EQ-5D-3L) and EQ VAS (visual analog score), and PROMIS 29 assessments. Recurrent symptomatic hemorrhages (SH) were verified. We assessed the hemorrhage rate, change in scores/scales over time, and determined which scales/scores were most affected in patients with SH. Results: 123 patients completed baseline assessments and were prospectively followed (60.9% female; mean age 44.1 years; 39.1% familial; 41.5% brainstem location). As of reporting date, 89 of 104 (86%) and 58 of 76 (76%) cases have completed assessments. There were 16 SHs during 147 patient-years of follow up (10.9% per patient year). At baseline, 111 (90.2%) of patients had NIHSS 0-2 and 77 (62.6%) patients had mRS baseline 0-1 which improved to 40 (95.2 %) and 43 (75.4%) at 2 years. At baseline, 54 (43.9%) of patients had at least one moderate to severely abnormal PROMIS-29 domain compared to 13 (26.5%) at 2 years. EQ-5D-3L baseline demonstrated a moderate proportion of patients reporting some difficulty with pain (48.8%), anxiety (41.5%), and activity (41.5%) which improved to 28%, 30%, and 34% respectively at 2 years. Patients with a prospective SH were more likely than those without recurrent SH at 2 years to display mild functional decline (0.5 SD) of most PROMIS-29 domains, a mRS of 2 or higher and an abnormal EQ- self-care (P<0.05). Other scales/scores were not significantly different between the groups. Conclusion: We report SH rate and functional outcome at baseline and rates of changes during 2 years in trial eligible CASH patients. The PROMIS-29, mRS, and EQ-self-care domains were most discriminative in patients with recurrent SH. No test was 100% specific.

Background: Quantitative susceptibility mapping (QSM) is a measure of iron content, and ≥6% increase in QSM has been correlated with new hemorrhage in previously stable cavernous angiomas. Longitudinal changes in QSM are not known in cavernous angiomas with symptomatic hemorrhage (CASH) with high rates of rebleeding and are the targets of novel pharmacotherapies. In a prospective multisite Trial Readiness project (clinicaltrials.gov NCT03652181), QSM is longitudinally assessed. Methods: Trial eligible subjects with CASH in the prior year and not undergoing lesion resection or radiation, were enrolled. Mean QSM of CASH lesion was acquired at baseline and at 1 and 2 year planned follow-ups. Relative change in mean lesional QSM during each follow-up year was assessed, and any symptomatic hemorrhage (SH), asymptomatic changes (AC; defined as subclinical bleed or growth) in the lesion during the same epoch. Results: Paired QSM assessments were completed to date in 99 CASH lesions (67 year 1, and 32 year 2) and are reported herein. Four SH and 6 ACs occurred during 1 st follow-up year, and 3 SH during 2 nd year. QSM increased in 54 lesion-years, decreased in 44, and remained stable in 1. The % lesional QSM change in year 1 was significantly higher than that observed in year 2 (mean +9.33, SD 37.52 vs. +5.20; SD= 24.86; p=0.05; Spearman correlation ρ -0.36). CAs with clinical SH or AC had a significantly higher % QSM change than lesions without (mean +28.03, SD 17.77 vs. +4.97, SD= 34.75; p=0.0014). All 13 lesions with SH/AC demonstrated a QSM increase ≥6% while 31 of 86 (36%) lesions with no clinical events had a ≥6% QSM increase. Conclusion: QSM change of ≥6% is present in every CASH lesion manifesting a new SH or AC (100% specificity), and is more common than clinical events (3.4X higher sensitivity). The biomarker can hence be used as a more sensitive categorical outcome than SH or AC in clinical trials of novel therapies aimed at bleeding in CASH lesions. Effect of an intervention on % QSM change may also be proposed as a time-averaged difference between 2 arms using a repeated measures analysis implemented as an unadjusted linear mixed model. These results are the basis of application for certification by the U.S F.D.A. of QSM as a monitoring biomarker of drug effect in CASH.

Background and Purpose: Brain cavernous angiomas with symptomatic hemorrhage (CASH) have a high risk of neurological disability from recurrent bleeding. Systematic assessment of baseline features and multisite validation of novel magnetic resonance imaging biomarkers are needed to optimize clinical trial design aimed at novel pharmacotherapies in CASH. Methods: This prospective, multicenter, observational cohort study included adults with unresected, adjudicated brain CASH within the prior year. Six US sites screened and enrolled patients starting August 2018. Baseline demographics, clinical and imaging features, functional status (modified Rankin Scale and National Institutes of Health Stroke Scale), and patient quality of life outcomes (Patient-Reported Outcomes Measurement Information System-29 and EuroQol-5D) were summarized using descriptive statistics. Patient-Reported Outcomes Measurement Information System-29 scores were standardized against a reference population (mean 50, SD 10), and one-sample t test was performed for each domain. A subgroup underwent harmonized magnetic resonance imaging assessment of lesional iron content with quantitative susceptibility mapping and vascular permeability with dynamic contrast-enhanced quantitative perfusion. Results: As of May 2020, 849 patients were screened and 110 CASH cases enrolled (13% prevalence of trial eligible cases). The average age at consent was 46±16 years, 53% were female, 41% were familial, and 43% were brainstem lesions. At enrollment, ≥90% of the cohort had independent functional outcome (modified Rankin Scale score ≤2 and National Institutes of Health Stroke Scale score <5). However, perceived health problems affecting quality of life were reported in >30% of patients (EuroQol-5D). Patients had significantly worse Patient-Reported Outcomes Measurement Information System-29 scores for anxiety ( P =0.007), but better depression ( P =0.002) and social satisfaction scores ( P =0.012) compared with the general reference population. Mean baseline quantitative susceptibility mapping and permeability of CASH lesion were 0.45±0.17 ppm and 0.39±0.31 mL/100 g per minute, respectively, which were similar to historical CASH cases and consistent across sites. Conclusions: These baseline features will aid investigators in patient stratification and determining the most appropriate outcome measures for clinical trials of emerging pharmacotherapies in CASH.

Background: Patients with cerebral cavernous angiomas with symptomatic hemorrhage (CASH) have high risk of disability from recurrent bleeding. Candidate medications to prevent rebleeding in CASH lesions will require multisite clinical trials with standardized data collection. Objective: To report the prevalence and baseline cohort features in CASH patients and establish a research network infrastructure for trials. Methods: This prospective observational cohort study includes adults with radiologically verified CASH lesion within 1-year of consent. Exclusions include prior or planned surgical intervention, spinal location, or prior brain irradiation. Six sites enrolled patients into the screening and clinical assessment portion of the study starting July 2018. Patients also had the option to participate in the follow up biomarker validation at 4 sites. Baseline demographics, clinical and imaging information, and outcomes (mRS, PROMIS-29, NIHSS, and EuroQol-5D) were collected. Biomarker imaging included dynamic contrast enhanced quantitative perfusion (DCEQP) and quantitative susceptibility mapping (QSM) that correlated with symptomatic bleeding. Descriptive statistics were performed and one-sample t-test was used to compare whether mean T-scores for PROMIS-29 domains differed significantly from a reference population. Results: As of May 2020, 849 CASH patients were screened of whom 110 (13%) were eligible and enrolled; 73 also enrolled into the biomarker validation study. Average age at enrollment was 46±16 years at a mean of 4.4 months after symptom onset; 53% were female, 41% were familial, and 43% of CASH lesions were brainstem location. At enrollment, 90% of the cohort had independent functional outcome (mRS ≤ 2 and NIHSS <5). Perceived health problems affecting QoL were reported in >30% (EuroQol-5D). CASH cases had significantly worse anxiety but better depression and social satisfaction scores compared to a general population (all P<0.01). Baseline DCEQP and QSM measures did not differ significantly across sites. Conclusion: We demonstrate feasibility of multisite recruitment of CASH patients and report prevalence of baseline features that will aid in design of clinical trials and inclusion of appropriate outcome measures.