MINocyclinE to Reduce inflammation and blood brain barrier leakage in small Vessel diseAse - results of the MINERVA randomised controlled trial

Abstract

Introduction: Cerebral small vessel disease (SVD) is a major cause of cognitive impairment and stroke. Neuroinflammation and blood-brain barrier (BBB) leakage may play a role in pathogenesis, but no definitive causal link has been established. In a rodent SVD model, minocycline treatment reduced brain lesions, neuroinflammation and BBB permeability. We tested whether these processes can be altered in SVD. Methods: MINERVA was a phase II, double-blind, randomised controlled trial in moderate-to-severe symptomatic SVD. Participants with lacunar stroke and confluent white matter hyperintensities underwent simultaneous dynamic contrast-enhanced MRI to measure BBB permeability and 11C- PK11195 positron emission tomography (PET) to quantify microglial signal (figure 1). They were randomised to either minocycline 100mg bd or placebo for three months, after which PET-MRI was repeated. The co-primary outcomes were volumes of ‘hotspots’ of increased BBB permeability and 11C-PK11195 binding in the normal appearing white matter above the 95th percentile of healthy control reference values. A sample size of 44 allowed detection of a 20% reduction in these metrics (power = 80%, α=0.05). Results: 44 patients were recruited from September 2019 - June 2022 at 23.1±24.7 months after lacunar stroke. Mean age was 69.9±10.8 years and 28/44 (63.6%) were male. 86.4% had a history of hypertension, 75.0% of hypercholesterolaemia and 18.2% were diabetic. Participants had mean white matter lesion volume of 31.3±26.0cc and median 2 (IQR 1–3) lacunes. The BBB permeability ‘hotspot’ tissue was 4.08±3.69% in the treatment group at baseline and 6.19±5.09% at follow-up; ‘hotspot’ tissue was 8.49±8.45% in the placebo group at baseline and 13.04±9.24% at follow-up (relative risk of treatment 0.97, 95% CI 0.91–1.03). The 11C-PK11195 ‘hotspot’ tissue was 10.71±4.04% in the treatment group at baseline and 9.97±5.50% at follow-up; ‘hotspot’ tissue was 10.11±4.67% in the placebo group at baseline and 7.79±5.67% at follow-up (RR 1.01, 95% CI 0.98–1.04; figure 2). Discussion: Minocycline does not alter BBB permeability or microglial activity (measured using DCE-MRI and 11C-PK11195 respectively) in SVD patients. Secondary outcomes include changes in a panel of serum inflammatory biomarkers, and one-year progression of MRI white matter damage and cognitive performance.International Clinical Trials Registry Platform reference: ISRCTN15483452 (http://isrctn.com/ISRCTN15483452)