Minocycline for Aneurysmal Subarachnoid Hemorrhage (MASH)

Overview

Previous work has demonstrated patients presenting with ruptured aneurysms that develop radiographic and clinical vasospasm have a higher permeability of the blood brain membrane. Matrix metalloproteinase 9 (MMP9) has been studied and recently implicated in both the pathogenesis of the blood brain barrier breakdown and vasogenic edema of ischemia strokes, and is suggested to be an accurate biomarker to predict the onset of cerebral vasospasm after subarachnoid hemorrhage. The therapeutic benefit of minocycline, an MMP9 inhibitor, has been investigated in ischemic stroke population, however its role in the treatment of cerebral vasospasm from ruptured aneurysms remains unknown. Our project has two main goals: to further confirm MMP9 has a reliable biomarker for the onset of cerebral vasospasm, and secondarily to investigate any possible therapeutic benefit that minocycline has in the vasospasm population. Vasospasm continues to be one of the major contributors of morbidity and mortality in the ruptured aneurysm population, and close monitoring of the neurologic exam during the 'vasospasm window' usually requires two weeks in the intensive care unit in most academic settings. As such, if we are better able to predict which patients are at risk of developing vasospasm based on MMP9 levels, we will be better able to anticipate the need for intervention and therefore mitigate the risk of vasospasm induced ischemic strokes, ultimately resulting in better outcomes in the ruptured aneurysm population. Further, if we are able to identify minocycline as a therapeutic agent to deter, or lessen the severity of vasospasm, we can possibly improve neurologic outcomes, decrease hospital stays, ultimately providing an improved and more cost-effective treatment strategy to our patients.

The incidence of symptomatic cerebral vasospasm following aneurysmal subarachnoid hemorrhage (aSAH) is approximately 30%.8 The resulting ischemic complications contribute substantially to the overall morbidity and mortality of brain aneurysm patients.1,13 Clinical, translational, and laboratory SAH studies suggest that extravascular inflammation (serum and cerebrospinal fluid) is a potent effector of cerebral vessel reactivity.2,16,21,26 Dhar et al. demonstrated a correlation between the systemic inflammatory response syndrome on admission and delayed cerebral vasospasm.6 Provencio and colleagues have shown that high cerebrospinal fluid (CSF) neutrophil content three days following aSAH is predictive of subsequent vasospasm onset.21 While the intact blood brain barrier (BBB) prohibits egress of cytokines, chemokines, and cellular traffic, aneurysmal rupture allows blood and toxic blood breakdown products to rapidly enter the cerebral cisterns and extravascular space.7,11 Subarachnoid blood activates leukocyte transmigration by cellular margination, adhesion, and diapedesis.2 In a clinical study, Dr. Mack's team (mentor on this grant proposal) previously demonstrated that elevated serum levels of intercellular adhesion molecule -1 (ICAM-1) during the vasospasm risk period correlated with poor functional outcome.17 The team further found that increased rates of serum soluble ICAM-1 elevations predicted the onset of angiographic vasospasm. While this and other data suggests that inflammation plays a critical role in the pathogenesis of vasospasm, the mechanisms responsible for BBB dysregulation that allows inflammatory mediators to permeate the extravascular space are not clear. The current proposal focuses on the relationship between matrix metalloproteinase 9 (MMP9) and BBB permeability. The study leverages off-label use of an approved therapeutic agent, Minocycline, to target this proximal step in the SAH- inflammation-vasospasm axis. Matrix metalloproteinases are membrane bound proteases known to be involved in remodeling of the extracellular matrix through interactions with laminins, collagenases, and proteoglycans.10 Previous studies have established that MMPs contribute to inflammatory conditions and BBB breakdown in the central nervous system.24 The majority of MMP studies related to CNS dysfunction and SAH/ vasospasm have focused on MMP9.19 In a rat SAH model, Sebha et al. demonstrated colocalization of upregulated MMP-9 and collagen IV degradation in the basal lamina of cerebral blood vessels.23 Clinical studies have demonstrated that elevated serum levels of MMP9 predict subsequent delayed cerebral vasospasm, implicating BBB permeability as a critical mediator.27,28 BBB permeability is measurable in a clinically relevant model system. Our group has utilized MR permeability (DCE-MR) imaging to, non-invasively examine the integrity of the blood brain barrier in the setting of aSAH. Data suggest that increased permeability can reliably predict subsequent delayed cerebral ischemia (DCI) \[Appendix B\].22 Minocycline, a tetracycline antibiotic and potent MMP9 inhibitor, was initially studied as a neuro-protective agent in the setting of ischemic stroke.9,20,27,30 A recent investigation by Vellimana et al. demonstrated that MMP9 deletion attenuated cerebral vasospasm and resulted in less neurobehavioral deficits in a murine SAH model.27 The investigators demonstrated that Minocycline decreased rates of SAH-induced vasospasm in both murine and rabbit experimental models. Our group conducted a pilot clinical study (n=20 patients, 10 placebo, 10 Minocycline) of high dose intravenous Minocycline treatment in the setting of ICH.4 Minocycline was associated with a significant decrease in MMP9 levels between days 1 and 5. No serious adverse events or complications were associated with the Minocycline infusions. Clinicians would benefit from identification of aSAH patients most likely to be afflicted by cerebral vasospasm prior to its radiographic/ clinical onset. This could facilitate preventative and treatment strategies focused on early pathophysiological correlates. Current benchmarks such as the Fisher scale, which relate the extent of subarachnoid blood at the time of aneurysm rupture to the likelihood of developing clinical vasospasm, are coarse and imprecise. If BBB dysfunction reliably precedes neuroinflammation and subsequent vasospasm, then DCE-MRI could serve as a sensitive radiographic biomarker and a relevant assay for therapies targeting the BBB. However, we do not yet know the effects of minocycline on BBB permeability. The issue of the establishment of the measure as a biomarker is to be an aim of a future work. We propose to use this advanced MR imaging technique to assess the ability of Minocycline, and MMP9 inhibition, to mitigate BBB dysfunction following aSAH. The primary outcome of this study is to assess the effect of minocycline on BBB permeability and serum MMP9 levels. We hypothesize that minocycline infusion will lead to a lesser increase in serum MMP9 levels compared to the control group. Further, we hypothesize that infusion of minocycline will correlate to a less permeable BBB on the MRI permeability imaging. Secondary and tertiary outcomes include the ultimate effect of minocycline on clinical vasospasm and safety profile within the aSAH population. We hypothesize that the MMP9 inhibition via minocycline will lead to a less permeable BBB, therefore mitigating shifts in Ktrans. This translates clinically into a lower incidence of clinical vasospasm, and thus DCI in the minocycline treatment cohort. We hypothesize that the administration of minocycline at the mentioned dosage will be both safe and effective in the minocycline population.