Hematoma clearance as a therapeutic target in intracerebral hemorrhage: From macro to micro

Abstract

Despite the absence of an intervention shown to improve outcomes in intracerebral hemorrhage, preclinical work has led to a greater understanding of the pathologic pathways of brain injury. Methods targeting hematoma clearance through both macroscopic (surgical) and microscopic (endogenous phagocytosis) means are currently under investigation, with multiple clinical trials ongoing. Macroscopic methods for removal involve both catheter- and endoscope-based therapies to remove the hematoma through minimally invasive surgery. Microscopic methods targeting hematoma clearance involve augmenting endogenous clearance pathways for red blood cells and altering the balance between phagocytosis and red blood cell lysis with the release of potentially harmful constituents (e.g. hemoglobin and iron) into the extracellular space.

Intracerebral hemorrhage (ICH) remains the only stroke subtype for which interventional therapies remain unproven. The failure of recent trials of intensive blood pressure control to show clinically meaningful benefit^1–3 may indicate that while early blood pressure control is important to prevent hematoma expansion, efforts to treat the acute hypertensive response to ICH may be of less importance to long-term outcomes than treatment of its underlying etiology. Because of the adverse physical effects of the hematoma (‘mass effect’) and the secondary injury induced by clot-derived factors such as hemoglobin and iron,⁴ surgical hematoma removal has long been a proposed therapeutic strategy for ICH. However, except in the case of surgery for posterior fossa hemorrhage that is considered life saving, its efficacy has yet to be proven.

Recently, greater effort has been focused on animal modeling in ICH, including predominantly pig models for studies of clot evacuation, and rodent models for pharmacologic strategies. Preclinical work thus far has yielded a number of potential new therapies, and efforts are underway to improve modeling, including the surgical approach, generalizability of models, and outcome endpoints, in the hope that better modeling may lead to more success in translating basic work to the clinical sphere.⁵ Although imperfect, this previous preclinical work has led to the investigation of a number of new strategies which target removal of the hematoma using both macroscopic (surgical) and microscopic (modulating endogenous resolution) methods. The latter includes both upregulating phagocytosis of the hematoma but also potentially modulating hemolysis.

Macroscopic removal methods

Open surgery gives way to minimally invasive clot removal strategies

After the original STICH trial failed to show a survival benefit to open surgical evacuation of all supratentorial ICH,⁶ the STICH II trial also failed to show efficacy for preventing death or disability in a more selective group of only superficial hematomas in conscious patients, though a trend towards clinically significant survival benefit was noted.⁷ Following preclinical studies showing decreased brain edema with minimally invasive clot removal,⁸ and with concerns that the surgical trauma of open surgery may negate the beneficial effects of clot removal, a number of strategies for minimally invasive methods of clot removal are being examined.

Catheter-based

MISTIE II was a phase-II trial examining minimally invasive clot removal via an introducer placed by imaged guidance, with aspiration of clot followed by tissue plasminogen activator (tPA) infusion through a catheter for thrombolysis. Notably, entry into the trial was possible only after blood pressure control and 6-h CT scan demonstrated stability. It showed equivalent clinical safety outcomes between interventional and medical management arms,⁹ though the rate of asymptomatic radiographic hemorrhage was higher in the interventional arm. ICH volume after treatment was significantly lower in the interventional arm, with the mean reduction in hematoma volume of 57% in the interventional arm versus 5% in the medical arm, and though the study was not powered to examine clinical outcomes, there was a non-significant trend towards improved modified Rankin Scale (mRS) in the interventional arm. These promising results supported performance of MISTIE III, a phase-III trial powered to examine efficacy measured by clinical endpoints. This trial has completed enrollment, with results expected soon (NCT01827046). Additionally, therapies to optimize direct tPA delivery are being pursued, including modifications to the tPA molecule which have been shown in preclinical models to reduce tPA-induced neurotoxicity and perihematomal edema while achieving similar efficacy at thrombolysis.¹⁰ Not all results from trials of catheter-based therapies have been positive however. In the CLEAR trial, a phase-III trial of tPA infusion through ventricular drains to lyse ventricular hemorrhage in the subset of ICH patients with intraventricular hemorrhage, intervention failed to improve the number of patients with an mRS of 0–2, although it did reduce mortality.¹¹

Endoscope-based

Minimally invasive removal via image-guided endoscopic removal is another related strategy, differing from catheter-based techniques in that the intent is to remove more of the hematoma during the initial procedure by using a larger endoscope, allowing for prompt removal of nonliquefied clot, as contrasted to the more gradual thrombolysis and irrigation approach of MISTIE III. In one method, an endoscopic sheath with space sufficient to allow microscopic irrigation and aspiration is placed along the long-axis of the hematoma, and using endoscopic microsurgical technique, the hematoma is evacuated, with a goal of 75–80% volume reduction, and hemostasis is achieved. An early-phase, randomized trial compared endoscopic surgery with best medical management and found no difference in clinical safety outcomes between the two groups.¹² As with MISTIE, a 6-h CT scan demonstrating stability was required prior to enrollment, along with BP control in both interventional and medical arms. ICH volume was reduced by a mean of 68% within 29 h of onset in the surgical group, and though not powered to assess clinical outcomes, a non-significant trend was seen favoring the interventional arm that had 43% with a good neurological outcome at six months versus 24% in the medical arm.

Microscopic removal methods

The toxic effects of blood products have been well described,⁴ and strategies to attenuate these effects with drugs such as iron chelators deferoxamine, minocycline, or VK-28 have been studied extensively.^13,14 Deferoxamine is currently in a phase-II clinical trial.¹⁵ Additional adjunct efforts to mitigate secondary brain injury due to ICH have included hypothermia, though effects have been mixed, and preclinical data suggest the rate of late rebleeding may be higher,^16,17 with two clinical trials ongoing currently.^18,19 While immune modulation with potential downregulation of hematoma clearance is a theoretical negative effect of induced hypothermia, augmentation of hematoma clearance through other methods may negate this effect, allowing competing potentially beneficial effects of hypothermia, such as reduced inflammation and excitotoxicity, to predominate. Beyond mitigating the effects of toxic blood products that have already been released, however, a greater understanding of the body’s own clearance mechanisms has led to strategies for enhancing red blood cell (RBC) removal to prevent release of damaging blood products in the first place. Clearance of extravasated RBCs through the body’s own phagocytic system may avoid the release of cytotoxic components during RBC lysis (hemolysis) which have been shown to cause neuronal death and edema. The following methods to enhance endogenous clearance of ICH or limit hemolysis are under investigation:

Enhancement of RBC phagocytosis by peroxisome proliferator-activated receptor-gamma agonists

Proliferator-activated receptor-gamma (PPAR-γ) is a nuclear hormone receptor thought to play an important role as a negative regulator of microglia/macrophage activation.²⁰ PPAR-γ agonists have been shown to downregulate pro-inflammatory responses in preclinical models of other diseases, including a dose-dependent reduction of infarct volume in models of ischemic stroke.²¹ Recently, its role in ICH has been examined, and upregulation of PPAR-γ has been shown to enhance clearance and improve deficits in a mouse model of ICH.^22,23 Its primary mechanism of promoting hematoma clearance is through upregulation of CD36, a scavenger receptor expressed on microglia/macrophages. Additional mechanisms include downregulation of proinflammatory markers²⁴ and increasing antioxidative enzymes,²⁵ both of which are thought to contribute to improved outcomes in preclinical models. The use of PPAR-γ agonists to improve hematoma clearance and clinical outcomes has been examined in an early phase clinical trial using the FDA-approved antidiabetic drug pioglitazone, a known PPAR-γ agonist²¹ (NCT00827892). This trial is complete but results have not yet been reported.

Although the effects of PPAR-γ agonists on hematoma resolution have received the most attention, there may be alternative pathways to accelerate resolution. Thus, interleukin-10 enhances phagocytosis by microglia/macrophage and accelerates hematoma resolution.²⁶ In contrast, toll-like receptor 4 activation, interleukin-1β and tumor necrosis factor-α inhibit phagocytosis and hematoma resolution,²⁷ suggesting inhibition of those targets may enhance resolution.

Inhibition of CD-47 integrin-associated protein “don’t eat me” signal on RBCs

CD-47 is an integrin-associated protein expressed on numerous cells, including RBCs, malignancies and other cells involved with regulation of hematoma clearance. CD-47 serves as an inhibitor of phagocytosis, and its absence in some cells may result in more M2-differentiated microglia/macrophages,²⁸ which are “pro-clearance” and enhance removal of cells. Inhibition of CD-47 using antibodies to enhance autophagy of several malignancies has been shown to inhibit tumor growth, improve survival, and even potentially cure some small tumors in preclinical models.^29,30 CD-47 expression has been examined in preclinical models of ICH, and as CD-47 levels decrease over time in the hematoma, phagocytosis increases and RBCs are cleared more rapidly, though cytotoxic effects may have already occurred.²⁸ Methods of CD-47 blocking have been proposed as a possible therapeutic intervention to enhance hematoma clearance and reduce the amount of toxic blood products released. CD-47 inhibition for malignancy is already under investigation clinically (NCT02367196), though whether such a strategy can be translated to enhance removal of RBCs from the brain remains to be seen.

CD-47 is one of several “don’t eat me”, as well as “eat me”, signals that cells can express, and there are complimentary signaling pathways on phagocytic cells.³¹ Manipulating these other signaling pathways may be a fruitful avenue for ICH research.

Augmenting heme-scavenging systems

Low-density lipoprotein receptor-related protein-1 (LRP1) is a transmembrane receptor with a role in inducing systemic heme clearance via binding to hemopexin-heme complexes and facilitating endocytosis and removal. It plays an important role in heme scavenging after subarachnoid hemorrhage, where it was positively correlated with iron deposition in brain tissue. Augmentation of LRP1 activity with exogenous recombinant LRP1 has been shown to promote heme resolution and improve outcomes after ICH in a preclinical model,³² and holds promise as another strategy for enhancing heme removal pharmacologically.³³

Modulating erythrocyte lysis

RBCs that are not phagocytosed by microglia/macrophages may lyse, releasing potentially harmful components (e.g. hemoglobin and iron) into the extracellular space. The mechanisms that regulate such lysis in ICH are largely unknown. Systemically, complement activation and insertion of the membrane attack complex into the plasma membrane are important in erythrocyte lysis.³⁴ Whether this is the case in ICH is unknown and merits investigation as it could be a method of shifting the balance between RBC lysis and phagocytosis. The current human hematoma evacuation trials present an opportunity to sample hematoma and examine the mechanisms involved in hematoma resolution (e.g. phagocytosis/hemolysis) and their regulation. On MRI, hematomas appear heterogeneous even very early,³⁵ suggesting that the processes involved in resolution may vary regionally and probably temporally.

Conclusions

Proposed strategies to improve clinical outcomes in ICH include enhanced hematoma clearance via macroscopic and microscopic methods. Results of ongoing trials in both strategies are eagerly awaited. A combination of methods also deserves consideration as surgical removal will likely remove the bulk of the hematoma quicker but be incomplete, while enhanced endogenous clearance may be slower but allow complete removal. While this commentary focuses on primary ICH, the utility of these approaches in other forms of cerebral hemorrhage (e.g. hemorrhage following traumatic brain injury) also merits investigation.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors were supported partially by grants NS-007222, NS-090925, NS-091545, NS-093399 and NS-096917 from the National Institutes of Health and the Joyce & Don Massey Family Foundation.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.