Tackling Neuroinflammation After Traumatic Brain Injury: Complement Inhibition as a Therapy for Secondary Injury

Traumatic brain injury (TBI) is a leading cause of mortality, sensorimotor morbidity, and neurocognitive disability. Neuroinflammation is one of the key drivers causing secondary brain injury after TBI. Therefore, attenuation of the inflammatory response is a potential therapeutic goal. This review summarizes the most important neuroinflammatory pathophysiology resulting from TBI and the clinical trials performed to attenuate neuroinflammation. Studies show that non-selective attenuation of the inflammatory response, in the early phase after TBI, might be detrimental and that there is a gap in the literature regarding pharmacological trials targeting specific pathways. The complement system and its crosstalk with the coagulation system play an important role in the pathophysiology of secondary brain injury after TBI. Therefore, regaining control over the complement cascades by inhibiting overshooting activation might constitute useful therapy. Activation of the complement cascade is an early component of neuroinflammation, making it a potential target to mitigate neuroinflammation in TBI. Therefore, we have described pathophysiological aspects of complement inhibition and summarized animal studies targeting the complement system in TBI. We also present the first clinical trial aimed at inhibition of complement activation in the early days after brain injury to reduce the risk of morbidity and mortality following severe TBI. Supplementary Information The online version contains supplementary material available at 10.1007/s13311-022-01306-8.


Background
Traumatic brain injury (TBI) presents a great challenge to public health worldwide. TBI is responsible for over a third of all traumatic deaths, and each year, 80-90.000 new cases of long-term disability due to TBI occur in the USA [1]. The dynamic pathophysiology that evolves over time after trauma to the central nervous system (CNS), consisting of primary injury by the direct traumatic impact, followed by secondary brain tissue injury driven substantially by host responses, makes it a highly complex problem to tackle compared to trauma in other organs [1]. Primary damage develops due to direct and contrecoup mechanical forces on the brain, including damage to neurons, axons, and glial cells and shearing of blood vessels causing hemorrhage [2]. The damage causes a breakdown of the blood-brain barrier (BBB), and changes in blood supply result in mitochondrial and subsequent energy production impairment, release of neurotransmitters and free radicals, immune cell activation and infiltration, apoptosis, and cytokine release [2,3]. This is the initiation of the secondary brain-injury period, which occurs after a latency interval of minutes to several hours, and is probably the most important to focus on. Neuroinflammation develops over hours to days after the trauma and results in edema formation and subsequent increased intracranial pressure (ICP). Increased ICP causes additional impairment of cerebral blood flow and oxygen delivery and may contribute to brain herniation, requiring additional neuro-interventions that complicate the hospital course and final recovery [4].
TBI-induced neuroinflammation has been hypothesized to contribute very substantially to the pathological progression of brain injury, in addition to the primary injury itself [5,6].
It is a complex interaction between the cellular components of the CNS (neurons, astrocytes, microglia), cytokines, and chemokines, in concert with influx of peripheral immune cells. Neuroinflammation is beneficial to promote clearance of debris and regeneration, but it can also cause collateral damage when dysregulated and excessive, leading to secondary brain injury. The fact that the intracranial space is inherently non-compliant, being enclosed by the rigid skull, with the evident advantage to primarily protect the brain, is a clear and unique disadvantage compared with trauma to other vital organs, when swelling occurs. Every organ develops edema in response to significant trauma, but when brain tissue develops edema facilitated by neuroinflammation, the rigid and protective skull bone is a barrier to allow for this swelling. Progressive brain edema and subsequent tissue swelling, that is refractory to ICP-lowering medical ICU treatments, will ultimately elicit brain ischemia due to impeded cerebral blood flow caused by high ICP, unless decompressive craniectomy (DC) is performed in the secondary injury phase after TBI, as has only recently been studied in a well-performed clinical trial [7].
After the (sub)acute phase of injury, a prolonged state of chronic inflammation may linger for years after TBI and predispose patients to develop other neurodegenerative disorders, such as dementia [8,9]. Probably chronic traumatic encephalopathy is caused by a similar pathophysiology which has been shown to occur after recurrent mild trauma, like sport-related injuries [10].
Early attenuation of neuroinflammation is therefore considered an important target for TBI treatment, especially in the early in-hospital phase. Despite the vast amount of research performed to improve our understanding of the pathophysiology in TBI, the field has repeatedly experienced collective failures to translate research from animals to successful therapeutic application in humans [11,12]. This review will summarize the most important drivers of neuroinflammation in TBI and previous trials aiming to attenuate these drivers. The main focus of the review will be the role of the complement system in post-traumatic neuroinflammation and future directions in research on complement inhibition.

Neuroinflammation in the Clinical Setting
The primary effects of moderate and severe TBI include diffuse injuries such as diffuse axonal injury (DAI) and focal brain damage, such as epidural and subdural hematomas (ASDH) and intracerebral hematomas/contusions (tICH). In the first hours after head trauma, expansion of hematomas is the main threat, whereas during the following days, the pathophysiological consequences of neuroinflammation may subsequently increase ICP [13]. International guidelines recommend monitoring of ICP in all patients with severe TBI at high risk of secondary injuries and abnormalities on computed tomography (CT) [14]. Currently, invasive monitoring with an intraparenchymally placed sensor is the most reliable and most applied method to monitor ICP. If ICP can be maintained below a threshold of between 20 and 25 mm Hg with general supportive intensive care treatments, including appropriate pain control, ventilator support, careful fluid, and temperature management, this portends a better prognosis [15]. Management of ICP has evolved towards a "staircase" approach with an escalating treatment intensity, including cerebrospinal fluid (CSF) drainage, deeper sedation, hyperosmolar therapy to dehydrate the brain and prevent a rise in ICP, evacuation of hematomas by craniotomy and, in case of raised ICP refractory to medical managements, including barbiturates, and in the end: DC [13,14]. High mortality has been related to the occurrence of increased ICP, and although a lower mortality has been reported when treating ICP with a DC, these patients will have higher rates of vegetative state and severe neurological disability [7]. Importantly, this current clinical practice has concentrated on trying to mitigate ICP to minimize the extent of secondary injury once this process has already started, rather than focused on managing neuroinflammatory pathways leading to a rise in ICP and thus to prevent secondary injury.

Molecular Mechanism of Neuroinflammation After TBI
Cerebral ischemia and direct traumatic apoptosis after TBI lead to disruption of cerebral energy metabolism due to depletion of cellular adenosine triphosphate (ATP) stores. In this state, cells produce energy via the less efficient pathway of anaerobic glycolysis. Cell metabolic impairment is followed by membrane depolarization and release of excitatory glutamate [16]. Excess accumulation of extracellular glutamate causes neuronal excitotoxicity, which is a major contributor to post-traumatic neurodegeneration. It results in excessive calcium influx within the neuronal cytoplasm due to increased activation of the N-methyl-D-aspartate (NMDA) receptors and voltage-dependent ion channels [17]. The increased intracellular Ca 2+ levels lead to mitochondrial damage, lipid peroxidation, production of free radicals, and activation of caspases and other proteases involved in membrane and nucleosomal DNA changes. This results in synapse elimination and neuronal apoptosis [17,18].
Cell injury in the brain results in the release of intracellular components such as ATP, other damage-associated molecular patterns (DAMPs), and complement components which activate pattern recognition receptors (PRRs) on glial cells. As such, injury of brain cells initiates and perpetuates a post-traumatic neuroinflammatory response [19,20].
Microglia are brain cells acting as support cells for neurons and other metabolic and immunological processes and play a critical role in neuroinflammation as the first line of defense. Microglial cell activation is prominent at the perilesional area and in ipsilateral and contralateral regions of the contusion area [21]. In the post-traumatic phase, microglial cells undergo morphological transformations and changes in their gene expression repertoire to produce a wide spectrum of pro-or anti-inflammatory cytokines (Fig. 1). A proinflammatory M1 phenotype characterized by the production of IL-1β and TNF-α, nitric oxide and reactive oxygen species (ROS), and an anti-inflammatory M2 phenotype characterized by the production of IL-10, IL-13 and transforming growth factor (TGF)-β have been previously described for the activated microglia [22]. However, nowadays is it well recognized that the M1 and M2 states are the extremes of a continuum of activation states and intermediate phenotypes are present [23]. In TBI, microglial polarization is being skewed towards a proinflammatory state varying with increasing time after the trauma [24]. Similar to microglia, a proinflammatory and an anti-inflammatory state have been described for activated astrocytes [25]. These proinflammatory astrocytes secrete many cytokines and other factors that may greatly enhance the inflammatory response [25]. Astrocytes may respond to trauma with proliferation followed by assembly of a dense barrier, known as the glial scar, aiming to protect healthy tissue from nearby areas of neuroinflammation [26].
Important factors in the promotion of neuroinflammation are proinflammatory cytokines. Well known is IL-1β, involved in oligodendrocyte damage and early microglia activation. Levels of IL-1β correlate with Glasgow Coma Scale (GCS) scores, ICP, and outcome [27,28]. Production of IL-1β by glia requires an activated NLRP3 inflammasome, which catalyzes the cleavage of pro-interleukins into their active forms [29]. Deactivation of inflammasomes results in the alleviation of brain edema, reduction of lesion volume, and improvement of long-term motor and cognitive function in experimental TBI in animals [30,31]. TNFα is another key cytokine in post-traumatic cerebral neuroinflammation. Human carriers of two TNF alleles, resulting in higher TNFα production in response to TBI, had a higher probability of poor outcome after TBI [32,33].
Besides their role in neuroinflammation, proinflammatory cytokines may also challenge the integrity of the BBB vasculature. This could lead to vasogenic brain edema and penetration of serum proteins into brain interstitium, such as complement components and fibrin which can further activate glial cells [34]. The BBB consists of endothelial cells, astrocytic endfeet, and pericytes, and its integrity results from the selectivity of the tight junctions between the endothelial cells to restrict the passage of solutes [35]. Disruption of the BBB integrity is primarily caused by damage of these tight junction proteins, especially occludin and claudin-5, and it is further challenged by post-traumatic systemic inflammation which promotes leukocyte chemotaxis and transendothelial migration [36,37].
Leukocyte recruitment into the CNS is mediated by upregulation of endothelial and leukocyte adhesion molecules and activated complement fragments, as discussed later on [38]. Neutrophils are the first-line transmigrated immune cells and can be found as early as 2 h after injury and peak within 24-48 h, before rapidly declining over the following days [39]. Cerebral accumulation of neutrophils has been associated with increased secondary brain damage and adverse outcome [40]. The early neutrophil recruitment is followed by infiltration of lymphocytes and monocyte-derived macrophages. Of note, the neutrophilto-lymphocyte ratio has been reported as an objective, low-cost, and early predictor of inflammation and clinical outcome in TBI patients [41].

The Complement System in TBI-Pathophysiology and Animal Models
The complement system functions to eliminate foreign pathogens and substances and remove debris and apoptotic cells [42]. Complement activation can be mediated in three distinct pathways: the classical, lectin, and alternative pathways ( Fig. 2), all resulting in the formation of the membrane attack complex (MAC). The host is protected against overactivation of the complement cascade by the complement regulatory proteins. These include the C1-inhibitor (C1-INH), which inactivates the C1r, C1s, and MASPs, the decay-accelerating factor (DAF or CD55) which accelerates the breakdown of C3 and C5 convertases, factor H (fH) leading to breakdown of the C3 convertase, membrane cofactor protein (MCP or CD46) aimed to cleave C3b, and CD59 protein which prevents MAC formation.
Complement has a key contribution to pathophysiological events that sequester the initial impact of injury and may radiate damage from the contusion core to the penumbra after TBI. For instance, glial synthesis of complement C3 and clusterin, a regulator of complement, was found in the vicinity of experimental brain contusion [43]. Further, C3d and C9 were localized on neurons in the vicinity of experimental brain contusion, suggesting that these neurons are targets of complement proteins [43].
High-quality preclinical evidence suggests that inhibition of various complement factors can improve neurological performance and reduce inflammation (Table 1). For Fig. 1 Pathophysiology of neuroinflammation after TBI example, our data on the closed head injury model of TBI showed that pharmacological inhibition of MAC formation improved neurological performance in mice by reducing inflammasome activation and preventing microglial activation and axonal damage [44]. In line with these data, mice with a genetic deletion in the CD59a gene, being deficient for a major regulator of MAC formation and therefore resulting in excessive MAC formation, showed increased neuronal cell death and brain tissue destruction following head trauma [45,46]. Facilitation of phagocytosis is another important contribution of complement activation to secondary damage in TBI. Cerebral biosynthesis of most complement proteins is induced in response to trauma and adds to the pool of complement proteins which penetrate the compromised BBB [47]. Intracerebral complement activation leads to the generation of complement opsonins which facilitate the clearance of debris at the site of injury by microglia and macrophages bearing the CR3 receptor [48]. In addition, elevated levels of C1q within the cerebral parenchyma The classical pathway, often referred to as "antibody-dependent," is activated by C1q binding to IgG or IgM antibodies bound to antigens. It triggers autoactivation of the proteolytic activity of C1r, which then cleaves C1s. The C1s component cleaves C4 to C4a and C4b and C2 to C2a and C2b. C4bC2a, known as C3 convertase, cleaves C3 to anaphylatoxin C3a and opsonin C3b. Anaphylatoxins recruit and activate immune cells, whereas opsonins tag target cells or cell compartments such as stressed and salvageable neuronal synapses for phagocytosis by microglia bearing the CR3 receptor The lectin pathway is initiated by binding of Mannan-binding lectin (MBL), ɴ-acetylglucosamine, or ficolins to carbohydrates on the surface of target cells. MBL and ficolins form a complex with Mannan-binding lectin-associated serine proteases (MASPs). These cleave C4 and C2, generating the C4bC2a C3 convertase. Both the classical and lectin pathway are powerful sensors of danger signals and dying cells The alternative pathway is initiated by hydrolyzing inert C3 to C3(H O), which exposes new binding sites for factor B. Factor B is then cleaved by factor D to generate complex C3bBb, the C3 convertase of the alternative pathway. C3 convertases recruit further C3b molecules to form the C5 convertase. The C5 convertase cleaves C5 into the anaphylatoxin C5a and the membrane bound fragment C5b. C5b binds to C6, C7, C8, and C9 to form the membrane attack complex (MAC) which is the end product of the complement cascade. MAC forms pores in the target membrane, inducing lysis or proinflammatory responses  [99] promote the transformation of microglia to the proinflammatory phenotype [49]. Lastly, C1q and C3 play a key role in microglia-mediated synapse elimination which is a prominent neurodegenerative mechanism after head trauma [50]. Activated microglial cells induce a proinflammatory phenotype to astrocytes via secretion of C1q, IL-1α, and TNFα, contributing to the rapid death or functional disability of neurons and oligodendrocytes [25]. Proinflammatory astrocytes, in turn, secrete many complement components, such as C3, that enhance the neuroinflammatory response [25,51].
In addition to the complement components which are synthesized by neurons and glial cells within the CNS [52,53], there is an influx of complement factors from the blood due to breakdown of the BBB after TBI [54,55]. The anaphylatoxins C3a and C5a act as chemoattractants for immune cells expressing the relevant receptors such as the granulocytes which are present within TBI lesions [56]. In addition, in vitro studies showed that C5a activates the expression of beta2-integrin on neutrophils promoting their adhesion to the inflamed endothelium of the BBB and stimulates the secretion of the proinflammatory mediators TNFα and IL1 by human mononuclear cells [57][58][59]. Notably, animal studies showed that genetic deletion of the genes encoding for the C3 or the C5 component or pharmacological inhibition of C3 or C5 resulted in a reduction of neutrophil infiltration, injury size, microglial activation. and brain edema leading to significantly improved neurological outcomes [60][61][62][63][64].

The Interaction of the Complement System with the Coagulation System
In TBI, hemostasis is often derailed, either leading to a hypo-coagulopathic state on one end of the spectrum, causing cerebral bleeding disorders leading to progression of contusions into growing t-ICHs and ASDHs, and to a hyper-coagulopathic state that contributes to ischemic lesions due to (micro)vascular thrombosis in lesioned areas [65]. The intimate interaction and co-evolution of the coagulation system together with the complement system is widely appreciated within the basic science research field. The complement system has been found to increase tissue factor activity, thereby activating the extrinsic coagulation pathway, and form activated thrombin from prothrombin. Moreover, complement factors increase platelet activity and aggregation and prothrombinase activity, including von Willebrand factor and P-selectin. Classical and lectin pathway activation has been reported to be associated with increased odds of venous thromboembolism in the clinical setting [66,67]. Moreover, in sepsis patients, disseminated intravascular coagulation (DIC) was correlated to the degree of complement activation [68]. It is also known that MAC attenuates endothelium-dependent relaxation leading to a hypertensive state [69]. This evidence, Primary injury results in brain tissue destruction, including shearing of blood vessels causing hemorrhage. Within the CNS, complement factors are activated, resulting in activation of both microglia and astrocytes (to M1 and A1 phenotype). Activation of microglia and astrocytes leads to activation of more complement factors and cytokines. Finally, disruption of the blood-brain barrier arises, causing infiltration of neutrophils, lymphocytes, and macrophages from the peripheral hemodynamic system to the CNS interstitium reviewed in [70], suggests that complement overactivation shortly after TBI could potentially lead to increased coagulation activity, i.e., a prothrombotic state often seen days to weeks following TBI. Nevertheless, the correlation between trajectories of complement activity and markers of hemostasis and platelet function, specifically in TBI, warrants more research.

Complement Activation in TBI-Human Studies
In TBI patients, high levels of C4, C3, and MAC have been found in serum [54,[71][72][73][74][75][76], and upregulation of factor B, C3, and MAC was detected in the CSF of severe TBI patients [55,77]. Moreover, increased immunoreactivity was found in resected contused tissue for C1q, C3b, C3d, and MAC within/on neurons located in the penumbra area [78,79]. Intracerebral deposition of MBL, ficolin-2 and 3, and MASP-2 and 3 was found after TBI within the vasculature and in the injured perivascular tissue [79,80]. High levels of complement proteins were strongly associated with lower GCS scores and independently predict mortality or unfavorable clinical outcomes in TBI [73,75]. A proteomics study using human frontotemporal cortex samples showed a consistent overexpression of C4a, C4b, C3, C7, and C9 [81]. More recently, microvesicles and exosomes were analyzed in the CSF of TBI patients and mass spectrometry-based proteomic identification of proteins indicated presence of complement C1q [82]. Furthermore, plasma astrocyte-derived exosomes (ADEs) protein levels of C4b, factor D, Bb, MBL, C3b, and MAC were significantly higher and those of the regulatory proteins CR1 and CD59 lower in the first week of TBI compared to controls [83].
The complement system is further triggered by secondary insults [84]. Expression of complement proteins C3, C8a, and C9 is still increased in the plasma of TBI patients at 1, 3, and 6 months after injury compared to controls, suggesting persistent complement activation during the subacute and chronic phase [85]. Prolonged complement activation has been linked to early-onset cognitive decline, behavior disorders, and predisposition to dementia syndromes like Alzheimer's disease [86].

Clinical Trials to Control Neuroinflammation: What Has Been Tried So Far?
Studies on the dysregulated inflammatory response are important to serve as a roadmap for future clinical trials aiming at "targeted" pharmacological neuroprotection and improved neurological recovery after TBI. Although all of the described interventions within trials up to now (Table 2) have shown to be effective in preclinical and small single center phase II trials, successful translation to phase III clinical trials showing efficacy of these treatments has not yet been accomplished. Multiple explanations have been proposed to explain these failures. First, the heterogeneity of the TBI population and the large treatment variation in the management of TBI indicate that large sample sizes are warranted to achieve any statistical significant difference between groups [87,88]. Second, there is a growing recognition of the problem of age and sex bias on the outcomes in neurotrauma research, as for example, fewer women than men are recruited in clinical trials (Table 2) [89,90]. Third, most trials focused on "delayed" outcome metrics, such as the Glasgow Outcome Scale Extended (GOSE) at 6 months, as primary endpoint, whereas animal models focus on the direct impact of a therapy on microglial activation, edema formation, or neuronal death.
Last and most importantly, most trials did not focus on a targeted dysregulated part of neuroinflammation after the first TBI impact. The steroid trials, as described in Table 2, show that broad inhibition of the immune response can be deleterious after TBI, and therefore a more targeted approach focused on a specific neuroinflammatory pathway and during a limited period of time may be more successful in improving outcomes [91,92]. Current and future clinical trials aiming to reduce secondary brain injury should focus on targeting well-defined specific pathways with a closely related endpoint to the therapeutic mechanism of action to test efficacy. This should be based on both thorough and sufficient preclinical testing in multiple injury models (including different age ranges and sex), together with a detailed insight into the most important drivers of TBI pathophysiology for each individual patient.

Complement Inhibition in the Clinical Setting-Future Directions
Currently, no clinical trials are present in literature aiming to inhibition complement activation in brain injury. Only a few drugs, C1-esterase inhibitors (C1-INH), Cinryze, Berinert and Ruconest, and C5-inhibitors, eculizumab and ravulizumab, are approved complement inhibitory drugs, but many others are in clinical development. The indications for the current approved drugs are hereditary angioedema (C1-INH) and paroxysmal nocturnal hemoglobinuria, atypical hemolytic uremic syndrome, and neuromyelitis optica (C5). A trial with C5-antibodies in patients with aneurysmal subarachnoid hemorrhage (SAH) is now recruiting participants [93]. As it has been reported that inhibition more upstream in the complement cascade is necessary to prevent the amplification of a feedforward mechanism of neuroinflammation that persists    [171] throughout the chronic phase [94], C1-INH might be more effective to attenuate complement overactivation. C1-INH is a potent multi-target serpin, which effectively inhibits activation of the classical, lectin, and alternative pathways [95][96][97]. Administration of C1-INH in animal models showed reduced contusion volume and brain water content and improvement of cognitive and motor function [98,99]. In addition to its role in complement inhibition, C1-INH is a known inhibitor of FXIIa, FXIa, FXII, thrombin, kallikrein, HMWK prekallikrein complexes, and plasmin which inhibit fibrinolysis, contact activation, and coagulation [100,101]. Efficacy and an excellent safety profile of high doses of C1-INH have been reported in off-label trials in sepsis and ischemiareperfusion injury patients [102,103]. Therefore, we are currently recruiting TBI patients in the Complement Inhibition: Attacking Overshooting inflammation @fter Traumatic Brain Injury (CIAO@TBI) trial to assess the safety and efficacy of C1-INH in this patient population [104]. In this trial, patients will be randomized to either receive one dose of 1600 IU C1-INH or a placebo injection. The primary outcome is the therapy intensity level scale that measures all ICP-directed interventions. This study will provide insight in the promising role of complement inhibition in brain injury. In the meantime, more research is warranted towards defining the inflammatory phenotypes of our patients based on injury characteristics (e.g., age, sex, and injury severity), imaging, and biomarkers to eventually being able to target inflammation with personalized immunomodulatory treatments,

Conclusion
Neuroinflammation is one of the "nonsurgical" key drivers causing secondary brain injury after TBI. Attenuation of the inflammatory response is a potential therapeutic target. This review covers the most important neuroinflammatory drivers resulting from TBI and summarizes the clinical work performed to date directed to attenuate neuroinflammation. The complement system play an important role in the pathophysiology of TBI, and therefore therapies targeting this pathway might contribute to future targeted therapy, currently evaluated in a clinical trial.

Acknowledgements
The authors would like to thank Fleur Fisher for designing the figures. The authors would like to thank Jan Schoones for his help with conducting the literature search.

Declarations
Ethics Approval and Consent to Participate Not applicable.

Consent for Publication Not applicable.
Competing Interests FB is founder of Complement Pharma BV, and KF is paid consultant for Complement Pharma BV. All other authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.