Abstract
Traumatic brain injury (TBI) is a major contributor to disability and death worldwide, and manifests in cognitive, behavioral, and motor impairment. Although there have been numerous pre-clinical studies that have identified promising pharmacologic treatments, to date, all Phase III clinical trials have failed. Thus, this is a priority area for ongoing research and development. Treatment strategies have traditionally focused on neuroprotection of the injured brain to reduce secondary injury, neuronal death, and lesion size. The aim of this minireview is to describe the secondary injury pathophysiology of TBI and give an examination of key targets of neuroprotection, select Phase III trials that have been undertaken, and future possibilities for successful drug development.
SIGNIFICANCE STATEMENT This minireview provides an up-to-date summary of the key Phase III clinical trials that have been undertaken in the development of a neuropharmacological treatment for traumatic brain injury. The article discusses the key targets for treatment, the potential reasons for the lack of translation of promising pre-clinical compounds, and the most promising avenues for future development.
Introduction
Traumatic brain injury (TBI) is defined as a disruption in the normal function of the brain that can be caused by head impact or penetrating head injury (National Center for Injury Prevention and Control, 2003). TBI is a major contributor to disability and death worldwide (Chauhan, 2014), with injuries most commonly sustained from falls and motor vehicle accidents (Faul et al., 2010), and also associated with other groups, including contact sport athletes, military personnel, victims of physical violence, or individuals with mental disorders. Injuries are triaged into mild, moderate, or severe categories based on medical imaging pathology and clinical tests, such as the Glasgow Coma Scale (GCS). While moderate and severe injuries can be defined based on macroscopic medical imaging characteristics, mild TBI (mTBI) may be asymptomatic and undetectable in standard clinical imaging such as computed tomography (CT), or magnetic resonance imaging, leading to an elusive diagnosis (Asken et al., 2018).
Injury Mechanisms
Injury mechanisms are heterogenous and may result from blunt force trauma or whiplash style injuries in the absence of direct contact. Regardless of the injury mechanism, the resulting acceleration and compression forces on the brain initiate multiple pathologic cascades. A construct for describing the pathophysiology of injury differentiates the phases into primary and secondary damage. Primary injury refers to the macroscopic damage that happens instantaneous to injury (Saatman et al., 2008). Focal neuronal damage could manifest as a hemorrhage, shear, or tear as a result of the physical force of insult (Park et al., 2008). These injuries may result in systemic alterations, such as edema and elevated intracranial pressure that restrict cerebral blood flow and compromise metabolic processes, ultimately resulting in ischemia. These processes are the initiating factors that lead to secondary injury.
Secondary injury mechanisms occur in the acute and chronic phases, and initiate concurrent and self-exacerbating molecular and cellular processes that affect brain function. These responses are triggered by rapid axonal stretching, which leads to membrane deficits known as excitotoxicity, involving rapid synaptic influx and inhibited reuptake of neurotransmitters and structurally related amino acids (Blaylock and Maroon, 2011). In response to this chemical disruption, unregulated ion shifts occur across lipid membranes, with potassium leaking out and calcium moving into the neuron (Giza and Hovda, 2014). This results in a damaging accumulation of calcium within the axon, which initiates the release of excitatory neurotransmitters, most notably glutamate. As excess glutamate binds to N-methyl-D-aspartate (NMDA) receptors further depolarization occurs, resulting in a greater influx of calcium ions, leading to suppressed neuronal glucose metabolism (Giza and Hovda, 2014). Membrane pumps increase activity to attempt to restore the balance of ions, which increases the consumption of glucose and depletes these energy stores, leading to mitochondrial calcium influx and a switch from oxidative metabolism to anaerobic glycolysis with excessive production of lactate, which can result in acidosis and edema (Barkhoudarian et al., 2011). In response to the increase in agonists in the synaptic cleft, excessive activation of ionotropic glutamate receptors attempt to clear these metabolites, leading to neurotoxicity and the characteristic breakdown and loss of postsynaptic structures, including dendrites and cell bodies (Lewerenz and Maher, 2015). Calcium influx also triggers the release of calpains, which begin the proteolytic breakdown of neurofilament proteins and microtubule disassembly (Blennow et al., 2012). This breakdown of structural proteins results in swelling, degeneration, reduced axonal transport ability and the subsequent accumulation of organelles at end bulbs (Chen et al., 2004). In focal lesions within white matter, for example in situations of microhemorrhage, a high number of axons and oligodendrocytes are damaged and may be lost due to caspase-3 apoptosis (Flygt et al., 2013), leading to the degradation and collapse of myelin sheaths. A significant component of white matter injury also involves demyelination of intact axons (Mierzwa et al., 2014), which leaves the axons at reduced function and vulnerable to further damage. In concert with these secondary injury processes are inflammatory cascades triggered by microglia due to mechanical and chemical neurologic injury (Loane and Kumar, 2016). The mechanism of neuroinflammation involves the production of proinflammatory cytokines as well as reactive oxygen species (ROS) (Schmidt et al., 2005; Glass et al., 2010), which perpetuate neuronal cell death that drives neurodegenerative diseases. Through these actions, inflammation has been shown to have a strong relationship in enhancing excitotoxicity in axons (Morimoto et al., 2002), and is also intrinsically linked with oxidative and nitrosative stress mechanisms. These post-TBI cascades result in blood brain barrier damage, hemorrhage, increased intracranial pressure, abnormal cerebral flow, hypoxia/ischemia, impaired metabolism, apoptosis, demyelination, and progressive atrophy of white and gray matter. These features contribute to collective cell death, neurodegeneration, and functional impairment seen following TBI.
Development of Agents
TBI manifests in cognitive, behavioral, and motor impairment. Current research has been unable to develop effective treatments of TBI, which is an urgent clinical priority (Ojo et al., 2016). Pharmacological treatment provides an attractive avenue for therapy; however, there are no agents that have been approved by regulatory bodies for acute or chronic recovery following TBI, despite several compounds showing promise in preclinical trials and reaching Phase II or III human trials. Treatment strategies have traditionally focused on neuroprotection of the injured brain to reduce secondary injury, neuronal death, and lesion size (Xiong et al., 2013). It is in the protection of injured tissue that many clinical trials over the last four decades have sought to provide effect, however no treatment options have been shown to effectively mitigate these negative neurologic outcomes. Table 1 provides a summary of Phase III and IV clinical trials with results reported in the literature. This list was compiled by searching clinicaltrials.gov for the term ‘traumatic brain injury’ with the modifiers: interventional studies; completed, suspended, terminated, unknown status studies; and phase 3 and phase 4 studies. This search returned 115 results, with 61 studies excluded due to being interventions other than pharmacotherapy, and 19 studies excluded due to no available results. Thirty-five final studies were included in the summary.
Phase 3 TBI pharmacology trials
In the absence of TBI treatments, current clinical practice is focused on symptom management. Presently, there is limited evidence in the efficacy of strategies to reduce symptoms following TBI. However, there are selected studies that have shown benefit, and Table 2 provides a list of agents that are used for varied purposes. It is outside the scope of this review to provide a thorough overview of these symptom management strategies, and the reader is directed to reviews on this topic (Talsky et al., 2010; Scher et al., 2011; Diaz-Arrastia et al., 2014; Bhatnagar et al., 2016).
Pharmacological agents for symptom management
Conventional de novo production of novel pharmacotherapies is a costly and time-consuming process, and despite the enormous investment made in TBI agent discovery, to date, there have been none approved for clinical practice. Several factors have limited the development of agents for TBI. Key among these is the complexity and heterogeneity of structures that are affected by injury, limited prognostic biomarkers and criteria to assess injury severity and treatment response, and differing time scales of rodent and human pathophysiology following injury (Agoston et al., 2019). A detailed description of pharmacology optimization for TBI has been comprehensively described recently (Poloyac et al., 2020). An attractive alternative to discovery of new treatment compounds is drug repurposing, which is the use of existing compounds with known mechanisms of action in a new treatment context (Papapetropoulos and Szabo, 2018). As existing drugs have well-defined safety profiles in humans, the time taken for approval and initiation of human trials may be reduced, and the potential for adverse events can be predicted and minimized. There has been an increasing number of drugs targeted for repurposing in neurodegenerative disease in recent years, and many of the compounds that are discussed in this review are promising repurposing candidates.
For neuroprotective compounds that act upon secondary injury processes, there are several agents that have been investigated in preclinical and clinical studies. This review aims to provide detail of agents that have progressed to human trials and their mechanisms of action. The following is a description of key targets of secondary injury neuroprotection and drugs directed at these actions that have undergone human trials.
Excitotoxicity
The neurotransmitter glutamate has the principle excitatory role in the central nervous system. Glutamate’s mechanism of action involves binding to metabotropic or ionotropic glutamate receptors, which have classes NMDA, kainite, and AMPA (α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) receptor subtypes. NMDA receptors are most prominent in the hippocampus and have been shown to be highly influential in long-term potentiation, which is hypothesized to be the molecular foundation for learning and memory (Miyamoto, 2006). Following TBI, increased glutamate concentrations initiate the excitotoxic cascade that leads to overload of calcium and neuronal injury. Human TBI studies have described glutamate increase in the CSF peaking at 48 hours, with raised levels persisting to 9 days (Zhang et al., 2001). However, glutamate CSF levels do not show correlation with severity of injury, GCS, or survival (Brown et al., 1998).
Several studies have described the role of NMDA antagonists in reducing the effects of excitotoxity following head injury; however, early clinical trials of midafotel, cerestat, and selfotel did not show clinical benefit (Ikonomidou and Turski, 2002). Amantadine is an NMDA receptor antagonist that has undergone Phase III investigation in severe TBI, in which functional recovery was accelerated in the amantadine group compared with placebo during the treatment period (Giacino et al., 2012). Two previous RCTs have also suggested that amantadine may be beneficial, although small sample sizes and other methodology limitations prevented applicability of findings (Schneider et al., 1999; Meythaler et al., 2002). The synthetic cannabinoid derivative dexanabinol also has NMDA antagonist actions but was ineffective in a Phase III clinical trial (Maas et al., 2006).
While the neurotransmitter serotonin (5-hydroxytryptamine) is well-researched in psychiatric diseases, it has also been implicated in TBI pathophysiology, with several studies examining the effect of serotonin modulation in a neurotrauma capacity. Human TBI studies have demonstrated that extracellular serotonin levels peak at 10 minutes and are elevated at 90 minutes after injury (Busto et al., 1997), and it may be that pharmacological antagonism may reduce excitotoxity. Serotonin receptor activation stimulates calcium influx and neurotransmission by gamma amino butyric acid (GABA) (Turner et al., 2004). By administering serotonin antagonists, GABA inhibition may protect against neuronal damage in the hippocampus and improving cognitive function (Kline et al., 2001; Cheng et al., 2008). Serotonin antagonists are widely used in clinical areas in the treatment of nausea and vomiting, with Food and Drug Administration-approved options, such as ondansetron. With a well-defined safety profile and pharmacokinetics, and a multifaceted mechanism of action, serotonin receptor antagonists may hold promise as TBI treatments. Serotonin’s role in psychiatric diseases also makes it a valuable agent in the treatment of TBI-induced behavioral complications (Fann et al., 2009). Inhibition of AMPA receptors may be another avenue for therapeutic effect. Following TBI, AMPA receptor affinity for calcium is altered, and neuroprotective effects have been shown following downregulation of these receptors following injury (Belayev et al., 2001; Furukawa et al., 2003). These receptors provide attractive therapeutic targets due to their wide expression and varied roles, and continued development of these concepts is warranted.
Calcium Channel Receptors
Many of the secondary injury mechanisms of TBI are influenced by neuronal calcium influx, contributing to a toxic intracellular environment (Giza and Hovda, 2014). The hypothesis driving this response involves the increase in glutamate and acetylcholine binding to calcium-permeable ligand-gated ion channels and the uninhibited calcium flow that overloads and damages the neuron. Also implicated in the calcium overload process is the role of intracellular stored calcium release in response to G-protein coupled metabotropic receptor stimulation and ineffective calcium buffering systems that fail due to overload (Mattson, 2007). These calcium accumulation mechanisms persist to seven days following injury (Deshpande et al., 2008), and ineffective calcium buffering response has been shown to last to 30 days (Sun et al., 2008).
There have been several clinical trials examining the therapeutic benefit of calcium channel blockade strategies following TBI. The most common agent that has been investigated is nimodipine; however, a Cochrane review concluded this agent has no appreciable benefit in TBI treatment (Langham et al., 2003). This review did report that a subgroup of patients with subarachnoid hemorrhage showed some improvement; however, a more recent meta-analysis including additional studies did not find beneficial effects of nimodipine in the subarachnoid hemorrhage group (Vergouwen et al., 2006). From these results, development of nimodipine and similar classes of drugs are not likely to be pursued in development.
Apoptosis Inhibitors
The combination of secondary injury processes results in cellular and physiologic disturbances involving brain edema, impaired cerebral perfusion, and increased intracranial pressure (ICP), which cause neuronal apoptosis (Nortje and Menon, 2004). Progesterone is a female reproductive hormone that has demonstrated neuroprotection by downregulating apoptosis and may also have effect in reducing edema, dampening inflammation, removing free radicals and products of lipid peroxidation, and inhibiting mitochondrial damage (Sayeed et al., 2009). In humans, an initial Phase II clinical trial reported that progesterone-treated patients displayed a reduced mortality rate compared with the group receiving placebo (Xiao et al., 2008). However, two Phase III clinical trials have been conducted with disappointing results. The ProTECT III trial involved intravenous administration of progesterone to patients with a GCS below 12. This study was terminated early due to a lack of treatment effect; however sub-analysis showed for the group with GCS of 9 to 12, there were improved outcome scores and disability rating scores (Wright et al., 2007). The SyNAPSe study found no clinical improvement in severe TBI patients administered progesterone (Skolnick et al., 2014). Erythropoietin is another endogenous hormone that has demonstrated neuroprotective effects which may be enacted through apoptosis inhibition via downregulation of caspase-3 expression (Yatsiv et al., 2005). Two Phase III clinical trials have investigated the neuroprotective effect of erythropoietin, with no demonstrable benefit in mortality in moderate and severe TBI patients (Nichol et al., 2015) and neurologic outcomes in severe TBI patients (Robertson et al., 2014).
Inflammation Modulators
Microglia provide defense in situations of neuroinflammation with the release of pro- and anti-inflammatory cytokines (Hiskens et al., 2021b). However, repetitive or excessive trauma can perpetuate pathologic microglial involvement that negates potential reparative functions and leads to chronic neurodegeneration (Hiskens et al., 2021c). Therefore, microglial inhibition is a strategy that has been hypothesized to have neuroprotective effects in these situations. Minocycline is a second-generation tetracycline that has been shown to inhibit microglial activation and proliferation. Minocycline has well-known anti-inflammatory properties (Shochat and Abookasis, 2015), but has also demonstrated anti-apoptotic and antioxidant effects (Garrido-Mesa et al., 2013) and the ability to prevent excitotoxic damage to neurons (Alano et al., 2006; Plane et al., 2010). The neuroprotective effects of minocycline in a range of neurologic conditions have been linked to reduced inflammation and oxidative stress that have improved pathologic tau formation (Garrido-Mesa et al., 2013; Budni et al., 2016; Cankaya et al., 2019), and these effects provide impetus for exploring minocycline as a treatment following TBI. Clinical TBI studies have found minocycline treatment reduced serum markers of inflammation (Casha et al., 2012), while in moderate-to-severe TBI minocycline improved microglial activation but increased neurodegeneration evidenced by brain atrophy and neurofilament light levels (Scott et al., 2018). Another tetracycline derivative, doxycycline, has been investigated in a Phase III clinical trial of moderate to severe TBI (Mansour et al., 2021). Doxycycline administration for the first seven days following injury was associated with a significant reduction in neuron-specific enolase level, a biomarker of neuronal damage. Additionally, the doxycycline group had favorable short-term GCS score outcomes. Considering these positive findings and the recognized safety of these tetracycline derivatives in clinical practice, future clinical trials with larger cohorts and long-term outcomes may be warranted.
Prostaglandin Inhibitors
Activated microglia produce prostaglandins, which are propagators of inflammation and secondary injury mechanisms in TBI. Prostaglandins are derived from the cyclooxygenase (COX) family of enzymes, which exist in two isoforms, COX-1 and COX-2 (Kelso et al., 2009). Due to basal levels of COX-2 in the glutamate neurons of the amygdala, cortex, and hippocampus, it has been hypothesized that prostaglandins derived from COX-2 may be involved in NMDA receptor synaptic signaling (Chen et al., 2002). Prostaglandin production is reduced with administration of two classes of anti-inflammatory compounds, COX inhibitors and glucocorticoids, which also reduce the production of free radicals, thromboxanes, and prostacyclins (Shapira et al., 1988). In exacting a neuroprotective effect, COX inhibition has been shown to decrease damage following focal ischemia (Candelario-Jalil et al., 2002), and provide improvement in conditions of neurotoxicity, neurodegeneration, and demyelination (Blasko et al., 2001; Pompl et al., 2003; Salzberg-Brenhouse et al., 2003; Shibata et al., 2003; Hiskens et al., 2021a). COX-2 is elevated in the cortex and hippocampus following TBI (Strauss et al., 2000). Interestingly, COX-2 inhibition influences a variety of diverse secondary injury mechanisms that are induced by repetitive mTBI, such as modulation of microglial cells and astrocytes (Dehlaghi Jadid et al., 2019). While these studies demonstrate promising therapeutic effects in TBI, other studies have found no benefit in cognition (Dash et al., 2000) or brain edema and lesion size (Hickey et al., 2007). Therefore, there are a wide range of pathways that may be protected by COX inhibition, and ongoing research is required to determine the capacity of COX inhibition as a neuroprotective strategy in TBI and the optimal circumstance for use.
COX-2 expression is also blocked by glucocorticoids without involving COX-1 expression. However, glucocorticoids may produce serious side effects, and there may be limited efficacy in alleviating inflammation following head injury (O'Callaghan et al., 1991) with clinical trials showing no effectiveness (Alderson and Roberts, 1997). Glucocorticoids, such as methylprednisolone and dexamethasone, have undergone investigation as TBI treatments. Dexamethasone has shown edema-mediating action at early administration related to aquaporin-1 regulation (Tran et al., 2010) and microglial inhibition (Zhang et al., 2007). Clinical trials have not shown success with corticosteroid treatment, and the MRC CRASH trial was prematurely ceased due to increased mortality in the first two weeks following TBI in the methylprednisolone group. Secondary to this evidence, corticosteroids are not currently being pursued as treatment of TBI (Carney et al., 2017).
Acetylcholinesterase Inhibitors
Acetylcholine (Ach) is rapidly metabolized by acetylcholinesterase (AChE), and AChE inhibitors prohibit breakdown in the synaptic cleft and are used as Food and Drug Administration-approved treatments for Alzheimer’s disease-related dementia. AChE does not rely on production of ATP, and therefore, unlike glutamate uptake during TBI, is not similarly susceptible to the energy crisis perpetuated by secondary injury cascades. Despite this, studies have identified an alteration in AChE action following TBI, with increased expression in the forebrain and decreased expression throughout the hippocampus and motor cortex in the first hours following injury that return to baseline levels by 72 hours (Donat et al., 2007). AChE has been targeted in three Phase 3 clinical trials of the drug rivastigmine in moderate and severe TBI. The initial study assessed drug safety and cognitive function after 12 weeks, with rivastigmine treatment no different from placebo (Silver et al., 2006). The follow-up 26-week extension found rivastigmine was safe for this duration of intake, but measures of cognitive function were not different between treatment and control (Silver et al., 2009). A recent Phase III trial of a 12-week rivastigmine transdermal patch therapy also found no difference in outcomes between treatment and placebo groups (Brawman-Mintzer et al., 2021). Other AChE inhibitors galantamine, memantine and donepezil have also been investigated as treatment following TBI (Tenovuo, 2005). Donepezil has also shown benefit in TBI patients in the sub-acute and chronic phase of recovery, with improved attention and short-term memory (Walker et al., 2004; Khateb et al., 2005). A recent systematic review of memantine in TBI patients cited evidence of improved serum neuron-specific enolase levels and improved GCS; however, overall, there was a lack of high quality RCTs demonstrating direct cognitive improvement (Khan et al., 2021). The potential shown in the early studies of these other AChE-inhibiting compounds suggests value in continued exploration.
Mitochondrial Stabilizers
TBI pathophysiology reduces the integrity of the mitochondria, resulting in calcium buffering deficits, production of ROS, and apoptosis initiation. Damage to the mitochondria occurs in the immediate aftermath of injury, which leads to oxidation of mitochondrial proteins and impaired electron transport chain processes, which ultimately result in systematic energy depletion (Opii et al., 2007). The key process is the increased permeability of the inner mitochondrial membrane to small solutes, which results in the rupturing of the outer mitochondrial membrane. This rupture causes leaking of toxic proteins and cellular components into the cytosol, which can precipitate necrosis. The mitochondrial permeability transition pore influences the movement of these solutes between the inner and outer mitochondrial membranes. Therefore, protection of mitochondrial bioenergetics may be possible with the inhibition of the mitochondrial permeability transition pore. The drug cyclosporine facilitates this inhibition (Scheff and Sullivan, 1999), and there is evidence for decreased production of ROS and cortical tissue sparing in treating TBI (Sullivan et al., 2000b; Alessandri et al., 2002). Human trials have established cyclosporine’s safety profile and have shown potential for a positive treatment effect (Hatton et al., 2008; Mazzeo et al., 2009). However, some studies have not demonstrated cyclosporine efficacy (Abdel Baki et al., 2010), and research is ongoing.
Oxidative Stress
The reactive ROS and RNS free radicals produced by mitochondrial dysfunction after TBI contain oxygen or nitrogen atoms with a high affinity for accepting electrons from other compounds. The formation of ROS and RNS occurs as a part of the standard processes of cellular respiration, whereby homeostasis is maintained by the scavenging and inactivation of antioxidants, including superoxide dismutase, catalase, glucose-6-phosphate dehydrogenase, glutathione, glutathione-S-transferase, glutathione peroxidase, and glutathione reductase (Ansari et al., 2008). However, TBI triggers reactive species production beyond the clearing capacity of CNS antioxidants. The first step in this process is the formation of superoxide radicals from the reduction of oxygen, which initiates a biochemical cascade leading to development of hydroxyl radicals. There is an immediate increase in hydroxyl radicals after TBI that tapers off within the first 60 minutes of injury (Hall et al., 1993). This increase reflects the alteration of mitochondrial respiration, involving oxidation of mitochondrial proteins and impaired electron transport chain activity of pyruvate dehydrogenase and complex I and complex IV (Opii et al., 2007). The endogenous antioxidant melatonin has been shown to increase in the CSF at 48 hours, indicating an attempt to upregulate endogenous antioxidant defenses to compensate for the increase in free radicals post-TBI. However, clinical trials of compounds to scavenge free-radicals and inhibit lipid peroxidation have not been successful. A recent clinical trial of children with persistent post-concussion symptoms found that melatonin was not more effective than placebo for improving symptoms (Barlow et al., 2020). Formulations that introduce the antioxidant superoxide dismutase also did not show benefit in multicenter clinical trials (Young et al., 1996; Marshall et al., 1998), although nanotechnology delivery mechanisms of superoxide dismutase and catalase into the brain have shown potential (Reddy and Labhasetwar, 2009; Brynskikh et al., 2010). Superoxide dismutase and catalase show prolonged decreases in the cortex and hippocampus following TBI (Ansari et al., 2008), and the optimal time for treatment remains to be confirmed.
Treatment via lipid peroxidation inhibition is another attractive option, with the goal of preventing cellular membrane damage by oxidative products. Early clinical trials of this strategy were unsuccessful, however the compound U-83836E has shown the ability to protect cellular proteins and lipids from ROS and RNS while assisting mitochondria function and calcium buffering (Mustafa et al., 2010). Future work will examine if lipid peroxidation inhibition is the optimal strategy in combating free-radical driven damage (Hall et al., 2010).
Dietary Compounds
Dietary options are an attractive treatment alternative due to the ease of delivery and the reduced potential for side effects compared with xenobiotic compounds. Most dietary treatment options involve the administration of a supplement that is endogenously produced. This section examines the dietary compounds that have received the most research attention.
Creatine is endogenously synthesized and primarily found in the nervous system and skeletal muscle, where phosphorylation of creatine provides an instant reserve of ATP for use when energy stores are diminished. Creatine supplementation is commonly used by athletes as a strategy for decreasing fatigue during high-intensity bouts of exercise. As mentioned earlier, mitochondrial damage induced by TBI leads to impaired energy metabolism and reduced production of ATP, leading to destruction of cells dependent on ATP. Assisting ATP availability and production using creatine supplementation has shown potential in pre-clinical studies (Sullivan et al., 2000a). A clinical trial of children and adolescents who had experienced a TBI showed improved clinical functioning (Sakellaris et al., 2006).
The omega-3 polyunsaturated fatty acids eicosapentaenoic acid and docosahexaenoic acid (DHA) are primarily consumed as fish oils, however western diets are often deficient. They are involved in maintaining cell membrane structural integrity by regulating membrane fluidity and thickness, cell signaling, and mitochondrial function (Salem et al., 2001). In the brain, DHA is present in very high levels, particularly in neuronal synaptic membranes and vesicles (Dyall and Michael-Titus, 2008). Omega-3 fatty acids regulate neurologic activity both directly and indirectly through processes including: 1) adjusting release of dopamine (Zimmer et al., 2000), serotonin (Kodas et al., 2004), and acetylcholine (Aïd et al., 2003); 2) inhibiting some signal transduction pathways (Mirnikjoo et al., 2001); 3) changing expression rates of genes affecting membrane proteins, synaptic proteins, regulatory kinases, receptors, energy metabolism, signal transduction, and lipid metabolism (Hamilton et al., 2000); 4) neuronal anti-apoptotic effects due to phosphatidylserine accumulation (Dyall and Michael-Titus, 2008); and 5) production of eicosanoids to be used in the inflammatory response (Calder, 2006). Therefore, supplementation may provide an avenue for neuroprotection in these processes. Despite omega-3 demonstrating positive effects in preclinical models of TBI (Bailes and Mills, 2010), the benefit of omega-3 supplementation has not been demonstrated in human injury. Only one clinical trial, involving American football athletes, has evaluated the effect of DHA on concussion (Oliver et al., 2016). This study measured the response of serum neurofilament light weekly throughout the season in groups receiving differing doses of DHA. While the athletes taking placebo had elevated neurofilament light levels toward the end of the season, indicative of cumulative neurologic damage (Hiskens et al., 2020), the groups receiving higher dosage of DHA sustained more concussions. Examination of the pathways of this injury mechanism are warranted before progression to future clinical trials.
Discussion
The TBI-induced dysfunction of neuronal membranes and ischemic injury leads to disruption of calcium homeostasis, stress to the endoplasmic reticulum, inflammation, and neuronal cell death. As highlighted in this review, these compounds directed at these mechanisms have undergone positive preclinical studies that have been followed up with Phase II and III clinical trials resulting in no therapeutic effect in treated patients. This disconnect between promising therapeutic effects shown in preclinical studies and a lack of effect in human clinical trials is highlighted by high profile negative studies involving progesterone and erythropoietin (Robertson et al., 2014; Skolnick et al., 2014). There is also evidence that symptom management drug therapies may have a deleterious effect on recovery following TBI. The Na+ channel blocker dilantin is used in the initial days following injury for antiseizure prophylaxis; however it has been shown as ineffective in decreasing seizure rates and may inhibit functional recovery (Bhullar et al., 2014). The use of the antipsychotic haloperidol has also been shown to provide limited improvement of outcomes and to result in further impaired cognitive function (Bellamy et al., 2009). Another example is thioridazine, which has anticholinergic properties, and discontinuation of this drug has resulted in improved cognitive function in some TBI patients (Stanislav, 1997). Despite these setbacks, concerted research effort continues toward the discovery of drug treatments.
In addition to the drugs assessed in this review, many more compounds have been investigated in animal models. These preclinical studies allow investigation of drug therapeutic window, dosing, pharmacodynamics, pharmacokinetics, and ability to cross the blood brain barrier. They provide a platform for controlling and analyzing variables to screen intervention response to identify options for development. Animal models allow detailed examination of injury pathophysiology, and as TBI involves heterogeneous mechanisms which can confound analysis, animal studies provide greater control over impact variables to assist in identification of injury and disease processes by controlling parameters such as genetics, injury site, and degree of injury (Hiskens et al., 2019). However, despite the potential demonstrated in animal model TBI studies several factors may limit the translation of this data. These include differing injury methodology in the animal and clinical studies, inadequate evaluation of pre-clinical data, and design limitations of the clinical trial. Obstacles in the development of TBI pharmacotherapy include the selection of species, sex and age of the animal, and the choice of injury model that replicates parameters of impact, recovery, and functional outcomes. These animal models primarily examine the effects of site-specific contusion or diffuse axonal injury, and do not involve the concurrent hypoxic, ischemic, or systemic injuries that are seen in human TBI. Additionally, animal TBI studies primarily use short-term measures of recovery and function for evaluation (Gold et al., 2013), while clinical trials frequently measure disability or mortality at 3 or 6 months. In these ways, there are fundamental differences in animal and human TBI trials that may underpin the difficulty in translating to successful treatments. To address these translational limitations, ample preclinical data should be generated from complimentary TBI models to ensure that an adequate understanding of administration dose, route, and timing is gained prior to initiating clinical trials. A novel approach to this is Operation Brain Trauma Therapy (OBTT), a multicenter pre-clinical severe TBI therapy and biomarker screening consortium (Kochanek et al., 2016). OBTT uses three rat TBI models, the fluid percussion injury, controlled cortical impact, and penetrating ballistic brain injury models to assess a range of corresponding clinical phenotypes, and progresses promising findings to a gyrencephalic animal model for further evaluation. This approach maximizes the rigor of the pre-clinical testing phase by adhering to a consistent protocol of blinding, randomization, testing, and data analysis to reduce the risk of inter-site variability that can occur from one study to another. OBTT targets two hypotheses, that a therapy showing beneficial outcomes in a variety of pre-clinical models will have enhanced potential for benefit in human clinical trials, and that individual TBI models represent distinct clinical phenotypes and will benefit from individual therapeutic approaches. While OBTT focuses on the development of acute therapies for severe TBI, this pre-clinical consortium approach would benefit other branches of TBI research, such as repetitive mild TBI, individual TBI phenotypes, or therapies for chronic treatment. Successful progression in clinical trials will depend on a coordinated, comprehensive approach in the pre-clinical phase.
In addition to the limitations of preclinical studies, the failure of Phase III clinical trials may relate to the heterogeneous TBI patient population. While secondary injury pathophysiology has been described in this review, there is not currently the ability to identify which of these cascades are active or dominant in individual TBI patients. This information will require the development of biomarkers with the sensitivity and specificity to identify nuanced components of injury within the TBI diagnosis. This information will guide the development of therapeutics to allow translation of effective agents in a mechanistic strategy in which TBI patients with a specific secondary injury pathway are matched to the compound of action, and to monitor the impact of therapy and the biologic response in addition to conventional clinical outcomes.
Combination therapy could be another strategy that will move the field forward by enhancing the effectiveness of individual compounds. Therapeutic development for neuroprotection following TBI has historically aimed at identifying a single mechanism of action. However, the consensus view of the U.S. Department of Veterans Affairs and the National Institute of Neurologic Disorders and Stroke is that TBI pathophysiology involves such complexity that a single therapeutic regimen is not sufficient (Margulies et al., 2009). A combination of drug treatments should therefore be undertaken to maximize efficacy of response. This could be approached as a combination addressing a single mechanism or targeting distinct mechanisms that offer complementary or synergistic effect. The regulatory limitations inherent with combination therapy make it difficult for clinical trials to be undertaken. While full details of combination therapy are beyond the scope of this article, this topic has been reviewed previously (Kline et al., 2016).
While clinical trials have not achieved success, there has been considerable progression in the understanding of injury processes, which may precipitate a paradigm shift in treatment philosophy. Until recently, it was accepted that the brain did not have the ability to regenerate axons and form new synapses following injury (Hall and Traystman, 2009). However, it has now been recognized that while the brain has limited capacity for structural plasticity following injury, there may be an ability for significant change involving functional recovery (Chopp et al., 2008). In line with this, an alternative approach has recently been proposed involving neurorestoration through the promotion of regeneration of neurovascular functions involving neurogenesis, oligodendrogenesis, and angiogenesis (Xiong et al., 2013). While human clinical trials have not been undertaken targeting oligodendrocyte and myelin repair (Huntemer-Silveira et al., 2021), preclinical studies suggest that targeting the corresponding neural stem cells, oligodendrocyte progenitor cells, and cerebral endothelial cells may improve neuroplasticity and functional outcomes following TBI (Xiong et al., 2010). Exploring these mechanisms is a priority for future TBI treatment development.
In conclusion, there are several secondary injury neurochemical targets and pathways that are influenced by TBI and may be modulated by pharmacological intervention. This review has presented trial data for 35 Phase III interventions registered at clinicaltrials.org, with no interventions gaining regulatory approval as TBI treatment. Looking forward, robust pre-clinical consortia approaches should be implemented to improve the translation from laboratory to clinical setting. With the high cost and extended timeframes of novel drug development, a method of identifying effective treatments that has received support is the repurposing of agents that are already in use in the clinical setting (Margulies et al., 2009). Several intervention compounds in this review have been identified for their relevant mechanisms of action, and further work is needed to ascertain their efficacy in clinical scenarios.
Acknowledgments
The author would like to thank Dr. Andrew Fenning (Central Queensland University, School of Health, Medical, and Applied Science) for conceptualization discussions.
Authorship Contributions
Participated in research design: Hiskens
Wrote or contributed to the writing of the manuscript: Hiskens.
Footnotes
- Received November 17, 2021.
- Accepted May 6, 2022.
This work received no external funding.
The author has no actual or perceived conflict of interest with the contents of this article.
Abbreviations
- Ach
- acetylcholine
- AchE
- acetylcholinesterase
- AMPA
- (α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate
- COX
- cyclooxygenase
- CT
- computed tomography
- DHA
- docosahexaenoic acid
- GCS
- Glasgow coma scale
- ICP
- intracranial pressure
- mTBI
- mild traumatic brain injury
- NMDA
- N-methyl-D-aspartate
- OBTT
- Operation Brain Trauma Therapy
- ROS
- reactive oxygen species
- TBI
- traumatic brain injury
- Copyright © 2022 by The American Society for Pharmacology and Experimental Therapeutics
