CCR5 antagonists as neuroprotective and stroke recovery enhancing agents: a preclinical systematic review and meta-analysis

Background C-C chemokine receptor type 5 (CCR5) antagonists may improve both acute stroke outcome and long-term recovery. Despite their evaluation in ongoing clinical trials, gaps remain in the evidence supporting their use. Methods With a panel of patients with lived experiences of stroke, we performed a systematic review of animal models of stroke that administered a CCR5 antagonist and assessed infarct size or behavioral outcomes. MEDLINE, Web of Science, and Embase were searched. Article screening and data extraction were completed in duplicate. We pooled outcomes using random effects meta-analyses. We assessed risk of bias using the Systematic Review Centre for Laboratory Animal Experimentation (SYRCLE) tool and alignment with the Stroke Treatment Academic Industry Roundtable (STAIR) and Stroke Recovery and Rehabilitation Roundtable (SRRR) recommendations. Results Five studies representing 10 experiments were included. CCR5 antagonists reduced infarct volume (standard mean difference −1.02; 95% confidence interval −1.58 to −0.46) when compared to stroke-only controls. Varied timing of CCR5 administration (pre- or post-stroke induction) produced similar benefit. CCR5 antagonists significantly improved 11 of 16 behavioral outcomes reported. High risk of bias was present in all studies and critical knowledge gaps in the preclinical evidence were identified using STAIR/SRRR. Conclusions CCR5 antagonists demonstrate promise; however, rigorously designed preclinical studies that better align with STAIR/SRRR recommendations and downstream clinical trials are warranted. Registration Prospective Register of Systematic Reviews (PROSPERO CRD42023393438)


Introduction
C-C chemokine receptor type 5 (CCR5) is expressed across a variety of leukocyte subtypes, endothelial cells, and cell types in the brain (e.g., neurons, microglia, astrocytes), and is thought to play crucial roles in post-stroke neuroinflammation, blood-brain barrier repair, and neuronal survival / repair processes. 13][4][5][6] However, no CCR5 antagonist drug has an approved indication in the stroke context, necessitating studies to establish safety and efficacy this population.This has led to an ongoing clinical trial to investigate efficacy of CCR5 antagonists in combination with post-stroke rehabilitation. 7Assessment of the preclinical evidence supporting CCR5's role in stroke is needed to identify areas of potential benefit, and knowledge gaps, that should be addressed by future preclinical research. 8roke neuroprotection and recovery communities have advocated for alignment of preclinical and clinical study parameters through publication of consensus recommendations for preclinical research, in an effort to enhance the translation of new stroke therapies.[9][10][11] Examples include identification of more sensitive and clinically relevant preclinical outcome measures and incorporation of potentially important effect modifiers of treatment efficacy, such as age, sex, and stroke-related comorbidities (hypertension, diabetes, etc.).[9][10][11] These recommendations aim to improve the translation of novel stroke therapeutics from preclinical to clinical populations, but the degree to which preclinical evidence for CCR5 antagonists satisfy these recommendations is unknown.
We sought to comprehensively evaluate the preclinical evidence for CCR5 antagonist drugs as both neuroprotective and stroke recovery-promoting agents. 9,11Both perspectives are necessary to fully understand the therapeutic potential of stroke-related treatments, as distinct biological principles, time windows for treatment, and outcomes of interest underpin each of these treatment domains. 12,13We conducted a systematic review and meta-analysis of the CCR5 literature in conjunction with a panel of individuals with lived experiences of stroke.

Materials and Methods
We registered the review protocol on the International Prospective Register of Systematic Reviews (CRD42023393438). 14The findings are reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (Table S1) 15 and Guidance for Reporting the Involvement of Patients and the Public (Table S2). 16A panel of eight patients and caregivers with lived and living experience of stroke informed project development and were actively involved throughout research conduct (see Table S2 for more information).

Eligibility criteria
-Animals: Any preclinical in vivo animal models of adult stroke were included.Human, invertebrate, in vitro, ex vivo, and neonatal animal studies were excluded.
-Model: Focal ischemic or intracerebral hemorrhagic stroke models were included, while animal models of global ischemia were excluded.
-Comparator: Vehicle-treated control groups where stroke was induced.CCR5 antagonist control groups without stroke were excluded.
-Outcome: Studies reporting at least one of the following: infarct size, behavioral tests, mortality, adverse events, and spasticity were included.
-Study design and publication characteristics: Controlled interventional studies (randomized, pseudo-randomized, or non-randomized) published as full journal articles in any language or year were included.Abstracts, review articles, opinion-based letters/editorials, and unpublished grey literature were excluded.

Information sources and search strategy
An information specialist with experience in systematic searches of the preclinical literature developed a comprehensive search strategy on October 25, 2022, based on a previously published strategy for identifying animal experimentation studies. 17The search strategy underwent peer-review using the Peer Review of Electronic Search Strategies (PRESS) checklist. 18We searched MEDLINE (OVID interface, including In-Process and Epub Ahead of Print), Web of Science, and Embase (OVID interface).See the Supplemental Methods for the full search strategy and exact search dates for each database.

Selection process and data collection
We deduplicated citations and uploaded them into DistillerSR® (Evidence Partners, Ottawa, Canada).Two reviewers independently screened citations by title and abstract using an accelerated method (one reviewer required to include, two reviewers required to exclude).We then screened and extracted data from full-text articles in duplicate.Graphical data was extracted using Engauge Digitizer. 19A third reviewer with content expertise in preclinical stroke studies audited all data extraction.Conflicts between reviewers were resolved by consensus discussion.
See Supplemental Methods for the complete list of data extraction elements.

Effect measures and data synthesis
We performed quantitative analyses using the R (version 4.1.2)"metafor" package (version 4.0.0) 20with inverse variance random effects modelling.We expressed continuous outcome measures as standardized mean differences (SMDs) with 95% confidence intervals (CIs) and assessed statistical heterogeneity of effect sizes using the Cochrane I² statistic. 21This was necessary due to the variety of outcome measures and measurement scales used across studies, which is a common feature of preclinical systematic reviews.Sensitivity analyses were performed using original measurement scales where possible.From patient partner input, subgroups were analyzed based on timing / dose / route of intervention, stroke model, stroke type, species, type of behavioral outcome (i.e., motor, cognitive), and comorbidities. 22Our planned subgroup analyses on study quality, specific regions/areas of the brain, and post-stroke rehabilitation paradigms were not performed due to insufficient number of studies.We did not assess publication bias using Egger plots due to fewer than 10 studies being included in the analysis, as per Cochrane recommendations. 23

Risk of bias assessment
Two independent reviewers used the Systematic Review Centre for Laboratory Animal Experimentation (SYRCLE) risk of bias tool to assess each study as having a "Low Risk", "Unclear Risk", "Some Concerns of Risk", or "High Risk" across domains such as randomization, blinding and outcome reporting. 24"Some Concerns of Risk" indicated reporting of a domain (i.e., randomization), but lacking methodological details.This differed from "Unclear Risk" where there was no mention of the domain in the study.

Comprehensiveness of preclinical evidence and alignment with clinical trials
We assessed comprehensiveness of the overall body of preclinical evidence for CCR5 antagonists as a neuroprotective treatment using The Stroke Treatment Academic Industry Roundtable (STAIR) I, VI, and XI Consolidated Recommendations. 11,25,26These recommendations encompass "candidate treatment qualification" (e.g., dose, timing of dose, outcomes) and "preclinical assessment and validation" (e.g., age, sex, sample size, animal type). 11We excluded domains that were redundant with risk of bias (e.g., randomization, blinding, etc.) and included additional items relevant to stroke recovery studies from the Stroke Recovery and Rehabilitation Roundtable (SRRR) Translational Working Group consensus recommendations. 9Two reviewers extracted data to determine if the overall evidence across studies satisfied each of the STAIR and SRRR recommendations.A third reviewer audited this analysis.We then assessed alignment of existing preclinical evidence with an ongoing clinical trial of CCR5 antagonists for stroke (CAMAROS, NCT04789616). 7

Protocol deviations
We incorporated an additional assessment using the PRIMED 2 tool for assessing the readiness of stroke neuroprotection therapies to be translated to clinical trials based on the body of preclinical evidence. 27Additionally, a list of outcomes used to determine CCR5's potential mechanisms of action were extracted (Figure S1, Table S3) based on feedback from individuals that reviewed the initial manuscript drafts.

Study selection
Our search identified 263 citations, which was reduced to 166 unique studies after deduplication. 5][6]

Intervention characteristics
Maraviroc was used in six of the experiments (n=6/10), TAK-779 in three of the experiments (n=3), and D-Ala-Peptide T-Amide (DAPTA) in one (n=1; Table 2).These CCR5 antagonists were delivered intraperitoneally (n=4), intranasally (n=2), intracerebroventricularly (n=2), subcutaneously (n=1), and intravenously (n=1) at a dose range from 0.01 to 100 mg/kg.Most studies delivered a single dose of the drug (n=6); experiments with multiple administrations (n=4) ranged from 3-63 doses.Time of initial treatment administration varied widely.4][5][6] One study (1 experiment) was conducted in the late sub-acute/early chronic period beginning at three to four weeks post stroke, which would be oriented towards recovery, rather than neuroprotective, effects. 6Patient partners had identified several a priori interests (physical therapy alongside CCR5 administration, spasticity), which were also aligned with SRRR recommended considerations for preclinical stroke recovery studies (see Supplemental Methods). 9These were not reported in any included studies.

Meta-analysis of infarct volume
Infarct volume was reported in six experiments (n=6/10) from four different studies with an overall pooled analysis demonstrating marked neuroprotection with CCR5 antagonists (SMD -1.02, 95% CI -1.58 to -0.46, p < 0.0001, I 2 =34%; Figure 2).Five of these experiments measured infarct volume at 1-3 days post-stroke, and one experiment measured infarct volume at a delayed time point of 63 days post-stroke.No significant differences between pre-or post-stroke administration were observed (P=0.47).Post-hoc sensitivity analysis removing one experiment with extreme values 2 demonstrated that neuroprotection in the remaining two experiments remained statistically significant while reducing heterogeneity (SMD -0.81, 95% CI -1.25 to -0.37, p < 0.001, I 2 =0%).A second sensitivity analysis excluded one study that measured infarct volume in mm 3 so that all other studies could be meta-analyzed using mean differences on the percentage scale (Figure S2; MD -9.1%, 95% CI -11.6 to -6.7%, p < 0.001, I 2 =0%).This demonstrated a similar neuroprotective effect as the other analyses.Further sub-group analyses by route of administration, time of administration, stroke model, species, CCR5 antagonist, dose, and whether behavior tests were assessed are described in Figure S3-S5.

Synthesis of behavioral outcomes without meta-analysis
Motor behavioral outcomes were reported in six experiments from three studies and represented seven different behavioral tasks.An additional study reported motor behavioral outcomes without standard deviations or standard errors, and thus could not be included (authors did not respond to email requests for data). 4A cognitive outcome (Morris Water Maze) was measured in one study.Overall, CCR5 inhibition was effective in 11 of 16 behavioral outcomes tested (Figure 3).Meta-analysis and planned subgroup analysis were contra-indicated due to an inadequate number of studies for each given outcome measure, which necessitated the synthesis without meta-analysis presented below. 28oral outcomes are presented by time of CCR5 antagonist administration, as discussed in the Intervention Characteristics section above, as administration time directly relates to the treatment context and patient population to which the results apply (i.e., acute neuroprotection vs. late sub-acute/early chronic neural repair).Li et al. reported that pre-stroke administration (relevant to surgical contexts with high risk of thrombosis) of DAPTA did not result in significantly greater performance on the neurological deficit score. 2 Regarding acute post-stroke administration times with the potential for neuroprotective effects (up to 24 hours post-stroke based on observed infarct reductions in Figure 2), Yan et al. observed that CCR5 antagonist (maraviroc) administration one-hour following stroke resulted in greater motor performance on the corner test, limb placement, modified Garcia, foot fault, and rotarod tasks compared to vehicle-treated controls.Cognitive outcomes in the Morris Water Maze task (proportion of time spent in the probe quadrant) were also improved by this one-hour post-stroke administration.Significantly improved motor performance was not observed if CCR5 antagonist was administered 24 hours post-stroke, using this model of intracerebral hemorrhage (Figure 3). 5 In contrast, using a focal ischemia model, Joy et al. observed that CCR5 antagonist (maraviroc) administration initiated 24 hours post-stroke and continued daily for 3 weeks (which could potentially represent a mix of neuroprotective and recovery-induced effects) resulted in greater performance on the cylinder and grid walk tasks compared to vehicle-treated controls.
Joy et al., also performed one experiment with late sub-acute/early chronic administration (a potentially critical time for neural repair) initiated at three to four weeks post-stroke and continued daily for 11 weeks, that demonstrated significantly improved performance for the grid walk, but not cylinder, task. 6This experiment was initiated outside the plausible range for a neuroprotective effect, implying behavioral improvement involved recovery-promoting mechanisms.However, equivalent infarct volumes were not demonstrated between the treated and control groups in this cohort, which could potentially lead to confounding effects.

Risk of bias
All articles had a 'high' risk of bias in at least one domain of the SYRCLE tool. 24Most domains within each study demonstrated an 'unclear' risk of bias.All studies reported randomizing animals; however, as commonly observed in the preclinical literature, [29][30][31] only one of these studies provided sufficient detail to ensure that the randomization method had a low risk of bias.
The SYRCLE domains with the highest risk of bias were incomplete outcome data, with 80% of studies (n=4/5) failing to provide complete data for all animals initially included in the study, as well as selective outcome reporting, with 60% (n=3/5; all studies of maraviroc) not providing complete data for all expected outcomes discussed in the methods (Figure S6).

Comprehensiveness of preclinical evidence and alignment with clinical trials
We assessed comprehensiveness of the preclinical evidence using the STAIR and SRRR recommendations 9,11,25,26 as well as alignment with study parameters of the CAMAROS trial. 7A summary of assessment items from STAIR XI is provided in Table 3, with additional items from STAIR I, VI, and SRRR recommendations included in Table S4.
For CCR5 antagonists as a post-stroke neuroprotectant, the overall body of evidence satisfied all five STAIR XI domains assessing 'candidate treatment qualification' (Table 3).Overall, a range of doses and clinically relevant administration times for neuroprotection were evaluated across a variety of motor and cognitive behavioral domains.All studies tested both behavioral and histological outcomes and demonstrated neuroprotective effects, but most studies failed to measure and control post-stroke temperature, which could potentially confound the observed neuroprotection (Table S4). 32Most histological measurements were also assessed at <72 hours, which could confound the observed neuroprotective effects if cell death was merely delayed. 32CCR5 antagonists as a post-stroke recovery-inducing treatment, one experiment assessed the effects of initiating CCR5 administration in a similar post-stroke phase as the CAMAROS trial.This experiment (Joy et al.) 6 did not demonstrate that each treatment group had equivalent baseline stroke volumes, which may potentially confound observed behavioral effects.
Furthermore, the maximum dose used in mice (100 mg/kg) was not sufficient to attain cerebrospinal fluid levels of maraviroc observed in humans using the CAMAROS dosing regime (mice: 13.8±5.4ng/mL; humans 33.6-60 ng/mL). 6of concern were identified in all STAIR XI domains assessing 'preclinical assessment and validation' (Table 3).Although adult animals were used in most of the preclinical studies, the reported weights and ages of these animals corresponded to young adults, rather than the aged adults that better represent the stroke population.All experiments used only male rodents that were free of common stroke comorbidities.It was also unclear if behavioral testing was performed during the inactive circadian phase or active (dark) phase, which could result in confounding if CCR5 antagonists affect arousal of animals during their inactive period. 33hermore, none of the studies had their protocols or results directly replicated by an independent laboratory, or across multiple sites in a multilaboratory study. 9,346][37] Both elements are highly relevant to the CAMAROS trial, as one of the primary outcomes of this trial is upper extremity performance on the Fugl-Meyer and maraviroc administration will be paired with an 8-week exercise program.These findings were supported by the PRIMED 2 tool, which resulted in a Readiness for Translation Score of "medium" (on a scale of "low", "medium", "high").This tool highlighted similar promising elements, as well as weaknesses, as our analysis above (Table 3, Table S4), identifying the limited preclinical evidence for the effects of CCR5 antagonists in clinically relevant sexes, ages, species, and disease comorbidities, without sufficient dose-response information to inform trials (Table S5).Overall, our assessments highlight a variety of knowledge gaps and areas of misalignment between the preclinical evidence and clinical trial parameters that could be improved with further preclinical experimentation.

Discussion
The overall body of preclinical evidence for CCR5 antagonists in stroke demonstrates potential acute neuroprotection with corresponding impairment reduction, as well as improved functional recovery in the sub-acute/early chronic phase.Our systematic review also highlights evidence gaps that could impact successful clinical translation of CCR5 antagonists.Our analysis of 10 independent experiments, identified that acute administration of CCR5 antagonists within the first 24 hours post-stroke was associated with a marked reduction in infarct volume.This neuroprotective reduction of infarct volume did not significantly vary based on treatment dose or any other experimental characteristics. 3,4,6Overall, the majority of behavioral effects appeared to be in a positive direction, but the low number of included studies precluded meta-analysis of these results.Indeed, no individual behavioral experiment included more than ten CCR5-treated animals, and given the wide range of dosages, timings, routes, stroke models, and rodent strains involved, the certainty of these findings is limited and should be interpreted cautiously.Further investigation is warranted to determine the optimal timing of administration and behavioral domains under which CCR5 antagonists exhibit the strongest post-stroke neuroprotective and recovery-inducing effects.
Despite the positive direction of treatment effects across all studies of CCR5 antagonists, we found a substantial risk of bias in the underlying studies. 24 Comprehensiveness of the preclinical evidence for CCR5 antagonists was assessed in relation to STAIR and SRRR consensus recommendations. 9,11,25,26These recommendations aim to provide investigators and regulators with "assurance that the candidate treatment shows signals of efficacy and safety, before embarking on an expensive clinical development program". 11The included studies provide good initial evidence for acute neuroprotection, as well as mechanistic and behavioral evidence for enhanced recovery in the late sub-acute/early chronic post-stroke phase.However, demonstration of efficacy under a wider range of conditions, such as in aged animals, females, animals with stroke-related comorbidities, more relevant timing of dose administrations, or in conjunction with rehabilitative therapies are necessary to provide further confidence in these findings.In addition, all studies used unique doses, timings, and outcomes, so independent replication of the most promising study parameters would further increase certainty in the evidence. 9,11ation to the ongoing CAMAROS trial assessing maraviroc in the subacute post-stroke phase, 7 the most relevant preclinical evidence comes from one experiment within the Joy et al. study 6 where maraviroc was initially administered at three to four weeks post-stroke and continued daily for 11 weeks.This experiment demonstrated that administration of maraviroc in the late sub-acute/early chronic post-stroke phase improved functional recovery on the grid walk, but not cylinder, task.These experimental conditions could potentially align with the putative therapeutic window and outcomes of interest being assessed in CAMAROS (e.g.10-minute walk test co-primary outcome). 7However, caution is warranted as this pivotal supporting preclinical evidence is based on a low sample size (n=9).Moreover, potential differences in dosing, severity of infarct, concomitant rehabilitative therapy, and other factors discussed above could influence the degree to which these results successfully translate to the clinical environment.Nevertheless, this study provides a plausible biological mechanism and "proof of concept" for how CCR5 antagonism might enhance neuroplasticity that improves functional recovery after stroke.
Our present synthesis of the preclinical evidence for CCR5 antagonists used novel approaches to increase its utility for assessing certainty of the findings and identification of knowledge gaps.
First, we engaged patients with lived experiences of stroke throughout the review process to ensure that our research questions, outcomes, and interpretations aligned with the priorities of the ultimate end-user of stroke research.Second, we incorporated consensus recommendations for both preclinical neuroprotection and recovery research as an evidence evaluation tool, which we found often aligned with the priorities of our patient panel. 9,11,25,26This guided assessment of the alignment of preclinical evidence with parameters of ongoing clinical trials, as well as appraisal of comprehensiveness of the preclinical evidence with a focus on translational validity 41 rather than only internal validity and risk of bias. 24We also used this method to provide concrete avenues for future preclinical studies to close knowledge gaps and improve certainty in the effects of CCR5 antagonists under clinically relevant experimental conditions.Similar approaches should be considered by future preclinical systematic reviews to improve interpretation of the preclinical evidence from a translational perspective.
In conclusion, CCR5 antagonists show promise in preclinical studies for stroke neuroprotection, corresponding reduction in impairment, as well as improved functional recovery related to neural repair in the late sub-acute/early chronic phase.However, high risk of bias and the limited (or no) evidence in clinically relevant domains underscore the need for more rigorous and transparent preclinical research to further strengthen the overall evidence base.Addressing these concerns will not only enhance the reliability of preclinical evidence but also better inform the design and execution of clinical trials of CCR5 antagonists, such as the ongoing CAMAROS trial. 7The integration of expert recommendations, such as STAIR and SRRR, should guide future preclinical investigations and synthesis of the body of preclinical evidence in stroke recovery research. 9,11Our present approach serves as a template by which the preclinical evidence supporting translation to clinical trials can be weighed when justifying early clinical trials of novel interventions for stroke neuroprotection and recovery.

Figure 2 .
Figure 2. CCR5 antagonists reduce infarct volume.Data is presented as a forest plot with standardized mean differences and 95% confidence intervals.Effect sizes <0 favours drug treatment and >0 favours control.Data is stratified by timing of CCR5 antagonist administration (pre-or post-stroke induction).The 'RE Model for All Studies' represents a pooled estimate of the CCR5 antagonist drug effect on infarct volume from all studies combined.Separate pooled estimates are also reported for post-stroke and pre-stroke CCR5.

Figure 3 .
Figure 3. Meta-analysis for all included preclinical CCR5 antagonist studies that reported motor and/or cognitive behavioral outcomes.Data is presented as a forest plot with a standardized mean difference and 95% confidence intervals.Effect sizes <0 favours drug treatment and >0 favours control.

Table 1 .
Summary of study and animal model characteristics of included articles.

Table 2 .
Summary of intervention characteristics.