Impact of cannabidiol on myocardial recovery in patients with acute myocarditis: primary results of the ARCHER study
Heart and Vascular Institute, University of Pittsburgh Medical Center, 200 Lothrop Street, Pittsburgh, PA 15213, USA
Department of Cardiovascular Medicine, Mayo Clinic College of Medicine and Science, Jacksonville, FL, USA
Departments of Medicine and Diagnostic Radiology, Research Institute of the McGill University Health Centre, McGill University, Montreal, QC, Canada
Department of Cardiology, Universitätsklinikum Heidelberg, Heidelberg, Germany
Departments of Cardiac Sciences and Radiology, University of Calgary, Calgary, AB, Canada
Sourasky Medical Center, Tel Aviv University, Tel Aviv, Israel
DeBakey Heart & Vascular Center, Houston, TX, USA
J.C. Walter Jr. Transplant Center, Houston Methodist Hospital, Houston, TX, USA
Instituto do Coração Hospital das Clinicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, Brazil
Cardiol Therapeutics Inc., Oakville, ON, Canada
Instituto de Cardiologia de Santa Catarina, São José, Brazil
Pitié Salpêtrière Hospital, Sorbonne University, Paris, France
ACTION Study Group, Paris, France
University of Ottawa Heart Institute, Ottawa, Canada
London School of Hygiene and Tropical Medicine, London, UK
Cleveland Clinic, Heart Vascular and Thoracic Institute, Cleveland, OH, USA
Instituto de Cardiologia, Hospital Zambrano-Hellion, Escuela de Medicina y Ciencias de la Salud, Tecnologico de Monterrey, Monterrey, Mexico
Deutsches Herzzentrum der Charité (DHZC), Clinict of Cardiology, Angiology and Intensive Medicine at Campus Virchow (CVK), Berlin, Germany
Berlin Institute of Health (BIH) at Charité—Center for Regenerative Therapies (BCRT), Berlin, Germany
German Centre for Cardiovascular Research (DZHK) Partner Site Berlin, Charité University, Berlin, Germany
Corresponding author. Email: mcnamaradm@upmc.eduAbstract
Introduction
Cannabidiol has been shown to exert significant anti-inflammatory effects and has demonstrated efficacy in murine models of autoimmune myocarditis, pericarditis, and heart failure. The ARCHER Study assessed whether a pharmaceutically produced cannabidiol formulation showed beneficial effects on cardiac magnetic resonance (CMR) endpoints known to predict prognosis in this patient population.
Methods
In a multicentre international double-blind placebo-controlled phase 2 study, we randomly assigned 109 patients within 10 days of CMR confirmed diagnosis of acute myocarditis to 12 weeks of pharmaceutically produced oral cannabidiol (active) or placebo. Dose was titrated up to 10 mg/kg of body weight twice daily. Primary endpoints were the difference in extracellular volume (ECV) and global longitudinal strain (GLS) measured by CMR at week 12. Other CMR endpoints included left-ventricular ejection fraction (LVEF), LV mass, intracellular volume (ICV), LV end-diastolic and end-systolic volumes (LVEDV, LVESV), and left-atrial end-systolic volume (LAESV).
Results
All randomized patients (56 active/53 placebo) completed the study with no loss to follow-up. Study drug appeared safe and well tolerated. Baseline mean ECV 38.9 ± 10.9 ml, GLS −15.3 ± 3.6%, and LVEF 60.6 ± 9.9% were consistent with mild to moderate myocarditis and predominantly intact LV function. Week 12 mean ECV was 33.6 ml in the active group and 37.3 ml in the placebo group, a difference of −3.7 ml, confidence interval (CI): −7.4 to 0.1; P = .0538; GLS was −16.0% in the active group and −15.9% in the placebo group, difference of −0.1, CI: −1.2 to 1.1; P = .90. Left ventricular mass was significantly reduced in the active group at 121.1 g compared to placebo 130.3 g, a difference of −9.2, CI: −16.4 to −2.1; P = .0117. In terms of remodelling, LAESV was significantly reduced in the active group (−8.1 ml; P = .0376) while the reduction in LVEDV failed to reach significance (−7.4 ml; P = .098).
Conclusion
In mild-to-moderate acute myocarditis, treatment with pharmaceutically manufactured cannabidiol was not associated with a statistically significant change in myocardial ECV or GLS, although a trend towards reduction in ECV was observed. In addition, improvement in other potential markers of myocardial recovery, including a significant reduction in LV mass, was seen in the active treatment group. Further investigation of the therapeutic potential of this therapy in inflammatory cardiac conditions is warranted.
Teaser
In a multicentre study we randomly assigned 109 patients within 10 days of CMR confirmed diagnosis of acute myocarditis to 12 weeks of oral cannabidiol or placebo. Treatment with cannabidiol was associated with improvements in potential markers of myocardial recovery, including a significant reduction in LV mass. No significant change was evident in myocardial ECV or GLS, although a trend toward reduction in ECV was observed. Further investigation of the therapeutic potential of cannabidiol in inflammatory cardiac conditions is warranted
Graphical Abstract
Introduction
There is no disease-specific therapy universally endorsed for patients who present with acute myocarditis. It remains a therapeutic dilemma despite multiple tested approaches, reflecting the tension between suppressing harmful immune-mediated injury and avoiding unchecked viral replication in the myocardium. Intense inflammation may facilitate early viral clearance and limit tissue destruction; yet, persistent inflammation after viral elimination can perpetuate myocardial damage, underscoring the need for targeted strategies that modulate, but not ablate, immune activity.1,2
Patients with acute myocarditis show significant activation of inflammatory and cell death pathways in the heart, which can be attenuated by cannabidiol3 and suggests potential as a targeted therapy for acute myocarditis. In a murine model of experimental autoimmune myocarditis (EAM) induced by immunization with a cardiac myosin peptide, treatment with cannabidiol reduced the infiltration of the myocardium by inflammatory cells, decreased levels of inflammatory cytokines, chemokines, markers of oxidative stress and reduced myocardial fibrosis.4
The ARCHER study assessed whether a pharmaceutically produced cannabidiol formulation, by modulating the immune response in the setting of acute myocarditis, showed beneficial effects on cardiac magnetic resonance (CMR) endpoints known to predict prognosis in this patient population.
Methods
The ARCHER study was a Phase-2 international, multi-centre, randomized, double-blind, parallel-group, placebo-controlled study, evaluating the effects of a pharmaceutically produced cannabidiol (active) formulation (CardiolRx™, Cardiol Therapeutics Inc.) on CMR parameters of left-ventricular (LV) function and tissue composition in patients with acute myocarditis (Figure 1). A more detailed description of the study rationale and design was published previously.5
The study was conducted in accordance with the Declaration of Helsinki, International Council for Harmonisation, Good Clinical Practiceguidelines, and all relevant country-specific regulations. The conduct of the study was approved by an institutional review board for each participating centre, and all participants provided written informed consent before any study-related procedures were performed. The study has been registered on ClinicalTrials.gov: NCT05180240.
Patient population
Patients enrolled in the ARCHER study were women and men, ≥18 years of age with acute myocarditis confirmed by CMR (Revised Lake Louise Criteria)6 within 10 days prior to randomization or endomyocardial biopsy showing cellular inflammation either by haematoxylin and eosin or immunohistochemistry. Key exclusion criteria included presence of coronary artery disease, severe valvular heart disease, inability to undergo CMR assessment, estimated glomerular filtration rate <30 ml/min, elevated liver function parameters, sepsis, severe LV dysfunction requiring inotropic support, LV assist device or other circulatory assist devices, or the urgent need for transplantation. In addition, documented biopsy evidence of giant cell or eosinophilic myocarditis, prior history of sustained ventricular arrhythmia, acute coronary syndrome or percutaneous coronary intervention within the prior 30 days, or a history of QT interval prolongation or having a QTc interval >500 ms were also exclusion criteria. For a complete list of all eligibility criteria, please refer to the previous design paper.5
CMR analysis
Cardiac magnetic resonance was obtained in all subjects at study entry and at 12 weeks post-randomization. All studies were reviewed by the CMR Core Laboratory (MUHC, Montreal, Canada). Standard Operating Procedures7 were applied according to the protocol of the Study and MUHC standards. All analyses were blinded to the readers and image data and accompanying information was kept anonymized/de-identified. Cardiac magnetic resonance readers had to have a minimum of 12 months experience in the evaluation procedure. All analyses of the primary endpoints were performed by a single reader with more than 20 years of experience in reading CMR images. For all CMR analyses, certified software was used (cvi42 version 5.13; Circle CV Imaging Inc., Calgary, AB, Canada). Before evaluation, the submitted image data were assessed for confirmation of myocarditis (revised Lake Louise criteria), completeness and image quality. The quality of images and derived maps [T1, T2, extracellular volume (ECV)] was deemed adequate if the anatomical structures were clearly identifiable and the images and the regions of interest (especially in the myocardium) were not significantly affected by field inhomogeneities, motion artifacts, or other artifacts.
Efficacy assessments
The primary efficacy outcome of the ARCHER study was comprised of two endpoints: the difference in the means of global longitudinal strain (GLS) as a sensitive and reproducible measure of LV function and the difference in the means of ECV as an estimate of oedema and/or fibrosis, quantified by CMR at 12 weeks post-randomization between the cannabidiol and the placebo groups. The secondary efficacy outcome was the difference in the means of LVEF, as measured by CMR, at 12 weeks post-randomization between the two groups. Exploratory efficacy outcomes included additional CMR parameters at 12 weeks and markers of remodelling.
Safety parameters consisted of the number of adverse events (AEs) and serious adverse events (SAEs), changes in ECG parameters, as well as changes in laboratory parameters, including liver function parameters, and INR associated with treatment.
Statistical considerations
Sample size calculations were based on published data for GLS and ECV.8 Assuming 80% power, a 5% alpha error boundary, and estimated 25% missing CMR follow-up data at week 12, 100 patients were required to demonstrate the desired treatment effect of 18%. For both primary outcomes, the means in GLS and ECV at 12 weeks were compared between the active and the placebo groups, applying an ANCOVA procedure, adjusted for baseline values. The two primary outcomes were considered statistically significant following the Hochberg Procedure9 in order to preserve the overall type 1 error rate at a level of 0.05 (two-sided).
The same analytical approach was used to evaluate the effect of study medication on the change in LVEF and other continuous outcomes. The differences in proportions experiencing binary outcomes were estimated, along with 95% confidence intervals. The primary analysis population was a modified intention-to-treat (mITT) population, which included all patients who were randomized and had the confirmed (revised Lake Louise criteria) of myocarditis by the CMR Core Laboratory. All analyses were done using SAS Statistical Software, version 9.4.
Results
Between 2 August 2022 and 4 November 2024, 166 patients were consented in 29 sites in Brazil, France, Israel and the United States, 57 patients did not meet the eligibility criteria, 109 patients were randomized (56 to active treatment and 53 to placebo). The patient disposition is shown in Figure 2. All randomized patients started study treatment: 6 (10.7%) patients in the active group and 5 (9.4%) in the placebo group discontinued study treatment prematurely. For 10 patients, the CMR Core Laboratory was not able to confirm that the patients fulfilled the revised Lake Louise criteria for myocarditis; these patients were excluded from the mITT analysis.
Presentation at baseline
Baseline demographic and clinical characteristics were generally well balanced between treatment arms (Table 1). The majority of participants were male (79.6% in the active group, 82.0% in the placebo group), with mean ages of 35.8 and 39.6 years, respectively. Most patients were White (77.6% vs 82.0%), and approximately half identified as Hispanic or Latino (44.9% vs 54.0%). The median time from acute myocarditis diagnosis to randomization was comparable between groups (8.0 vs 7.0 days).
| Baseline characteristics | ||
|---|---|---|
| Cannabidiol (N = 49) | Placebo (N = 50) | |
| Male/female | 39 (80%)/10 (20%) | 41 (82%)/9 (18%) |
| Age, years; mean (SD) | 35.8 (15.35) | 39.6 (15.47) |
| Race | ||
| White | 38 (78.6%) | 41 (82.0%) |
| Black or African American | 5 (10.2%) | 2 (4.0%) |
| Other/not reported | 6 (12.2%) | 7 (12.0%) |
| Ethnicity | ||
| Hispanic | 22 (44.9%) | 27 (54.0%) |
| Not Hispanic | 21 (42.9%) | 16 (32.0%) |
| Unknown/not reported | 6 (12.3%) | 7 (14.0%) |
| CMR to randomization—days, mean (SD) | 7.1 (2.63) | 6.4 (2.76) |
| Clinical symptoms/signs | ||
| Elevated troponin within 90 days | 47 (95.9%) | 47 (94.0%) |
| Chest pain | 46 (93.9%) | 47 (94.0%) |
| History of viral-like illness | 36 (73.5%) | 36 (72.0%) |
| Shortness of breath | 20 (40.8%) | 23 (46.0%) |
| Arrhythmia | 3 (6.1%) | 3 (6.0%) |
| NYHA class | ||
| I | 31 (63.3%) | 33 (66.0%) |
| II | 18 (36.7%) | 14 (28.0%) |
| III/IV | 0 | 3 (6.0%) |
| Systolic BP (mmHg), mean (SD) | 120.29 (14.770) | 118.80 (12.367) |
| Diastolic BP (mmHg), mean (SD) | 70.39 (10.498) | 71.54 (8.908) |
| Heart rate (beats/min), mean (SD) | 71.82 (11.733) | 71.46 (12.061) |
| Weight (kg) mean (SD) | 80.74 (16.797) | 83.07 (15.993) |
| Body mass index (kg/m²), mean (SD) | 27.30 (5.388) | 27.82 (5.337) |
Nearly all were hospitalized for the index myocarditis event (95.9% vs 96.0%), with median hospital stays of 5.0 days in both groups and similar ICU durations. Only two patients underwent an endomyocardial biopsy procedure. Chest pain was the most common presenting symptom occurring in 93.9% (active) and 94.0% (placebo), with shortness of breath reported by 41% (active) and 46% (placebo) and a history of viral-like illness in 73.5% (active) and 72.0% (placebo). Elevated troponin levels within 90 days of symptom onset were present in almost all patients (>94%). Baseline glucose, high-sensitivity troponin I, and high-sensitivity troponin T were similar between groups, though values showed wide variations. Biomarker analyses will be presented in a subsequent publication.
The prevalence of prior acute myocardial infarction, arrhythmia, heart failure, and other cardiac history was low and comparable between groups. Traditional cardiovascular risk factors were uncommon: hypertension, 18.4% (active) vs 20.0% (placebo), hyperlipidaemia, 12.2% (active) vs 16.0% (placebo), and diabetes mellitus, 4.1% (active) vs 8% (placebo). Most patients did not have symptoms of heart failure at the time of enrolment with NYHA Class I at baseline of 63.3% (active) vs 66.0% (placebo), with the remainder primarily in Class II; three placebo patients were classified as Class III. Mean baseline BMI was 27.3 kg/m² (active) vs 27.8 kg/m² (placebo), with similar blood pressure, heart rate, and height distributions in both groups.
Study drug treatment
The mean duration of study treatment was 80.3 days (SD 9.85) in the active group and 79.2 days (SD 15.96) in the placebo group. Planned and actual total doses of the investigational product and compliance rates were generally similar across arms. Mean compliance exceeded 89% in both groups, with most patients achieving the 10 mg/kg twice daily dose: 91.8% (active) vs 96.0% (placebo). Premature discontinuation before week 12 occurred in 8.2% and 8.0% of patients in the active and placebo groups, respectively.
Safety results
Most AEs in both groups were mild. In the cannabidiol group, 117 events were reported in 43 patients (76.8%) who experienced at least one treatment-emergent adverse event (TEAE), compared with 82 events in 32 patients (60.4%) in the placebo group (Table 2). SAEs in active group included recurrence of myocarditis, ventricular tachycardia, alanine aminotransferase increase, and transaminase increase in one patient each. One patient had two SAEs of gastritis and tremors, and another had chest pain on two occasions. Events in the placebo group included palpitations and hypocalcaemia in one patient each and the third patient had both aspartate aminotransferase and alanine aminotransferase increased.
| Treatment emergent adverse events (TEAE) reported during the study by medDRA preferred term (safety population) in >5% of patients | |||||||
|---|---|---|---|---|---|---|---|
| Cannabidiol (N = 56) | Placebo (N = 53) | ||||||
| MedDRA Preferred term | No. of events reported | No. of patients with at least one event | Percentage of patients (%) | No. of events reported | No. of patients with at least one event | Percentage of patients (%) | P value |
| Cardiac disorders | |||||||
| Palpitations | 2 | 1 | 1.8% | 4 | 4 | 7.5% | .198 |
| Gastrointestinal disorders | |||||||
| Diarrhoea | 20 | 18 | 32.1% | 11 | 11 | 20.8% | .200 |
| Nausea | 5 | 5 | 8.9% | 2 | 2 | 3.8% | .439 |
| General disorders and administration site conditions | |||||||
| Asthenia | 1 | 1 | 1.8% | 7 | 7 | 13.2% | .029 |
| Chest pain | 11 | 9 | 16.1% | 5 | 4 | 7.5% | .239 |
| Infections and infestations | |||||||
| Nasopharyngitis | 0 | 0 | 0.0% | 3 | 3 | 5.7% | .112 |
| Investigations | |||||||
| Alanine aminotransferase increased | 5 | 5 | 8.9% | 4 | 3 | 5.7% | .717 |
TEAEs leading to study drug discontinuation occurred in three patients in both active and placebo groups. No clinically important change in the QTc interval was observed between baseline and follow-up visits. Two patients in the active group reported non-specific suicidal thoughts during the study; however, both patients were amongst the five active group patients who reported lifetime suicidal ideation and behaviour on C-SSRS at baseline. Overall, cannabidiol was well tolerated, with AE patterns comparable to placebo for most systemic events and laboratory abnormalities.
Efficacy analysis
There were no significant differences between groups at 12 weeks in the difference in the means of GLS or ECV, however a trend towards reduction in ECV was observed (P = .0538).
Treatment with cannabidiol was associated with a significantly lower LV mass at 12 weeks (Figure 3), LV mass adjusted means at week 12 being 121.10 g (active) vs. 130.33 g (placebo), with a mean difference of −9.23 g (95% CI −16.36 to −2.11) P = .0117. Due to limitations in recording quality, only 71 patients had both baseline and week 12 ECV and ICV recordings analyzable. All other CMR endpoint measurements were feasible in 95 to 99 patients (Table 3).
| Cannabidiol patients at baselineb N = 49 | Placebol patients at baselineb N = 50 | Cannabidiol patients at 12 weeksc (95% CI) | Placebo patients at 12 weeksc (95% CI) | Observations available | Mean treatment difference at week 12 (95% CI) | P-value | |
|---|---|---|---|---|---|---|---|
| Myocardial tissue characteristics (mean ± SD) | |||||||
| ECV (ml) | 37.74 ± 9.93 | 40.13 ± 11.75 | 33.61 (30.88–36.34) | 37.28 (34.74–39.83) | 71 | −3.67 (−7.41 to 0.06) | .0538 |
| LV Mass (g) | 130.75 ± 29.28 | 134.59 ± 40.70 | 121.10 (116.04–126.16) | 130.33 (125.32–135.34) | 99 | −9.23 (−16.36 to −2.11) | .0117 |
| ICV (ml) | 89.14 ± 19.63 | 91.69 ± 30.79 | 85.59 (80.82–90.36) | 91.16 (86.71–95.60) | 71 | −5.57 (−12.09 to 0.95) | .0928 |
| ECV (%)a | 29.74 ± 4.47 | 30.83 ± 4.31 | 28.25 (27.13–29.36) | 29.12 (28.08–30.15) | 71 | −0.87 (−2.40 to 0.66) | .2609 |
| LV function (mean ± SD) | |||||||
| GLS (%) | −15.35 ± 3.10 | −15.34 ± 4.06 | −16.03 (−16.83 to −15.23) | −15.96 (−16.76 to −15.15) | 98 | −0.07 (−1.21 to 1.07) | .9021 |
| LVEF (%) | 61.90 ± 8.88 | 59.35 ± 10.79 | 62.09 (60.39–63.79) | 61.54 (59.87–63.22) | 99 | 0.55 (−1.85 to 2.94) | .6520 |
| LV and LA volumes (mean ± SD) | |||||||
| LVEDV (ml) | 138.07 ± 29.60 | 137.70 ± 40.15 | 134.48 (128.21–140.76) | 141.91 (135.70–148.12) | 99 | −7.43 (−16.25 to 1.40) | .0981 |
| LVESV (ml) | 53.14 ± 17.40 | 56.43 ± 23.14 | 51.30 (47.87–54.74) | 54.23 (50.83–57.63) | 99 | −2.92 (−7.77 to 1.92) | .2339 |
| LAESV (ml) | 51.28 ± 18.59 | 54.85 ± 26.91 | 49.95 (44.48–55.41) | 58.04 (52.75–63.33) | 95 | −8.09 (−15.71 to −0.47) | .0376 |
The ECV adjusted means at week 12 were 33.61 mL (95% CI 30.88 to 36.34) in the active group vs. 37.28 mL (95% CI 34.74 to 39.83) in the placebo group. The mean treatment difference in ECV at week 12, adjusted for baseline values, was −3.67 mL (95% CI −7.41 to 0.06); P = .0538. The means in GLS were similar for both the active and placebo groups and both showed minimal changes from baseline: −16.03% (95% CI −16.83 to −15.23) for the active group and −15.96% (95% CI −16.76 to −15.15) for patients taking placebo (mean difference: −0.07, 95% CI 1.21 to 1.07); P = .9021 (Table 3 and Figure 3).
The mean baseline LVEF was 61.90% ± 8.88 for the active group and 59.35% ± 10.79 for placebo. At week 12, the adjusted mean LVEF was maintained or slightly higher in both groups at week 12, being 62.09% (95% CI 60.39 to 63.79) for active and 61.54% (95% CI 59.87 to 63.22) for the placebo group, with a mean difference of 0.55 (95%CI −1.85 to 2.94) P = .6520.
Intracellular volume (ICV) trended lower similar to ECV in the active group (mean difference: −5.57 mL, 95%CI-12.09 to 0.95; P = .0928). The ECV fraction (%ECV) appeared similar in both groups and showed a mean of 28.25% (95% CI 27.13 to 29.36) in the active group versus 29.12% (95% CI 28.08 to 30.15) in the placebo group; P = .2609.
With respect to remodelling, no significant differences was seen in LV end-diastolic volume (LVEDV for active 134.5 mL (CI 128.21 to 140.8 mL) versus placebo 141.9 ml (CI 135.7 to 148.1 mL) with mean difference of −7.4 mL, (CI −16.2 to 1.4); P = .098), or LV end systolic volume (LVESV for active 51.3 mL (CI 47.9 to 54.7) versus placebo 54.2 (CI 50.8 to 57.6) with mean difference of −2.9 mL (CI −7.8 to 1.9); P = .234). However, for the left-atrial end-systolic volume (LAESV) less remodelling was evident in the active group as the adjusted means were 49.95 mL (active) vs. 58.04 mL (placebo), mean difference −8.09 mL (95% CI −15.71 to −0.47); P = .0376).
Discussion
For patients with acute myocarditis, previous attempts to develop targeted therapies to reduce cardiac inflammation and improve outcomes have not been successful.10 In the ARCHER study, although the primary endpoints were not met, treatment with pharmaceutically produced cannabidiol was associated with a significant reduction in LV mass which was reflective of trends towards reduction in both ECV and ICV. This may reflect improvement in cardiac inflammation evident in preclinical models. Myocardial oedema in myocarditis is strongly correlated with LV mass.11 We observed a change of LV mass which was not correlated with similar changes in myocardial T2. This is however not surprising, given the mild degree of myocardial inflammation in our population. Milder inflammation is characterized by primarily intracellular oedema, as opposed to more extracellular oedema in more severe forms.12 T2 relaxation times in MRI are more sensitive to changes in the extracellular space, where water mobility is less restricted, compared to the intracellular compartment, where water is more tightly bound and less mobile. Therefore, reduction in intracellular oedema results in smaller changes in myocardial T2 values compared to reduction in extracellular oedema. This has been demonstrated in preclinical studies.12,13 Therefore, our observation of a pronounced decrease in LV mass likely reflects a recovery from primarily intracellular oedema as consistent with mild myocarditis. Cannabidiol therapy was also associated with less left atrial remodelling and a trend towards lower LVEDV. While the mode of action of cannabidiol therapy has yet to be fully elucidated, the observed changes may have resulted from attenuation of inflammation and immune cell infiltration, limitation of fibrosis, and mitigating hypertrophic signalling. These reductions in LV mass and remodelling should be considered as potential markers for improved outcomes in future clinical trials. In addition, targeting cardiac inflammation with cannabidiol in ARCHER was shown to be safe with no concerning adverse effects associated with treatment.
The rationale for the ARCHER study was based upon the known anti-inflammatory and anti-fibrotic properties of cannabidiol. A murine model of EAM induced by immunization with the cardiac myosin peptide (αMHC334–352) resulted in T cell infiltration into the myocardium, T cell–mediated inflammation, cardiomyocyte cell death, fibrosis and myocardial dysfunction. In this model, chronic treatment with cannabidiol (10 mg/kg, ip for 46 days) reduced the infiltration of the myocardium by inflammatory cells, decreased myocardial inflammation as reflected by lowered levels of inflammatory cytokines and chemokines (IL6, IL1-beta, IFN-gamma, MCP1), reduced markers of oxidative stress and reduced myocardial fibrosis.4 Cannabidiol has also been shown to attenuate a number of measures of potential importance in the treatment of heart failure, including cardiac dysfunction, oxidative stress, fibrosis, and inflammatory and cell death signalling pathways in vitro and in a mouse model of diabetic cardiomyopathy.3 Cannabidiol has been shown to be protective against doxorubicin-induced cardiotoxicity, including reducing pro-inflammatory responses in the heart in two rodent models.14,15 A recent important observation suggests that the mitochondrial calcium uniporter (MCU) is a key regulator of the hypertrophic response in mice and in humans.16 In a murine model of angiotensin II–induced cardiomyopathy, cannabidiol administered subcutaneously prevented myocyte cell hypertrophy, fibrosis and preserved myocardial function; changes that occur in part by modulating MCU function and expression.17 Overall, the effectiveness of cannabidiol therapy to reduce cardiac inflammation in multiple preclinical models indicates a potential role for cannabidiol in other cardiomyopathy states where hypertrophy is a common mediator of injury and appears consistent with findings observed in the ARCHER study.
The enrolment of patients with normal LVEF and acute myocarditis may explain why ARCHER did not meet the primary endpoint of an improved mean GLS at week 12. These results are not unexpected in a cohort where LV function was predominantly intact. With a normal overall mean baseline LVEF of 60.6% and slightly reduced baseline GLS of −15.3%, a significant increase would also be difficult to interpret.
The second primary endpoint of ECV at 12 weeks also failed to reach significance; however, it demonstrated a trend towards a reduction with cannabidiol therapy (P = .054). Depending on image quality, assessment of ECV can be technically challenging and ECV was not available for ∼30% of the subjects in ARCHER, limiting study power. While absolute ECV demonstrated a trend towards reduction with cannabidiol, % ECV did not. The formula to calculate is % ECV = ECV/(ECV + ICV)×100, and in chronic conditions where ICV is relatively stable, interventions which reduce ECV should reduce % ECV. However, this is not necessarily the case for treatments in acute myocarditis, where resolution of interstitial and intracellular oedema may reduce both ECV and ICV, therefore, diminishing the impact on %ECV. This was indeed evident in ARCHER, where both ECV and ICV were reduced in parallel with cannabidiol treatment resulting in significant reductions in LV mass but less impact on the %ECV.
ARCHER is the first placebo-controlled pharmaceutical intervention study utilizing changes in CMR parameters as endpoints in patients with acute myocarditis. The primary endpoints of ECV18–20 and GLS8,21 are known independent predictors of prognosis in patients with acute myocarditis. However, minimal data exist on the appropriate CMR measures to evaluate therapies intended to improve long-term clinical outcomes in acute myocarditis. In the ITAMY (ITAlian study in MYocarditis) registry,19 CMR was performed within the first week after symptom onset and repeated after 6 months. During a median clinical follow-up of 7 years cardiac events occurred in 22 patients. Patients with increased extent of LGE at 6 months had a worse prognosis than those with decreased/unchanged LGE. In the recently presented ARAMIS trial (NCT03018834) evaluating an IL-1 inhibitor in acute myocarditis patients, the vast majority of patients included had a preserved LVEF, and similar characteristic compared to the ARCHER study. The overall rate of events was low and LVEF improved spontaneously across patients, regardless of treatment allocation with no difference observed, suggesting that, LVEF is not a sufficiently sensitive surrogate marker to capture the risk of adverse outcomes in a low-risk population, as spontaneous recovery can mask potential differences between treatment groups.22
The main limitations of our study include the enrolment of patients with only mild to moderate acute myocarditis with predominantly intact LV function. In addition, suboptimal cardiac MRI quality reduced the number of patients with analysable ECV at both baseline and week 12 to 71, significantly weakening the statistical power for this endpoint compared with LV mass, which was measurable on baseline and week 12 in all 99 patients included in the primary analysis. The required weekly titration schedule from initially 2.5 mg/kg twice daily and reaching 10 mg/kg twice daily at the week 3 visit meant exposure to cannabidiol early in the 12-week treatment period may have been suboptimal and limited the potential for cannabidiol to benefit the patients.
In summary, in the ARCHER study, treatment with cannabidiol in patients with myocarditis was associated with a decrease in LV mass consistent with the reduction of cardiac inflammation evident in preclinical models. These results provide sound rationale for advancing the clinical development of this novel therapy in other conditions of the myocardium characterized by oedema, fibrosis, and remodelling, including heart failure and perhaps the growing challenge of immune checkpoint inhibitor-induced myocarditis.23 This study also provides reassuring safety data that the use of pharmaceutically produced cannabidiol appears safe in patients with myocarditis. This is important given the increasing recognition of the overlap between myocarditis and pericarditis, the ‘inflammatory myopericardial syndrome’2 and the currently ongoing Phase 3 MAVERIC trial in recurrent pericarditis (NCT06708299).
Acknowledgements
All CMRs were analysed centrally by the Imaging Core Laboratory at McGill University, Montreal, Canada. SOCAR Research SA., Switzerland was responsible for the overall data management as well as management of local Contract Research Organizations. Local CROs: PharmaClinical, Nes Ziona, Israel (Israel), Atlantis, São Paulo, Brazil (Brazil), ActionCoeur, Paris, France (France), Ozmosis, Toronto, Canada (US). TMC Pharma, Hants, United Kingdom, was responsible for pharmacovigilance. We would like to thank all patients, the study coordinators, the investigators and all the investigative site personnel for their participation in the study.
Declarations
Disclosure of Interest
A.B.: research grant and scientific advisory committee for Cardiol Therapeutics Inc. L.T.C.: scientific advisory committee for Cardiol Therapeutics Inc. M.G.F.: institution receives research funding from Cardiol Therapeutics Inc., shareholder of Circle CVI Inc. G.T.: institution receives research funding from Cardiol Therapeutics Inc., receives payment as director of Cardiol Therapeutics Inc. A.H., A.B.P., G.T.: employee of Cardiol Therapeutics Inc. With the exception of A.H., A.H.H, A.B.P., and G.T., all authors received payment from Cardiol Therapeutics Inc. for service on the study steering committee.
Data Availability
The data underlying this article were provided by Cardiol Therapeutics Inc. by permission. Data will be shared on reasonable request to the corresponding author with permission, subject to applicable law, of Cardiol Therapeutics Inc.
Funding
This study was sponsored by Cardiol Therapeutics Inc., Oakville, Ontario, Canada; no additional funding was used to support this manuscript.
Ethical Approval
This investigation was approved by the institutional review boards at all particpating institutions.
Pre-registered Clinical Trial Number
ClinicalTrials.gov Study Details | NCT05180240 |
Appendix Group
Appendix
ARCHER Study Group
Site Investigators
Brazil: Artur Haddad Herdy, Hospital Regional de São José, São José; Paulo Ricardo Avancini Caramori, Hospital São Lucas, Porto Alegre; Edimar Bocch, Instituto do Coração – InCor, São Paulo; Lídia Moura, PUCtrials, Curitiba; Ariane Vieira Scarlatelli Macedo, Irmandade da Santa Casa de Misericórdia de São Paulo, São Paulo; Maria da Consolação Vieira Moreira, Hospital Felicio Rocho - Fundação Felice Rosso, Belo Horizonte; Denilson Campos de Albuquerque, Instituto D´Or de Pesquisa e Ensino, Rio de Janeiro; Eduardo Dytz, Hospital Moinhos de Vento, Porto Alegre; Fernando Neuenschwander, Nupec-Orizonti, Belo Horizonte; Luís Beck da Silva, Hospital de Clínicas de Porto Alegre (HCPA), Porto Alegre.
France: Benoit Lattuca, Centre Hospitalier Universitaire de Nîmes, Nîmes; Claire Bouleti, Centre Hospitalier Universitaire de Poitiers, Poitiers; Theo Pezel, Hôpital Lariboisière, Paris; Mathieu Kerneis, Institut de Cardiologie hopital Pitié Salpêtrière, Paris; Jeremie Abtan, Hopital Bichat Claude Bernard, Paris; Etienne Puymirat, Hôpital européen Georges-Pompidou, Paris; Clément Delmas, CHU RANGUEIL, Toulouse; Florent Huang, Hôpital Foch, Suresnes; François Roubille, CHU de Montpellier, Montpellier; Thomas Bochaton, Hopital Louis Pradel Hospices Civils de Lyon, Bron.
Israel: Gil Moravsky, Shamir Medical Center (Assaf Harofeh), Zriffin, Be'er Yakov; Xavier Alejandro Piltz, Barzilai Medical Center, Ashkelon; Yaron Arbel, Tel Aviv Sourasky Medical Center (Ichilov), Tel-Aviv; Amir Orlev, Shaare Zedek Medical Center, Jerusalem.
USA: Pavan Bhat, Cleveland Clinic, Cleveland; Daniel Zlotoff, Massachusetts General Hospital, Boston; Georgia K. Thomas, Virginia Commonwealth University, Richmond; Mark Hofmeyer, MedStar Heart and Vascular Institute, Washington; David Lin, Minneapolis Heart Institute Foundation, Minneapolis; Brittany Palmer, University of Pittsburgh Medical Center, Pittsburgh.
Steering Committee: Dennis M. McNamara, (Chair), University of Pittsburgh Medical Center, Pittsburgh, PA , USA; Leslie T. Cooper (Co-chair), Mayo Clinic, Jacksonville, FL, USA; Matthias G. Friedrich, McGill University, Montreal, QC, Canada; Yaron Arbel, Tel Aviv University, Tel Aviv, Israel; Arvind Bhimaraj, DeBakey Heart & Vascular Center, Houston, TX, U.S.A; Edimar Bocchi, Universidade de São Paulo, São Paulo, Brazil; Mathieu Kerneis, Pitié Salpêtrière Hospital, Sorbonne University, Paris, France; Peter P. Liu, University of Ottawa Heart Institute, Ottawa, Canada; W. H. Wilson Tang, Cleveland Clinic, Cleveland, OH, USA; Guillermo Torre-Amione, Escuela de Medicina y Ciencias de la Salud, Tecnologico de Monterrey, Monterrey, Mexico; Carsten Tschöpe, German Centre for Cardiovascular Research (DZHK) Partner Site Berlin, Charité University, Berlin, Germany.
Cardiol Therapeutics, Oakville, ON, Canada: Andrew Hamer; Andrea B. Parker; Eldon R. Smith.
Study Statistician: Stuart J Pocock, London School of Hygiene and Tropical Medicine, London, United Kingdom.