Oxford, England, and New York, United States, October 28, 2026
Brainomix and Remedy Pharmaceuticals have announced promising findings from an artificial intelligence (AI)-powered imaging analysis of the Phase 3 CHARM clinical trial, identifying potential treatment-response signals in patients with large hemispheric infarction (LHI), a severe form of ischemic stroke. Presented at the 18th World Stroke Congress, held in Seoul from October 21–23, 2026, the post-hoc analysis evaluated whether automated non-contrast computed tomography (NCCT) imaging could help identify patients more likely to benefit from CIRARA (intravenous glyburide), an investigational treatment being developed by Remedy Pharmaceuticals. The findings suggest that AI-assisted imaging may support more precise patient selection in future stroke trials. However, the results are exploratory and do not establish the treatment’s efficacy or regulatory approval.
AI Imaging Analysis Identifies Potential Treatment-Response Signals
The analysis examined the relationship between acute infarct core volume and treatment outcomes among patients enrolled in the CHARM Phase 3 study. An infarct core represents brain tissue that has already sustained significant damage following an interruption of blood supply. Earlier post-hoc analyses of CHARM had suggested that patients with smaller imaging-defined infarct cores, particularly those with lesion volumes below 125 mL measured through CT perfusion or magnetic resonance imaging, might experience a greater treatment effect. Building on those observations, Brainomix used its Brainomix 360 Stroke software and research capabilities from its in-house AI Core Lab to assess imaging data. The company analyzed baseline and follow-up NCCT, CT perfusion, and diffusion-weighted MRI scans alongside clinical outcomes. Automated NCCT analysis was used to measure infarct volume in 428 patients, with the objective of assessing whether a widely available imaging method could help identify potential treatment responders.
Smaller Infarct-Core Subgroup Shows Favorable Outcomes
In a per-protocol analysis involving 362 patients, after excluding patients enrolled using NCCT with an e-ASPECTS score above 5, CIRARA showed a borderline association with improved functional outcomes measured using the modified Rankin Scale (mRS). More encouraging findings emerged in a subgroup of 191 patients with an acute infarct core volume below 125 mL. In this subgroup, CIRARA treatment was significantly associated with improved functional outcomes, including a favorable shift in mRS scores and a greater likelihood of achieving an mRS score of 0–3 at 90 days. A broader modified intention-to-treat analysis of 428 patients also identified a significant non-linear interaction between infarct-core volume and treatment allocation, suggesting that both lower and upper infarct-volume thresholds may influence treatment response. These findings provide a potential direction for further investigation into imaging-based patient selection. Nevertheless, because the analysis was conducted after the original trial, the results should be interpreted as hypothesis-generating rather than definitive evidence of clinical benefit.
Advertisement
Implications for Future Stroke Trials and Treatment Development
The findings highlight the potential role of AI-powered non-contrast CT analysis in improving the design of future clinical trials for severe stroke. NCCT is widely available in hospitals and is commonly used during the initial assessment of patients with suspected stroke. Using automated analysis of this imaging method could make it easier for more hospitals to identify suitable participants for clinical research without relying exclusively on advanced imaging techniques. This may help broaden access to clinical trials and support more consistent assessment of patients across different healthcare settings. The CHARM study was a double-blind, randomized, placebo-controlled Phase 3 trial that evaluated CIRARA in adults aged 18–85 with large hemispheric infarction. Treatment was initiated within 10 hours of symptom onset. The trial enrolled 535 participants, while its efficacy analysis population included 431 patients aged 18–70. The study was sponsored by Biogen and was terminated prematurely following a strategic realignment of resources; the programme was subsequently returned to Remedy Pharmaceuticals. CIRARA remains investigational and has not been approved by any regulatory authority.
For Brainomix, the analysis demonstrates how AI-based imaging tools may contribute to precision medicine by translating complex medical images into information that can support clinical research. For Remedy Pharmaceuticals, the findings provide additional exploratory evidence to investigate as it continues advancing the CIRARA programme for patients with large hemispheric infarction. Further research and prospective clinical evaluation will be needed to confirm whether the imaging characteristics identified in this analysis can reliably predict treatment response and guide clinical decisions. The collaboration underscores the growing importance of combining medical imaging, artificial intelligence, and pharmaceutical development to improve research strategies for serious neurological conditions.
Source: Brainomix press release
