Intracerebral hemorrhage (ICH) is a devastating form of acute stroke with poor clinical outcomes. Although ICH accounted only for 28.8% of incident strokes, it was responsible for nearly half of the long-term burden of stroke measured in disability-adjusted life years . In contrast to the improving outcomes seen in ischemic stroke with advances in reperfusion therapy, outcomes of patients with ICH have shown little progress over the past two decades. Current standard care focuses primarily on blood pressure control and supportive management, yet rate of functional independence remained modest. Randomized evidence suggested that fewer than half of the ICH patients achieved independent activities of daily living even with intensive blood pressure lowering.

A cascade of pathophysiological events is thought to determine the prognosis of ICH. First, the mass effect of the hematoma and its expansion with uncontrolled blood pressure cause primary neuronal injury. Second, neuroinflammation involving blood-brain barrier (BBB) dysfunction, activated microglia and astrocytes, together with neutrophil infiltration in response to extravascular blood, propagates neuronal injury, leading to perihematomal edema. Third, the direct neurotoxicity of blood breakdown products and oxidative stress may further amplify neuroinflammation. Mitigating hematoma expansion through intensive blood pressure control therefore only addresses one of these three pathophysiological processes, and is constrained by a short treatment window, mostly within 6 hours. Therapeutic strategies targeting secondary neuroinflammation should therefore be actively pursued. In addition, multimodal studies incorporating longitudinal imaging and omics markers are needed to elucidate the key pathways mediating neuroinflammation following ICH.

Recent preclinical evidence suggests that glucagon-like peptide-1 receptor agonists (GLP-1RA) may offer neuroprotective benefits in ICH. In animal models of ICH, intracerebroventricular liraglutide suppressed neuroinflammation, prevented brain edema, and reduced neurologic deficits. Previous work from our group has also shown that, across several animal models, GLP-1RA attenuates BBB dysfunction and suppresses neuroinflammatory signaling via microglial modulation. Importantly, a recent translational clinical trial by our team has also provided preliminary evidence that GLP-1RA exerts neuroprotective effects in patients with large vessel occlusion strokes. We therefore hypothesize that compared to standard therapy, administration of GLP-1RA in patients with primary ICH may limit perihaematomal edema, reduce secondary brain injury, and improve neurological outcomes.

In this phase 2, randomized, open-label pilot study with blinded endpoint assessment, we aim to determine the safety and signals for efficacy of GLP-1RA in patients with primary ICH.

GLP-1RA in Primary Intracerebral Hemorrhage (GLICH) trial is an investigator-initiated, phase 2, multicenter, randomized, open-label trial with blinded endpoint assessment, conducted across centers in Hong Kong Special Administrative Region (HKSAR), Wenzhou and Hefei, China. We aim to recruit 200 patients with primary ICH in the putamen (10-30mL) or thalamus (5-15mL) who present within 24 hours of last-known-well and meet all eligibility criteria. Participants will be randomized in a 1:1 ratio to receive additional semaglutide or just standard therapy after a written informed consent. Patients in the semaglutide group will receive subcutaneous injections of semaglutide on baseline (D0), Day 7 (D7), Day 14 (D14) and Day 21 (D21) after enrollment.

Serial imaging with non-contrast computed tomography (CT) will be performed at baseline, 3+/-1 days, and 7+/-1 days post-enrollment to assess hematoma volume, hematoma expansion, and perihematomal edema (PHE). Brain magnetic resonance imaging (MRI) will be arranged on Day 5-9 (D30) post-ICH to assess the degree of neuronal injury. Blood sampling for omics analysis will be collected at baseline, D3, D7, D14 and D30. Other investigations will be arranged according to clinical requirement. Clinical assessments including National Institutes of Health Stroke Scale (NIHSS), modified Rankin Scale (mRS), Montreal Cognitive Assessment (MoCA) and Mini Mental State Examination (MMSE) will be obtained on D0, Day 90 (D90) and Day 180 (D180). Recruitment targets are 100 patients from Wenzhou, 75 from Hefei and 25 patients from HKSAR according to catchment area size.

Inclusion Criteria:

• 1. Primary spontaneous ICH with hematoma location in putamen (10-30mL) or thalamus (5-15mL) on admission CT imaging. Both locations are selected because they are strongly associated with hypertensive arteriopathy-related ICH and there is limited evidence supporting neurosurgical intervention compared with posterior fossa hemorrhages. The thalamic and putaminal volume cutoffs are based on a recent observational study showing that restricting enrolment to 5-15 mL (thalamus) and 10-30 mL (putamen) enriches for patients with substantial but potentially modifiable prognosis while avoiding extremes with ceiling or floor effects. For patients with hematoma involving both putamen and thalamus, a volume cutoff of 5-30mL will be used.
• 2. National Institutes of Health Stroke Scale (NIHSS) score ≥ 6 AND ≤ 25 at presentation
• 3. Glasgow Coma Scale (GCS) score ≥ 10
• 4. Last-known-well (LKW) to presentation time ≤ 24 hours
• 5. Pre-stroke modified Rankin Scale (mRS) ≤ 2
• 6. Patients deemed not suitable for acute neurosurgical intervention at the time of randomization
• 7. Informed consent obtained from patient (if mentally competent) or legal representative, as per national laws, regulations, and applicable ethics committee requirements

Exclusion Criteria:

• 1. Secondary ICH: ICH due to macrovascular abnormalities (e.g., arteriovenous malformation, aneurysm, arterial dissection, cavernous malformation), coagulopathy, anticoagulant use, antiplatelet overdose, or thrombocytopenia.
• 2. ICH involving locations other than putamen or thalamus (e.g., lobar, brainstem, cerebellar, isolated intraventricular hemorrhage). Extension of hematoma into other structures is allowed if the hematoma centroid is within the thalamus or putamen, and the hematoma volume does not exist the respective thresholds as stated in Inclusion Criterion 1.
• 3. ICH with planned neurosurgical procedure prior to randomization, including hematoma evacuation, external ventricular drainage and decompressive craniectomy.
• 4. Estimated or known body mass index (BMI) < 18 kg/m².
• 5. Pregnancy, lactation, or positive urine or serum beta human chorionic gonadotropin (β-hCG) test. β-hCG testing should be guided by clinical need.
• 6. Creatinine clearance < 30 mL/min (estimated by Cockcroft-Gault equation or measured)
• 7. Severe or fatal comorbid illness with life expectancy < 3 years (e.g., terminal malignancy, advanced organ failure)
• 8. Participation in another clinical trial investigating a drug, medical device, or medical procedure within 30 days preceding trial inclusion.
• 9. Known history of allergy or hypersensitivity to GLP-1RA.
• 10. Family or personal history of multiple endocrine neoplasia (MEN), medullary thyroid carcinoma, or pancreatic carcinoma
• 11. Active sepsis at time of randomization, defined as a body temperature of ≥ 38.5C, or suspected or documented infection and acute organ dysfunction, operationalized as an increase in SOFA score ≥ 2 points from baseline (baseline assumed 0 if no pre-existing organ dysfunction)
• 12. Contraindications to proposed imaging studies (e.g., pacemaker incompatibility with MRI where applicable)