Mitochondrial-targeting drugs may represent an underexplored approach to managing immune effector cell-associated neurotoxicity syndrome following CAR T-cell therapy, with leflunomide identified as a prototype candidate for further evaluation.
The investigators proposed that mitochondrial dysfunction may amplify neuroinflammation through a feedforward loop not adequately addressed by current steroid-based and cytokine-blocking approaches.
In the perspective, the authors reviewed the pathophysiology and current management of immune effector cell-associated neurotoxicity syndrome (ICANS) and developed a framework for identifying mitochondrial-targeting drugs for potential repurposing. Rather than enrolling patients or analyzing a new clinical dataset, the researchers evaluated candidates across four domains: mechanistic relevance, central nervous system (CNS) bioavailability, oncologic safety, and translational precedent.
Applying the framework, investigators identified several candidate classes, including dihydroorotate dehydrogenase (DHODH) inhibitors such as leflunomide and teriflunomide, mitochondrial-targeted antioxidants, NLRP3 inflammasome inhibitors with mitochondrial activity, AMPK activators such as metformin, and cGAS-STING pathway inhibitors. Leflunomide was highlighted as a prototype that satisfied all four selection criteria. None of the identified candidates, to the researchers' knowledge, are currently under formal investigation for ICANS prophylaxis or treatment.
Leflunomide's active metabolite, teriflunomide, inhibits DHODH, an enzyme on the inner mitochondrial membrane required for de novo pyrimidine synthesis. Zahedi and colleagues proposed that this mechanism has downstream effects on mitochondrial reactive oxygen species, NLRP3 inflammasome priming, and macrophage metabolic polarization. They also pointed to teriflunomide's established use in relapsing multiple sclerosis and measured cerebrospinal fluid concentrations as evidence supporting CNS exposure.
The rationale for the approach centers on mitochondrial involvement in neuroinflammation. ICANS is conceptualized as a two-phase process, beginning with systemic cytokine elevation and followed by inflammatory amplification within the CNS. Neuroinflammatory cytokines can impair mitochondrial electron transport, increase reactive oxygen species generation, and trigger the release of mitochondrial damage-associated molecular patterns. These signals can further activate inflammatory pathways, including the NLRP3 inflammasome and cGAS-STING. Zahedi and colleagues proposed that this bidirectional relationship creates a feedforward loop that may sustain and worsen ICANS independently of systemic cytokine levels.
The proposed strategy addresses limitations of current ICANS management. Corticosteroids remain the mainstay of treatment for established ICANS and are effective in most cases, but some patients develop steroid-refractory disease. Corticosteroid treatment can also be associated with hyperglycemia, acute myopathy, and psychosis. High-grade ICANS has been reported in up to 10% of cases in pooled analyses cited in the perspective, although incidence varies by CAR T-cell construct, disease indication, pretreatment conditioning, and patient-specific factors.
A key consideration is whether leflunomide could interfere with CAR T-cell activity. Because DHODH inhibition can suppress rapidly proliferating lymphocytes, Zahedi and colleagues identified potential impairment of CAR T-cell expansion as a concern. They noted that DHODH inhibition is reversible and can be pharmacologically rescued with uridine supplementation. Careful timing of leflunomide initiation relative to CAR T-cell infusion and temporary dosing interruptions during the expansion window were also proposed as potential elements of a clinical protocol.
Additional support for leflunomide came from experience in other inflammatory conditions. Beyond the clinical precedent provided by teriflunomide in multiple sclerosis, the authors noted that leflunomide has been explored in neuropsychiatric systemic lupus erythematosus, graft-vs-host disease, and viral encephalitis. Zahedi and colleagues cited these settings as translational precedent and characterized leflunomide as a mechanistically coherent, clinically tractable prototype for the broader class of mitochondrial-targeted ICANS candidates.
The investigators proposed a staged evaluation pathway beginning with preclinical studies and progressing through assessment of CNS exposure to early clinical evaluation. The pathway would ultimately lead to a randomized clinical trial comparing leflunomide with standard-of-care approaches.
The perspective has important limitations. It provides no direct evidence that leflunomide or other mitochondrial-targeting drugs prevent or treat ICANS. No patients were enrolled, clinical outcomes were not measured, and treatment effect estimates and subgroup analyses were not reported. The rationale instead draws on ICANS pathobiology and evidence from adjacent neuroinflammatory and immune-mediated conditions. Whether leflunomide can preserve CAR T-cell expansion and antitumor activity while providing the proposed neuroinflammatory effects remains to be established.
Investigators concluded that mitochondrial-targeting drugs warrant further experimental investigation, with leflunomide serving as a prototype for the broader repurposing strategy. “A staged evaluation pathway from preclinical validation to randomized clinical trial is feasible and warranted,” wrote lead study author Atena Zahedi, of the University of California Irvine, and colleagues.
Disclosures: Dean reported unlicensed patents related to cellular immunotherapy held by Moffitt Cancer Center and consultancy with Kite Pharma. The other authors reported no conflicts of interest. No financial support was received for the work or publication. Generative artificial intelligence was used to help create the manuscript figures.
Source: Frontiers in Pharmacology
