A 72-hour checkpoint for CAR T-cell neurotoxicity treatment

The weeks following CAR T-cell therapy carry both promise and uncertainty. As clinicians look for signs that the treatment is working, they must also remain alert for potentially serious toxicities.  Even subtle behavioral changes including altered handwriting or difficulty finding the right words can be early warning signs of immune effector cell-associated neurotoxicity syndrome, or ICANS, a common and potentially life-threatening complication of CAR T-cell therapy that can progress from mild changes in cognition and communication to severe neurologic dysfunction.

Predicting how ICANS unfolds remains challenging. Patients can experience a wide spectrum of symptoms and disease severity, and the biology driving ICANS is still not fully understood. Clinicians may have to make high-stakes treatment decisions while both the patient’s condition and the evidence guiding treatment are evolving.

Corticosteroids are the standard first-line treatment for ICANS. But for an estimated 30% to 50% of patients, steroids alone are not enough, and prolonged corticosteroid exposure carries risks of its own. The anti-inflammatory drug anakinra is increasingly used as a second-line treatment, but clinicians have lacked validated guidelines and benchmarks for determining whether anakinra is working for a particular patient, and when it might be time to change course.

A recent study led by Dr. Emily Liang, with researchers from the Clinical Research Division’s Clinical Biostatistics group and Oregon Health & Science University, suggests that one relatively simple measure could help guide treatment decisions: how much a patient’s neurological symptoms improve within 72 hours of starting anakinra.

The researchers retrospectively studied 101 patients who received anakinra for ICANS following CAR T-cell therapy, including 90 whose ICANS had persisted or worsened despite corticosteroid treatment. They examined patients’ clinical outcomes and looked for an early signal that could distinguish those who were responding to anakinra from those who might benefit from another treatment strategy.

They defined significant neurological improvement, or SNI, as an improvement of at least two ICANS grades without the need for another ICANS-directed therapy. Patients who achieved SNI within 72 hours recovered from ICANS more quickly and were discharged from the hospital sooner than patients whose symptoms had not improved significantly by that point. None of the patients who achieved early SNI experienced treatment-related mortality, while about 20% of patients who did not achieve early SNI experienced treatment-related mortality.

“For clinicians accustomed to reassessing severely ill patients day by day, or sometimes hour by hour, the finding offers the beginnings of a more concrete decision point”, explained Liang. Rather than continuing the same treatment indefinitely while waiting for improvement, the 72-hour mark could eventually help identify patients who are unlikely to respond adequately to anakinra and may benefit from an alternative or investigational therapy.

The study also questioned another common escalation strategy. When ICANS does not improve, clinicians may add anakinra while simultaneously switching to a more potent corticosteroid. But in this cohort, patients whose corticosteroid treatment was intensified did not have better neurological response or faster ICANS resolution.

That finding is important because higher-dose corticosteroids can bring additional risks, including risk of infection and muscle weakness. If escalating steroid intensity does not improve neurological outcomes, clinicians may eventually be able to spare some patients additional steroid exposure while considering other approaches.

The study also highlights that there is considerable room for improvement in ICANS treatment. Across the entire cohort, it took a median of eight days to resolve ICANS after anakinra was started, and treatment-related mortality at 28 days was 13%. The researchers also identified several characteristics, including older age, CAR T-cell product and markers of systemic inflammation, that were associated with a lower likelihood of early neurological improvement and could eventually contribute to more individualized approaches to treatment.

For now, the 72-hour checkpoint will need to be validated, but Liang sees this retrospective analysis as a foundation for designing future studies. One possibility, she noted, would be a clinical trial that uses 72-hour SNI to guide what treatment patients receive next and tests whether doing so ultimately improves outcomes.

CAR T-cell therapy has transformed treatment for certain cancers that have resisted conventional therapies. But as its use expands, ICANS remains an important barrier to making these treatments safer and more broadly accessible. Managing a rapidly evolving complication often means making decisions before every question has been answered. Establishing clearer signals of when a treatment is working, and when it is time to change course, could help turn some of that uncertainty into evidence-based action.

Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium Members Drs. Jennifer Huang, Andrew Portuguese, Alexandre Hirayama, Mazyar Shadman, Qian Wu, and Jordan Gauthier contributed to this research.

The spotlighted research was funded by the National Institutes of Health and Swim Across America.

Liang EC, Kungwankiattichai S, Wu X, Voutsinas J, Macero D, Khouderchah C, Qiao Y, Jeon Y, Basom R, Huang JJ, Portuguese AJ, Kimble EL, Hirayama AV, Shadman M, Chen AI, Desai A, Hayes-Lattin B, Meyers G, Bastola S, Rai M, Leonard J, Trussell J, Stomberg C, Spurgeon SE, Okada C, Shree T, Silbermann R, Medvedova E, Ratterree B, Bixler R, Williamson S, Phan C, Ahrenkiel R, Wu Q, Maziarz RT, and Gauthier J. 2026. Early neurologic response predicts outcomes after anakinra for immune effector cell–associated neurotoxicity syndrome. Blood Advances. DOI: 10.1182/bloodadvances.2026020103