The brain complication oncologists watch for after CAR-T cell therapy is called ICANS — Immune Effector Cell-Associated Neurotoxicity Syndrome — and it is more common than many patients and families expect. Pooled data across clinical studies show that roughly 26.9 percent of all CAR-T recipients develop some grade of ICANS, while severe cases (grades 3 through 4) affect about 10.5 percent. In some treatment cohorts, neurological toxicity of any kind affects over half of all recipients. For a therapy celebrated as a breakthrough against blood cancers, these numbers demand serious attention — particularly from anyone already living with concerns about cognitive health or neurodegenerative disease. ICANS is not a single event but a spectrum.
A patient may notice tremor or difficulty finding words five days after infusion, or may progress within hours to seizures and life-threatening cerebral edema. Fatal neurotoxicity occurs in an estimated 3 to 9 percent of patients receiving CD19-directed CAR-T cells, and fatal cerebral edema specifically accounts for 1 to 2 percent of cases. Patients have been documented deteriorating from neurologically normal to death from brain herniation within 24 hours. These are not distant statistical abstractions — they are the reason oncologists keep CAR-T recipients under close neurological monitoring in specialized treatment centers. This article covers what ICANS actually looks like at the bedside, how it is graded and tracked, what drives the blood-brain barrier breakdown behind it, and the emerging concern about delayed neurological complications that fall outside the ICANS definition entirely — including a form of parkinsonism now recognized after certain CAR-T products. We also address current treatment approaches, what the FDA’s recent regulatory shifts mean for patient safety, and what families focused on brain health should understand about long-term monitoring.
Table of Contents
- What Brain Complication Do Oncologists Watch for After CAR-T Therapy?
- How CAR-T Cells Breach the Blood-Brain Barrier
- The Delayed Brain Complications That Fall Outside ICANS
- How ICANS Is Treated and What Options Are Emerging
- FDA Regulatory Shifts and What They Signal About CAR-T Safety
- What CAR-T Brain Risks Mean for Patients With Existing Cognitive Concerns
- Where CAR-T Neurotoxicity Research Is Heading
- Conclusion
- Frequently Asked Questions
What Brain Complication Do Oncologists Watch for After CAR-T Therapy?
The formal name is Immune Effector Cell-Associated Neurotoxicity Syndrome, and the American Society for Transplantation and Cellular Therapy defines it as a pathological process involving the central nervous system following immunotherapy that results in the activation or engagement of endogenous or infused T cells and other immune effector cells. In plainer terms, the engineered immune cells that hunt cancer also trigger an inflammatory cascade that can cross into the brain and cause real neurological damage. ICANS typically appears around five days after CAR-T cell infusion and lasts approximately ten days in most patients, though the timeline varies depending on disease type, tumor burden, and the specific CAR-T product used. Oncologists grade ICANS using the ICE score — the Immune Effector Cell-Associated Encephalopathy assessment — a 10-point bedside tool. A patient scoring 7 to 9 has Grade 1 ICANS, which may look like mild confusion or slightly altered handwriting. Grade 2 (ICE score 3 to 6) involves more noticeable somnolence and cognitive impairment.
Grade 3 (ICE score 0 to 2) includes seizures, and Grade 4 means the patient is unarousable or in a coma, or has developed life-threatening cerebral edema. The grading system exists because ICANS is not a yes-or-no diagnosis. It is a sliding scale, and knowing where a patient sits on that scale determines how aggressively the medical team intervenes. What makes ICANS different from many treatment side effects is the speed at which it can escalate. A patient might have a mild tremor one morning and be in status epilepticus by evening. This volatility is why all six FDA-approved CAR-T therapies — Yescarta, Kymriah, Tecartus, Breyanzi, Abecma, and Carvykti — carry boxed warnings for neurotoxicity. It is also why, until recently, patients were required to remain near the treatment center for four weeks after infusion under a strict Risk Evaluation and Mitigation Strategy.
How CAR-T Cells Breach the Blood-Brain Barrier
The mechanism behind ICANS centers on the blood-brain barrier — the tightly regulated boundary that normally keeps inflammatory molecules and immune cells out of the central nervous system. After CAR-T infusion, the engineered cells begin attacking cancer and secreting a storm of cytokines, including IL-1, IL-2, IL-6, IL-10, interferon-gamma, and GM-CSF. These signaling molecules activate the endothelial cells lining blood vessels in the brain, and that activation weakens the barrier’s integrity. Once the barrier is compromised, inflammatory proteins and immune cells flood into brain tissue. Research has identified elevated serum levels of IL-6, interferon-gamma, and MIP-1α as the markers most closely associated with ICANS development. This is worth understanding because it explains why ICANS often occurs alongside or shortly after cytokine release syndrome (CRS), the other major CAR-T side effect. Both are driven by the same inflammatory surge.
However — and this is a critical distinction — the treatment that works for CRS does not work for ICANS. Tocilizumab, the IL-6 receptor blocker that is standard therapy for CRS, does not resolve ICANS and may actually worsen it. This counterintuitive reality catches some families off guard. The inflammatory pathways damaging the brain are not identical to those causing the fevers and blood pressure drops of CRS, even though they share upstream triggers. The practical limitation here is that predicting which patients will develop ICANS remains imperfect. Higher tumor burden, certain pre-existing neurological conditions, and the specific CAR-T product all influence risk. Severe ICANS (grades 3 to 4) occurs in up to 31 percent of patients receiving anti-CD19 CAR-T cells but only about 9 percent of those receiving anti-BCMA CAR-T cells used for multiple myeloma. If you are evaluating CAR-T therapy for a family member who already has cognitive vulnerability — mild cognitive impairment, early dementia, or a history of stroke — this risk stratification matters enormously, and it should be part of a direct conversation with the treating oncologist before infusion.
The Delayed Brain Complications That Fall Outside ICANS
ICANS is the complication oncologists have learned to expect, manage, and in most cases resolve. What is generating new concern in the neuro-oncology community are the delayed neurological complications that do not fit the ICANS pattern — different in timing, different in mechanism, and in some cases far more persistent. The most striking example is delayed-onset parkinsonism following BCMA-directed CAR-T therapy. In the CARTITUDE-1 study of ciltacabtagene autoleucel (marketed as Carvykti), five patients developed parkinsonism with a median onset of 43 days after infusion — well past the typical ICANS window. Their symptoms included rigidity, bradykinesia, a mask-like facial expression (hypomimia), and postural instability. For anyone familiar with Parkinson’s disease or Lewy body dementia, these symptoms are immediately recognizable.
The difference is that these patients did not have a pre-existing movement disorder. The parkinsonism appeared to be a direct consequence of the therapy. This finding is still being studied, but it has sharpened attention on the basal ganglia and dopaminergic pathways as potential targets of CAR-T-related immune damage. Beyond parkinsonism, clinicians have documented Guillain-Barré syndrome, cranial nerve palsies, immune-mediated myelitis, and peripheral neuropathies occurring after CAR-T infusion. These non-ICANS neurological complications were poorly understood until recently — there were no formal guidelines for managing them. That changed in April 2025, when the European Society for Blood and Marrow Transplantation (EBMT) published the first-ever guidelines specifically addressing non-ICANS neurological complications after CAR-T therapy. The existence of these guidelines is itself an acknowledgment that the brain risks of CAR-T extend well beyond the acute neurotoxicity window.
How ICANS Is Treated and What Options Are Emerging
The mainstay of ICANS treatment is corticosteroids, not the tocilizumab used for cytokine release syndrome. For mild ICANS (Grade 1), oncologists often take a watchful waiting approach with supportive care. For Grade 2 and above, dexamethasone is typically the first-line steroid. Severe cases — Grade 3 or 4 — may require high-dose methylprednisolone at doses up to 1,000 milligrams. The tradeoff with aggressive steroid therapy is real: corticosteroids suppress the immune system broadly, which can potentially blunt the anti-cancer effect of the CAR-T cells themselves. Oncologists must balance the immediate threat of brain swelling against the longer-term goal of cancer remission. For the delayed, non-ICANS neurotoxicities, the treatment landscape is even less certain.
Early evidence suggests that cyclophosphamide may help mitigate non-ICANS neurotoxicities following ciltacabtagene autoleucel, but this approach is still being evaluated. Meanwhile, several clinical trials are testing preventive strategies: dasatinib (a kinase inhibitor that can act as a CAR-T “off switch”), siltuximab (an IL-6 antagonist that works differently from tocilizumab), and interferon-beta-1a (FP-1201) are all under investigation for their ability to prevent or reduce CAR-T neurotoxicity before it starts. The comparison that matters for families weighing these risks is this: ICANS is usually reversible with prompt treatment. Most patients who develop Grade 1 or 2 ICANS recover fully. But the delayed complications — parkinsonism, neuropathies, myelitis — may not follow the same reassuring trajectory. There is not yet enough long-term data to know whether CAR-T-associated parkinsonism is progressive, stable, or potentially reversible. That uncertainty is the honest answer, and it should be part of any informed consent discussion.
FDA Regulatory Shifts and What They Signal About CAR-T Safety
All six FDA-approved CAR-T therapies carry boxed warnings — the most serious warning category — for both cytokine release syndrome and neurotoxicity. In 2024, the FDA added a third boxed warning for secondary T-cell malignancies, meaning the label now flags the risk that the treatment itself could, in rare cases, lead to new cancers. These are not cautionary footnotes. Boxed warnings represent the FDA’s determination that a risk is serious enough to influence prescribing decisions. However, in June 2025, the FDA made a move that may seem contradictory at first glance: it eliminated the REMS requirement for all six approved CAR-T therapies. Under the previous REMS, treatment centers had to be specially certified, and patients were required to stay near the facility for four weeks after infusion. The new policy shortens the mandatory observation period from four weeks to two. The FDA’s rationale is that real-world evidence now shows CAR-T toxicities are predictable and manageable within the first two weeks.
For patients and families, this is a double-edged signal. On one hand, it reflects genuine progress in recognizing and treating acute neurotoxicity. On the other, the removal of REMS does not change the underlying biology. ICANS still occurs. Delayed neurotoxicities still occur beyond the two-week window. The regulatory relaxation means more of the vigilance burden shifts from mandated institutional protocols to the clinical judgment of individual treatment teams — and, inevitably, to patients and caregivers themselves. Lifetime monitoring for secondary malignancies is now recommended for all CAR-T recipients. Families should understand that “lifetime monitoring” is not a formality — it means ongoing follow-up visits, blood work, and attentiveness to new symptoms for years and potentially decades after treatment.
What CAR-T Brain Risks Mean for Patients With Existing Cognitive Concerns
For families navigating dementia, mild cognitive impairment, or other neurodegenerative conditions, the prospect of a therapy that can disrupt the blood-brain barrier and cause brain inflammation carries a particular weight. Consider a patient in their seventies with both relapsed lymphoma and early-stage Alzheimer’s disease. CAR-T therapy might offer the best chance at cancer remission, but the 26.9 percent baseline risk of ICANS — with its potential for seizures, cerebral edema, and rapid cognitive deterioration — intersects with a brain already under neurodegenerative stress.
There is limited published data on how pre-existing cognitive impairment modifies ICANS risk or severity, which is itself a gap that needs attention. The practical takeaway is that any discussion of CAR-T eligibility for a patient with cognitive vulnerability should involve both the oncologist and a neurologist, ideally one with experience in neuro-oncology or neurotoxicity. Baseline cognitive testing before infusion is essential — not just for clinical grading of ICANS if it occurs, but for establishing a reference point against which delayed changes can be measured months or years later.
Where CAR-T Neurotoxicity Research Is Heading
The trajectory of CAR-T neurotoxicity research is moving in two directions simultaneously. The first is prevention — designing next-generation CAR-T constructs that are less prone to triggering the cytokine cascades that breach the blood-brain barrier. Some experimental designs incorporate suicide genes or molecular switches (like the dasatinib-responsive systems now in trials) that allow clinicians to dial down CAR-T activity if neurotoxicity begins. The second direction is understanding the long-term neurological footprint of CAR-T therapy.
As survival rates improve and more patients live years beyond their infusion, the question shifts from “did they survive ICANS” to “what is the lasting effect on their brain.” The EBMT’s 2025 guidelines for non-ICANS complications are a first step, but longitudinal neurological follow-up studies — tracking cognition, motor function, and neuroimaging changes over five, ten, and twenty years — are what will ultimately define whether CAR-T therapy carries a chronic neurological cost. For the brain health community, this is a space worth watching closely. CAR-T therapy is expanding beyond blood cancers into solid tumors, autoimmune diseases, and even early investigations in neurological conditions. Each new application will bring its own neurotoxicity profile, and the lessons learned from ICANS and delayed parkinsonism will shape how those risks are managed.
Conclusion
CAR-T cell therapy represents one of the most significant advances in cancer treatment in a generation, but its neurological side effects are neither rare nor trivial. ICANS affects roughly one in four recipients, with severe forms occurring in about one in ten. Fatal neurotoxicity, while uncommon, is real — estimated at 3 to 9 percent for CD19-directed therapies. Beyond the acute ICANS window, delayed complications including parkinsonism, neuropathies, and other immune-mediated neurological conditions are now recognized as distinct risks requiring their own management frameworks. The FDA’s boxed warnings on all six approved products reflect the seriousness of these concerns, even as the recent elimination of REMS requirements signals growing confidence in early detection and treatment.
For patients, caregivers, and anyone invested in brain health, the key action is informed engagement. Understand the specific CAR-T product being recommended, its associated neurotoxicity rates, and what monitoring protocols the treatment center has in place — both during the acute period and for long-term follow-up. If the patient has any pre-existing cognitive or neurological condition, insist on neurological consultation before and after infusion. Ask about baseline cognitive testing. Ask about the plan if delayed symptoms appear weeks or months later. CAR-T therapy can be life-saving, but the brain deserves the same careful advocacy that drives the decision to pursue the treatment in the first place.
Frequently Asked Questions
What is the most common early sign of ICANS after CAR-T therapy?
The earliest signs are often subtle — altered handwriting, mild tremor, word-finding difficulty, and slight confusion. These typically appear around five days after infusion. Because these symptoms can seem minor, caregivers and patients should report any cognitive or motor changes to the medical team immediately, even if they seem insignificant.
Can tocilizumab treat ICANS the way it treats cytokine release syndrome?
No. This is a common misconception. Tocilizumab is effective for cytokine release syndrome but does not resolve ICANS and may actually worsen it. Corticosteroids — particularly dexamethasone and, in severe cases, high-dose methylprednisolone — are the standard treatment for ICANS.
How long does ICANS typically last?
In most patients, ICANS lasts approximately ten days. However, the severity varies enormously. Some patients experience only mild confusion that resolves on its own, while others develop seizures or cerebral edema requiring intensive care. Most patients with Grade 1 or 2 ICANS recover fully, but outcomes for severe cases are less predictable.
Is the delayed parkinsonism seen after CAR-T therapy the same as Parkinson’s disease?
The symptoms overlap — rigidity, slow movement, facial masking, balance problems — but CAR-T-associated parkinsonism appears to have a different cause, likely related to immune-mediated damage rather than the progressive neurodegeneration seen in idiopathic Parkinson’s disease. Whether it is progressive, stable, or potentially reversible is still under investigation.
Are patients with pre-existing dementia or cognitive impairment at higher risk for ICANS?
There is limited published data specifically addressing this question, which is itself a concern. Clinically, a brain already under neurodegenerative stress may be more vulnerable to the inflammatory insult of ICANS, but this has not been rigorously studied. Any patient with cognitive vulnerability should have a neurologist involved in the treatment planning process.
Why did the FDA remove the REMS requirement if CAR-T neurotoxicity is still a real risk?
The FDA’s decision, made in June 2025, was based on real-world evidence showing that most CAR-T toxicities are predictable and manageable within the first two weeks. The observation period was shortened from four weeks to two. This does not mean the risks have diminished — it means clinicians have become better at recognizing and treating them early. Patients and caregivers should remain vigilant beyond the two-week window, particularly for delayed neurological symptoms.
