April 2026 Research Update

This month added a few early signals that certain immune-focused treatments can be followed by short-term improvement in some autoimmune encephalitis situations, but it also highlighted that controlling inflammation does not always stop longer-term brain changes. The strongest treatment-focused data were in children receiving plasma exchange and adults with anti-LGI1 encephalitis receiving efgartigimod, where patients improved after treatment in these reports. At the same time, an extended follow-up case of anti-CD19 CAR T-cell therapy showed stable control of relapses and antibodies while neurodegeneration still progressed, showing a possible gap between calming immune activity and preventing ongoing damage. [1] [2] [3] [4]

In pediatric intensive care settings, therapeutic plasma exchange (a procedure that filters immune factors out of blood) was followed by clinical improvement in most children across immune-mediated neurologic disorders, including autoimmune encephalitis, and no life-threatening complications were reported. Earlier start (within 7 days of symptom onset in that cohort) was linked to a higher likelihood of improvement, which supports the idea that timing may matter for recovery in acute illness. A separate pediatric cohort comparing two plasma exchange methods reported that 6-month outcomes were similar between the two approaches, while side effects were reported in about 20% of sessions and became less likely with increasing age. [1] [2]

For anti-LGI1 encephalitis, a small prospective multicenter report of efgartigimod (a medicine that lowers IgG antibodies) described improved function and cognition by 2 weeks, along with lower total IgG and lower anti-LGI1 antibody levels. This is encouraging but still early, because the group was small, there was no randomized comparison group, and follow-up was short. A separate single-patient report of anti-KLHL11 encephalitis also described improvement after a combination that included efgartigimod along with steroids and IVIG, but one person cannot show how much each treatment contributed. [3] [5]

A few case reports described improvement after immunotherapy in different settings, including seronegative autoimmune encephalitis and anti-NMDA receptor encephalitis after infection or with tumor removal. Another case report described restarting nivolumab after a prior immune-checkpoint-inhibitor-related encephalitis without recurrence over 9 months, which is an early safety observation rather than proof of what will happen for others. In contrast, one suspected seronegative autoimmune encephalitis case did not improve despite multiple immune treatments, showing that some severe, fast-moving illnesses can remain difficult to reverse even with aggressive care. [6] [7] [8] [9] [10] [11]

The most novel treatment report was a single patient with refractory anti-DAGLA antibody-mediated autoimmune encephalitis treated with anti-CD19 CAR T cells, where relapses and autoantibodies stayed controlled on follow-up. The same report described continued neurodegeneration despite that immune control, which matters because it suggests that for some people, the disease process may include injury pathways that do not fully stop when antibody-driven inflammation is suppressed. Because this is one patient, it is best read as an important signal to study rather than a general expectation. [4]

For patients and caregivers, these studies add to a practical picture: immune-based treatments can be followed by improvement in some situations, but responses vary and may be incomplete. The pediatric plasma exchange studies are based on real-world hospital experience and help clarify expected patterns such as frequent improvement and mostly manageable side effects, although they do not prove which approach is best for autoimmune encephalitis specifically. The efgartigimod report in anti-LGI1 encephalitis gives an early human signal that lowering IgG and antibody levels can line up with short-term symptom improvement, but larger and longer studies are still needed. The CAR T-cell case is mainly early research; it helps clarify that relapse control and antibody suppression may not always prevent ongoing brain changes, and it leaves open what might be needed to protect the brain over the long term. [1] [2] [3] [4]

This month reinforced that autoimmune encephalitis diagnosis is often not straightforward, even when antibody tests are positive. In a referral-center review, about half of people evaluated ultimately received another explanation for their symptoms, including some who had neural antibodies. Several case-based reports also showed how autoimmune encephalitis can look like primary psychiatric illness, seizures-only presentations, or other neurologic diseases, and how early MRI or spinal fluid tests can be normal. [12] [13] [14]

A key message from the referral-center cohort was that antibody results did not guarantee autoimmune encephalitis. Among the alternative diagnoses, some patients had antibodies considered low clinical relevance, but a smaller group had antibodies usually considered highly relevant, showing that context still matters. The most common alternatives included other immune-mediated disorders, somatic symptom disorder, and primary psychiatric conditions, which helps explain why evaluation often has to stay broad rather than focusing on one label. [12]

Two men with seizures and mass-like temporal brain lesions were initially treated as anti-NMDA receptor encephalitis but were later found on biopsy to have glioblastoma, highlighting that an antibody result can mislead when imaging looks like a tumor. Another case described a long course initially thought to be immunotherapy-responsive seronegative autoimmune encephalitis that later proved to be chorea-acanthocytosis on genetic testing, again showing how some progressive disorders can mimic autoimmune disease early on. These reports add to the bigger picture that repeated reassessment is sometimes needed when the clinical course does not fit. [15] [16]

Several reports focused on autoimmune encephalitis beginning with symptoms that could be mistaken for psychiatric illness or for other seizure disorders. Examples included anti-NMDA receptor encephalitis starting with acute psychosis and seizures, anti-GABA(B) receptor encephalitis presenting with a new seizure while early tests were unrevealing, and anti-NMDA receptor encephalitis presenting with a reversible lesion on MRI that first looked like a psychiatric condition. Another case described pilomotor seizures (goosebumps) most often coming from the temporal lobe and being linked to causes that can include autoimmune encephalitis, which helps patients recognize why unusual seizure features sometimes trigger an autoimmune workup. [13] [14] [8] [17]

Beyond classic autoimmune encephalitis categories, case reports showed overlap with other immune or medical contexts, such as suspected IgG4-related pachymeningitis occurring alongside anti-NMDA receptor antibodies, and new seizures with tissue-based autoimmune encephalitis test positivity after a complicated postoperative course. These cases do not prove a single shared cause, but they illustrate how the immune system can be involved in complex ways that make diagnosis and labeling challenging. [18] [19]

For patients and caregivers, these studies help explain why getting to a confident diagnosis can take time and may involve revisiting earlier assumptions. The referral-center study suggests that alternative explanations are common, so uncertainty during evaluation is not unusual, and antibody tests are only one piece of the puzzle. The tumor and genetic mimic case reports are reminders that when symptoms or imaging do not match the expected autoimmune encephalitis pattern, other diagnoses may still be possible even with antibody positivity. Taken together, this body of work matters most for understanding the diagnostic journey; it does not point to a single new test or rule that settles diagnosis for everyone. [12] [15] [16]

This month added research clues about measuring inflammation, understanding risk, and mapping possible mechanisms, but most of it is not yet ready to guide individual decisions. A meta-analysis supported CSF soluble CD27 as a general marker of neuroinflammation across conditions that include autoimmune encephalitis, while noting that test methods differed between studies. Other work linked certain HLA gene types to anti-LGI1 encephalitis risk and to earlier or later symptom onset, and animal research explored how anti-NMDA receptor antibodies might trigger brain changes through specific signaling pathways. [20] [21] [22]

Across 10 studies, CSF soluble CD27 was higher in neuroinflammatory diseases than in controls, with strong overall diagnostic performance reported. For patients, the practical meaning is that this marker may reflect immune activation inside the brain and spinal fluid, but the review also reported substantial differences between studies in how the test was done and reported. That variability is a reminder that a promising biomarker still needs standardization before it can be compared cleanly across hospitals. [20]

In a cohort of 80 people with anti-LGI1 encephalitis, certain HLA gene variants were much more common than in controls, with DRB1*07:01 reported as a major risk allele. The same study reported differences between people with onset before age 50 and those with later onset, including more generalized tonic-clonic seizures at onset in the younger-onset group and lower disability scores at onset. These findings do not change diagnosis by themselves, but they help clarify that immune genetics may shape who gets anti-LGI1 encephalitis and how it first shows up. [21]

In mice, researchers created an anti-NMDA receptor encephalitis model by transferring human anti-NMDA receptor antibodies into the brain and observed temporary problems with memory-like tasks and depressive-like behaviors, along with changes in synaptic support proteins. They also reported activation of a PI3K-AKT-HIF-1alpha pathway and microglial activation, which are clues about possible steps between antibodies and brain dysfunction. This is laboratory research, so it cannot predict a person s course, but it can guide future work on protective strategies. [22]

For patients and caregivers, this set of studies mostly adds understanding rather than changing what happens in clinic today. The soluble CD27 review supports that immune activity in spinal fluid can sometimes be captured with measurable markers, but it remains uncertain how best to use this test for autoimmune encephalitis specifically. The anti-LGI1 HLA study suggests that inherited immune factors may influence risk and presentation, but genetic findings do not tell an individual what will happen or replace antibody testing and clinical evaluation. The mouse study matters for future research because it offers a possible chain of events from antibodies to brain changes, but it is not a direct test of treatments in people. [20] [21] [22]

This month also highlighted that autoimmune encephalitis can carry a heavy hospital-care burden, and that access to diagnostic testing varies widely. A single-center cost study reported long hospital stays were a major driver of direct hospital costs, especially when intensive care was needed. A global clinician survey reported financial barriers and limited access to neural antibody testing in many regions, which was linked to more empiric treatment without confirmatory testing. [23] [24]

In the cost cohort of 35 patients, about 31% required ICU care and the median hospital stay was 26 days, with higher costs in ICU-treated patients. Costs did not differ by antibody-positive versus antibody-negative status overall, but NMDA receptor antibody positivity was linked to higher costs in that center. The study also reported that delays in diagnosis and treatment were among factors linked to longer stays, which helps explain why early diagnostic clarity is often discussed as a goal in this disease. [23]

In the global survey of 157 clinicians, about 69% reported familiarity with updated definitions for autoimmune-related seizures, but many still reported difficulty identifying autoimmune seizures clinically. Testing access and cost barriers were common, and many respondents reported treating empirically without confirmatory antibody results. This does not describe what patients should do, but it helps explain why evaluation pathways can look different depending on location and resources. [24]

For patients and caregivers, these studies help set expectations about the practical side of autoimmune encephalitis care. The cost data underline that ICU care and long hospital stays can be a major part of the illness for some people, which can affect family logistics and stress even when medical outcomes are not described. The clinician survey helps clarify that differences in testing access can shape how quickly a clear diagnosis is reached and why treatment may sometimes start before all test results are available in some settings. What remains uncertain is how these system factors translate into differences in long-term outcomes, because neither study directly tested patient recovery strategies. [23] [24]