Understanding Vivid Animal Hallucinations
Peter Jones, an academic psychiatrist at the University of Cambridge, once treated a patient who experienced remarkably vivid hallucinations of animals. During his visits, the patient engaged with these creatures “as realistically as he did with me,” Jones recalls. Traditional antipsychotic medications failed to alleviate these Snow White-like visions. However, after undergoing a plasma exchange—a treatment typically reserved for malfunctioning immune systems—the patient’s hallucinations vanished completely.
This case did not represent a typical instance of schizophrenia. Instead, the patient’s symptoms were attributed to autoimmune encephalitis, a rare form of brain inflammation that can lead to hallucinations and other psychotic symptoms. When treated with medications that inhibit the immune system, these patients often see a reduction in their psychotic manifestations.
Autoimmune encephalitis is usually diagnosed through blood tests or analysis of cerebrospinal fluid to identify specific types of antibodies. A positive result may suggest that the immune system is erroneously attacking the brain. Other diagnostic indicators include bright areas of inflammation on MRI scans, rapidly deteriorating psychotic symptoms, and, crucially, unmistakable neurological symptoms such as seizures.
However, the patient seen by Jones did not exhibit these evident neurological symptoms. Had he not suspected an autoimmune cause, the case could have easily gone unnoticed.
Uncovering Overlooked Cases in Psychiatry
Experts express concern that the field of psychiatry might be overlooking a small number of such patients. Efforts are underway to identify more individuals who could benefit from autoimmune testing by expanding the range of psychiatric patients screened and investigating new antibodies that may be associated with autoimmune encephalitis.
Simultaneously, there is apprehension about potentially creating false hope and diverting patients away from the psychiatric treatments that the majority genuinely require, should there be an overdiagnosis of autoimmune encephalitis.
Michael Zandi, a neurologist at University College London, asserts that the “artificial division” between the body and mind separating psychiatry from neurology is beginning to crumble. Psychosis can stem from a mix of mental and biological factors, including psychological trauma, substance abuse, and physical illnesses or infections that disrupt brain activity. Approximately 3% of individuals will experience a psychotic episode in their lifetime.
The Historical Context of Autoimmune Encephalitis
Brain inflammation caused by the body attacking itself, as seen in autoimmune encephalitis, has been recognised for a couple of decades. In the early stages of this condition, many patients display core features of psychosis—hallucinations, delusions, and agitation—and are often referred to psychiatric services as a result.
A landmark research article published in 2007 detailed cases involving twelve women with autoimmune encephalitis. Nine of these women initially displayed symptoms resembling psychosis, while the others primarily experienced memory loss. Except for one, all eventually suffered from seizures. Researchers discovered that their condition was linked to antibodies that bind to NMDA receptors in the brain, which are crucial for numerous cognitive processes, including learning and memory formation.
The triggering factor for the autoimmune encephalitis in these patients appeared to be tumours, with 90% of them being ovarian. The five tumours examined contained nervous tissue with NMDA receptors, misleading the immune system into recognising the NMDA receptor as foreign, resulting in increased production of antibodies targeting it.
Linking Infections to Autoimmune Responses
However, these antibodies also bound to healthy NMDA receptors in the brain, disrupting normal function. Other immune cells flocked to the area, causing inflammation that, in many instances, led to psychosis. Beyond tumours, experts believe that classical autoimmune encephalitis can also be triggered by prior infections, such as the herpes simplex virus affecting the brain. Nevertheless, the underlying causes often remain a mystery.
Only a small fraction of psychosis cases is attributable to autoimmune causes, with most experts estimating that it affects between 1% and 2% of patients. Without treatment, these individuals risk falling into a coma or even dying, which is why experienced psychiatrists look for additional neurological symptoms, particularly seizures, along with other indicators such as catatonia or mouth spasms.
Identifying Neurological Symptoms
When signs of brain inflammation emerge, blood or cerebrospinal fluid can be analysed for autoantibodies. If these tests, often combined with findings from MRI scans or electroencephalograms, suggest autoimmune encephalitis, patients are urgently treated with immunotherapy—medications designed to curb an overactive immune system. Many experience a full recovery, with Zandi asserting, “We can cure them completely.”
Autoimmune encephalitis is commonly linked to antibodies that target NMDA receptors, known as NMDAR antibodies. However, researchers have identified dozens of additional types of antibodies associated with encephalitis, some of which also appear to induce psychiatric symptoms.
A notable example is LGI1 antibodies, discovered in 2010, which bind to LGI1 proteins in the brain, hindering their ability to transmit signals between nerve cells. This form of autoimmune encephalitis often results in symptoms akin to dementia, leading to initial misdiagnosis. Zandi has treated patients whose apparent dementia vanished following immunotherapy, restoring their independence.
Transformative Outcomes Through Antibody Testing
Belinda Lennox, a psychiatrist at the University of Oxford, has witnessed similarly remarkable transformations. Around the time the 2007 article was published, she began testing psychiatric patients for antibodies when their symptoms appeared to overlap with signs of autoimmune encephalitis, such as rapidly changing psychotic symptoms or neurological features like seizures.
One of Lennox’s early patients who tested positive for NMDAR antibodies was a young librarian who had rapidly descended into paranoia, ceased sleeping, and eventually became mute. Antipsychotic medications exacerbated her condition. After detecting the antibodies, she received treatment to eliminate them from her blood. Within about a week, she made a complete recovery. “It was one of those astonishing moments,” Lennox recalls.
Currently, Lennox is intrigued by the possibility that there exists a broader population of patients with psychosis caused by autoimmune problems that remain unnoticed. This could explain why some individuals do not respond well to psychiatric treatments. “It’s the tip of the iceberg yet to be explored,” she notes.
Investigating Undiagnosed Populations
In her quest to uncover undiagnosed patients, Lennox has been leading a clinical trial called SINAPPS2 for the past decade, investigating whether antibodies might be causing psychosis even in patients without other evident neurological symptoms, such as seizures or comas. Jones from the University of Cambridge is also involved in this trial.
The researchers are evaluating patients who have experienced psychotic episodes for less than two years, looking for four types of antibodies, including NMDAR and LGI1. At the time of testing, none of the participants exhibited severe neurological issues. Lennox reports that the team has identified antibodies in the blood or cerebrospinal fluid of approximately 5% of the nearly 5,000 psychiatric patients analysed.
The Challenges of Blood Testing
However, blood tests are not always reliable. Some of these detections may yield false positives, indicating the presence of autoantibodies that the body does not actually contain. Others may have some circulating antibodies that are harmless and unrelated to their illness. Yet, a third group may be in a situation similar to that of Jones’s memorable patient.
So far, 70 participants have been recruited to receive either immunotherapy or a placebo treatment. Following the treatment, the researchers will monitor their progress for up to a year and a half (Jones emphasises that if any patient begins to experience seizures or other evident neurological symptoms, they will be treated with immunotherapy, even if they were part of the placebo group).
Future Directions in Research and Treatment
Results expected in 2027 are anticipated to clarify the risks and benefits of immunotherapy for patients with psychosis who have tested positive for autoantibodies, guiding future care, according to Jones. A pending and contentious issue is whether testing for these antibodies should be extended to a broader group of patients with early-stage psychosis, a practice already common in Germany and Australia.
Researchers continue to seek new antibodies that affect the brain, hoping to understand patients who do not exhibit known antibodies but present symptoms reminiscent of autoimmune encephalitis. Detecting these new antibodies does not prove they are responsible for the symptoms—establishing causality takes time—but it does provide clues for researchers.
Continuous Discovery of Antibodies
“Every year, a few new antibodies are discovered, and some are more relevant to psychosis than others,” states Jonathan Rogers, a neuropsychiatry researcher at University College London. For instance, in 2024, researchers identified an antibody that binds to a brain protein called CAMKV, detecting this uncommon antibody in five cancer patients, three of whom exhibited psychosis or other psychiatric changes.
Moreover, various scientists are employing methods used in basic research to examine patients suspected of having an autoimmune disease. This