The novel schizophrenia drug showing renewed potential for better outcomes

The green light for xanomeline-trospium chloride (Cobenfy; Bristol Myers Squibb) in September 2024 breaks a 70-year drought in novel medicines for schizophrenia. However, its approval by regulators in the United States has been met with fanfare from some experts and caution from others.

Xanomeline-trospium chloride’s action on muscarinic acetylcholine receptors stands apart from current antipsychotics, which all modulate dopamine D2 receptors.

“For the past 70 years, all antipsychotic drugs have shared this mechanism of action,” explains Robert McCutcheon, associate professor and consultant psychiatrist at the University of Oxford. “That’s really been the only game in town, up until the recent approval of Cobenfy.”

While dozens of antipsychotics are already available, a new option could be good news. Current antipsychotics come with diverse side effects, such as weight gain, involuntary motor spasms or rigidity, and sedation. These can prompt people to switch or discontinue their medicines, which complicates their recovery and contributes to shortened life expectancy^​1​. Despite these options, around one in three people are deemed ‘treatment-resistant’ when their symptoms are not relieved by at least two antipsychotics^​2​.

Xanomeline-trospium chloride offers a different way, by working on the brain’s acetylcholine system. It combines two drugs in an oral capsule: xanomeline, a muscarinic receptor agonist, and trospium chloride, which blocks xanomeline’s actions outside of the brain, subsequently limiting unwanted side effects.

Over the past five years, research has highlighted the potential of this combination: it not only reduces psychosis in people with schizophrenia, but it also shows some efficacy for other symptoms, such as the disabling lack of motivation, social withdrawal and cognitive impairment that remain untouched by current antipsychotics. While xanomeline-trospium chloride has some gastrointestinal side effects, these may prove more tolerable to more people than currently available antipsychotics.

“This is really the first kind of novel, effective treatment we’ve had for decades. There is a potential not only to improve treatment, but to also tell us more about the disorder,” McCutcheon says.

Whether these possibilities pan out with wider clinical use remains to be seen. Clinical trials have so far compared the drug with placebo, which leaves unclear how xanomeline-trospium chloride fares against current antipsychotics and which patients would be good candidates for it. Questions remain about its long-term efficacy, side effects and patient acceptability. The jury is still out as to whether the drug truly addresses negative and cognitive symptoms, or whether the improvements detected so far are a by-product of quelling psychosis.

“What everybody wants is for this drug to be a major step forward in the treatment of schizophrenia,” says David Taylor, professor of psychopharmacology at King’s College London and director of pharmacy and pathology at the Maudsley Hospital in south London.

“It has potential, but we know very little about it so far, and we should withhold judgement until we’ve seen comparative trials, and until we ask patients what they think,” he adds.

Serendipity for schizophrenia

Schizophrenia comes with a range of symptoms that usually begin in young adulthood. Psychosis is the most prominent, since it disconnects a person from reality through hallucination and delusions. Other symptoms include ‘negative symptoms’, which are marked by social withdrawal and apathy, and ‘cognitive impairment’, which makes it difficult to remember, learn or pay attention.

Negative and cognitive symptoms are considered to be more debilitating than psychosis because they make working, living independently and participating in society difficult and often persist when psychosis is managed^​3​. Although schizophrenia has an estimated prevalence of just 7 people per 1,000 in England, the disability associated with it results in high use of medical services and significant burdens to family and society^​4​.

The first drug treatment for schizophrenia was discovered serendipitously in 1952. Originally synthesised as an antihistamine, chlorpromazine was found to relieve psychosis owing to antagonism at D2 dopamine receptors. Newer antipsychotics have since been developed and these all have D2 dopamine receptor modulation in common. In the UK, commonly prescribed antipsychotics for schizophrenia include olanzapine, aripiprazole and risperidone^​5​.

However, dopamine-modulating drugs also produce a thicket of side effects: movement disorders; cardiac and metabolic effects; hyperprolactinaemia; and sedation.

“Because there aren’t huge differences in efficacy in terms of how well the drugs work, a lot of treatment decisions are based on the side effect profile of a medication,” McCutcheon says.

Like chlorpromazine, serendipity also had a hand in xanomeline-trospium chloride’s development. Xanomeline alone was initially developed in the 1990s as a cognition booster for people with Alzheimer’s disease; the drug selectively activates the M1 and M4 subtype of muscarinic receptors, which are highly expressed in the brain’s cortex, hippocampus and striatum — places likely to mediate acetylcholine’s effects on cognition.

A clinical trial of xanomeline in people with probable Alzheimer’s disease bore this out in 1997, with increases in cognition scores. Unexpectedly, the study also noted improvements in psychosis-like behaviours associated with dementia, such as agitation, hallucination, delusions and vocal outbursts^​6​.

“It was really quite dramatic, and I think that’s turning out to be a seminal paper, because a lot of people in academia and industry started working on muscarinic receptors to try to come up with a novel antipsychotic,” says Steven Paul, a professor of psychiatry and neurology at Washington University School of Medicine, Missouri. Paul has also worked in the biotech industry to develop psychiatric drugs and was involved in the development of xanomeline-trospium chloride as head of research and development at Eli Lilly.

In 2008, a small trial in 20 people with schizophrenia followed, which revealed similar antipsychotic effects; however, the side effects owing to activation of muscarinic receptors in the peripheral nervous system — nausea, vomiting, and gastrointestinal distress — prevented Eli Lilly from moving forward with the drug^​7​.

Eventually, scientists thought to combine xanomeline with an agent that would block its unwanted effects outside of the brain. This combination of xanomeline and trospium chloride, a known muscarinic receptor antagonist that does not cross the blood-brain barrier, was then developed as KarXT by Karuna Therapeutics after xanomeline was licensed to the company in 2012.

“The thinking was that maybe this combination therapy would give the beneficial effects that are mediated in the brain, but avoid those peripheral side effects that were problematic,” explains Paul, who joined the board of Karuna Therapeutics in 2018.

Emerging trials

This prediction was confirmed in ‘EMERGENT-1’, a double-blind, placebo-controlled phase II trial of the combination xanomeline-trospium chloride^​8​. Using data from 182 people diagnosed with schizophrenia who had been hospitalised owing to worsening psychosis, the researchers found that those receiving xanomeline-trospium chloride for five weeks showed significantly greater symptom improvements than those receiving a placebo.

Unlike xanomeline alone, the side effects seemed tolerable, with similar numbers of participants in the drug and placebo groups discontinuing the trial or reporting adverse events. Among those taking the drug, the most common side effects were constipation, nausea, vomiting and dry mouth, with the latter three declining over time.

Two larger phase III trials with similar designs followed (EMERGENT-2 and EMERGENT-3)^​9,10​. These inpatient studies also lasted five weeks and reported both positive and negative symptom reductions with xanomeline-trospium chloride over placebo.

Again, the drug was safe and well-tolerated, although it did cause some gastrointestinal side effects, such as nausea, indigestion, constipation and vomiting. The rate of discontinuation did not differ from placebo, and measurement of side effects common to dopamine-related antipsychotics were also not different from placebo.

“If you had to put your kid on an antipsychotic drug, this is the kind of drug you want to put them on, not a heavy-duty D2 blocker. There’s no sedation, no somnolence, no extrapyramidal symptoms, no weight gain, and no likely risk of tardive dyskinesia,” Paul says.

These trials have turned into one-year open-label extension trials that started in early 2021 (‘EMERGENT-4’ and ‘EMERGENT-5’), with preliminary meeting reports indicating stable or improved metabolic health and safety^​11​. Pooled analysis of the first three EMERGENT studies have pointed to similar improvements across subgroups of people, as defined by age, sex, race, body mass index and baseline symptoms^​12​.

Notably, based on computerised assessments of visual memory, attention, verbal recall and recognition, a new analysis of data collected in the two phase III studies found that xanomeline-trospium chloride improved cognition in the subgroup that showed cognitive impairment at baseline — around half of participants — whereas improvement was not found in the participants as a whole^​13​.

While these findings need to be confirmed, they suggest that xanomeline-trospium chloride can address all of schizophrenia’s symptom domains.

“This may be the very first drug that treats the syndrome we call schizophrenia — not just the positive, psychotic symptoms, but the most disabling negative and cognitive symptoms,” Paul says.

Others have had muted reactions to the studies, citing a history of promising psychiatric drugs that have come and gone. Taylor notes how effect sizes tend to diminish with repeated studies, how side effects can inadvertently unblind a study, and how unknowns about long-term tolerability and patient acceptance could detract from xanomeline-trospium’s potential usefulness.

“I’m not joining in the excitement because I’ve been in this game too long, and I’ve seen it before,” he says.

Who benefits?

Amid the steady stream of positive reports, questions remain about who stands to benefit most from xanomeline-trospium chloride. This requires information about how it compares with current antipsychotics.

McCutcheon and colleagues are planning such a head-to-head trial at the University of Oxford, set to begin in 2025. The trial will enrol people who are unhappy about their treatment, either because it is not working or owing to side effects. They will be randomised to receive either another dopamine receptor antipsychotic or xanomeline-trospium chloride and followed for six weeks. The main outcome measure will be cognition, but standard psychosis metrics will also be used.

The results will help distinguish effects on negative and cognitive symptoms, as well as clarify who might be the best candidates for the drug. For example, xanomeline-trospium chloride might be particularly useful for those who do not respond to dopamine-based antipsychotics, or for those whose symptoms are mainly negative and cognitive in nature. Such a subgroup may have a fundamentally different biology underlying their schizophrenia, which makes them more sensitive to muscarinic modulation.

However, McCutcheon advises some caution with this idea: “We ought to be a bit careful in taking the biology of a disorder and getting it to neatly map onto the kind of drugs that we’ve discovered, by chance,” he says.

Other trials are exploring whether xanomeline-trospium chloride could work as an adjunct therapy for people to take in addition to their current antipsychotic. Having acquired Karuna Therapeutics in March 2024, Bristol Myers Squibb is testing this in a phase III trial expected to finish in 2025, with an open-label extension trial to test long term efficacy and safety^​14,15​.

Other muscarinic receptor-targeting drugs are also in the pipeline, including one that targets M1 receptors and another focused on M4 receptors^​16,17​. Emraclidine, a novel M4 receptor-selective positive allosteric modulator, did not show improvements in schizophrenia symptoms in two phase II trials, despite earlier positive results^​18​.

This raises the question of whether action at both M1 and M4 receptors is important for therapeutic effects, or whether the emraclidine studies failed for reasons that challenge all clinical trials, such as recruiting the correct patients or dosing.

“There’s tremendous debate now about what is the secret sauce of Cobenfy,” Paul says. “Nobody knows.”

In the meantime, McCutcheon anticipates it will be quite a few years before xanomeline-trospium chloride is available through the NHS. Even if it is licensed in the UK, it will take time to make a convincing case for its cost effectiveness. In the United States, Cobenfy has been priced at US$1,850 per month, but cheaper generic antipsychotics will compete.

As scientific, regulatory and cost issues are worked out for xanomeline-trospium chloride, a lot more can be done with current medicines. Some argue that clozapine and long-acting injectable antipsychotics are underused in schizophrenia despite their demonstrated efficacy, owing to “psychopharmacologic laziness”^​19​. These drugs require careful monitoring and paperwork when prescribed, which has led to reluctance in using them in practice.

“We could do so much better with what we’ve got,” Taylor says.