Pimavanserin May Hold Promise for Alzheimer's Psychosis

Pauline Anderson

February 20, 2018

Pimavanserin reduces psychosis related to Alzheimer's disease (AD), psychosis new research suggests. But at least one expert questions whether the study's findings warrant progression to a phase 3 trial.

Results of a phase 2 randomized, double-blind trial show the drug was significantly more effective than placebo in a group of nursing home residents with possible or probable AD and psychotic symptoms. In addition, researchers found the drug was well tolerated, with minimal side effects, and had no negative effect on cognition on motor function.

"The study showed that the drug was more than twice as effective as current atypical antipsychotics with fewer adverse events, so potentially a big step forward," lead author, Clive Ballard, MD, dean of the medical school and professor of age-related diseases, University of Exeter, United Kingdom, told Medscape Medical News.

A phase 3 study is already underway, said Ballard.  "We will know in about 18 months whether this is a new treatment of psychosis in people with Alzheimer's disease."

The study was published in the March issue of Lancet Neurology.

Approved for Use in Parkinson's

While most antipsychotics bind to dopamine, acetylcholine, histamine, and 5-HT2A receptors, pimavanserin targets only 5-HT2A receptors. This, said Ballard, means it leads to fewer symptoms, such as parkinsonism and sedation.

The drug is marketed in the United States for hallucinations and delusions associated with Parkinson's disease. No antipsychotics are approved in the United States for agitation or psychosis associated with dementia, although these drugs are commonly used in patients with AD.

The aim of the current study was to include patients with hallucinations or delusions occurring after the onset of dementia that continued for more than 4 weeks and required antipsychotic treatment.

To ensure that only patients who required pharmacologic treatment were included in the study, prospective participants completed a 3-week brief psychosocial therapy session.

These sessions allowed family members and care staff to interact with the person with dementia experiencing psychotic symptoms for 10 to 15 minutes a day "in a shared enjoyable activity," said Ballard.

Some who had been initially screened improved during this period and were no longer eligible for the study.

The investigators randomly assigned 181 residents of London-area nursing homes who experienced psychosis and had probable or possible AD to receive placebo (two tablets) or pimavanserin (two 17-mg tablets) administered daily for up to 12 weeks. This dose of pimavanserin is the same as that used to treat psychosis in patients with Parkinson's disease.

The primary outcome was change from baseline to 6 weeks in the Neuropsychiatric Inventory-Nursing Home version (NPI-NH) psychosocial score (hallucinations and delusions).

Of those randomly assigned, 178 were included in the efficacy analysis (87 in the treatment group and 91 in the placebo group); most were female, and the mean age was about 86 years. Of these, 160 completed 6 weeks of treatment and 140 completed 12 weeks of treatment.

The adjusted mean change in the NPI-NH psychosis score at week 6 was –3.76 for pimavanserin and –1.93 for placebo (mean difference, –1.84; 95% confidence interval [CI], –3.64 to –0.04]; P = .045).

Frail Patient Population

Response, defined as  a 30% or greater improvement, was observed in 55.2% of pimavanserin recipients vs 37.4% of placebo recipients (P = .016).

A subgroup analysis showed that the size of benefit in people with severe psychosis (NPI-NH psychosis score ≥ 12) was "even more striking," said Ballard. In this group, the adjusted mean change in score at week 6 was –10.15 (95% CI, –12.50 to –7.80) for pimavanserin and –5.72 (–8.14 to –3.30) for placebo (mean difference, –4.43; 95% CI, –7.81 to –1.04; P = .011).

In participants with mild psychotic symptoms (NPI-NH psychosis score < 12), the mean difference between treatment and placebo groups was –0.42 (95% CI, –2.52 to 1.68; P = .694).

The drug might not work as well in people with less severe psychosis because milder symptoms are more likely to have multiple causes, such as environmental triggers, said Ballard.

"This probably means that the people with the strongest biological risk for psychosis do better."

Very frail older people with AD living in nursing homes "are at highest risk of psychosis and the group where a treatment is most urgently needed," added Ballard.

NP-NH score for agitation at week 6 or week 12 did not significantly differ between pimavanserin and placebo. However, the authors noted that this was a secondary outcome and that these symptoms were generally mild and below the usually recognized threshold of clinical significance.

By week 12, no significant advantage for pimavanserin vs placebo was observed for the overall study population or for the severe subgroup. While this might indicate the absence of sustained benefit, Ballard noted that the study was powered only for effects at 6 weeks.

And this finding might not be surprising given the remitting and relapsing nature of psychosis in people with AD, he said.

"There is a lot of spontaneous resolution of psychotic symptoms over 3 months, which introduces a lot of noise and makes ongoing benefit difficult to measure."

Falls, Urinary Tract Infections

He added that most people with spontaneous recovery will relapse again over 12 months.

"This is why the phase 3 study is evaluating sustained benefit by examining whether pimavanserin can prevent relapse."

There was minimal change at 12 weeks in cognition as measured by the Mini-Mental State Examination or in motor function as assessed by the Unified Parkinson's Disease Rating Scale part III. The lack of detrimental effects on cognitive and motor symptoms could be a unique advantage in elderly patients with AD and psychosis, said the authors.

At least one adverse event was reported in 96% of study participants. The most common of these in both groups were falls, urinary tract infections (UTIs), and agitation.

The incidence of falls and UTIs was higher in patients with agitation receiving the treatment drug.

There is "no clear mechanism" that might explain this, but because these are important symptoms, they should be carefully monitored in future studies, said Ballard.

There were no notable differences between groups with regard to physical examination results, vital signs, or clinical laboratory tests.

At week 12, mean change in body weight was –0.7 kg with pimavanserin and –0.1 kg with placebo. Ballard pointed out that weight loss occurred only in a small number of study patients and has not been seen in other trials in people with Parkinson's disease.

"It is a potentially important symptom in people with Alzheimer's disease, though, so it needs to be carefully monitored in further studies.

After 12 weeks, the mean change in the corrected QT interval in patients receiving pimavanserin was 9.4 milliseconds ms versus –2.0 milliseconds in patients receiving placebo.

"This did not lead to any adverse cardiac events in this study or in previous studies of people with Parkinson's disease," commented Ballard. However, he stressed that patients need to have electrocardiography before starting treatment.

Chance Finding?

In an accompanying editorial, Lon S Schneider, MD, professor of psychiatry, neurology, and gerontology, Keck School of Medicine, University of Southern California, Los Angeles, and director, California Alzheimer's Disease Center, argued that this phase 2 trial "did not yield the evidence of proof of concept that would inform a subsequent phase 3 confirmatory trial."

In an interview with Medscape Medical News, Schneider elaborated and noted the absence of effect for pimavanserin at all but one timepoint.

"There was no effect for pimavanserin at weeks 2, 4, 9, and 12, just at week 6," he said.

He questioned whether the remitting and relapsing course of psychosis in AD could lead "by play of chance" to a statistical difference only at week 6.

He also noted that "there were no significant secondary or exploratory outcomes that might support the 6-week finding."

Only one subgroup analysis — baseline psychosis severity score — was notable, and here, only about 30% of the more severe group showed a significant effect, said Schneider.

"There was absolutely no effect for the milder group, not even a trend."

Schneider pointed out that hallucinations are relatively uncommon in AD and that the more severe group in the study had hallucinations and the less severe group had very few hallucinations, while both groups had delusions.

If there was any true effect at week 6, "it probably would have been in the most severe group and at improving hallucinations," he said.

While testing the drug in patients not residing in nursing homes might be "reasonable," no evidence from the current study would support — or not support — focusing on outpatients or younger patients, said Schneider.

"To the extent that outpatients may be less severely psychotic, they might be less likely to benefit, but you'd have to test this," he said.

The study raises questions about whether psychosis or agitation should be targeted, said Schneider.  He noted that while the trial targeted "psychosis of AD," patients in the study typically also had agitation.

The distinction between psychosis of dementia and agitation is not clear because delusions and agitation are common and substantially overlap, he added.

"Given the negative outcome on the NPI agitation/aggression scale, there's no evidence that pimavanserin helps agitation. It might, but there is no evidence."

Unfortunately, concludes Schneider in his editorial, "experimental drugs often are advanced to larger phase 3 trials without evidence from phase 2 trials."

Ballard has received grants and personal fees from ACADIA and Lundbeck, personal fees from Heptares, Roche, Lilly, Otsuka, Orion, GlaxoSmithKline, and Pfizer. Schneider reports grants from the National Institute on Aging,the State of California, and the University of Southern California; grants from Eli Lilly, Lundbeck, Novartis, and Biogen; grants and personal fees from Merck, Roche/Genentech, and TauRx; and personal fees from AC Immune, Avraham, Boehringer Ingelheim, Cerespir, Cognition, Neurim, Stemedica, Takeda, vTv, Toyama/FujiFilm, Heptares, Allergan, Axovant, and Impel NeuroPharma.

Lancet Neurol. 2018:17:213-222, 194-195. Abstract, Editorial

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.