March 28, 2012 (Washington, DC) — Discontinuing risperidone (Risperdal, Janssen Pharmaceuticals) in patients with Alzheimer's disease (AD) who have psychosis or agitation/aggression significantly increases the risk for relapse in these behavioral symptoms, new research shows.
In a study funded by the National Institutes of Health (NIH) of patients with AD who were treated for 4 months previously with risperidone, there was a significantly greater rate of relapse in psychosis and agitation symptoms in patients who were switched to placebo compared with those who continued receiving the active drug. The placebo-treated group also had a shorter time to relapse.
Treatment-related adverse events did not differ significantly between the 2 groups up to 32 weeks later.
Presented here at the American Association for Geriatric Psychiatry (AAGP) 2012 Annual Meeting, the findings appear to be contrary to current guidelines, which recommend that clinicians try to ease AD patients off of antipsychotics as soon as possible.
"This is one of the first studies to really address this issue directly," co-investigator Davangere P. Devanand, MD, professor of clinical psychiatry and neurology at Columbia University in New York City and director of the Division of Geriatric Psychiatry at the New York State Psychiatric Institute, told meeting delegates.
"So our way of thinking is that federal regulations urging early antipsychotic discontinuation may need reconsideration. Our findings certainly raise questions," said Dr. Devanand.
FDA Warning
According to the investigators, psychosis or agitation/aggression in patients with AD can lead to more rapid cognitive decline, future institutionalization, and increased healthcare costs.
However, there are currently no medications approved by the US Food and Drug Administration (FDA) to treat these symptoms in this patient population, said Dr. Devanand.
"Behavioral interventions may be useful, but well-controlled research evidence supporting these approaches is limited," he said.
Although previous research has shown that antipsychotics are the only medications that are superior to placebo in this patient population, the effect size has not been robust. In addition, the agents' efficacy "needs to be weighed against sedation, neurological side effects, and metabolic syndrome."
Further, antipsychotics currently carry a black box warning from the FDA about increased mortality risk, based on pooled clinical trials.
"In other words, federal regulations urge discontinuation after a few months," said Dr. Devanand.
He noted that although previous antipsychotic discontinuation trials have shown mixed results, most did not establish that a treatment response had been achieved before discontinuation. In addition, most of the trials included patients who did not have baseline symptoms of psychosis or agitation and had been receiving antipsychotics "for several years" and/or looked at the discontinuation of several different medications.
"Patients in the US get a bunch of psychotropics and antipsychotics frequently. So discontinuation becomes very complicated. And it can be difficult to interpret which drug is doing what," he said.
Significant Relapse Rate
The Antipsychotic Discontinuation in Alzheimer's Disease Trial had an initial open-label phase in which 180 patients older than 49 years who had probable AD and symptoms of psychosis or agitation/aggression received risperidone (0.25 - 3 mg/day).
Baseline symptoms of psychosis were determined with a score of 4 or higher on delusions or hallucinations on the Neuropsychiatric Inventory (NPI) rating scale, and symptoms of behavioral dyscontrol were determined by a score of 4 or higher on agitation/aggression on the NPI. Exclusion criteria included stroke, transient ischemic attack, or uncontrolled atrial fibrillation.
Participants who responded to treatment at the 16-week mark (n = 110; mean age, 79.9 years; ≥30% decline in NPI core scores) were than randomly assigned to continue receiving risperidone (mean dose, 0.97 mg/day) or were switched to a matching placebo.
All patients were then examined 16 weeks and 32 weeks after randomization. Relapse was defined as a 30% or greater increase in NPI core score and a significant change on psychosis/agitation scores.
Results showed that in the first 16 weeks after randomization, "the time to relapse and risk of relapse favored risperidone over placebo," reported Dr. Devanand.
The relative hazards ratio for relapse for the placebo group was 1.94 (95% confidence interval [CI], 1.09 - 3.45; P = .022).
In addition, 60% of the patients receiving placebo experienced relapse compared with 32.9% of those receiving risperidone (P = .004).
"In other words, the placebo group was almost twice as likely to relapse as the patients on risperidone," said Dr. Devanand.
At the 32-week post-randomization point, the relative hazards ratio for those receiving placebo vs those receiving risperidone was 4.88 (95% CI, 1.08 - 21.98; P = .023). The relapse rates were 48% vs 15% for placebo and risperidone, respectively (P = .017).
Throughout the entire post-randomization period, there were 11 serious adverse events (SAEs) reported, including 3 deaths (2 patients were receiving risperidone and 1 was receiving placebo). There were no significant between-group differences in SAEs for either 16-week examination period.
There were also no significant differences between the 2 groups in adjusted scores on the Treatment Emergent Symptom Scale, the Mini-Mental State Examination, the Alzheimer's Disease Assessment Scale-Cognitive subscale, and the Physical Self-Maintenance Scale, nor were there significance differences in extrapyramidal symptoms on the Simpson-Angus Scale or in body weight.
FDA Response
"Looking at it in context of current federal regulations, if someone actually has these symptoms, you treat with an antipsychotic for 4 months. If they're doing better in terms of the symptoms and side effects are not terrible, then you're supposed to discontinue use of the medication," said Dr. Devanand.
"But our findings suggest that discontinuing means you're more likely to relapse. Although we did not see any difference in side effects, that is still a somewhat open question. So you have to weigh the risk of relapse with risk of side effects after long-term treatment for your individual patient."
During the question-and-answer session after his presentation, Dr. Devanand was asked to speak more about the current FDA black box warning.
"I think we should think of this warning as a starting point. My understanding is that the FDA based this on specific data from certain studies. And perhaps more research is needed to address that," he answered.
That set the stage for a brief discussion by Ni Khin, MD, medical team leader in the Division of Psychiatry Products at the FDA, who told attendees that that she was speaking "on my views and not the official views of the FDA."
"The efficacy of antipsychotics for symptoms of psychosis and behavioral symptoms in Alzheimer's patients has not been established yet in FDA-registration trials," said Dr. Khin.
"In 2005, we put a box warning out based on 17 trials that showed an increase in mortality and cardiovascular effect."
She noted that in the current study, only the initial treatment responders were randomly assigned into the discontinuation phase of the trial. She also pointed out that although 60% of the patients receiving risperidone did not relapse, over 30% of those receiving placebo also did not relapse.
"The main question we often face is: if patients are responding to treatment, how long should we go on treating them? Also, when looking at results from this study, it's important to ask if there was any patient selection bias," said Dr. Khin.
Individual Approach
"Science looking to prove the effectiveness of this drug might be missing some important variables. But I can tell you this: I give Depakote [divalproex sodium]. I give Risperdal [risperidone]. And 70% to 80% of my patients do well, even though you scientifically can't seem to demonstrate how I'm doing that," said Gerald Plovsky, MD, adult and geriatric psychiatrist in private practice in Greensboro, North Carolina, to thunderous applause from the audience during the session's discussion period.
Dr. Devanand replied that it is much harder to find a significant response when comparing a medication against placebo than it is with open-label treatment.
Dr. Plovsky later told Medscape Medical News that "researchers care about what happens to a population. In clinical practice, we care about the individual."
"I think this study is great because it justifies why geriatric psychiatrists are so important. We take the time to get to know the patients, and we do things slowly and cautiously," said Dr. Plovsky, who is also the former director of the Public Mental Health Clinic for Geriatrics in Greensboro.
"The bottom line is, I think when you take people off these drugs, you should go slowly. If you go quickly, then people get sick. And it's pros and cons when it comes to the decisions we make. Over the years, I have used Risperidol, and I think it's saved many people a lot of disaster. And I've rarely seen an adverse side effect from it as long as you give 1 or 2 mg," he concluded.
This study was supported by grants from the NIH and by the Department of Veterans Affairs. The medications were provided by Janssen Pharmaceuticals, "which had no other role in the study," report the investigators. Dr. Devanand reported having received research support from Eli Lilly. Dr. Plovsky has disclosed no relevant financial relationships.
In a study funded by the National Institutes of Health (NIH) of patients with AD who were treated for 4 months previously with risperidone, there was a significantly greater rate of relapse in psychosis and agitation symptoms in patients who were switched to placebo compared with those who continued receiving the active drug. The placebo-treated group also had a shorter time to relapse.
Treatment-related adverse events did not differ significantly between the 2 groups up to 32 weeks later.
Presented here at the American Association for Geriatric Psychiatry (AAGP) 2012 Annual Meeting, the findings appear to be contrary to current guidelines, which recommend that clinicians try to ease AD patients off of antipsychotics as soon as possible.
"This is one of the first studies to really address this issue directly," co-investigator Davangere P. Devanand, MD, professor of clinical psychiatry and neurology at Columbia University in New York City and director of the Division of Geriatric Psychiatry at the New York State Psychiatric Institute, told meeting delegates.
"So our way of thinking is that federal regulations urging early antipsychotic discontinuation may need reconsideration. Our findings certainly raise questions," said Dr. Devanand.
FDA Warning
According to the investigators, psychosis or agitation/aggression in patients with AD can lead to more rapid cognitive decline, future institutionalization, and increased healthcare costs.
However, there are currently no medications approved by the US Food and Drug Administration (FDA) to treat these symptoms in this patient population, said Dr. Devanand.
"Behavioral interventions may be useful, but well-controlled research evidence supporting these approaches is limited," he said.
Although previous research has shown that antipsychotics are the only medications that are superior to placebo in this patient population, the effect size has not been robust. In addition, the agents' efficacy "needs to be weighed against sedation, neurological side effects, and metabolic syndrome."
Further, antipsychotics currently carry a black box warning from the FDA about increased mortality risk, based on pooled clinical trials.
"In other words, federal regulations urge discontinuation after a few months," said Dr. Devanand.
He noted that although previous antipsychotic discontinuation trials have shown mixed results, most did not establish that a treatment response had been achieved before discontinuation. In addition, most of the trials included patients who did not have baseline symptoms of psychosis or agitation and had been receiving antipsychotics "for several years" and/or looked at the discontinuation of several different medications.
"Patients in the US get a bunch of psychotropics and antipsychotics frequently. So discontinuation becomes very complicated. And it can be difficult to interpret which drug is doing what," he said.
Significant Relapse Rate
The Antipsychotic Discontinuation in Alzheimer's Disease Trial had an initial open-label phase in which 180 patients older than 49 years who had probable AD and symptoms of psychosis or agitation/aggression received risperidone (0.25 - 3 mg/day).
Baseline symptoms of psychosis were determined with a score of 4 or higher on delusions or hallucinations on the Neuropsychiatric Inventory (NPI) rating scale, and symptoms of behavioral dyscontrol were determined by a score of 4 or higher on agitation/aggression on the NPI. Exclusion criteria included stroke, transient ischemic attack, or uncontrolled atrial fibrillation.
Participants who responded to treatment at the 16-week mark (n = 110; mean age, 79.9 years; ≥30% decline in NPI core scores) were than randomly assigned to continue receiving risperidone (mean dose, 0.97 mg/day) or were switched to a matching placebo.
All patients were then examined 16 weeks and 32 weeks after randomization. Relapse was defined as a 30% or greater increase in NPI core score and a significant change on psychosis/agitation scores.
Results showed that in the first 16 weeks after randomization, "the time to relapse and risk of relapse favored risperidone over placebo," reported Dr. Devanand.
The relative hazards ratio for relapse for the placebo group was 1.94 (95% confidence interval [CI], 1.09 - 3.45; P = .022).
In addition, 60% of the patients receiving placebo experienced relapse compared with 32.9% of those receiving risperidone (P = .004).
"In other words, the placebo group was almost twice as likely to relapse as the patients on risperidone," said Dr. Devanand.
At the 32-week post-randomization point, the relative hazards ratio for those receiving placebo vs those receiving risperidone was 4.88 (95% CI, 1.08 - 21.98; P = .023). The relapse rates were 48% vs 15% for placebo and risperidone, respectively (P = .017).
Throughout the entire post-randomization period, there were 11 serious adverse events (SAEs) reported, including 3 deaths (2 patients were receiving risperidone and 1 was receiving placebo). There were no significant between-group differences in SAEs for either 16-week examination period.
There were also no significant differences between the 2 groups in adjusted scores on the Treatment Emergent Symptom Scale, the Mini-Mental State Examination, the Alzheimer's Disease Assessment Scale-Cognitive subscale, and the Physical Self-Maintenance Scale, nor were there significance differences in extrapyramidal symptoms on the Simpson-Angus Scale or in body weight.
FDA Response
"Looking at it in context of current federal regulations, if someone actually has these symptoms, you treat with an antipsychotic for 4 months. If they're doing better in terms of the symptoms and side effects are not terrible, then you're supposed to discontinue use of the medication," said Dr. Devanand.
"But our findings suggest that discontinuing means you're more likely to relapse. Although we did not see any difference in side effects, that is still a somewhat open question. So you have to weigh the risk of relapse with risk of side effects after long-term treatment for your individual patient."
During the question-and-answer session after his presentation, Dr. Devanand was asked to speak more about the current FDA black box warning.
"I think we should think of this warning as a starting point. My understanding is that the FDA based this on specific data from certain studies. And perhaps more research is needed to address that," he answered.
That set the stage for a brief discussion by Ni Khin, MD, medical team leader in the Division of Psychiatry Products at the FDA, who told attendees that that she was speaking "on my views and not the official views of the FDA."
"The efficacy of antipsychotics for symptoms of psychosis and behavioral symptoms in Alzheimer's patients has not been established yet in FDA-registration trials," said Dr. Khin.
"In 2005, we put a box warning out based on 17 trials that showed an increase in mortality and cardiovascular effect."
She noted that in the current study, only the initial treatment responders were randomly assigned into the discontinuation phase of the trial. She also pointed out that although 60% of the patients receiving risperidone did not relapse, over 30% of those receiving placebo also did not relapse.
"The main question we often face is: if patients are responding to treatment, how long should we go on treating them? Also, when looking at results from this study, it's important to ask if there was any patient selection bias," said Dr. Khin.
Individual Approach
"Science looking to prove the effectiveness of this drug might be missing some important variables. But I can tell you this: I give Depakote [divalproex sodium]. I give Risperdal [risperidone]. And 70% to 80% of my patients do well, even though you scientifically can't seem to demonstrate how I'm doing that," said Gerald Plovsky, MD, adult and geriatric psychiatrist in private practice in Greensboro, North Carolina, to thunderous applause from the audience during the session's discussion period.
Dr. Devanand replied that it is much harder to find a significant response when comparing a medication against placebo than it is with open-label treatment.
Dr. Plovsky later told Medscape Medical News that "researchers care about what happens to a population. In clinical practice, we care about the individual."
"I think this study is great because it justifies why geriatric psychiatrists are so important. We take the time to get to know the patients, and we do things slowly and cautiously," said Dr. Plovsky, who is also the former director of the Public Mental Health Clinic for Geriatrics in Greensboro.
"The bottom line is, I think when you take people off these drugs, you should go slowly. If you go quickly, then people get sick. And it's pros and cons when it comes to the decisions we make. Over the years, I have used Risperidol, and I think it's saved many people a lot of disaster. And I've rarely seen an adverse side effect from it as long as you give 1 or 2 mg," he concluded.
This study was supported by grants from the NIH and by the Department of Veterans Affairs. The medications were provided by Janssen Pharmaceuticals, "which had no other role in the study," report the investigators. Dr. Devanand reported having received research support from Eli Lilly. Dr. Plovsky has disclosed no relevant financial relationships.
Comments
Post a Comment