Side-Effect Risk Often Tempers Antipsychotic Use for Dementia

A significant number of elderly patients with Alzheimer's disease who are prescribed second-generation antipsychotic medications to treat psychosis, aggression, or agitation associated with the progressive disorder appear to improve while taking the medications. However, significant adverse effects often cause physicians, patients, and their caregivers to decide to stop using the drugs, regardless of possible benefits.

These two take-home points, which came from the National Institute of Mental Health's CATIE-AD (Clinical Antipsychotic Trials of Intervention Effectiveness-Alzheimer's Disease) study, were largely reduced by popular press reports to the single conclusion that “the drugs don't work.” The actual conclusions arising out of the complex, multiphase study protocol, however, are more complicated.

CATIE-AD is a 42-site, randomized, double-blind, placebo-controlled trial on which NIMH has spent $16.9 million. The study is the largest to date to compare the real-world effectiveness of three of the most widely used antipsychotic medications with placebo by taking into consideration the reason for drug therapy failure.

The first round of data from the study appeared in the October 12 New England Journal of Medicine

“Antipsychotic medications have been used extensively for Alzheimer's patients without enough solid evidence of whether or not they are effective,” said NIMH Director Thomas Insel, M.D., in a press release.“ The study has vital public health implications because it provides physicians and patients with information to more accurately weigh the medications' benefits against their drawbacks, with the needs and unique reactions of individual patients.”

Some Findings Unexpected

The team of CATIE-AD researchers followed 421 elderly patients (average age 78) with Alzheimer's disease who were experiencing delusions, hallucinations, aggression, or agitation that disrupted their daily functioning. They were ambulatory and still living at home or in an assisted-living facility. For each patient, a family member or other caregiver who had regular contact with the patient also took part in the study.

During phase 1 of the study, each patient was randomly assigned to take olanzapine (average dose 5.5 mg a day), quetiapine (average dose 56.5 mg a day), risperidone (average dose 1 mg a day), or placebo for up to 36 weeks. Study physicians were free to adjust dosing throughout the protocol.

The primary outcomes in CATIE-AD were the time to discontinuation of the study drug for any reason and the number of patients with at least minimal improvement on the Clinical Global Impression of Change (CGIC) scale at week 12 of the study. Secondary outcomes included time to discontinuation due to lack of efficacy and time to discontinuation due to adverse effects, intolerability, or death.

“We thought that overall the drugs would show their effectiveness,” Lon Schneider, M.D., the principal investigator for CATIE-AD, told Psychiatric News. “And the answer turned out to be, yes, the drugs are somewhat effective. But overall, that effectiveness is offset by adverse events that resulted in discontinuation of the medications. It was somewhat surprising really, given that the expert opinion upon which this study was built considered these drugs to be particularly useful in treating these very difficult symptoms.”

Schneider is a professor of psychiatry, neurology, and gerontology at the Keck School of Medicine of the University of Southern California.

Different Outcomes = Different Results

After a minimum period of two weeks, a medication could be discontinued if the patient was not benefiting from it or was experiencing intolerable side effects. This “all cause discontinuation” or “time in treatment” outcome was meant to measure an aggregate of patients', caregivers', and clinicians' judgments of efficacy, safety, and tolerability into an overall measure of effectiveness that reflected therapeutic benefits in relation to undesirable effects.

In phase 1, Schneider and his colleagues reported, there were no differences among the four treatment groups regarding the length of time that patients stayed on their medication (see chart). The average length of time across the four groups was eight weeks.

When the researchers examined the reasons for discontinuation, they found some statistically significant differences between the medications and placebo. Those taking olanzapine and risperidone were less likely to cite lack of benefit as a reason to discontinue use than those taking placebo or quetiapine. However, those taking any of the three antipsychotic medications were more likely to discontinue use because of intolerable side effects than those taking placebo.

Schneider noted that the dosage levels of olanzapine and risperidone used in this study were similar to those used in previous studies, but the quetiapine dose was lower than that used in previous studies. The lower dose could have biased the results, if it was too low to have a therapeutic effect.

Nonetheless, Schneider pointed out, “the physicians were free to raise it, and the dose was still high enough to cause side effects.”

Balancing Benefits and Risks

Memory loss and disorientation are the most common symptoms of Alzheimer's disease, but many people with the disease experience symptoms such as delusions, hallucinations, aggressive behavior, or agitation. These symptoms are associated with a rapid worsening of the illness and often result in the patient's being placed in a nursing home or other specialized care institution.

“Clearly some of these patients were benefiting from these three antipsychotics,” Schneider told Psychiatric News. “But there is also a large group of patients who found the medications to be intolerable.”

All three antipsychotic medications were more likely to cause sleepiness and weight gain than placebo (see table). In addition, there were some differences among the three medications with respect to extrapyramidal signs (less likely with quetiapine), confusion (more likely with olanzapine), and psychotic symptoms (more likely with olanzapine). As expected, prolactin levels were markedly elevated in the patients taking risperidone, but not the patients taking the other two drugs. Patients taking any of the three antipsychotics tended to gain about one-half to one pound a month compared with those taking placebo, who tended to lose one pound a month. Overall, there were no significant differences between the three drugs and placebo regarding patients' glucose, total cholesterol, or triglyceride levels.

Taken together, Schneider said, the results from this first phase of the study indicate that the overall benefit of these medications is offset by intolerability to the associated side effects.

“In deciding whether to prescribe atypical antipsychotic medications for their Alzheimer's patients, clinicians need to consider the risks, benefits, and individual needs of a given patient,” Schneider concluded in the study report. “Although some patients may benefit greatly from these medications, the evidence from this study suggests these medications hold limited value for the majority of patients.”

The CATIE-AD protocol also included a second phase in which patients who discontinued their phase 1 medication were randomly assigned to either one of the other atypical antipsychotics or to the SSRI antidepressant citalopram. Researchers hope to report the phase 2 results sometime next year.

An abstract of “Effectiveness of Atypical Antipsychotic Drugs in Patients With Alzheimer's Disease” is posted at<http://content.nejm.org/cgi/content/short/355/15/1525>. The accompanying editorial, “Alzheimer's Disease—Clinical Trials and the Logic of Clinical Purpose,” is posted at<http://content.nejm.org/cgi/content/full/355/15/1604>.▪