Investigational Agents and Methodologies at NCDEU

(This is the second installment in a series on the 41st Annual NCDEU meeting. Part I was published in the October issue of Psychiatric Times, p40 -- Ed.)

Promising investigational agents and innovative methodologies for clinical investigation were considered at the 41st annual New Clinical Drug Evaluation Unit (NCDEU) meeting sponsored by the National Institute of Mental Health in Phoenix, May 28-31.

Agents in Clinical Trials

Tomoxetine is a novel compound for attention-deficit/hyperactivity disorder (ADHD) that enhances noradrenergic function by selectively blocking the pre-synaptic norepinephrine transporter. This non-stimulant, non-dopaminergic agent was found to be well-tolerated through meta-analysis of safety data from over 1,000 children receiving doses as high as 3.5 mg/kg/day. Anorexia was the most frequently reported adverse event, occurring in 23% of patients and associated with a mean weight loss of approximately 0.5 kg over nine weeks of treatment.

Summarizing three efficacy studies in children and two in adults, Eli Lilly and Company researchers reported that tomoxetine was superior to placebo in reducing ADHD symptoms assessed with both hyperactive and inattentive subscales of the ADHD rating scale. University of Nebraska researchers reported long-term open trial data with pediatric patients indicating that 19 out of 35 patients who participated for one year maintained their initial decrease in symptom severity ratings over the course of treatment.

The abuse liability of tomoxetine was considered by Eli Lilly researchers and colleagues at University of Vermont in a study of subjective medication responses of 16 healthy volunteers who had previously used recreational drugs. The subjects were provided three different doses of tomoxetine, two of methylphenidate (Ritalin) and placebo in randomized, double-blind cross-over design on separate days. Their response to each drug and dose was assessed using a seven-item visual analog scale (VAS), the Addiction Research Center Inventory (ARCI) short-form and Adjective Rating Scale (ARS) questionnaires.

The researchers reported that mean scores on the stimulant subscales of the VAS and ARCI, as well as the amphetamine and euphoria subscales of the ARCI, were all greater for methylphenidate 40 mg daily than for placebo, while scores for tomoxetine did not differ from placebo on these measures. On the other hand, tomoxetine 90 mg daily was associated with significantly greater increases on the sick and bad subscales of the VAS and the sedative subscale of the ARCI, compared with placebo. Tomoxetine 90 mg daily and methylphenidate 40 mg daily both significantly differed from placebo but not from each other on the ARCI dysphoria group.

"The profile associated with tomoxetine differed from that typically observed with stimulants," the researchers concluded, "and significant scores on non-pleasurable subscales suggest tomoxetine is unlikely to have significant abuse liability."

Another noradrenergic agent, reboxetine (Vestra), which is a selective norepinephrine reuptake inhibitor now under review by the U.S. Food and Drug Administration for treating depression, was evaluated for depression, anxiety and panic disorder in three separate studies.

Researchers at University of California, Los Angeles, conducted a nine-week, single-blind study with 25 patients with major depression. Thirteen patients who completed four weeks of treatment with reboxetine 4 mg twice daily without adequate improvement were provided 5 mg twice daily for the remainder of the trial. Using response criterion of final Hamilton Rating Scale for Depression (HAM-D) score

10, 11 (65%) of 17 patients completing the nine-week trial were deemed responsive. Applying several assessment instruments, the researchers reported that treatment with reboxetine was effective in improving symptoms of depression as well as social functioning and psychomotor activity.

In the evaluation of effectiveness for anxiety associated with major depression, 381 patients were randomized to receive eight weeks of treatment with either reboxetine 8 mg/day, fluoxetine (Prozac) 20 mg/day to 40 mg/day or placebo. The two antidepressants were found comparable and significantly superior to placebo in the number of patients improving on the HAM-D anxiety items 10 and 11. In a separate comparison of reboxetine 6 mg/day to 8 mg/day to placebo for panic disorder, reboxetine was significantly more effective in reducing the number of major panic attacks per week and in reducing phobic symptomatology.

Two investigational antipsychotics appeared well-tolerated in separate reports. Iloperidone exhibits a characteristic atypical antipsychotic pharmacologic profile in blocking serotonin as well as dopamine receptors. Aripiprazole is an agonist at presynaptic dopamine receptors, in addition to postsynaptic blockade of serotonin and dopamine.

The safety of iloperidone was evaluated for psychotic and behavioral symptoms in elderly patients with dementia in flexible daily doses of 0.5 mg to 6 mg in a placebo-controlled study sponsored by Novartis Pharmaceuticals. With recent attention to the cardiac effects of atypical antipsychotics, the researchers pointed out that none of the 10 patients receiving active medication developed clinically significant electrocardiogram abnormalities. They also noted that no patient experienced a QT_C interval prolongation of greater than 480 msec, the approximate threshold for risk of torsades de pointes ventricular arrhythmia.

The efficacy and safety of aripiprazole was compared to that of haloperidol (Haldol) in 404 patients with schizophrenia or schizoaffective disorder in a study sponsored by Otsuka America Pharmaceuticals. Efficacy measures distinguished both drugs from placebo, but not from each other in this four-week study of two aripiprazole daily doses (15 mg and 30 mg) and haloperidol 10 mg. Safety monitoring revealed "no clinically meaningful increases" in QT_C prolongation, and the extrapyramidal symptoms (EPS) in patients receiving aripiprazole were comparable to those in patients receiving placebo.

Duloxetine, an investigational antidepressant progressing in the Eli Lilly development pipeline as the fluoxetine patent expires, is a dual reuptake inhibitor of serotonin and norepinephrine. In an eight-week, double-blind study, 173 patients with depression were randomized to receive either duloxetine, fluoxetine or placebo. Duloxetine was titrated from 20 mg twice daily to 60 mg twice daily, and fluoxetine was maintained at 20 mg daily. The researchers reported comparable efficacy from the active agents, and both were superior to placebo.

Escitalopram is the single, active s-isomer of citalopram (Celexa), which is a racemic mixture of the active s- and the non-antidepressant r-isomers. In March, Forest Laboratories, which now markets citalopram, filed a New Drug Application for escitalopram.

A fixed-dose study of escitalopram 10 mg/day, escitalopram 20 mg/day, citalopram 40 mg/day or placebo randomly assigned to 491 outpatients for eight weeks showed the superiority of both 10 mg and 20 mg doses to placebo on depression efficacy measures. Researchers at University of Nebraska found significant improvement in patients receiving escitalopram compared to those on placebo on the Montgomery-Asberg Depression Rating Scale (MADRS) beginning at week 2 and on the Clinical Global Impression (CGI) beginning at week 1.

Gepirone is a serotonin 5-HT_1A receptor partial agonist being investigated for treatment of depression. A dose of 20 mg to 90 mg was administered in open-label design to 134 patients for six weeks. The 70 patients whose symptoms improved then continued into a placebo-controlled phase for an additional six weeks, allowing comparison of relapse rates between those continuing on active agent and those switched to placebo. The reported 26% relapse rate among patients receiving gepirone was significantly lower than the 55% relapse rate with placebo. The active agent was well-tolerated according to the report, although four patients receiving the medication dropped from the double-blind phase due to side effects.

Method and Methodologies

A lack of standardization in administering and scoring the HAM-D has been a long-standing concern and was addressed at the 1999 NCDEU meeting with the formation of the Depression Rating Scale Standardization Team (DRSST), comprising researchers from academia and industry. The DRSST recommendations presented at the 2001 meeting included developing a consensus scoring system in which intensity and frequency of a symptom are rated separately and yield an overall severity score for each item. In addition, the DRSST clarified item anchors and provided clinical examples for each severity level.

Telecommunication technologies were embraced to facilitate clinical studies and diagnostic screening at several U.S. sites. Researchers from Hillside Hospital in Glen Oaks, N.Y., reported successfully administering neurocognitive tests by telephone to 42 patients with schizophrenia. The researchers noted that, despite recent elucidation of neurocognitive correlates with functional disability in schizophrenia, application of standardized neurocognitive tests has been limited by the requirement of in-person administration.

"This can lead to studies with sampling bias caused by patients withdrawing from studies because they are unwilling or unable to travel to the testing site," they observed. "Also, researchers may choose to use samples of convenience or less frequent follow-up assessments, which restricts [generalizing] the study results."

To ascertain whether neurocognitive testing can reliably be administered over the telephone, the researchers selected a small battery of tests and adapted them to telephone administration. These tests involved both oral administration and oral response and required no presentation of materials to the patient. The battery included tests of short-term memory, working memory, ideational fluency and verbal learning. These were administered twice, once by telephone and once in person, approximately one month apart.

The researchers found no significant differences between in-person and telephone administration of the tests in scores for working memory, semantic fluency and short-delay free recall. There were, however, different results in phonemic fluency and immediate attention and recall between these two administration methods.

University of South Florida researchers found the Telephone Interview for Cognitive Status-Modified (TICS-M) of limited use as the sole instrument in screening for Alzheimer's disease or mild cognitive impairment. Although the researchers promoted a no-cost, telephone-based memory screening based on the TICS-M to the community, they improved screening efficiency by adding other instruments. They remain advocates, however, of screening procedures that can overcome the cost, transportation and confidentiality concerns that can be barriers to early medical evaluation of memory.

An interactive voice response (IVR) system from Healthcare Technology Systems was used for screening patients with dementia and, in another study, patients with generalized anxiety disorder. In the first study, 155 subjects were assessed by clinical interview and scores from the Clinical Dementia Rating (CDR) scale to serve as the criterion standard. A battery of neuropsychological tests were also administered by IVR with a standard touch-tone telephone.

The researchers reported that six of eight domains in the IVR testing correlated with CDR status and concluded that computer-automated telephone screening for early dementia is feasible. "Such systems could provide wide-scale, cost-effective screening, education, and referral services to patients and caregivers," they suggested.

The second IVR study, assessing its utility in screening for anxiety, follows others that have shown discrepancies between the IVR patient self-report of anxiety symptoms and clinician ratings. Open-label data were obtained from an ongoing anxiety disorder clinical trial, and patients received the IVR version of the Hamilton Rating Scale for Anxiety (HAM-A) in addition to traditional measurements at screening, baseline and after an eight-week intervention phase. The IVR HAM-A scores were found to be significantly lower than clinician scores at screening and baseline but not after intervention.

Several reports examined the issue of patients with psychotic disorders providing informed consent to participate in clinical trials. Researchers at University of California, San Diego, compared an enhanced informed-consent presentation with slides to a standard written form. They found that, even in older patients with chronic psychotic disorders, "Understanding of protocols may be enhanced when educational interventions are utilized."

They further advocated that, rather than assume patients have limited capacity to make decisions about their participation, "procedures for improving patients' comprehension of informed consent should be actively sought."