Novel Antipsychotics for Treatment-Resistant Depression
Novel Antipsychotics for Treatment-Resistant Depression
Currently available antidepressant medications are effective for a large segment of the population with depression. However, recent data suggest that a sizable portion of patients with unipolar depression do not experience full therapeutic recovery. For example, Simon et al. (1999) found that only about half of depressed patients treated with antidepressants in a primary care setting achieve recovery over a two-year period. Since lack of full response is common, management strategies need to be established and implemented. Current approaches include: the use of combination antidepressant treatments (e.g., selective serotonin reuptake inhibitor and bupropion [Wellbutrin]), augmentation (e.g., the addition of lithium [Eskalith, Lithobid] or thyroid hormone to an antidepressant), the addition of psychotherapy or electroconvulsive therapy (Shelton, 2003, 1999). However, even with these approaches, a significant minority of patients do not experience a full therapeutic effect.
Recently, novel antipsychotics have shown some promise for the management of depressive disorders. From a mechanistic standpoint, the pharmacological properties of at least some of these drugs predict antidepressant properties. Novel antipsychotics act, to varying degrees, on a variety of dopamine, serotonin (5HT), glutamate and other receptors. The antagonism of 5HT2A receptors is common among these drugs, and blockade of this subtype is seen with other antidepressant agents such as mirtazapine (Remeron) and nefazodone (Serzone). Blocking of 5HT2C receptors has also been shown to enhance release of frontal dopamine and norepinephrine, which is thought to be a key antidepressant property (Shelton, 2003; Zhang et al., 2000).
Further, several lines of clinical evidence support the usefulness of novel agents for the management of depression. These include the use of novel antipsychotics for the mood symptoms of schizophrenia (Tollefson and Sanger, 1999) and both mixed and depressive states in bipolar disorder (Tohen et al., 2003). In fact, evidence supporting the effectiveness of novel antipsychotics in depressive disorders is not new. For example, clozapine (Clozaril) has been shown effective for the management of mood symptoms in schizophrenia, dysphoric mania and psychotic depression (Shelton, 2003).
A key question, however, is whether certain of these drugs are producing direct antidepressant effects or are augmenting actions when combined with antidepressants, or alternatively, whether the actions are related to the antipsychotic effects alone. Tohen et al. (2003) reported that olanzapine (Zyprexa) given alone or in combination with fluoxetine (Prozac) produced a greater antidepressant response than placebo in patients with bipolar I disorder depression. Both olanzapine and combined olanzapine and fluoxetine (Symbyax) produced beneficial effects on core symptoms of depression, including sadness, pessimistic thoughts, suicidal ideation, fatigue and poor concentration (Tohen et al., 2003). This suggests that the actions are not mediated via the impact of the antipsychotic drug on psychotic symptoms alone.
Based on this line of reasoning, olanzapine was tested in an augmentation trial with fluoxetine in 28 patients with treatment-resistant, nonpsychotic unipolar depression (Shelton et al., 2001). Treatment resistance was demonstrated retrospectively by a failure of a previous adequate trial of two different antidepressants and a prospective run-in of six weeks of fluoxetine up to 60 mg/day. Patients then were randomly assigned to eight weeks of double-blind treatment with continuation fluoxetine (plus placebo), olanzapine alone (plus placebo), or the combination of olanzapine and fluoxetine. The average dose of fluoxetine was 52 mg/day in both groups during the double-blind phase. The average olanzapine dose was 12.5 mg/day in the monotherapy condition and 13.5 mg/day in the combined group. As expected, continuation fluoxetine produced no additional benefit. Olanzapine given alone resulted in a transient effect at three weeks, but otherwise did not produce significant differences during this phase. The effect of the combination, however, was robust by the end of the first week. By the end of the blinded phase, the combination of olanzapine and fluoxetine was superior to either agent given alone.
Two large-scale follow-up studies attempted to establish the effectiveness of the olanzapine and fluoxetine combination in treatment-resistant patients. In both circumstances, the combined treatment was not superior to comparator agents (Dube et al., 2002a, 2002b). However, the basic designs were significantly different. In one study (Dube et al., 2002b), nortriptyline (Aventyl, Pamelor) was used as the run-in drug. Patients were randomly assigned to continuation nortriptyline, olanzapine alone, fluoxetine alone, or the combination of olanzapine and fluoxetine. Although the combination group showed a robust effect in the first week, there were no significant differences between nortriptyline and combined olanzapine and fluoxetine by the end of the double-blind phase. However, nortriptyline also produced a substantial response during this phase. Similar results occurred in a second study that used a venlafaxine (Effexor) run-in, with continuation venlafaxine during the double-blind phase (Dube et al., 2002a). The significant effect of the comparator drugs in the double-blind phases of these studies indicates that the patients who were randomized into the blinded treatment conditions were not treatment resistant and suggests that these were failed trials.
Small-scale case series also support the effects of novel drugs in treatment-resistant patients. For example, Ostroff and Nelson (1999) reported that risperidone (Risperdal) added to an SSRI produced a significant antidepressant response in treatment-resistant patients. Arnell et al. (2001) treated patients who had experienced two prior antidepressant treatment failures to the olanzapine-fluoxetine combination or venlafaxine for four weeks. There was a significantly greater effect on Montgomery-Asberg Depression Rating Scale (MADRS) scores in the combination group (12.9 points) versus the venlafaxine-alone treatment (9.1 points).