Efficacy and Safety of Lamotrigine Treatment for Bipolar Disorder. A Review of Randomized Controlled Trials of Lamotrigine Monotherapy. Emil ter Veer (6032885)
Tutor: Dr. Huib van Dis Bachelorthesis Brain & Cognition, Department of Psychology, University of Amsterdam Date: 13-07-2012
1
Summary Abstract
3
Introduction
4
-
Bipolar Disorder
4
-
Neuromodulator and Glutamate model
4
-
Pharmacological treatment of Bipolar Disorder
5
-
Anticonvulsants as Mood stabilizers
6
-
Lamotrigine as a new Moodstabilizer
7
-
The Lamotrigine Program
7
-
Publication Bias
8
Review of Acute studies of Lamotrigine treatment
9
-
Manic and Mixed episodes
9
-
Depressive episodes
9
-
Bipolar Disorder with Rapid Cycling pattern
11
Review of Maintenance studies of Lamotrigine treatment
12
-
Bipolar Disorder with Rapid Cycling pattern
12
-
Manic and Depressive episodes
14
Safety and Tolerability of Lamotrigine across all clinical trials
16
Conclusions and Discussion
18
Meta-analysis of maintenance studies of Lamotrigine monotherapy References
22
Appendix
35
31
2
Abstract Lamotrigine is a moodstabilizer that is nowadays frequently used as pharmacological treatment for Bipolar Disorder. To examine the efficacy and tolerability profile of Lamotrigine, all randomized controlled trials in which Lamotrigine monotherapy was compared to placebo were reviewed in this article. Earlier articles that were supposed to do the same created a misleading view of Lamotrigine as a very efficious and safe drug, because usually only the published studies of Lamotrigine are included in reviews. But since there is a large publication bias in this field of research, all unpublished trials will be included also this review. Therefore, the clinical trial register of GlaxoSmithKline, the sponsor of the trials, Clinicaltrial.gov and online databases were extensively searched. It appeared that only 2 out of eight acute studies and four out of five maintenance studies were published as a full article. The review of the acute studies indicated that Lamotrigine established very limited efficacy in treating manic (two negative studies) and depressive episodes (four negative and one positive studies). One acute study in treating rapid-cycling showed a positive effect, but due to the design of the study and the small sample size, these results should be interpreted with caution. The review of the maintenance studies indicated that Lamotrigine showed far more efficacy in preventing mood episodes than acutely treat them. These prevention effect was particularly visible for the depressive episodes (three studies) and rapid-cycling (one out of two studies) and to a lesser extent the manic episodes (one study). The re-analysis of the safety results of each study indicated that there occurred not significantly more adverse events in the Lamotrigine group compared to the placebo group, which means that Lamotrigine is generally well tolerated in these studies. Meta-analysis should further investigate the prevention effect of Lamotrigine systematically.
3
Introduction Bipolar Disorder Bipolar Disorder is a well known psychiatric syndrome that falls in the spectrum of affective disorders (Angst, 2007). Bipolar Disorder is an illness characterized by fluctuating episodes of both depression and mania. According to the Diagnostic and Statistical Manual of Mental Disorders (DSM), patients who are experiencing an episode of mania, show an abnormally elevated mood pattern for at least one week. Usually, someone becomes more talkative, impulsive and highly distractible. On the other hand, patients who are experiencing a major depressive episode, show a very depressed mood pattern for at least two weeks (DSM-IVTR). Bipolar I disorder affects 1% of the entire population worldwide (Belmaker, 2004). Bipolar Disorder is associated with social impairment, increased psychological burden and increased morbidity and mortality (Goodwin, 2004). These difficulties are mainly due to the depressive episodes of Bipolar Disorder, which usually last longer and are hard to treat (Saunders & Goodwin, 2011; Bourin & Prica, 2007). There are mainly two types of bipolar disorder. In Bipolar I Disorder, episodes of depression and episodes of mania are experienced in a ratio of 3:1. Bipolar II Disorder is characterized by episodes of major depression and episodes of hypomania, in which patients experience mania to a lesser extent (Bourin & Prica, 2007). Around 40% of Bipolar I patients also experience mixed Bipolar episodes, which are characterized by a mixture of both depressive and manic symptoms that are present in nearly every day for at least one week (Konstantinos et al., 2012). A mixed episode can be distinguished from another pattern called rapid cycling, which means that the patient experiences at least four mood episodes a year and is present in a quarter of all Bipolar patients (Calabrese et al., 2000). Both the presence of mixed episodes or a pattern of rapid cycling is associated with poorer prognosis than when these symptoms are absent (Konstantinos et al., 2012).
Neuromodulator and glutamate model There are many models of neuromodulator systems that can partly explain symptoms of Bipolar Disorder. A neuromodulator that is found to be involved in Bipolar Disorder is glutamate, the main fast chemical messenger in the brain (Gould, Chen & Manji, 2002). Glutamate is known to interact with the three monoamines Dopamine, Serotonin and Noradrenalin. For depressive disorder, the so-called trimonoamine-hypothesis predicts that a low concentration of these three monoamines reaching their post-synaptic receptors would cause the depressive mood (Kuala & Sanatoria, 2005). A paradigm of Bipolar Disorder that could be directly inferred from the trimonoamine-hypothesis would sound like: high neuromodulator concentrations cause mania as low concentrations cause depression. And because glutamate interacts with the three monoamines, it could be no coincidence that that excessive glutamate activity accounts for manic symptoms, as evidence suggests (Weisler et al., 2006). This paradigm is rather too simple, but since detailed paradigms of the Bipolar mechanism are missing, this paradigm could be temporally used to infer which treatment to choose. To create a more detailed paradigm, there has to be taken into account which
4
neuromodulator circuit is involved in each symptom of depression or mania and in what way it is dysfunctional. For many symptoms of both mania and depression, multiple circuits between multiple brain areas are involved, which makes it more difficult to infer which neuromodulator to target with psychoactive drugs.
Pharmacological treatment of Bipolar Disorder In the early days, all kinds of anti-depressants were used to treat Bipolar Disorder, for example Mono Amine Oxidase inhibitors (MAO-inhibitors), Tricyclic antidepressants (TCA’s) and Selective Serotonin Reuptake Inhibitors (SSRI’s) (Millan, 2004). But later there has been a controversy about using anti-depressants for treating the depressive episodes of Bipolar Disorder. Some research showed that the use of anti-depressants could increase the risk of switching into mania or rapid cycling (Leverich et al., 2006; Yatham, 2004; Ghaemi, 2003). Moreover, there is evidence that anti-depressants do not decrease the risk for suicide in patients with Bipolar Disorder (Gibbons et al., 2006), where it does for patients with Major Depressive disorder (Guzzetta et al., 2007). These critics have eventually led to a decrease of anti-depressant prescription for patients suffering from Bipolar Disorder. Nowadays antidepressants are not the first choice for Bipolar Disorder and yet are prescribed with caution (Millan, 2004). A class of drugs that is particularly developed for Bipolar Disorder is the class of mood stabilizers. For several reasons, the term ‘mood stabilizers’ is somewhat ambiguous (Rapoport, Basselin, Kim & Rao, 2009). First, the term ‘mood stabilizer’ indicates that there are drugs that can treat both mania and depression equally well (and literally stabilize mood) . But for so far, no drug has achieved this (Gould, chen & Manji, 2002). In reality, some drugs have more efficacy in ‘treating from above’ and ‘stabilize from above’ which means respectively reduce manic symptoms and prevent relapse into a manic episode. On the other hand, some drugs have more efficacy in ‘treating from below’ and ‘stabilize from below, which mean respectively reduce depressive symptoms and prevent relapse into a depressive episode (Stahl, 2008). Second, the class of mood stabilizers is made up of drugs from other classes, like anti-psychotics, anti-convulsants and anti-depressants. The first mood stabilizer that was approved for Bipolar Disorder by the Food and Drug Administration (FDA) is Lithium (Gould, Chen & Manji, 2002). Lithium was once used to treat Unipolar Depression and thus originally belongs to the class of anti-depressants. This drug is called the ‘classic’ mood stabilizer and is nowadays still frequently used for treating Bipolar Disorder. Lithium has shown efficacy in treating both the manic episodes and to a lesser extent the depressive episodes of Bipolar Disorder (Baldessarini, Tondo, Hennen & Viguera, 2002). The mechanism by which Lithium acts remains a kind of a mystery. Possible mechanisms are inhibiting second messengers or boosting the three monoamines (Kuala & Sanatoria, 2005). The downside of Lithium is that it has to be very precisely titrated to prevent intolerability (Licht et al., 2001). After the approval of Lithium as a mood stabilizer, many candidate drugs that were already approved for treatment for other classes of diseases were tested for efficacy of treating Bipolar Disorder. These drugs predominantly came from the classes of anti-psychotics and
5
anti-convulsants (Weisler et al., 2006). Many anti-psychotics have made their way to FDAapproval for Bipolar Disorder (Yatham et al., 2005). Especially atypical anti-psychotics show some efficacy in treating the manic episodes, with or without psychosis, and to a lesser extent the depressive episodes (Cipriani et al., 2011). The antagonism of the D2 Dopamine receptor and the Serotonin 2A (5HT2A) receptor which reduces glutamate activity by the drug may account for the reduction of manic symptoms and depression, depending on the neuromodulator circuits that are involved (McMahon et al., 2006). Some of these drugs that are frequently used are Olanzapine and Quetiapine. However, when these agents are used for long term treatment, the patient could experience tardive diskinesia as side-effect (Yathman et al., 2005). Therefore, one should be cautious about the use of these agents for too long.
Anticonvulsants as moodstabilizers As stated earlier in this review, excessive glutamate activity could be involved in manic symptoms of Bipolar Disorder. In Epilepsy, seizures are known to be the consequent of excessive glutamate activity throughout the brain (Brodie, 2010). A pharmacological mechanism by which this excessive activity could be decreased is the blockage of VoltageSensitive-Sodium-Channels (VSSC). For a long time so-called anti-convulsants, which primary effect is blocking the VSSC’s, have been used to treat the seizures (Semah et al., 1998). Psychiatrists began to recognize the possible similar mechanism in Epilepsy and Mania and speculated whether these anticonvulsants could be helpful for the treatment of Bipolar Disorder too (Weisler et al., 2009). Valproic Acid (patent name: Divalproex) and Carbamazepine (patent name: Carbymal), which main mechanisms is blocking the VSSC’s, were the first anticonvulsants that were tested in randomized controlled trials (RCT) for treating Bipolar Disorder (Weisler et al., 2006). Both drugs have shown efficacy in treating and preventing the manic episodes of Bipolar Disorder (Marson et al., 2002). Although the two drugs described above may share a common mechanism, there are some differences. First, the two have a different tolerability profile (Bourin & Prica, 2007). Where some major side-effects of Valproic Acid are hair loss and weight gain, some major side-effects of Carbamazepine include sedation and bone marrow toxicity. Second, except their efficacy in treating mania, they have a different efficacy profile in a broader sense. For example, were Valproic Acid has also shown efficacy in treating migraine, Carbamazepine has shown efficacy in neuropathic pain. (Nasrallah, Ketter & Kalali, 2006). Since Carbamazepine and Valproic Acid did only show efficacy in treating the manic symptoms. Therefore, there was a great need for a more tolerable moodstabilizer that could treat the depressive symptoms too (Weisler et al., 2008). In 1994, Lamotrigine (patent name: Lamictal), another anti-convulsant, was approved by the FDA for treating epilepsy in the United States. The proposed mechanism of Lamotrigine is also blocking VSSC’s (Weisler et al., 2009). After the first clinical trials of Lamotrigine in treating Epileptic seizures, it became clear that there was an indication of potential mood improving properties of Lamotrigine (Smith et al., 1993). And because of previously shown mood stabilizing effects of VSSC
6
blockers Valproate and Carbamazepine, experts speculated whether Lamotrigine could be efficious as a moodstabilizer.
Lamotrigine as a new moodstabilizer From the late 1980’s until the late 1990’s, experts started to prescribe Lamotrigine to some individuals who were suffering from Bipolar Disorder for a long time and who did not respond to any other pharmacological treatments (Weisler et al., 2009). The case reports of these individuals were surprising. A few case reports describe individuals who were suffering from Bipolar I Disorder with rapid cycling (Bowden, 1999; Calabrese et al., 1996; Fatemi et al., 1997; Fogelson and Sternbach, 1997; Kotler and Matar, 1998) and were treated with Lamotrigine monotherapy. These individuals received a dose between 100 and 300mg/day and all showed markable improvement on several screening instruments like the Hamilton Anxiety and Depression Scale (HAM-D) and Bipolar Disorder Inventory (BDI) compared to baseline and some showed complete remission of symptoms. There are also case reports about individuals who received Lamotrigine therapy added on to their normal medication therapy. Weisler described two patients at the 1994 APA congress, one suffering from Bipolar I and one suffering from Bipolar II, who had a long history several drug therapies and did not respond to any of them or did not tolerate them well. When Lamotrigine was added on to their current medication, their symptoms of Bipolar Disorder, especially the depressive episodes and rapid cycling were relieved (Weisler et al, 2009). The same pattern of response has been described in a case report of an individual with severe depressive symptoms and rapid cycling when Lamotrigine therapy was added on to Valproic Acid (Walden et al., 1996). Furthermore, a case series of 7 Bipolar Depression patients with rapid cycling or mixed episodes (Kusumakar and Yatham, 1997) and a case series of 9 Bipolar I and II patients (with current depressive episode) (Sporn and Sachs, 1997) showed that 10 of 16 patients responded to Lamotrigine added on to their current medication and showed remission of symptoms. After these case reports/series, two non-comparative open studies of Lamotrigine as add-on therapy were conducted. In one 6-week study, 22 Bipolar patients with current depression received Lamotrigine (50-100mg/day) as add on to Valproate therapy (1000-2250mg/day) (Kusumakar and Yatham, 1997b). As measured by the HAM-D-21, 63% of the patients were in remission of their symptoms after week 6. None of the patients developed serious rash, a well known side-effect of Lamotrigine. In the other open study, 41 patients with current depression received Lamotrigine monotherapy (25-600mg/day), or Lamotrigine in combination with Valpoate (25-200mg/day) or Carbamazepine (50-400 mg/day) (Calabrese et al., 1999a). The patients showed a significant improvement on the HAM-D-31 compared to baseline, 53% of the patients showed a remarkable response ( 18) for at least one week but no longer than 9 months (SCAA2008) and 3 months (SCAB2009), were included in the 3 week (SCAA2008) and 6 week (SCAB2009) study. In the 3 week SCAA2008 study, the patients (n=216) were randomized to either Lamotrigine 50 mg/day (n=84), Lithium 0.8-1.3 Eq/ l (n=36) or Placebo (n=95). In the 6 week SCAB2009 study, the patients (n=228) received either Lamotrigine 200 mg/day (n= 74), Lithium (n=77) or Placebo. In SCAA2008, the number of subjects that completed the study was 62% (Lamotrigine), 44% (Lithium)and 64% placebo. In SCAB2009 these numbers where respectively 50%, 73% and 60%. In both studies, Lamotrigine showed no significant reduction of scores on the primary outcome scale, the MRS-11, and all the secondary outcome scales compared to Placebo (MRS-16, Manic Syndrome Subscale of MRS, Ham-D-31, Brief Psychiatric Rating Scale, CGI-I and CGI-S) Lithium did only show a significant reduction in MRS-11 score in SCAB2009 but not in SCAA2008. These two studies did not show efficacy of Lamotrigine in acute treatment of manic episodes. Only lithium showed some efficacy in SCAB2009. But since Lithium did not show reduction of symptoms in SCAA2008, which was twice as short than SCAB2009, this could indicate that SCAA2008 was too short for drugs to reach a therapeutic level. Since the titration time of Lamotrigine should be long, because of possible development of serious skin rash, the duration of both studies could be too short for Lamotrigine to reach a therapeutic level. In SCAA2008, the Lamotrigine dose was 50 mg/day, which is relatively low compared to SCAB2009 and other studies, and is sometimes considered to be sub-therapeutic (Calabrese et al., 2003). So even if the duration of SCAA2008 was much longer, a potential treatment effect could have been remained undiscovered due to the low dose.
9
Depressive Episodes
Five acute studies have been conducted on Lamotrigine treatment in Bipolar patients who currently experience a depressive episode. These studies were part of the Lamotrigine Program too. Study SCAB2001 was the first randomized placebo-controlled trial conducted and is the only one that has been published from these five acute studies (Calabrese, 1999). These five studies have been reviewed multiple times (Calabrese et al., 2008 for example). In this review, the five studies will be reviewed once again but from a different perspective. In Studies SCAB2001, SCA40910 and SCA30924, only patients with Bipolar I Disorder were enrolled. Study SCAA2010 enrolled both patients with Bipolar I and II, whereas study SCA100223 enrolled only patients with bipolar II Disorder. At the beginning of all studies, the patients must have had a minimum score of 18 on the HAM-D-17 and have been experiencing a depressive episode with a minimum duration of 2 weeks (SCAB2001, SCAA2010, SCA40810) and 8 weeks (SCA100223, SCA30924). Patients were randomized to either Placebo or Lamotrigine for 8 weeks at a fixed dose of 200 mg/day (SCA40910, SCA100223 and SCA30924), for 7 weeks at a fixed dose of 200 mg/day or 50 mg/day (SCAB2001) and for 10 weeks at a flexible dose of 100-400 mg/day (SCAA2010). The primary outcome was a mean change score between baseline and end of study assessment on the HAM-D-17 (SCAB2001 and SCAA2010) or MADRS (SCA40910, SCA100223 and SCA30924). Several secondary outcomes which were identical for al studies were HAM-D-1, HAM-D-31, CGI-I and CGI-S. (For further information on number of subjects randomized, look at table). Study SCAB2001 did not reach a significant difference between Lamotrigine both 50 mg/day (n=64) and 200 mg/day (n=63) and placebo treatment (n=65) on the primary outcome measure (HAM-D-17). However, Lamotrigine 200 mg/day did show significant difference on several secondary outcomes (HAM-D-1, MADRS,CGI-I and CGI-S) . The percentage of patients who completed the study was equal (71%) across the lamotrigine and the placebo treatment group. The percentage of patients that were responding to their treatment should show 50% symptom reduction on the MADRS and CGI-I. The percentages of responders were significantly different between Lamotrigine group (MADRS: 54% and CGI-I:51%) and placebo group (MADRS 29% and CGI-I 26%). In the studies SCAA2010, SCA40910, SCA100223 and SCA30924 the numbers of patients who were randomized to Lamotrigine were respectively 102, 126, 109 and 123. The numbers of patients who received placebo were respectively 100, 117, 105 and 120. The Lamotrigine group did not reach any significant difference on any outcome assessments compared to the placebo group. The percentages of patients who completed the study were 66%, 61%, 73% and 60% respectively in the Lamotrigine group and 67%, 73%, 67% and 57% respectively in the placebo group. The percentages of responders in the Lamotrigine group were not significantly different compared to the percentages responders in the placebo group, as measured by the MADRS. The responders analysis with the CGI-I indicated that only in study
10
SCA100223 the percentage responders was significantly higher in the Lamotrigine group (61%) compared to the percentages responders in the placebo group (45%). As with acute studies, the slow titration time of Lamotrigine or insufficient subjects may explain the failure to detect a small treatment effect of Lamotrigine compared to placebo in all studies except SCAB2001. The relatively high placebo response rates found in these studies may account for this failure too, as in SCAB2001 (which did find significant differences), the placebo response was relatively low. A high placebo response rate has been found a common problem to detect a treatment effect in randomized clinical trials (Khan et al., 2003). A recent meta-analysis of these five studies indicated that there is a significant treatment effect of Lamotrigine over placebo only if there has been controlled for severity of baseline score (Geddes, Calabrese & Goodwin, 2009). The rationale for taking baseline severity as a moderator is that the difference between response rates of Lamotrigine and placebo tends to be higher when the patients show a more severe mood pattern at baseline assessment (HAMD > 24) then when they show a relatively moderate one (HAM-D
