University of Groningen. Lamotrigine in bipolar depression Loos, Marcus Lambertus Maria van der

Transcription

1 University of Groningen Lamotrigine in bipolar depression Loos, Marcus Lambertus Maria van der IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2011 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Loos, M. L. M. V. D. (2011). Lamotrigine in bipolar depression: a randomised placebo-controlled trial on the acute and long-term outcome of lamotrigine as add-on to lithium with the possibility of the addition of paroxetine Groningen: s.n. Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date:

2 Stellingen 1. De toevoeging van lamotrigine aan lithium bij patiënten met een bipolaire stoornis gedurende een depressieve episode is effectief en veilig (dit proefschrift). 2. Verschillende stappen van additie van medicatie tijdens een depressieve episode van de bipolaire stoornis is niet alleen care as usual, het is ook te onderbouwen (dit proefschrift). 3. Onderzoek naar additie-strategieën bij de behandeling van bipolaire stoornissen is nodig, nuttig en mogelijk (dit proefschrift). 4. Het is mogelijk om binnen één studie drie achtereenvolgende stappen in de behandeling van de bipolaire stoornis te onderzoeken (dit proefschrift). 5. De kwaliteit van het separatiebeleid in een psychiatrisch ziekenhuis afmeten aan een verplichte daling van het aantal separaties met 10 procent is absurd en doet geen recht aan de ernst van de situatie van de patiënt. 6. De angst voor samenwerking met de farmaceutische industrie is een rem op kwalitatief hoogwaardig onderzoek en evidence based medicine. 7. Marketeers zijn er in geslaagd mensen te doen geloven dat kamperen het optimum aan vrijheid en privacy betekent. 8. Het belang van het hebben van een eigen mening wordt in het algemeen schromelijk overschat. 9. Gelukkig zijn er nog steeds mensen die niet doen wat ze zeggen en vooral niet zeggen wat ze denken. 10. Hysterie is als diagnose niet verdwenen omdat het ziektebeeld niet meer bestaat maar omdat disproportioneel reageren normaal is geworden. 11. Social media is een contradictio in terminus. Marc van der Loos Lamotrigine in bipolar depression: a randomised placebo-controlled trial on the acute and long-term outcome of lamotrigine as add-on to lithium with the possibility of the addition of paroxetine 28 september 2011

3 LAMOTRIGINE IN BIPOLAR DEPRESSION a randomised placebo-controlled trial on the acute and long-term outcome of lamotrigine as add-on to lithium with the possibility of the addition of paroxetine Marc van der Loos

4 Cover design Koen Kloosterhuis, May 2011 Design and lay-out Goedbezig Naarden, the Netherlands Print Brügemann B.V. Hilversum, the Netherlands ISBN: Copyright 2011 M.L.M. van der Loos No part of this book may be reproduced in any form without permission of the author.

5 RIJKSUNIVERSITEIT GRONINGEN LAMOTRIGINE IN BIPOLAR DEPRESSION a randomised placebo-controlled trial on the acute and long-term outcome of lamotrigine as add-on to lithium with the possibility of the addition of paroxetine Proefschrift ter verkrijging van het doctoraat in de Medische Wetenschappen aan de Rijksuniversiteit Groningen op gezag van de Rector Magnificus, dr E. Sterken, in het openbaar te verdedigen op woensdag 28 september 2011 om uur door Marcus Lambertus Maria van der Loos geboren op 10 maart 1958 te Amsterdam

6 Promotor: Prof dr W.A. Nolen Beoordelingscommissie: Prof dr R.A. Schroevers Prof dr R.W. Kupka Prof J.R. Calabrese M.D.

7 Ah, si j avais pu travailler sans cette sacré maladie! Que de chôses j aurais fait isolé des autres selon que le pays m en dirait. Mais oui - c est bien fini ce voyage ci. Ah, if I d been able to work without this bloody illness! How many things I could have done, isolated from the others, according to what the land would tell me. But yes - this journey is well and truly finished To Theo van Gogh. Saint-Rémy-de-Provence, on or about Thursday, 1 May 1890 Letter 865, Vincent van Gogh

8

9 Contents Chapter 1 Introduction Chapter 2 Lamotrigine in bipolar disorder: an overview Chapter 3 Efficacy and Safety of Lamotrigine as Add-On Treatment to Lithium in Bipolar Depression: A Multicenter, Double-Blind, Placebo-Controlled Trial Chapter 4 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine Chapter 5 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design Chapter 6 Discussion Summary Samenvatting Curriculum Vitae M.L.M. van der Loos Dankwoord (Acknowledgments)

10 12

11 13 Introduction 1

12 14 Chapter 1 Foreword Bipolar disorder (manic-depressive illness) is one of the great and maybe the most intriguing of psychiatric diseases, not only because of its high prevalence but also because of its spreading through time and cultures. The change of mood over a short time has already been described by ancient Greeks as a disturbance of the equilibrium between yellow and black bile. Where an excess of yellow bile was responsible for (hypo)manic episodes, an excess of black bile was a cause of melancholia. The term manic-depressive illness (manisch-depressieve Irresein) was introduced by Kraepelin(1) in The modern term bipolar disorder (which gives also the less severe (hypo) manic episodes a place and differentiates between unipolar and bipolar disorders) was introduced in 1957 by Leonard(2). The core characteristic of bipolar disorder is the alternation in mood between high, low and normal. Moreover, there are changes in energy, self esteem, sexual desire, need of sleep, keeping relations with other people. In the most serious forms psychosis, rage, and suicide can destroy or even end lives. In this thesis we will use the classification of bipolar disorders according to the DSM-IV(3) (table 1, page 18, 19, 20 and 21). The instability of the core characteristics of the self of patients may be one of the heaviest burdens of this disease. Who am I? An energetic, attractive, confident outgoing big spender who rules the world or the slow, gloomy, tired, suicidal bunch of misery or may be something in between? How can I rely on myself, make appointments, live together, raise children, follow a study, find and keep a job, etcetera when there is no stable Me? Questions enough for patients but also for relatives and treating physicians. There can be a great discrepancy between the goals of the physician and the patient. The main problem is the definition of stability. Whereas patients may perceive stability as plain and boring, family members and doctors may

13 Introduction 15 regard such a situation as desirable. The frequently used questionnaires to assess severity of illness such as the Montgomery-Åsberg Depression Rating Scale (MADRS)(4), the Hamilton Depression Rating Scale (HAM-D)(5), and the Young Mania Rating Scale (YMRS)(6) are all designed to measure the presence and severity (or absence) of symptoms. They don t say much about functioning of the patient or his satisfaction with life. For many patients hypomania or even mania is the most sought after situation. As one patient once stated: I prefer being secluded once in a few years as long as I don t have to take lithium and feel nothing at all. 1 In our society there is a great pressure on being enthusiastic, active, outgoing, sexual active, etcetera. In fact these characteristics can also be regarded as symptoms of hypomania. For that reason it is not so strange that drugs like cocaine and amphetamines which can bring users in a euphoric (or hypomanic ) state, are so popular. And even the depression and insomnia which come after the euphoria, are comparable with the depressive episodes which comes so often after a (hypo)manic episode. In the beginning these drugs were mostly used in the glamorous world of art, film and fashion. But also in bipolar disorder there is always a hint of romanticism and artistry. In 1993 Kay Redfield Jameson wrote Touched with fire, a book about the association of the artistic temperament with bipolar disorder. The first sentence of this book is a quote from Lord Byron: We writers are all insane. This romantic view of disease is in the medical world exclusive for psychiatric diseases; Patients suffering from addiction are viewed as bohemien, schizophrenic patients are hidden geniuses (see: A beautiful Mind (Sylvia Nasar 1998)) and last but not least bipolar patients are artists. Especially hypomanic episodes were hardly seen as part of a disease but more as the holy grail of psychiatry. A state of mind which gives way to all kinds of artistic eruptions. The other side of this high mood; the depression is a far less sexy kind of mood disorder, a prize to be paid preferentially in silence. This silence is best demonstrated with 932 hits in Pubmed for the combination of the search terms bipolar depression and randomised controlled trial versus hits for the combination depression

14 16 Chapter 1 and randomised controlled trial. In this thesis we try to add some knowledge to the treatment of the bipolar depression; an understudied but very significant medical problem. Stabilizing mood and treatment of depression Although suffering and also mortality due to suicide is high in bipolar depression, the focus of research into the treatment of bipolar disorder has until recently been mostly on the acute treatment of manic episodes (e.g. with antipsychotics) and on maintenance treatment (e.g. with lithium). The treatment of bipolar depression was hardly studied as apparently the same treatments were used as for unipolar (i.e. non-bipolar) depression: initially since the 1940ies electroconvulsive therapy (ECT) and since the 1950ies and 1960ies the antidepressants, originally the tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs) and later the selective serotonin reuptake inhibitors (SSRIs) and other related compounds. Only after the 1980ies there became some interest for alternatives for these treatments, especially when it became evident that the regular antidepressants might also induce a switch to (hypo)mania and an increase of episode frequency; cycle acceleration(7;8). With the introduction of lithium in the 1970ies(9) there became also interest in treatments that were effective against both poles (mania and depression) of the illness, not only to reduce the symptoms of the mood episodes (as acute treatment) but also to prevent further episodes, as long-term or maintenance treatment. For the latter purpose the quest was for a mood stabilizer. The main problem with this concept is that the term mood stabilizer has many definitions. The most stringent definition is that a true mood stabilizer demonstrates efficacy in the acute treatment of both manic and depressive episodes and prevents against further manic and depressive episodes (fig 1).

15 Introduction 17 Figure 1. What is a mood stabilizer? 1. Effective in treating acute mania Both acutely and in continuation Without causing depression 1 2. Effective in treating acute bipolar depression Both acutely and in continuation Without causing mania 3. Effective in preventing recurrence of both mania and depression (prophylaxis) (And also in rapid cycling) Figure 2. Treatments from above and from below Treat from above Mania Symptoms Normalcy Stabilize from above Stabilize from both sides Symptoms Depression Stabilize from below Treat from below

16 18 Chapter 1 The loosest definition requires efficacy in only one pole without worsening of the other pole. Although it is in our opinion justified to use a strict definition, it makes sense to differentiate between mood stabilizing from below (i.e. to diminish depression) and mood stabilizing from above (i.e. to diminish mania)(10). Until recently lithium came closest to meeting the strictest definition as a mood stabilizer. Also valproic acid/valproate and carbamazepine claimed this term but unfortunately all three compounds were considered (and partly also found) more effective in treating and preventing manic episodes than depressive episodes. As a result, bipolar depressed patients were mostly treated with antidepressants which had largely been tested only in unipolar depression. As evidence grew that the efficacy of antidepressants is limited(11;12) and that there may be the risk of induction of (hypo) mania or cycle acceleration(7;8), the treatment of bipolar depression became an unmet need: for a mood stabilizer from below without inducing a (hypo)manic episode (fig 2). Lamotrigine After some case reports in the 1990ies(13;14) it became clear that lamotrigine might be such a drug. In 1999, the first randomized, placebocontrolled trial showed efficacy of lamotrigine in the acute treatment of bipolar depression, although not on the a priori chosen primary outcome measure (the HAM-D) but only on secondary outcome measures (e.g. the MADRS) (15). However, four subsequent studies could not replicate this finding(16). At the same time two other randomized placebo-controlled trials studying the long-term effect of lamotrigine showed a preventive effect against further depressive episodes but not against manic episodes, while in the same studies lithium was found effective in the prevention of manic episodes but not of depressive episodes(17;18). This brought us to the idea to start a double-blind randomized placebocontrolled trial into the effect of lamotrigine as add-on to ongoing treatment

17 Introduction 19 with lithium in patients with a bipolar depression. During Phase I of that study lamotrigine or placebo was added to ongoing treatment with lithium for 8 weeks. After 8 weeks responders continued their combination of medication whereas paroxetine was added open label in non-responders in both groups (lithium plus lamotrigine and lithium plus placebo) (Phase II). At 16 weeks responders continued their medication regime and were followed until a new depressive or (hypo)manic relapse or until the maximum duration of the study (68 weeks) (Follow-up). 1 Outline of this thesis In chapter II we will review the pharmacological profile and the place of lamotrigine in bipolar disorder. In chapter III the results obtained during the first 8 weeks of the study (phase I) will be discussed. In chapter IV the results obtained during week 9-16 (phase II) will be discussed. In chapter V the results obtained during follow-up (until 68 weeks) of responders at week 16 will be discussed. In chapter VI the overall results of our study and more specifically the place of lamotrigine among the other options in the treatment of bipolar depression and the advantages and disadvantages of the design of our study in comparison with other long-term studies that have been performed in bipolar disorder will be discussed. Finally, some suggestions for future research will be presented.

18 20 Chapter 1 Table 1 Bipolar II Disorder-Diagnostic Features The essential feature of Bipolar II Disorder is a clinical course that is characterized by the occurrence of one or more Major Depressive Episodes accompanied by at least one Hypomanic Episode. Hypomanic Episodes should not be confused with the several days of euthymia that may follow remission of a Major Depressive Episode. Episodes of Substance- Induced Mood Disorder (due to the direct effects of a medication, or other somatic treatments for depression, a drug of abuse, or toxin exposure) or of Mood Disorder Due to a General Medical Condition do not count toward a diagnosis of Bipolar I Disorder. In addition, the episodes are not better accounted for by Schizoaffective Disorder and are not superimposed on Schizophrenia, Schizophreniform Disorder, Delusional Disorder, or Psychotic Disorder Not Otherwise Specified. Criteria for Major Depressive Episode A. Five (or more) of the following symptoms have been present during the same 2-week period and represent a change from previous functioning; at least one of the symptoms is either (1) depressed mood or (2) loss of interest or pleasure Note: Do not include symptoms that are clearly due to a general medical condition, or mood-incongruent delusions or hallucinations. 1. depressed mood most of the day, nearly every day, as indicated by either subjective report (e.g., feels sad or empty) or observation made by others (e.g. appears tearful). Note: In children and adolescents, can be irritable mood. 2. markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day (as indicated by either subjective account or observation made by others) 3. significant weight loss when not dieting or weight gain (e.g., a change of more than 5% of body weight in a month), or decrease or increase in appetite nearly every day. Note: In children, consider failure to make expected weight gains. 4. insomnia or hypersomnia nearly every day 5. psychomotor agitation or retardation nearly every day (observable by others, not merely subjective feelings of restlessness or being slowed down) 6. fatigue or loss of energy nearly every day

19 Introduction feelings of worthlessness or excessive or inappropriate guilt (which may be delusional) nearly every day (not merely self-reproach or guilt about being sick) 8. diminished ability to think or concentrate, or indecisiveness, nearly every day (either by subjective account or as observed by others) 9. recurrent thoughts of death (not just fear of dying), recurrent suicidal ideation without a specific plan, or a suicide attempt or a specific plan for committing suicide 1 B. The symptoms do not meet criteria for a Mixed Episode. C. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning. D. The symptoms are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition (e.g., hypothyroidism). E. The symptoms are not better accounted for by bereavement, i.e., after the loss of a loved one, the symptoms persist for longer than 2 months or are characterized by marked functional impairment, morbid preoccupation with worthlessness, suicidal ideation, psychotic symptoms, or psychomotor retardation. Criteria for Manic Episode A. A distinct period of abnormally and persistently elevated, expansive, or irritable mood, lasting at least 1 week (or any duration if hospitalization is necessary). B. During the period of mood disturbance, three (or more) of the following symptoms have persisted (four if the mood is only irritable) and have been present to a significant degree: 1. inflated self-esteem or grandiosity 2. decreased need for sleep (e.g., feels rested after only 3 hours of sleep) 3. more talkative than usual or pressure to keep talking 4. flight of ideas or subjective experience that thoughts are racing 5. distractibility (i.e., attention too easily drawn to unimportant or irrelevant external stimuli)

20 22 Chapter 1 6. increase in goal-directed activity (either socially, at work or school, or sexually) or psychomotor agitation 7. excessive involvement in pleasurable activities that have a high potential for painful consequences (e.g., engaging in unrestrained buying sprees, sexual indiscretions, or foolish business investments) C. The symptoms do not meet criteria for a Mixed Episode. D. The mood disturbance is sufficiently severe to cause marked impairment in occupational functioning or in usual social activities or relationships with others, or to necessitate hospitalization to prevent harm to self or others, or there are psychotic features. E. The symptoms are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication, or other treatments) or a general medical condition (e.g., hyperthyroidism). Note: Manic-like episodes that are clearly caused by somatic antidepressant treatment (e.g., medication, electroconvulsive therapy, light therapy) should not count toward a diagnosis of Bipolar I Disorder. Criteria for Mixed Episode A. The criteria are met both for a Manic Episode and for a Major Depressive Episode (except for duration) nearly every day during at least a 1-week period. B. The mood disturbance is sufficiently severe to cause marked impairment in occupational functioning or in usual social activities or relationships with others, or to necessitate hospitalization to prevent harm to self or others, or there are psychotic features. C. The symptoms are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication, or other treatment) or a general medical condition (e.g., hyperthyroidism). Criteria for Hypomanic Episode

21 Introduction 23 A. A distinct period of persistently elevated, expansive, or irritable mood, lasting throughout at least 4 days, that is clearly different from the usual nondepressed mood. B. During the period of mood disturbance, three (or more) of the following symptoms have persisted (four if the mood is only irritable) and have been present to a significant degree: 1. inflated self-esteem or grandiosity 2. decreased need for sleep (e.g., feels rested after only 3 hours of sleep) 3. more talkative than usual or pressure to keep talking 4. flight of ideas or subjective experience that thoughts are racing 5. distractibility (i.e., attention too easily drawn to unimportant or irrelevant external stimuli) 6. increase in goal-directed activity (either socially, at work or school, or sexually) or psychomotor agitation 7. excessive involvement in pleasurable activities that have a high potential for painful consequences (e.g., engaging in unrestrained buying sprees, sexual indiscretions, or foolish business investments) 1 C. The episode is associated with an unequivocal change in functioning that is uncharacteristic of the person when not symptomatic. D. The disturbance in mood and the change in functioning are observable by others. E. The episode is not severe enough to cause marked impairment in social or occupational functioning, or to necessitate hospitalization, and there are no psychotic features. F. The symptoms are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication, or other treatment) or a general medical condition (e.g., hyperthyroidism). Note: Hypomanic-like episodes that are clearly caused by somatic antidepressant treatment (e.g., medication, electroconvulsive therapy, light therapy) should not count toward a diagnosis of Bipolar II Disorder.

22 24 Chapter 1 References 1. E.Kraepelin. Psychiatrie. Ein Lehrbuch fur Studirendeund Aerzte. II Band Sechste Auflage. Leipzig, Verlag von Johann Ambrosius Barth K.Leonhard. Aufteilung der Endogenen Psychosen. Berlin, Academie Verlag American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV). Washington DC, American Psychiatric Press Montgomery SA, Åsberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979 Apr;134: HAMILTON M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960 Feb;23: Young RC, Nysewander RW, Schreiber MT. Mania scale scores, signs, and symptoms in forty inpatients. J Clin Psychiatry 1983 Mar;44(3): Altshuler LL, Post RM, Leverich GS, Mikalauskas K, Rosoff A, Ackerman L. Antidepressant-induced mania and cycle acceleration: a controversy revisited. Am J Psychiatry 1995 Aug;152(8): Licht RW, Gijsman H, Nolen WA, Angst J. Are antidepressants safe in the treatment of bipolar depression? A critical evaluation of their potential risk to induce switch into mania or cycle acceleration. Acta Psychiatr Scand 2008 Nov;118(5): SCHOU M, JUEL-NIELSEN N, STROMGREN E, VOLDBY H. The treatment of manic psychoses by the administration of lithium salts. J Neurol Neurosurg Psychiatry 1954 Nov;17(4): Ketter TA, Calabrese JR. Stabilization of mood from below versus above baseline in bipolar disorder: a new nomenclature. J Clin Psychiatry 2002 Feb;63(2): Ghaemi SN, Wingo AP, Filkowski MA, Baldessarini RJ. Long-term antidepressant treatment in bipolar disorder: meta-analyses of benefits and risks. Acta Psychiatr Scand 2008 Aug 24.

23 Introduction Sachs GS, Nierenberg AA, Calabrese JR, Marangell LB, Wisniewski SR, Gyulai L, et al. Effectiveness of adjunctive antidepressant treatment for bipolar depression. N Engl J Med 2007 Apr 26;356(17): Smith D, Baker G, Davies G, Dewey M, Chadwick DW. Outcomes of add-on treatment with lamotrigine in partial epilepsy. Epilepsia 1993 Mar;34(2): Calabrese JR, Fatemi SH, Woyshville MJ. Antidepressant effects of lamotrigine in rapid cycling bipolar disorder. Am J Psychiatry 1996 Sep;153(9): Calabrese JR, Bowden CL, Sachs GS, Ascher JA, Monaghan E, Rudd GD. A doubleblind placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. Lamictal 602 Study Group. J Clin Psychiatry 1999 Feb;60(2): Calabrese JR, Huffman RF, White RL, Edwards S, Thompson TR, Ascher JA, et al. Lamotrigine in the acute treatment of bipolar depression: results of five double-blind, placebo-controlled clinical trials. Bipolar Disord 2008 Mar;10(2): Calabrese JR, Bowden CL, Sachs G, Yatham LN, Behnke K, Mehtonen OP, et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently depressed patients with bipolar I disorder. J Clin Psychiatry 2003Sep;64(9): Bowden CL, Calabrese JR, Sachs G, Yatham LN, Asghar SA, Hompland M, et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently manic or hypomanic patients with bipolar I disorder. Arch Gen Psychiatry 2003 Apr;60(4):

24 26

25 27 Lamotrigine in bipolar disorder: an overview Marc L.M. van der Loos (1) Joseph R. Calabrese, M.D. (2) Willem A. Nolen (3) David J. Muzina, M.D. (4) 1. Department of Psychiatry, Isala Klinieken, location Sophia, Zwolle, the Netherlands 2. Case Western Reserve University School of Medicine, University Hospitals of Cleveland, Cleveland, Ohio, USA 3. Department of Psychiatry, University Medical Center Groningen, University of Groningen, Groningen, the Netherlands 4. Department of Psychiatry and Psychology, Cleveland Clinic Foundation, Cleveland, Ohio This chapter is an adaptation from: Chapter 4 Pharmacological profile and clinical utility of lamotrigine in mood disorders Bipolar Psychopharmacotherapy: Caring for the Patient, 2nd Edition Hagop S. Akiskal (Editor), Mauricio Tohen (Editor) Wiley-Blackwell John Wiley & Sons, Ltd. 2

26 28 Chapter 2 Introduction There remains a pressing need for additional treatment options for mood disorders, particularly those refractory or difficult to treat disorders such as bipolar disorder. Estimates of the prevalence of bipolar I disorder across diverse cultures and ethnic groups are consistent, ranging between 0.4% and 1.6% in adults (Weissman 1996 and Regeer 2004). Recent estimates of the overall prevalence of bipolar disorder (types I and II as well as NOS) suggest that 4-5% suffers from this debilitating mental disorder (Hirschfeld.2003; Regeer 2004). Bipolar disorders carries a high burden, negatively affecting lives in many areas, most notably the performance of work-related, leisure, and interpersonal activities (Calabrese 2003). The greatest need exists in the treatment of depressive episodes associated with bipolar disorders as symptoms of depression are much more commonly experienced and more difficult to treat than manic symptoms (Post 2002; Judd 2002 and Kupka 2005). Reports of lamotrigine s beneficial effects on mood in epilepsy patients led to its use and study in affective disorders. Studies involving bipolar patients suggested that lamotrigine possessed a broad spectrum of therapeutic activity in bipolar disorder, with later descriptions of greater efficacy for depressive than for manic symptoms. Given the significant burden, prevalence, and recurrent nature of bipolar disorder - especially bipolar depression lamotrigine appears to begin to address an area of serious unmet public health need (Calabrese 2003). Clinical Pharmacology of Lamotrigine Pharmacodynamics Lamotrigine is an antiepileptic drug of the phenyltriazine class that has demonstrated efficacy as add-on treatment of partial seizures (Mikati 1989; Matsuo 1993) and as maintenance treatment of bipolar I disorder to delay the time of occurrence of mood episodes (Bowden 2003; Calabrese 2003). Although its mechanism of therapeutic action in humans is not definitively understood, lamotrigine is thought to possess modulatory and protective effects

27 Lamotrigine in bipolar disorder: an overview 29 on neurotransmission and intracellular signal transduction processes. Structurally distinct from other antiepileptic drugs, lamotrigine interacts preferentially on the slow inactivated state of presynaptic neuronal sodium and calcium channels to prolong inactivation of the neuron and promote stabilization of the neuronal membrane (Xie 1998). This effect is augmented by a usedependent action in which further inhibition by the drug develops during rapid, repetitive stimulation (i.e., epileptiform bursts). Consequently, the release of the excitatory amino acid glutamate is antagonized (Fitton 1995, Li and Ketter 2002). Lamotrigine has also been observed to inhibit cortical and amygdaloid kindling (Gilman 1995, Leach 1986 and Xie 1995). Lamotrigine has no substantial in vitro affinity for adenosine, adrenergic, dopaminergic, muscarinic and opioid receptors at clinically applicable concentrations and binds only weakly to inhibit serotonin 5HT3 receptors (Leach 1991). Lamotrigine lacks clinically meaningful activity at the 5HT1A receptor, where changes in 5HT1A receptor-mediated cyclic adenosine monophosphate pathway have been implicated in affective disorders (Shiah 1998 and Vinod 2002). In early epilepsy studies with lamotrigine it was noted that many patients reported improvement in mood and psychological well-being independent from reduction in seizure frequency, which led to its investigation for use in affective disorders (Smith 1993a; Smith 1993b). Significantly, lamotrigine has minimal negative effects on cognitive, memory, or psychomotor function and is not associated with sedative effects or weight gain (Goa 1993, Cohen 1985 and Ginsberg 2003). 2 Pharmacokinetics Lamotrigine is extensively absorbed demonstrating linear kinetics, resulting in 98% bioavailability (Garnett 1997). Peak plasma concentrations are attained after 1-3 hours with mean plasma protein binding of 55-68% (Cohen 1987, Ramsay, 1991 and Rambeck 1993). Lamotrigine readily crosses the placental barrier causing fetal blood concentrations similar to maternal levels and passes into breast milk reaching 40-80% of the maternal lamotrigine

28 30 Chapter 2 concentration (Ohman, 2000 and Pennell 2003). The rate-limiting step in the elimination of lamotrigine is N-glucoronidation by the liver with a plasma elimination half-life of hours (Cohen 1987). Autoinduction of its own metabolism does not occur and there are no active metabolites. Pregnancy increases lamotrigine clearance by more than 50% early during pregnancy and reverts quickly after delivery (Tran 2002). Clearance may be reduced in the elderly and in patients with moderate to severe hepatic dysfunction. Drug Interactions Lamotrigine administration affects hardly the serum concentrations of other drugs. However, enzyme-inhibiting drugs such as divalproex sodium increase lamotrigine concentrations by significantly competing for metabolism through glucoronidation effectively increasing the mean half-life of lamotrigine to about 70 hours (Yuen 1992 and Anderson 1996). Enzyme-inducing drugs, such as phenytoin, carbamazepine, and phenobarbital, reduce the lamotrigine s mean elimination half-life to about 12 hours and decrease lamotrigine concentrations (Hachad 2002). There are pharmacokinetic interactions between lamotrigine and anti-conceptive medication.(sidhu 2005). The clearance of levonogestrel is increased with a possible decrease of the anti-conceptive reliability. On the other hand is the clearance of lamotrigine enhanced by anti-conceptive medication, with a lowering of lamotrigine plasma levels. Lamotrigine does not significantly affect the pharmacokinetics of lithium (Chen 2000). Lamotrigine and Mood Disorders Acute treatment of bipolar depression The observations of Smith and colleagues (Smith 1993a, 1993b) that patients treated with add-on lamotrigine for partial seizures experienced improved mood apart from effects on epilepsy stimulated investigations of lamotrigine s efficacy to treat mood disorders. Early case reports involving

29 Lamotrigine in bipolar disorder: an overview 31 bipolar patients suggested that lamotrigine possessed a broad spectrum of therapeutic activity in bipolar disorder, including rapid cycling and mixed states (Weisler 1994, Calabrese 1996 and Walden 1996). A subsequent open-label study provided preliminary data that lamotrigine was effective for patients with refractory bipolar disorder, with 68% of depressed patients and 84% of manic / hypomanic / mixed patients showing moderate to marked response (Calabrese 1999a). An open naturalistic study of twenty-two depressed bipolar patients who were refractory to treatment with a combination of divalproex sodium and another mood stabilizer or divalproex sodium and an antidepressant for 6 weeks were treated with add-on lamotrigine. Sixteen out of 22 (72%) responded by the end of week 4, suggesting that lamotrigine may be useful in bipolar depression (Kusumakar 1997). In a double-blind placebo-controlled study of lamotrigine monotherapy in outpatients with non rapid-cycling bipolar I depression, treatment with lamotrigine over 7 weeks (50 or 200 mg/day) did not result in a significant difference versus placebo on the primary outcome measure (HAM-D), but did so on several secondary outcome measures including the MADRS, particularly in the 200 mg group, compared with placebo (Calabrese 1999b). Improvements were observed as early as week 3. After this first positive result for lamotrigine in a RCT, lamotrigine was tested against placebo in four other RCT s (Calabrese 2007 and Goldsmith 2003). None of these RCT s discriminated significantly between lamotrigine and placebo on the primary outcome criteria. Part of this negative result was because of a high placebo response in all four studies. Nevertheless, a meta-analysis of all five studies showed a significant although modest benefit for lamotrigine. (Geddes 2009) Thirty-one patients with refractory bipolar and unipolar mood disorders participated in a double-blind, randomized, crossover study of three 6-week monotherapy evaluations including lamotrigine, gabapentin, and placebo. Using the Clinical Global Impressions Scale for Bipolar Illness (CGI-BP) as the primary outcome variable, this study reported a 52% response rate (CGI-BP as much or very much improved) for lamotrigine, compared to 26% for gabapentin and 23% 2

30 32 Chapter 2 for placebo-treated patients (Frye 2000). The final study (n=410) compared lamotrigine (dose uptitrated to 200 mg/day) with olanzapine/fluoxetine (dose 6/25, 6/50, 12/25 or 12/50 mg/ day) (Brown.2006) and showed a small but significant advantage (improvement on the MADRS after 7 weeks of vs respectively p=0.002) for the olanzapine/ fluoxetine combination. Acute treatment of mania / mixed states The 1999 open-label study by Calabrese and colleagues observed a robust response to lamotrigine (primarily as add-on therapy) in 84% of manic / hypomanic / mixed patients (Calabrese 1999a). In the first double-blind, randomized, controlled study of lamotrigine in acute mania lamotrigine titrated to 100 mg/day over three weeks was as effective as lithium 800 mg/day (mean blood level 0.77 mmol/l) in reducing manic symptoms (Ichim 2000). However, lamotrigine has not shown anti-manic activity in placebo-controlled studies (Bowden 2003). Unpublished studies (data on file, GlaxoSmithKline 2003) in patients with acute manic or mixed exacerbations of bipolar I disorder have found no significant difference from baseline in the 11-item Mania Rating Scale (MRS) between lamotrigine and placebo (Goldsmith 2003). Maintenance therapy The initial lamotrigine maintenance study was a double-blind, placebocontrolled study in rapid cycling bipolar disorder (Calabrese 2000). Open label lamotrigine added to the treatment regimens of 324 patients meeting DSM-IV criteria for rapid cycling bipolar disorder resulted in 182 stabilized patients who were then randomly assigned to the double-blind maintenance phase after being stratified for bipolar I or II disorder. Other psychotropic agents were tapered and patients randomly assigned to either lamotrigine monotherapy or placebo in a 1:1 ratio for the six-month maintenance phase. Overall 49 placebo patients (56%) and 45 lamotrigine-treated patients (50%) required treatment of an emerging mood episode with additional pharmacotherapy. Although there was no statistical significance between these two treatment groups on the primary

31 Lamotrigine in bipolar disorder: an overview 33 outcome measure of time to additional pharmacotherapy (median of 12 weeks for placebo versus 18 weeks for lamotrigine), lamotrigine was significantly more effective than placebo in the survival in study analysis (median of 8 weeks for placebo versus 14 weeks for lamotrigine). In the sub-analysis according to disease type, lamotrigine was significantly more effective than placebo in delaying time to additional pharmacotherapy for bipolar II patients than bipolar I patients. Forty-six percent of bipolar II patients on lamotrigine monotherapy were stable without relapse after 6 months compared to only 18% of placebo-treated bipolar II patients. (Calabrese 2000). These results were confirmed in a post hoc analysis of the same patient sample with the Life chart method. Measured with the life chart at least once weekly patients taking lamotrigine were 1,8 times more likely to achieve euthymia than those taking placebo (Goldberg 2008). In an unpublished study (data on file, GlaxoSmithKline 2003), these findings were not replicated. However, significantly fewer lamotrigine treated rapid cyclers required intervention for a depressive episode. Collectively, these results suggest that lamotrigine may be a useful treatment for rapid cycling bipolar patients, especially for type II and for the prevention of depressive relapses. Two 18-month maintenance studies (Bowden 2003 and Calabrese 2003) comparing lamotrigine, lithium, and placebo provided further support for the use of lamotrigine as a mood stabilizer and led to its approval by the US FDA in June 2003 as a maintenance treatment for bipolar I disorder. These complementary studies enrolled patients in a double-blind phase of maintenance therapy after a recent depressed, hypomanic, or manic episode remitted (CGI-S score of < 3 for 4 consecutive weeks) during open-label stabilization during which lamotrigine was initiated as adjunctive or monotherapy and other psychotropics discontinued. In both studies, 50% of patients achieved stabilization criteria allowing for progression into double-blinded maintenance therapy with lamotrigine ( mg/day), lithium ( meq/l), or placebo. The primary efficacy endpoint used was time to intervention for any mood episode. In both of these studies, lamotrigine and lithium demonstrated effective prophylaxis against any emerging mood episode compared with placebo. There were statistically fewer relapsing mood episodes and longer median survival in 2

32 34 Chapter 2 lamotrigine and lithium treated patients compared with placebo. Median survival before intervention for any mood episode for lamotrigine-treated patients ranged from days, significantly better than placebo (85-93 days). Lithium also outperformed placebo with a median survival of days. However, there were important differences in the spectra of maintenance efficacy. In both studies, lithium, but not lamotrigine was superior to placebo at delaying the time to intervention for a manic or hypomanic episode. In total for both studies 123 hypomanic/manic events emerged during maintenance, with 25% of placebo treated patients experiencing a hypomanic, manic, or mixed episode (47 of 188); 21% of the lamotrigine group relapsed (58 of 273), but only 11% of patients in the lithium-treatment arm had breakthrough mania or hypomania (18 of 164). Lithium was clearly more effective than placebo and lamotrigine in preventing manic and hypomanic relapse in recently symptomatic bipolar I patients. In contrast, lamotrigine appears to be more effective than lithium in the prevention of depressive relapse in bipolar I patients. In both studies, lamotrigine but not lithium, was significantly better than placebo at prolonging time to intervention for a depressive mood episode. There were 209 depressive relapses in the patients observed in the two studies, with a higher rate of relapse in the placebo (68 of 188, or 36%) and lithium treatment groups (56 of 164, or 34%); lamotrigine was better at preventing recurrence of depression with a 31% rate of depressive relapse and a longer median survival before treatment of depression was necessary. The difference was particularly evident in the treatment of recently manic or hypomanic patients in whom only 14% of those receiving lamotrigine (8 of 58) needed intervention for emergent depression compared with 23% of patients receiving lithium (10 of 44) and 30% receiving placebo (21 of 69). In a post hoc analysis of the first 6 months of the bipolar I patients (index episode depression) there was no evidence for a greater risk for mood destabilization (emergent (hypo) manic symptoms) in the lamotrigine group versus the placebo group (Goldberg 2009).

33 Lamotrigine in bipolar disorder: an overview 35 Safety Rash Rashes requiring hospitalization and discontinuation of treatment have been reported with lamotrigine. The reported rash rate is 10% in adult patients with epilepsy treated with adjunctive lamotrigine and 7% in bipolar I adults treated with monotherapy; the placebo rate is 5% in both populations (lamotrigine package insert). Serious rash rates are considerably lower at 0.8% in pediatric epilepsy patients, 0.3% in epileptic adults as monotherapy and 0.13% as adjunctive therapy. A retrospective analysis conducted of rates of lamotrigine-related rash in 12 multicenter studies of lamotrigine for mood disorders (n=1955), including 1 open study, 7 randomized controlled acute trials, and 4 randomized controlled maintenance trials from 1996 to 2001 reported serious rash in 0% with lamotrigine, 0.1% (n = 1) with placebo, and 0% with comparators (Calabrese 2002). Across all bipolar disorder studies with 2,272 patients on lamotrigine, serious rash was observed in three patients for an overall rate of 0.1% (lamotrigine package insert). One of these three patients experienced a mild Stevens-Johnson syndrome on adjunctive lamotrigine and did not require hospitalization. In a randomized trial Usual Care Precautions were compared with Dermatologic Precautions (UCP plus additional specific precautions intended to decrease the risk of rash) for 12 weeks (Ketter 2006). Lamotrigine was added (target dose 200mg/day) to ongoing medication with a slow up titration. 2 Usual Care precautions were: 1. not exceeding the initial dose or dose-escalation schedule 2. if a rash developed during the study the patient contacted the investigator immediately 3. lamotrigine was stopped in case of rash. Patients in the Dermatologic Precautions group were, besides UCP, advised not to: 4. take new medications or foods, use new cosmetics, deodorants, detergents or

34 36 Chapter 2 fabric softeners 5. stimulate the immune system through excessive sun exposure 6. participate in activities that might lead to exposure to poison oak or poison ivy 7. receive any immunizations 1175 patients were included. In both groups there were no reports of serious rash. The rate of non-serious rash was low for UCP as well as for DP (8.8% and 8.6% respectively, OR 0.99). which is comparable with rates of rash in previous studies. Clinical response was 50% with 29% remission over both groups (according to the Clinical Global Impressions Bipolar version). A German registry that documents the incidences of serious rash has been in place since 1990 and ascertains hospitalized cases of Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) related to antiepileptic drug use (Rzany 1996). Reported cases are confirmed by a registry physician and reviewed by a dermatological expert committee ensuring high diagnostic accuracy. The incidence of serious rash rising to the level of SJS/TEN diagnosis in the German registry was 0.02% from (Messenheimer 2002). Risk of rash is noted to be higher in pediatric patients and may be increased by co-administration of divalproex sodium or exceeding the recommended initial dose or dose escalation schedule for lamotrigine. Women may be at increased risk for rash compared with men, with a relative risk of 1.8 (Wong 1999). Other side effects and tolerability Lamotrigine is known to be generally well tolerated in the treatment of epilepsy (Choi 2003). A great deal of data generated by the 18-month lamotrigine maintenance trials (Bowden 2003 and Calabrese 2003) confirms clinical experience and epilepsy management reports of lamotrigine s favorable side effect profile in the treatment of bipolar patients. In 280 patients on maintenance lamotrigine during double-blinded treatment in both trials combined, only 23 (8.2%) discontinued study participation prematurely due to an adverse event; 7.9% of placebo treated patients experienced adverse events that led to

35 Lamotrigine in bipolar disorder: an overview 37 end of study (15 of 191). Over the 18 months randomized phase, lamotrigine treated patients exhibited a 4.9 pound mean decrease in body weight, compared to a 2.6 pound mean weight gain in the placebo group. In a combined analysis of both studies, the incidence of mania/ hypomania/ mixed episodes reported as adverse events was 5% for patients treated with lamotrigine, 4% for patients treated with lithium, and 7% for patients treated with placebo. In general, lamotrigine exhibits placebo level rates of treatment-emergent adverse events and is better tolerated than lithium (Calabrese 2003). Table 1 lists treatment-emergent adverse events for the combined maintenance studies during randomized phase. 2 Table 1. Adverse events observed during randomized phase of lamotrigine maintenance studies (in percent)* Event Lamotrigine (n = 227) Placebo (n = 190) Nausea Insomnia 10 6 Somnolence 9 7 Back pain 8 6 Fatigue 8 5 Rhinitis 7 4 Benign rash 7 5 Abdominal pain 6 3 Dry mouth 6 4 Constipation 5 2 Vomiting 5 2 Cough exacerbation 5 3 Pharyngitis 5 4 * Adverse events listed had incidence > 5% and were numerically greater than placebo. [Bowden 2003; Calabrese 2003; Data on file, GlaxoSmithKline]

36 38 Chapter 2 Clinical applications for lamotrigine Bipolar depression For the management of acute bipolar depression, the APA Practice Guidelines of 2002 recommended lamotrigine as a first-line option with moderate clinical confidence (Hirschfeld 2002). Since then other options have become available, especially quetiapine and the combination of olanzapine plus fluoxetine. With the data that became available after the 2002 guideline, lamotrigine has a place in the treatment of acute bipolar depression, especially in patients experiencing a depressive episode despite adequate treatment with lithium. In these patients lamotrigine can be recommended as add-on therapy with substantial clinical confidence. Dosing should follow published titration schedules to minimize the risk of rash, with a minimum dose of 50 mg/day and a target dose of 200 mg/ day for most patients (table 2). There is no pharmacokinetic interaction with lithium, However, in bipolar patients experiencing a breakthrough depression on carbamazepine or valproate, lamotrigine should be added while keeping in mind pharmacokinetic interactions that may affect dosing: double dose of lamotrigine in combination with carbamazepine, half the dose in combination with valproate. Table 2. Recommended Lamotrigine dosing schedule Concomitant medication? Not on divalproex or carbamazepine On divalproex On carbamazepine or phenytoin Week 1 & 2 Week 3 & 4 Week 5 Week 6 Week 7 25 mg qd 50 mg qd 100 mg 200 mg 200 mg qd qd qd 25 mg 25 mg qd 50 mg qd 100 mg 100 mg qod qd qd 50 mg qd 100 mg 200 mg 300 mg 300 mg qd qd qd qd

37 Lamotrigine in bipolar disorder: an overview 39 Mania / mixed states Lamotrigine monotherapy is not recommended for the acute management of mania or mixed states associated with bipolar I disorder. Despite limited case reports and early data suggesting that lamotrigine may possess anti-manic properties, further study has led to the conclusion that lamotrigine has only mild to moderate efficacy in mania and should not be reliably used as monotherapy in these urgent above baseline mood disturbances. The addition of lamotrigine to other first-line antimanic agents, such as lithium, divalproex, olanzapine, or quetiapine may hasten or complete recovery from acute manic or mixed states. Lamotrigine should definitely be considered as an initial adjunctive treatment for these bipolar I patients to address the need to prevent future mood episodes after recovery from the current manic or mixed episode. Again, dosing guidelines should be followed as indicated previously and in the package insert to avoid increased risk of rash. In bipolar II patients experiencing acute hypomania, lamotrigine can be used as a monotherapy or in conjunction with another mood stabilizer such as lithium. One effective strategy is the prescription of lamotrigine plus a short-term atypical antipsychotic 2 Maintenance therapy Lamotrigine is indicated by the US FDA and the EMEA for the maintenance treatment of bipolar I disorder to delay the time to occurrence of mood episodes. It should therefore be considered in the treatment of all bipolar patients due to the inherent natural tendency for the illness to be chronic and recurrent. Maintenance monotherapy with lamotrigine can be considered in those individuals with non-bipolar I illness, milder forms of bipolar II disorder with predominant depressive course, and in other bipolar spectrum disorders such as cyclothymic or hyperthymic personalities. In theory, lamotrigine monotherapy could also be considered as a maintenance therapy after successful ECT for either unipolar or bipolar depression. In most patients, lamotrigine should be part of a combined pharmacological strategy to prevent relapse. Although lamotrigine

38 40 Chapter 2 has been demonstrated an effective maintenance treatment for bipolar I disorder, its ability to prevent manic episodes was considerably less robust than its antidepressant ability. A comprehensive bipolar maintenance medication treatment plan must address prevention of depressive and manic phases, and lamotrigine is only moderately effective in the delay of manic episodes necessitating use of other mood stabilizers with a complementary efficacy profile, such as lithium, divalproex.or atypical antipsychotics. Use in women There have been reports of lamotrigine serum concentrations decreasing in women after starting oral contraceptives and increasing after oral contraceptives were discontinued (Sidhu 2005). Dosage adjustments may have to be made in women taking lamotrigine during times of oral contraceptive initiation or termination, although no known link exists between blood levels and therapeutic effects for affective disorders. Lamotrigine carries a Category C FDA warning for use during pregnancy. Pregnant women maintained on lamotrigine during pregnancy should have blood levels of lamotrigine checked monthly as its clearance increases significantly as pregnancy progresses. A preconception determination of lamotrigine blood level in a stable bipolar woman may provide a target dose range to maintain during pregnancy. Clearance of lamotrigine drops quickly after delivery, requiring dosage reduction in most postpartum patients. Breastfeeding is not recommended to nursing mothers on lamotrigine since significant levels of lamotrigine are present in breast milk and long term effects of exposure to neonates is unknown. Although there are some reports of an increased risk for isolated cleft palate or lip deformity after exposure to lamotrigine monotherapy during the first trimester of pregnancy (Holmes 2008) the overall risk for major birth defects appear to be similar to that of the general population (Cunnington 2005).

39 Lamotrigine in bipolar disorder: an overview 41 Summary The discovery of lamotrigine s positive effects on mood in epilepsy patients led to its investigations in and eventual approval for treating bipolar I disorder. Although the exact mood stabilizing mechanism of action for this novel anticonvulsant is unknown, it is thought to work by inhibiting voltage-sensitive sodium currents, in this manner stabilizing neuronal membranes and as a result modulating the presynaptic transmitter release of excitatory amino acids such as glutamate. It is approved for maintenance treatment of bipolar disorder to prevent depressive episodes. In addition, it deserves a place in the treatment of acute bipolar depression, with the strongest evidence as add-on treatment to lithium (this thesis). Lamotrigine is well tolerated, the major risk is a severe rash, and it requires careful dose monitoring. 2

40 42 Chapter 2 References Anderson, G. D., Yau, M. K., Gidal, B. E., Harris, S. J., Levy, R. H., Lai, A. A., Wolf, K. B., Wargin, W. A. and Dren, A. T. (1996),Bidirectional interaction of valproate and lamotrigine in healthy subjects, Clin Pharmacol Ther, 60, Bowden, C. L. (2003), Acute and maintenance treatment with mood stabilizers, Int J Neuropsychopharmacol, 6, Bowden, C. L., Calabrese, J. R., Sachs, G., Yatham, L. N., Asghar, S. A., Hompland, M., Montgomery, P., Earl, N., Smoot, T. M. and DeVeaugh-Geiss, J. (2003), A placebocontrolled 18-month trial of lamotrigine and lithium maintenance treatment in recently manic or hypomanic patients with bipolar I disorder, Arch Gen Psychiatry, 60, Brown E.B., McElroy S.L., Keck P.E., Deldar A., Adams D.H., Tohen M.and Williamson D.J. (2006), Olanzapine/Fluoxetine Combination Versus Lamotrigine in the Treatment of Bipolar I Depression, J. Clin Psychiatry, 67, Calabrese, J. R., Bowden, C. L., McElroy, S. L., Cookson, J., Andersen, J., Keck, P. E., Jr., Rhodes, L., Bolden-Watson, C., Zhou, J. and Ascher, J. A. (1999), Spectrum of activity of lamotrigine in treatment-refractory bipolar disorder, Am J Psychiatry, 156, Calabrese, J. R., Bowden, C. L., Sachs, G., Yatham, L. N., Behnke, K., Mehtonen, O. P., Montgomery, P., Ascher, J., Paska, W., Earl, N. and DeVeaugh-Geiss, J. (2003), A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently depressed patients with bipolar I disorder, J Clin Psychiatry, 64, Calabrese, J. R., Bowden, C. L., Sachs, G. S., Ascher, J. A., Monaghan, E. and Rudd, G. D. (1999), A double-blind placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. Lamictal 602 Study Group, J Clin Psychiatry, 60, Calabrese, J. R., Fatemi, S. H. and Woyshville, M. J. (1996), Antidepressant effects of lamotrigine in rapid cycling bipolar disorder, Am J Psychiatry, 153, Calabrese, J. R., Hirschfeld, R. M., Reed, M., Davies, M. A., Frye, M. A., Keck, P. E., Lewis, L., McElroy, S. L., McNulty, J. P. and Wagner, K. D. (2003), Impact of bipolar disorder on a U.S. community sample, J Clin Psychiatry, 64, Calabrese, J.R.; Huffman, R.F.; White,R.L.; Edwards,S.; Thompson,T.R.and Ascher,J.A. (2007) Additional clinical trial data and a retrospective pooled analysis of response rates across all randomized trials conducted by GSK., Bipolar Disorders 8 (suppl. 1),

41 Lamotrigine in bipolar disorder: an overview 43 Calabrese, J. R., Sullivan, J. R., Bowden, C. L., Suppes, T., Goldberg, J. F., Sachs, G. S., Shelton, M. D., Goodwin, F. K., Frye, M. A. and Kusumakar, V. (2002), Rash in multicenter trials of lamotrigine in mood disorders: clinical relevance and management, J Clin Psychiatry, 63, Calabrese, J. R., Suppes, T., Bowden, C. L., Sachs, G. S., Swann, A. C., McElroy, S. L., Kusumakar, V., Ascher, J. A., Earl, N. L., Greene, P. L. and Monaghan, E. T. (2000), A double-blind, placebo-controlled, prophylaxis study of lamotrigine in rapid-cycling bipolar disorder. Lamictal 614 Study Group, J Clin Psychiatry, 61, Calabrese, J. R., Vieta, E. and Shelton, M. D. (2003),Latest maintenance data on lamotrigine in bipolar disorder, Eur Neuropsychopharmacol, 13 Suppl 2, S Chen, C., Veronese, L. and Yin, Y. (2000),The effects of lamotrigine on the pharmacokinetics of lithium, Br J Clin Pharmacol, 50, Choi, H. and Morrell, M. J. (2003),Review of lamotrigine and its clinical applications in epilepsy, Expert Opin Pharmacother, 4, Cohen, A. F., Ashby, L., Crowley, D., Land, G., Peck, A. W. and Miller, A. A. (1985),Lamotrigine (BW430C), a potential anticonvulsant. Effects on the central nervous system in comparison with phenytoin and diazepam, Br J Clin Pharmacol, 20, Cohen, A. F., Land, G. S., Breimer, D. D., Yuen, W. C., Winton, C. and Peck, A. W. (1987), Lamotrigine, a new anticonvulsant: pharmacokinetics in normal humans, Clin Pharmacol Ther, 42, Cunnington, M, Tennis P. and the International Lamotrigine Pregnancy Registry Scientific Advisory Committee (2005), Lamotrigine and the risk of malformations in pregnancy, Neurology, 64, Fitton, A. and Goa, K. L. (1995),Lamotrigine. An update of its pharmacology and therapeutic use in epilepsy, Drugs, 50, Frye, M. A., Ketter, T. A., Kimbrell, T. A., Dunn, R. T., Speer, A. M., Osuch, E. A., Luckenbaugh, D. A., Cora-Ocatelli, G., Leverich, G. S. and Post, R. M. (2000),A placebocontrolled study of lamotrigine and gabapentin monotherapy in refractory mood disorders, J Clin Psychopharmacol, 20, Garnett, W. R. (1997),Lamotrigine: pharmacokinetics, J Child Neurol, 12 Suppl 1, S10-5. Geddes, J.R., Calabrese, J.R, Goodwin, G.M.(2009) Lamotrigine for treatment of bipolar depression: independent meta-analysis and meta-regression of individual patient data from

42 44 Chapter 2 five randomised trials, Br. J. of Psychiatry 194, 4-9 Gilman, J. T. (1995),Lamotrigine: an antiepileptic agent for the treatment of partial seizures, Ann Pharmacother, 29, Ginsberg, L., Sachs, G. and Ketter, T. (2003) In American Psychiatric Association 2003 Annual MeetingSan Francisco, CA, pp Goa, K.L. Ross, S. R. and Chrisp, P. (1993),Lamotrigine. A review of its pharmacological properties and clinical efficacy in epilepsy, Drugs, 46, Goldberg, J.F., Bowden, C.L., Calabrese, J.R., Ketter, T.A., Dann, R.S, Frye M.A. Suppes, T and Post R.M, (2008) Six-Month Prospective Life Charting of Mood Symptoms with Lamotrigine Monotherapy Versus Placebo in Rapid Cycling Bipolar Disorder, Biological Psychiatry, 63, Goldberg, J.F., Calabrese, J.R. Saville, B.R., Frye, M.A. Ketter, T.A. Suppes, T, Post, R.M. and F.K.Goodwin, (2009), Mood Stabilization and Destabilization During Acute and Continuation Phase Treatment for Bipolar I Disorder With Lamotrigine or Placebo, J Clin Psychiatry 70, Goldsmith, D. R., Wagstaff, A. J., Ibbotson, T. and Perry, C. M. (2003),Lamotrigine: a review of its use in bipolar disorder, Drugs, 63, Hachad, H., Ragueneau-Majlessi, I. and Levy, R. H. (2002),New antiepileptic drugs: review on drug interactions, Ther Drug Monit, 24, Hirschfeld, R. M., Calabrese, J. R., Weissman, M. M., Reed, M., Davies, M. A., Frye, M. A., Keck, P. E., Jr., Lewis, L., McElroy, S. L., McNulty, J. P. and Wagner, K. D. (2003),Screening for bipolar disorder in the community, J Clin Psychiatry, 64, Holmes, L.B., Baldwin E.J., Smith C.R., Habecker E., Glassman L., Wong S.L. and Wyszynski D.F. (2008), Increased frequency of isolated cleft palate in infants exposed to lamotrigine during pregnancy, Neurology, 70, Ichim, L., Berk, M. and Brook, S. (2000),Lamotrigine compared with lithium in mania: a double-blind randomized controlled trial, Ann Clin Psychiatry, 12, Judd, L. L., Akiskal, H. S., Schettler, P. J., Endicott, J., Maser, J., Solomon, D. A., Leon, A. C., Rice, J. A. and Keller, M. B. (2002),The long-term natural history of the weekly symptomatic status of bipolar I disorder, Arch Gen Psychiatry, 59,

43 Lamotrigine in bipolar disorder: an overview 45 Kupka, R. W., Luckenbaugh,D.A,. Post,R.M, Suppes,T., Altshuler, L.L., Keck,P.E., Jr, Frye,M.A., Denicoff,K.D., Grunze,H.,Leverich,G.S., McElroy,S.L., Walden,J. and Nolen,W.A.(2005), Comparison of rapid-cycling and non-rapid-cycling bipolar disorder based on prospective mood ratings in 539 outpatients, Am.J Psychiatry, 162, Ketter, T.A., Greist J.H., Graham, J.A., Roberts, J.N., Thompson, T.R. and Nanry, K.P., (2006), The effect of Dermatologic Precautions on the Incidence of Rash With Addition of Lamotrigine in the Treatment of Bipolar I Disorder: A Randomized Trial, J Clin. Psychiatry 67, Kusumakar, V. and Yatham, L. N. (1997),An open study of lamotrigine in refractory bipolar depression, Psychiatry Res, 72, Leach, M. J., Baxter, M. G. and Critchley, M. A. (1991),Neurochemical and behavioral aspects of lamotrigine, Epilepsia, 32 Suppl 2, S4-8. Leach, M. J., Marden, C. M. and Miller, A. A. (1986),Pharmacological studies on lamotrigine, a novel potential antiepileptic drug: II. Neurochemical studies on the mechanism of action, Epilepsia, 27, Li, X., Ketter, T. A. and Frye, M. A. (2002),Synaptic, intracellular, and neuroprotective mechanisms of anticonvulsants: are they relevant for the treatment and course of bipolar disorders?, J Affect Disord, 69, Matsuo, F., Bergen, D., Faught, E., Messenheimer, J. A., Dren, A. T., Rudd, G. D. and Lineberry, C. G. (1993),Placebo-controlled study of the efficacy and safety of lamotrigine in patients with partial seizures. U.S. Lamotrigine Protocol 0.5 Clinical Trial Group, Neurology, 43, Messenheimer, J. A. (2002) In American Academy of Neurology Annual MeetingDenver, CO. Mikati, M. A., Schachter, S. C., Schomer, D. L., Keally, M., Osborne-Shafer, P., Seaman, C. A., Sheridan, P. H., Ashworth, M., Kupferberg, H., Valakas, A. and (1989), Long-term tolerability, pharmacokinetic and preliminary efficacy study of lamotrigine in patients with resistant partial seizures, Clin Neuropharmacol, 12, Ohman, I., Vitols, S. and Tomson, T. (2000),Lamotrigine in pregnancy: pharmacokinetics during delivery, in the neonate, and during lactation, Epilepsia, 41, Pennell, P. B. (2003),Antiepileptic drug pharmacokinetics during pregnancy and lactation, Neurology, 61, S35-42.

44 46 Chapter 2 Post, R. M., Denicoff, K. D., Leverich, G. S. and al., e. (2002),Presentations of depression in bipolar illness, Clin Neurosci Res, 2, Rambeck, B. and Wolf, P. (1993),Lamotrigine clinical pharmacokinetics, Clin Pharmacokinet, 25, Ramsay, R. E., Pellock, J. M., Garnett, W. R., Sanchez, R. M., Valakas, A. M., Wargin, W. A., Lai, A. A., Hubbell, J., Chern, W. H., Allsup, T. and (1991), Pharmacokinetics and safety of lamotrigine (Lamictal) in patients with epilepsy, Epilepsy Res, 10, Regeer, E. J., ten,have M., Rosso,M.L., Hakkaart-van,Roijen L., Vollebergh,W. and Nolen W.A.(2004), Prevalence of bipolar disorder in the general population: a Reappraisal Study of the Netherlands Mental Health Survey and Incidence Study. Acta Psychiatr. Scand. 110, Rzany B., Mockenhaupt, M., Baur, S., Schroder, W., Stocker, U., Mueller, J.,Hollander, N., Bruppacher, R. and Schopf, E. (1996),Epidemiology of erythema exsudativum multiforme majus, Stevens-Johnson syndrome and toxic epidermal necrolysis in Germany ( ): structure and results of a population-based registry, J Clin Epidemiol, 49, Shiah, I. S., Yatham, L. N., Lam, R. W. and Zis, A. P. (1998),Effects of lamotrigine on the 5-HT1A receptor function in healthy human males, J Affect Disord, 49, Sidhu J., Job S., Singh S. and Philipson R., The pharmacokinetic and pharmacodynamic consequences of the co-administration of lamotrigine and a combined oral contraceptive in healthy female subjects. Br. J. of Clin. Pharmacol. 61, Smith, D., Baker, G., Davies, G., Dewey, M. and Chadwick, D. W. (1993),Outcomes of add-on treatment with lamotrigine in partial epilepsy, Epilepsia, 34, Smith, D., Chadwick, D., Baker, G., Davis, G. and Dewey, M. (1993),Seizure severity and the quality of life, Epilepsia, 34 Suppl 5, S31-5. Tran, T. A., Leppik, I. E., Blesi, K., Sathanandan, S. T. and Remmel, R. (2002),Lamotrigine clearance during pregnancy, Neurology, 59, Vinod, K. Y. and Subhash, M. N. (2002),Lamotrigine induced selective changes in 5-HT(1A) receptor mediated response in rat brain, Neurochem Int, 40, Walden, J., Hesslinger, B., van Calker, D. and Berger, M. (1996),Addition of lamotrigine to valproate may enhance efficacy in the treatment of bipolar affective disorder, Pharmacopsychiatry, 29,

45 Lamotrigine in bipolar disorder: an overview 47 Weisler, R., Risner, M., Ascher, J. and Houser, T. (1994) In American Psychiatric Association 1994 Annual MeetingPhiladelphia, PA. Weissman, M. M., Bland, R. C., Canino, G. J., Faravelli, C., Greenwald, S., Hwu, H. G., Joyce, P. R., Karam, E. G., Lee, C. K., Lellouch, J., Lepine, J. P., Newman, S. C., Rubio- Stipec, M., Wells, J. E., Wickramaratne, P. J., Wittchen, H. and Yeh, E. K. (1996),Crossnational epidemiology of major depression and bipolar disorder, Jama, 276, Wong, I. C., Mawer, G. E. and Sander, J. W. (1999),Factors influencing the incidence of lamotrigine-related skin rash, Ann Pharmacother, 33, Xie, X. and Hagan, R. M. (1998),Cellular and molecular actions of lamotrigine: Possible mechanisms of efficacy in bipolar disorder, Neuropsychobiology, 38, Xie, X., Lancaster, B., Peakman, T. and Garthwaite, J. (1995),Interaction of the antiepileptic drug lamotrigine with recombinant rat brain type IIA Na+ channels and with native Na+ channels in rat hippocampal neurones, Pflugers Arch, 430, Yuen, A. W., Land, G., Weatherley, B. C. and Peck, A. W. (1992),Sodium valproate acutely inhibits lamotrigine metabolism, Br J Clin Pharmacol, 33,

46 48

47 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial Marc L. M. van der Loos, M.D.; Paul G. H. Mulder, Ph.D.; Erwin G. Th. M. Hartong, M.D., Ph.D.; Marc B. J. Blom, M.D.; Anton C. Vergouwen, M.D., Ph.D.; Herman J. U. E. M. de Keyzer, M.D.; Peter J. H. Notten, M.D.; Marijke L. Luteijn, M.D.; Manuela A. Timmermans, M.Sc.; Eduard Vieta, M.D., Ph.D.; and Willem A. Nolen, M.D., Ph.D., for the LamLit Study Group From the Department of Psychiatry, Isala Klinieken, location Sophia, Zwolle (Dr. van der Loos); the Department of Epidemiology & Biostatistics, Erasmus University Medical Center, Rotterdam (Dr. Mulder); the Department of Psychiatry, Canisius- Wilhelmina Hospital, Nijmegen (Dr. Hartong); Parnassia Psychiatric Institute, The Hague (Dr. Blom); the Department of Psychiatry, St. Lucas Andreas Hospital, Amsterdam (Dr. Vergouwen); Group Practice: Biological Psychiatry and Psychotherapy, Retranchement (Dr. de Keyzer); the Department of Psychiatry, St. Elisabeth Hospital, Tilburg (Dr. Notten); the Department of Psychiatry, Albert Schweitzer Hospital, Dordrecht (Dr. Luteijn); Clinical Operations, GlaxoSmithKline, Zeist (Ms. Timmermans); Bipolar Disorders Program, Department of Psychiatry, University Medical Center Groningen, University of Groningen (Dr. Nolen), the Netherlands; and Institute of Neuroscience, Hospital Clinic, University of Barcelona, IDIBAPS, CIBER-SAM, Catalonia, Spain (Dr. Vieta). The Journal of Clinical Psychiatry 2009; 70:

48 50 Chapter 3 Abstract Objective: Lamotrigine is one of the pharmacologic options for the treatment of bipolar depression but has only been studied as monotherapy. This study compared the acute effects of lamotrigine and placebo as add-on therapy to ongoing treatment with lithium in patients with bipolar depression. Method: Outpatients (n = 124) aged 18 years and older with a DSM- IV bipolar I or II disorder and a major depressive episode (Montgomery-Åsberg Depression Rating Scale [MADRS] score 18 and Clinical Global Impressions- Bipolar Version [CGI-BP] severity of depression score 4) while receiving lithium treatment ( mmol/l) were randomly assigned to 8 weeks of doubleblind treatment with lamotrigine (titrated to 200 mg/d) or placebo. The primary outcome measure was mean change from baseline in total score on the MADRS at week 8. Secondary outcome measures were response (defined as a reduction of 50% on the MADRS and/or change of depression score on the CGI-BP of much improved or very much improved compared to baseline) and switch to mania or hypomania (defined as a CGI-BP severity of mania score of at least mildly ill at any visit). Patients were included in the study between August 2002 (Spain started in October 2003) and May Results: Endpoint mean change from baseline MADRS total score was (SE = 1.32) points for lamotrigine and (SE = 1.36) points for placebo (t = 2.29, df = 104, p =.024). Significantly more patients responded to lamotrigine than to placebo on the MADRS (51.6% vs. 31.7%, p =.030), but not on the CGI-BP change of depression (64.1% vs. 49.2%, p =.105). Switch to mania or hypomania occurred in 5 patients (7.8%) receiving lamotrigine and 2 patients 3.3%) receiving placebo (p =.441). Conclusion: Lamotrigine was found effective and safe as add-on treatment to lithium in the acute treatment of bipolar depression.

49 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial 51 Introduction Patients with bipolar disorder spend more time depressed than manic or hypomanic (1 4), while most mood stabilizers (lithium, valproate, carbamazepine) are more effective against manic episodes than depressive episodes (5 7). The same appears true for atypical antipsychotics, although olanzapine (alone as well as in combination with fluoxetine (8)) and quetiapine (9) have demonstrated efficacy in the acute treatment of bipolar depression. As a consequence, many bipolar patients also receive antidepressants, mostly in combination with other therapies. The finding that the effect size for the combination of olanzapine with fluoxetine was superior to the effect size of olanzapine monotherapy underscores the importance of combination therapy studies(10). While some studies found that antidepressants, either as monotherapy or combination therapy with mood stabilizers, are effective in the treatment of acute bipolar depression (11), other studies found no effect of the addition of antidepressants to mood stabilizers (12,13). In addition, their use may be associated with the risk of switch to mania or hypomania (11,14) The anticonvulsant lamotrigine provides an additional treatment option. After a few case reports and some open label studies (15 17), the first doubleblind randomized controlled trial (18) on the efficacy of lamotrigine monotherapy in the acute treatment of bipolar depression found it more effective than placebo at doses of 50 and 200 mg/day on several secondary outcome measures, including the Montgomery-Åsberg Depression Rating Scale (MADRS), but not on the primary outcome measure, i.e., the Hamilton Rating Scale for Depression (HAM-D). Subsequently, its efficacy in the acute treatment of bipolar depression was studied in 4 other randomized controlled trials. Although those studies were all negative on the primary outcome criterion (19), a meta-analysis of the 5 studies showed a small but significant positive result (20,21). In another study, lamotrigine was slightly less effective but better tolerated than a combination of olanzapine and fluoxetine (22). Collaborative results were also obtained from a crossover study that compared lamotrigine and gabapentin to placebo (23). In 2 long-term studies of bipolar I patients with a current (or recent) manic or 3

50 52 Chapter 3 depressive episode (24,25), lamotrigine (titrated up to mg/day) was added to ongoing treatment with other psychotropic drugs over 8 to 16 weeks. Concomitant drugs were gradually withdrawn, after which stabilized patients were randomly assigned to continuation of lamotrigine or to lithium or placebo (each as monotherapy) for up to 18 months. In both studies, lamotrigine and lithium were statistically more effective than placebo on the primary outcome measure (time from randomization to intervention for any mood episode), but lithium was predominantly effective against manic episodes and lamotrigine predominantly against depressive episodes, suggesting that lithium and lamotrigine have differential and potentially complementary effects. In this article, we report the results of the first placebo controlled study addressing the efficacy and safety of lamotrigine as add-on to lithium in the acute treatment of patients with bipolar depression. Method Study Design This study is the first phase of an investigator-initiated (W.A.N.) randomized, double-blind, placebo-controlled trial, with 23 recruiting centers (of 25 selected) in the Netherlands and 3 recruiting centers (of 5 selected) in Spain. Recruitment took place between August 2002 and May 2005 (Spain started in October 2003). The study was approved by the ethical review board of the University Medical Center Utrecht, the Netherlands, and by local institutional review boards in both countries. All patients gave written informed consent prior to initiation of any study procedure. Patients Outpatients (men or women) aged 18 years or older could be included if they met criteria for DSM-IV bipolar I or II disorder, current major depressive episode confirmed by the Mini-International Neuropsychiatric Interview Plus (MINI-Plus)(26) with a score of 18 points on the MADRS (27) and a score of 4 (moderately ill) on the Clinical Global Impressions-Bipolar Version (CGI-BP)

51 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial 53 severity of depression rating (28). All patients had to be receiving treatment with lithium with a stable dose (plasma level, mmol/l) during at least 2 weeks prior to the study. Exclusion criteria were the following: the presence of psychotic features, a severe rapid cycling course with 10 episodes over the last 12 months (rapid cycling with 4 9 episodes in the previous 12 months was allowed), severe suicidality (score of 5 on item 10 of the MADRS), a history of alcohol or substance abuse within 1 month or dependence within 12 months of enrollment, a severe personality disorder suggesting noncompliance, or a severe neurologic or other somatic illness and the use of somatic medication that could affect mood. Women of child-bearing potential were only eligible for the study if they had a negative pregnancy test at screening and agreed to utilize an effective contraceptive method. Except for lithium, patients were not allowed to have used an antipsychotic or an antidepressant within 2 weeks (fluoxetine, 4 weeks) of randomization. Benzodiazepines were allowed at a maximum of 2-mg lorazepam equivalents per day throughout the study, i.e., for standing as well as rescue medication, for anxiety, agitation, or sleep problems. 3 Treatments After inclusion in the study, patients were randomly assigned 1:1 to either lamotrigine or placebo in blocks of 2 as adjunctive treatment to ongoing treatment with lithium according to a randomization list. Stratification was done by site. The code of each patient was kept in a sealed envelope at the pharmacy of each participating center and could be opened by the pharmacist only in case of medical emergency. The study blind was preserved until after the last patient evaluation. Study medication was administered once daily in the morning. Lamotrigine was started at 25 mg/day in weeks 1 and 2 and was increased to 50 mg/day in weeks 3 and 4, to 100 mg/day in weeks 5 and 6, and, finally, to 200 mg/ day in weeks 7 and 8. Placebo was administered by utilizing tablets identical in appearance and number. The lithium dosage remained stable during the study with monitoring of plasma levels (to be kept at mmol/l) at

52 54 Chapter 3 baseline and week 8. To establish medication compliance, a pill count was done at every visit. Patients returning more than 30% of their prescribed medication or reporting an interruption of medication for more than 3 days were withdrawn from the study. Patients who missed their study medication for 1, 2, or (maximum) 3 days restarted at the same dose. Outcome Measures At baseline, diagnosis was confirmed with the MINI-Plus sections depressive episode, manic episode, hypomanic episode, alcohol abuse and dependence, and substance abuse and dependence. In addition, clinicians completed the Clinical Questionnaire for Bipolar Illness (CQBP-C), with items on illness and treatment history, as used by the former Stanley Foundation Bipolar Network (29). Severity of symptoms was assessed at baseline and at weeks 2, 4, 6, and 8 by using the MADRS and the CGI-BP severity of depression and severity of mania. Adverse events were assessed at each visit by inquiring about any unpleasant feeling since the last visit and, if present, rated mild, moderate, or severe. The protocol as well as the case record form were written in English but included validated Dutch and Spanish translations of the MADRS and the CGI-BP. In the Netherlands, there were 4 research meetings, including MADRS training sessions; 27 raters scored at least 4 patients with an intraclass coefficient = After receiving an initial MADRS training in Spain, all Spanish researchers participated in a research meeting in the Netherlands with additional MADRS training. The primary outcome measure was the mean change from baseline in total MADRS score at week 8. Secondary outcome measures were response (defined as a reduction of 50% on the MADRS and/or CGI-BP change of depression score of much improved or very much improved compared to baseline) and switch into mania or hypomania (defined as a CGI-BP severity of mania score of at least mildly ill at any visit). Post hoc, we also analyzed on

53 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial 55 which items of the MADRS there were significant differences between lamotrigine and placebo at week 8. Switch to mania or hypomania is an undesirable outcome in the treatment of bipolar depression but can occur following recovery from depression. However, it often remains unclear in publications whether patients who switched were also counted as responders. Therefore, in a post hoc analysis, we assessed whether patients reached the criteria for both response (MADRS and/or CGI-BP change of depression) and no switch to mania or hypomania (CGI-BP), which is the ultimate goal in the treatment of bipolar depression. The study was initially aimed at comparing 2 groups of 110 patients in order to detect a mean difference of 4 points in MADRS scores between the 2 groups, with 80% power and α of.05 (2-sided). The standard deviation (SD) of the MADRS score was initially set at 11 points. Because of difficulties in recruiting the planned number of patients, the sample size was recalculated after inclusion of a total of 43 patients. The MADRS scores based on these actually observed patients, who were evaluated blindly to outcome, resulted in an SD of 8 points and a corresponding recalculated sample size of 60 patients per group. 3 Analysis Mixed-model analysis of variance (ANOVA) for repeated measures was prospectively chosen as method of analysis for MADRS scores. The independent variables in this analysis were time and treatment and their interaction. Time as categorical variable allowed estimating and testing mean differences in MADRS score between the 2 treatments at any visit, including the primary efficacy measure: score change from baseline at week 8. For this purpose the relevant contrasts were specified in the model for which the ANOVA produced the estimates, SEs, confidence intervals (CIs), and appropriate t tests per visit. In order to make inference about an overall treatment effect over all visits, time was considered a numeric trend variable in the model. This allowed estimating and testing a downward trend in time of the mean MADRS score in either treatment

54 56 Chapter 3 group and the difference in time trend between the 2 treatments as a single efficacy measure, for which the mixed-model ANOVA produced the estimate SE, CI, and appropriate t test. In accordance with the intent-to-treat principle, all randomly assigned subjects contributed to these analyses, missing values of MADRS being properly adjusted for by the restricted maximum likelihood estimation procedure used in the mixed-model ANOVAs. Clinical Global Impression-Bipolar Version severity scores were similarly analyzed by using mixed-model ANOVA, with time as categorical variable. Dichotomous outcome scores (response, switch, or presence of adverse effects) were made complete by using the last-observation-carried-forward method and then were analyzed by comparing percentages between the 2 treatment groups using Fisher exact test. Differences in the distribution of categorical nominal variables between the 2 treatment groups were tested using Fisher exact test; for categorical ordinal variables the exact χ2 trend test was used. For each test, a 2-sided p value below.05 was considered to denote statistical significance. Results Patients and Illness Characteristics In total, 128 patients were recruited (fig 1): most patients (n = 82) from the 23 centers in the Netherlands, 34 patients via advertisements in Dutch newspapers, and 12 patients from the 3 centers in Spain. The range of included patients per center was between 1 and 25. Four patients were not included in the analysis: at baseline, 1 patient appeared to use an antipsychotic; 1 patient had used an antidepressant in the 2 weeks before randomization; 1 patient had hypothyroidism; and 1 patient never took any medication. Thus, 124 patients were randomly assigned and included in the analysis: 64 patients to lithium plus lamotrigine and 60 patients to lithium plus placebo. There were no statistically significant differences between the 2 groups in patient characteristics and illness severity at baseline (table 1).

55 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial 57 Figure 1. Flow chart of patients randomly assigned to lamotrogine or placebo as add-on treatment to lithium Assessed for Eligibility (n = 128) Randomly assigned (n = 124) Excluded (n = 4) Not meeting inclusion Criteria (n = 3) Not Taking Study Medication (n = 1) Lamotrigine (n = 64) Placebo (n = 60) Lost to Follow-Up (n = 0) Discontinued (n = 12) Lack of Efficacy (n = 2) Adverse Effects (n = 4) Withdrawn Consent (n = 1) Protocol Violation (n = 2) Noncompliance (n = 1) Other (n = 2) Completed (n = 52) Analyzed (n = 64) Lost to Follow-Up (n = 1) Discontinued (n = 9) Lack of Efficacy (n = 3) Adverse Effects (n = 2) Withdrawn Consent (n = 1) Protocol Violation (n = 0) Noncompliance (n = 1) Other (n = 2) Completed (n = 50) Analyzed (n = 60) 3

56 58 Chapter 3 Table 1. Baseline patient and illness characteristics in patients randomly assigned to lamotrogine or placebo as add-on treatment to lithium Lamotrigine n=64 Placebo n=60 Total n=124 Statistic Value df P Female gender (n, %) Fisher s Age (years) (mean, SD) t-test Ilness characteristics Bipolar I disorder (n, %) Fisher s Rapid cycling course last 12 months (n, %) Fisher s MADRS (mean, SD) t-test CGI-BP depression severity (mean, SD) CGI-BP mania severity (mean, SD) Previous treatments for index depressive episode Lithium Baseline plasma level (meq) (mean, SD) Duration of treatment before randomisation Chi Sq. trend Fisher s t-test to 4 weeks (n, %) Fisher s - - N.S. 1-3 months (n, %) Fisher s - - N.S 3 months (n, %) Fisher s - - N.S. Unknown/missing (n, %) TCA (n, %) Fisher s SSRI/SNRI (n, %) 15 23, Fisher s MAO-inhibitor (n, %) Fisher s Antipsychotic, typical (n, %) Antipsychotic, atypical (n, %) Fisher s Fisher s Valproic acid (n, %) Fisher s Carbamazepine (n, %) Fisher s Benzodiazepine (n, %) Fisher s Other (n, %) Fisher s CGI-BP = Clinical Global Impressions-Bipolar Version, MADRS = Montgomery-Åsberg Despression Rating Scale, MAOI = monoamine oxidase inhibitor, SSRI = selective serotonin reuptake inhibitor, SNRI = serotonin-norepinephrine reuptake inhibitor, TCA = tricyclic antidepressant.

57 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial 59 In both groups, more than 80% of the patients had been receiving lithium maintenance treatment for at least 3 months. At baseline, lithium plasma levels were within the predefined range for 119 patients ( mmol/l), while 5 patients had marginally lower plasma levels (all above 0.53 mmol/l). As these levels were only marginally lower than the required level of 0.6 mmol/l, a decision to include these patients in the analysis was made prior to breaking of the treatment blind. There was no significant difference between lithium plasma levels in the placebo group versus the lamotrigine group at baseline (t = 0.48, df = 122, p =.63; 0.84 [SD = 0.16] mmol/l and 0.82 [SD = 0.16] mmol/l, respectively) nor at 8 weeks (t = 1.36, df = 100, p =.18; 0.80 [SD = 0.14] mmol/l and 0.76 [SD = 0.16] mmol/l, respectively). In the lamotrigine group, 52 patients (81%) completed the 8-week study period versus 50 patients (83%) in the placebo group. Of the 12 patients who discontinued the trial in the lamotrigine group, 2 discontinued because of lack of efficacy and 4 because of adverse effects. In the placebo group, 10 patients discontinued the trial: 3 because of lack of efficacy and 2 because of adverse effects. Other reasons in both groups for discontinuing were withdrawn consent, protocol violation, noncompliance, and lost to follow-up. 3 Efficacy Results are presented in Tables 2 and 3. On the primary outcome measure, change in the MADRS score from baseline to week 8, lamotrigine was significantly more effective than placebo (decrease of vs , respectively, p =.024) (table 2). The lamotrigine group also had significantly lower MADRS scores at weeks 6 and 8 compared to the placebo group (Figure 2). Starting from an overall average baseline level of 28.4 points, the MADRS score decreased on average by 3.77 points per 2 weeks (95% CI = 3.15 to 4.38; p <.0005) during the first 8 weeks of treatment in the lamotrigine group. In the placebo group, the decrement was on average 2.59 points per 2 weeks (95% CI = 1.96 to 3.22; p <.0005). Hence, the mean MADRS score in the lamotrigine group became increasingly lower than the mean score in the placebo

58 60 Chapter 3 group by, on average, 1.17 points per 2 weeks (95% CI = 0.33 to 2.02; p =.007) during the first 8 weeks of treatment. Response, defined as a reduction of the MADRS score of 50%, was statistically greater (p =.03) in the lamotrigine group (51.6%) than in the placebo group (31.7%) (table 3). Response according to CGI-BP change of depression scores 2 and response according to either criterion (MADRS and/or CGI-BP change of depression) showed no significant differences. Table 2. Efficacy Results (mixed model estimates): severity scales Lamotrigine (n = 64) Placebo (n = 60) Difference Measure Mean SE Mean SE Mean SE MADRS score change from baseline (primary outcome measure) t Test df P Week Week Week Week MADRS score at each visit Baseline Week Week Week Week CGI-BP severity of depression score at each visit Baseline Week Week Week Week CGI-BP change of depression score from baseline Week Week Week Week CGI-BP severity of mania score at each visit Baseline Week Week Week Week CGI-BP = Clinical Global Impressions-Bipolar Version, MADRS = Montgomery-Åsberg Depression Rating Scale. Bold values indicate significant levels < 0.05

59 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial 61 Because 12 patients (18.8%) receiving lamotrigine versus 4 patients (6.7%) in the placebo group had been rapid cyclers in the past year (difference approaching statistical significance [p =.06, Fisher exact test]), this subgroup was the subject of a post hoc analysis, revealing 8 (66.7%) lamotrigine and 1 (25%) placebo responders by MADRS and/or CGI-BP change of depression (p =.26, Fisher exact test). Four patients (25%) of the 16 rapid cycling patients (all 4 receiving lamotrigine) had a manic or hypomanic episode versus 3 patients (2.8%) of 108 non rapid cycling patients (1 patient receiving lamotrigine, 2 patients receiving placebo). Over both treatment conditions, the switch rate to mania or hypomania was significantly higher in the rapid cycling versus the non rapid cycling patients (p =.005, Fisher exact test). On the different MADRS items, there was a significant difference after 8 weeks between lamotrigine and placebo on apparent sadness, reported sadness, reduced appetite, lassitude, inability to feel, and suicidal thoughts. The sample size was too small to detect the following treatment-by-subgroup interactions: bipolar I versus II, rapid cycling versus non rapid cycling, and severe versus less severe depression at baseline. A site-by treatment analysis was not done as it was not foreseen in the protocol. 3

60 62 Chapter 3 Figure 2. Severity of depression according to the mean MADRS score at each visit for patients receiving placebo or lamotrigine MADRS Score, mean * ** 5 0 Lamotrigine Placebo Baseline Wk 2 Wk 4 Wk 6 Wk 8 *p =.031 **p =.006 MADRS = Montgomery-Åsberg Depression Rating Scale Table 3. Efficacy results: response and switch to (hypo)mania* Lamotrigine n=64 Placebo n=60 Response N % N % Reduction MADRS 50% CGI-BP change of depression 2 Statistic fisher ** fisher Either or both ** fisher Switch to (hypo)mania *** Response and no switch to (hypo)mania **** fisher fisher Bold values indicate significant levels < 0.05 * adjusted for gender, bipolar I/II, age, lithium and rapid cycling ** 1 missing *** CGI-BP severity of mania 3 **** Reduction MADRS 50% and/or CGI-BP change of depression 2 and CGI-BP severity of mania < 3 P

61 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial 63 Safety Five patients had a serious adverse event (SAE). One patient in the placebo group experienced a severe rash. As this was an unexpected finding, we checked the randomization code for possible mistakes and asked the laboratory to check the pills directly; both checks revealed that the medication was indeed placebo. So we decided that this was a chance finding. Four SAEs occurred in the lamotrigine group. One patient developed a manic psychotic episode requiring hospitalization. The episode resolved completely after stopping of the study medication and with additional treatment. The 3 other SAEs with lamotrigine involved 1 patient with severe hypertension that was considered to be unrelated to study drug, 1 patient with a non severe lithium intoxication (lithium level, 1.36 mmol/ L, unexplained and probably not the result of an auto intoxication), and 1 patient who had to be hospitalized due to deterioration of depressive symptoms. Study medication was continued in the patients with the lithium intoxication and the hypertension and was stopped in the other patients. Other adverse events occurring in more than 5% in either group are presented in table 4. Most of the adverse events were mild or moderate. There were no statistically significant differences between the 2 groups in occurrence of adverse events. In particular, there was no difference in the incidence of skin rash (4.7% for lamotrigine vs. 6.7% for placebo, p =.71). As mentioned before, 1 patient in the lamotrigine group developed a manic episode with psychotic features for which he was hospitalized (SAE). In addition to this patient, 4 patients receiving lamotrigine and 2 patients receiving placebo developed a hypomanic episode (CGI-BP severity of mania = 3). The number of manic or hypomanic switches was not statistically different. Combined response and no switch into mania or hypomania was obtained in 39 patients (60.9%) receiving lamotrigine and 28 patients (46.7%) receiving placebo (p =.149). 3

62 64 Chapter 3 Table 4. Adverse events occuring in 5% of patients in any group Discussion Lamotrigine (n = 64) Placebo (n = 60) Adverse Event N % N % P Headache Fatigue Nausea Flu-like symptoms Insomnia Tremor Skin problems/mild rash Dizziness Abdominal pain Rash Joint/muscle pain Back pain Agitation To our knowledge this is the first randomized con trolled trial to assess the efficacy of lamotrigine as add-on treatment to ongoing lithium therapy in patients with bipolar depression. The results show a clinically significant (> 4 points) difference in favor of lamotrigine on the primary outcome measure (change from baseline in MADRS score) and on several secondary outcome measures, including MADRS scores at weeks 6 and 8, the percentage of MADRS responders at endpoint, and 6 of 10 items on the MADRS, including core depressive symptoms as apparent sadness, reported sadness, inability to feel, and suicidal thoughts. Results obtained with the CGI-BP change of depression were equivocal, with weekly scores and overall response rates numerically favoring lamotrigine but not reaching statistical significance. This could be due to limitations in the sensitivity of the CGI-BP. Another possibility is that this resulted from a more optimistic impression among the physicians when assessing the effect of treatment since start of treatment, compared with the cross-sectionally

63 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial 65 assessed MADRS. However, previous lamotrigine studies did show significant change in CGI ratings of depressed bipolar patients (23 25). This study indicates efficacy for lamotrigine in the acute treatment of bipolar depression, which complements similar findings in 2 previous studies with lamotrigine monotherapy (17,23). However, 4 subsequent placebo controlled trials with lamotrigine monotherapy were all negative, possibly because of high placebo response (35% 47% as assessed by MADRS or HAM-D-17) or lack of intrinsic efficacy (20), while in another study lamotrigine was less effective but better tolerated than a combination of olanzapine and fluoxetine (22). Besides the possibility of a chance effect, there are several possible explanations for our positive finding compared with previous, mostly negative studies. First, this was an add-on study in which all patients received lithium, a treatment that appears to have some efficacy on its own in the treatment of bipolar depression (30). The combination of lithium and lamotrigine may have produced additive effects, or the efficacy of lamotrigine may have been enhanced by concurrent lithium treatment. Although this might be speculative, the potentially synergic action could come as a result of common final therapeutic pathways involving presynaptic serotonin release and increased neuroplasticity (31,32). A pharmacokinetic interaction seems unlikely in view of a previously conducted drug interaction study (33). On the other hand, the add-on design could have reduced the chance of positive findings, since, as a general rule, this type of trial is less likely to be successful in bipolar disorder (34). Second, the relatively low overall dropout rate (17.7%) in the current study permitted a robust efficacy assessment due to fewer missing data. Thus, it affirms the utility of add-on study designs in this patient population. The low dropout rate may be attributable to various reasons. First, all patients had the option to get an antidepressant during the second phase of the study, if they should not respond to the study drug at the end of the double-blind trial. Second, there is a difference in health care systems in the countries in which our study was conducted compared to the health care systems in which previous lamotrigine studies (21) were conducted. In the Netherlands and Spain all inhabitants have a mandatory health insurance program that pays for the treatment of bipolar 3

64 66 Chapter 3 disorder, whereas this is not the case in the United States, where all the negative acute trials were conducted. This means that most patients could stay with their own physician, even during their participation in the trial. The response to lamotrigine occurred rather late. In 2 recent trials comparing quetiapine with placebo in bipolar depression and in the trial with the combination of olanzapine plus fluoxetine versus placebo, there was a separation between medication and placebo after 1 week (8,9). In our study, this separation occurred at 6 weeks, which is comparable with the separation at 5 weeks in the prior positive study with lamotrigine monotherapy (18) This delayed effect of lamotrigine might be due to the slow titration of lamotrigine. Switch to mania or hypomania occurred in 7 patients: in 5 patients who were taking lamotrigine and in 2 patients taking placebo. Changing mood states form the core feature of bipolar disorder; placebo response is always a factor to consider, especially in rapid cycling patients who typically have frequent episodes of relatively short duration. Although severe rapid cycling (more than 10 episodes in the last year) was an exclusion criterion, nearly 13% of enrolled patients met DSM-IV criteria for rapid cycling (4 9 episodes), which is comparable with the overall frequency (15%) of rapid cycling reported in previous studies (3). In the quetiapine versus placebo study (9), 18.2% of the patients were rapid cyclers. In the olanzapine (plus fluoxetine) versus placebo study (8), more than 35% of the patients were rapid cyclers. Although the rapid cycling patients showed more switches than the non-rapid cycling patients, we did not detect a difference in efficacy, unlike 2 recent studies (35,36), which indicated a greater short-term response to olanzapine or clozapine in rapid cycling versus non-rapid cycling patients. However, detection of any such difference may have been hampered by lack of statistical power for this subgroup and confounded by a near-significant difference in the incidence of rapid cycling between the lamotrigine and placebo groups. In this study, the combination of lithium and lamotrigine was well tolerated. There were relatively few serious adverse events for this population and other adverse events were all mild or moderate and did not differ in frequency from placebo. The relatively low rate of skin rash observed in this

65 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial 67 study may be attributable to slow titration of lamotrigine in accordance with the manufacturer s recommendations and confirms that lithium does not increase the risk of lamotrigine-induced rash. The most important limitation of our study is the relatively small overall sample size (smaller than initially planned), which did not permit subgroup analyses for assessment of response predictors and moderators. Another limitation is that we have no reliable data regarding the number of bipolar patients in the study centers who developed a depressive episode while being treated with lithium and who were not selected for the study. Finally, we have no sufficient data on patients recent treatment histories to suggest where the combination of lithium and lamotrigine might fit within treatment algorithms. Because many bipolar depressed patients do not respond adequately to monotherapy, there is an urgent need for studies addressing the efficacy of other treatment regimens, including combinations with mood stabilizers. Currently published placebo-controlled studies have only addressed the combination of an antidepressant (paroxetine or imipramine) with lithium (13), of olanzapine plus fluoxetine (8) and of the addition of an antidepressant to various mood stabilizers and/or atypical antipsychotics (12). Our study is the first placebo-controlled study that addressed the combination of lamotrigine and lithium in bipolar depression. This combination has been considered interesting, especially since in the 2 long-term studies comparing both lamotrigine and lithium with placebo, lithium was especially effective in the prevention of manic episodes and lamotrigine especially effective in the prevention of depressive episodes (24,25). Regarding clinical consequences, several questions remain about the place of the combination of lamotrigine and lithium in the treatment of bipolar depression. Antidepressants remain another possible option in the treatment of bipolar depression in patients using lithium (8,11,37 39). However, their use may be associated with the risk of a switch to mania or hypomania (8,11,37 39). Further studies are needed to compare these and other options, such as (combinations with) atypical antipsychotics. Another question is whether the effects of lamotrigine and lithium will be maintained in the responders; this will be addressed in the not yet analyzed, 1-year continuation phase of this study. 3

66 68 Chapter 3 References 1. Judd LL, Akiskal HS, Schettler PJ, et al. A prospective investigation of the natural history of the long-term weekly symptomatic status of bipolar II disorder. Arch Gen Psychiatry 2003;60: Judd LL, Akiskal HS, Schettler PJ, et al. The long-term natural history of the weekly symptomatic status of bipolar I disorder. Arch Gen Psychiatry 2002;59: Kupka RW, Luckenbaugh DA, Post RM, et al. Comparison of rapid- cycling and nonrapid-cycling bipolar disorder based on prospective mood ratings in 539 outpatients. Am J Psychiatry 2005;162: Calabrese JR, Hirschfeld RMA, Frye MA, et al. Impact of depressive symptoms compared with manic symptoms in bipolar disorder: results of a US community-based sample. J Clin Psychiatry 2004;65(11) Ernst CL, Goldberg JF. Antidepressant properties of anticonvulsant drugs for bipolar disorder. J Clin Psychopharmacol 2003;23: Malhi GS, Mitchell PB, Salim S. Bipolar depression: management options. CNS Drugs 2003;17: Geddes JR, Burgess S, Hawton K, et al. Long-term lithium therapy for bipolar disorder: systematic review and meta-analysis of randomized controlled trials. Am J Psychiatry 2004;161: Tohen M, Vieta E, Calabrese J, et al. Efficacy of olanzapine and olanzapinefluoxetine combination in the treatment of bipolar I depression. Arch Gen Psychiatry 2003;60: Calabrese JR, Keck PE Jr., Macfadden W, et al. A randomized, double blind, placebo-controlled trial of quetiapine in the treatment of bipolar I or II depression. Am J Psychiatry 2005 Jul;16(7)2: Gao K, Gajwani P, Elhaj O, et al. Typical and atypical antipsychotics in bipolar depression. J Clin Psychiatry 2005;66(11): Gijsman HJ, Geddes JR, Rendell JM, et al. Antidepressants for bipolar depression: a systematic review of randomized, controlled trials. Am J Psychiatry 2004;161:

67 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial Sachs GS, Nierenberg AA, Calabrese JR, et al. Effectiveness of adjunctive antidepressant treatment for bipolar depression. N Engl J Med 2007;356: Nemeroff CB, Evans DL, Gyulai L, et al. Double-blind, placebo- controlled comparison of imipramine and paroxetine in the treatment of bipolar depression. Am J Psychiatry 2001;158: Altshuler LL, Post RM, Leverich GS, et al. Antidepressant-induced mania and cycle acceleration: a controversy revisited. Am J Psychiatry 1995;152: Sporn J, Sachs G. The anticonvulsant lamotrigine in treatment-resistant manicdepressive illness. J Clin Psychopharmacol 1997;17: Kusumakar V, Yatham LN. An open study of lamotrigine in refractory bipolar depression. Psychiatry Res 1997;72: Calabrese JR, Fatemi SH, Woyshville MJ. Antidepressant effects of lamotrigine in rapid cycling bipolar disorder. Am J Psychiatry 1996; 153(9): Calabrese JR, Bowden CL, Sachs GS, et al. A double-blind placebo- controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. Lamictal 602Study Group. J Clin Psychiatry 1999;60(2): Goldsmith DR, Wagstaff AJ, Ibbotson T, et al. Lamotrigine: a review of its use in bipolar disorder. Drugs 2003;63: Calabrese JR, Huffman RF, White RL, et al. Lamotrigine in the acute treatment of bipolar depression: results of five double-blind, placebo- controlled clinical trials. Bipolar Disord 2008;10: Geddes J, Huffman RF, Paska W, et al. Lamotrigine for acute treatment of bipolar depression: additional clinical trial data and a retrospective pooled analysis of response rates across all randomized trials conducted by GSK. Bipolar Disord 2007;8(suppl 1): Brown EB, McElroy SL, Keck PE, Jr., et al. A 7-week, randomized, double-blind trial of olanzapine/fluoxetine combination versus lamotrigine in the treatment of bipolar I depression. J Clin Psychiatry 2006;67(7): Frye MA, Ketter TA, Kimbrell TA, et al. A placebo-controlled study of lamotrigine and gabapentin monotherapy in refractory mood disorders. J Clin Psychopharmacol 2000;20:

68 70 Chapter Calabrese JR, Bowden CL, Sachs G, et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently depressed patients with bipolar I disorder. J Clin Psychiatry 2003;64(9): Bowden CL, Calabrese JR, Sachs G, et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently manic or hypomanic patients with bipolar I disorder. Arch Gen Psychiatry 2003;60: Sheehan DV, Lecrubier Y, Sheehan KH, et al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59(suppl 20): Montgomery SA, Åsberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979;134: Spearing MK, Post RM, Leverich GS, et al. Modification of the clinical global impressions (CGI) scale for use in bipolar illness (BP): the CGI-BP. Psychiatry Research 1997;73: Leverich GS, Nolen WA, Rush AJ, et al. The Stanley Foundation Bipolar Treatment Outcome Network. I. Longitudinal methodology. J Affect Disord 2001;67(1 3): Bhagwagar Z, Goodwin GM. The role of lithium in the treatment of bipolar depression. Clin Neurosci Res 2002;2: Consoni FT, Vital MA, Andreatini R. Dual monoamine modulation for the antidepressant-like effect of lamotrigine in the modified forced swimming test. Eur Neuropsychopharmacol 2006;16: Corbella B, Vieta E. Molecular targets of lithium action. Acta Neuropsychiatrica 2003;15: Chen C, Veronese L, Yin Y. The effects of lamotrigine on the pharmacokinetics of lithium. Br J Clin Pharmacol 2000;50: Vieta E, Carne X. The use of placebo in clinical trials on bipolar disorder: a new approach for an old debate. Psychother Psychosom 2005;74: Vieta E, Calabrese JR, Hennen J, et al. Comparison of rapid-cycling and non-rapidcycling bipolar I manic patients during treatment with olanzapine: analysis of pooled data. J Clin Psychiatry 2004(10);65:

69 Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial Suppes T, Ozcan ME, Carmody T. Response to clozapine of rapid cycling versus noncycling patients with a history of mania. Bipolar Disord 2004;6: Post RM, Altshuler LL, Leverich GS, et al. Mood switch in bipolar depression: comparison of adjunctive venlafaxine, bupropion and sertraline. Br J Psychiatry 2006;189: Vieta E, Martinez-Aran A, Goikolea JM, et al. A randomized trial comparing paroxetine and venlafaxine in the treatment of bipolar depressed patients taking mood stabilizers. J Clin Psychiatry 2002;63(6): Leverich GS, Altshuler LL, Frye MA, et al. Risk of switch in mood polarity to hypomania or mania in patients with bipolar depression during acute and continuation trials of venlafaxine, sertraline, and bupropion as adjuncts to mood stabilizers. Am J Psychiatry 2006;163:

70 72

71 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine M. L. M. van der Loos 1, P. Mulder 2, E. G. Th. M. Hartong 3, M. B. J. Blom 4, A. C. Vergouwen 5, M. S. van Noorden 6, M. A. Timmermans 7, E. Vieta 8, W. A. Nolen 9, for the LamLit Study Group. 1 Department of Psychiatry, Isala Klinieken, Location Sophia, Zwolle, 2 Department of Biostatistics, Erasmus University Medical Center, Rotterdam, 3 Department of Psychiatry, Canisius-Wilhelmina Hospital, Nijmegen, 4 Parnassia Psychiatric Institute, The Hague, 5 Department of Psychiatry Sint Lucas Andreas Hospital, Amsterdam, 6 Department of Psychiatry, Leiden University Medical Center, Leiden, 7 Praktijkondersteuning ZuidOost Brabant, Veldhoven, the Netherlands, 8 Bipolar Disorders Program, CIBERSAM Institute of Neuroscience, Hospital Clinic, University of Barcelona, IDIBAPS, Barcelona, Catalonia, Spain and 9 Department of Psychiatry, University Medical Center Groningen, University of Groningen, Groningen, the Netherlands 4 Acta Psychiatrica Scandinavica 2010; 122:

72 74 Chapter 4 Abstract Objective: In a previous paper, we reported about the efficacy of the addition of lamotrigine to lithium in patients with bipolar depression. In the second phase of this study paroxetine was added to ongoing treatment in nonresponders. Method: Bipolar depressed patients (n = 124) treated with lithium were randomized to addition of lamotrigine or placebo. In nonresponders after 8 weeks, paroxetine 20 mg was added for another 8 weeks to ongoing treatment. Results: After 8 weeks the improvement in patients treated with lamotrigine vs. patients treated with placebo was significant. After addition of paroxetine this difference disappeared as a result of greater further improvement in the non-responders to placebo. Conclusion: Addition of lamotrigine to lithium was found effective in bipolar depressed patients. Further addition of paroxetine in nonresponders to lithium plus lamotrigine did not appear to provide additional benefit, while it appeared to do so in non-responders to lithium plus placebo. Significant outcomes Addition of lamotrigine is more effective than the addition of placebo in bipolar depressed patients already using lithium. Paroxetine addition in non-responders to the former treatment appeared to provide benefit for nonresponders to lithium plus placebo but not for nonresponders to lithium plus lamotrigine. The combination of lithium plus lamotrigine plus paroxetine was well tolerated and safe. Tremor was the only adverse event that separated between lamotrigine and placebo. There were no differences between lamotrigine and placebo in serious adverse events.

73 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine 75 Limitations The sample size of the non-responders to lithium plus lamotrigine or placebo was relatively small. The addition of paroxetine in the second phase of the study was open and uncontrolled. The duration of this study was short in regard to the lifelong pattern of mood episodes in bipolar disorder. Efficacy and safety of two treatment algorithms in bipolar depression. Introduction Since 1993 case reports on the effectiveness of lamotrigine in the acute treatment of bipolar depression have been published (1 3). The first placebocontrolled randomized controlled trial (RCT) showing efficacy of lamotrigine monotherapy was negative on the primary outcome criterion the Hamilton Rating Scale for Depression (HAMD) (4) but positive on several secondary outcome measures including the Montgomery-Åsberg Depression Rating Scale (MADRS) (5, 6). Four monotherapy studies thereafter (7) ( ) were all negative on the primary outcome criterion (the MADRS), although a meta-analysis of all these five studies showed a small but statistically significant positive result (8). Paroxetine is a selective serotonin re-uptake inhibitor (SSRI) which is very broadly used for the treatment of major depressive disorder, and which has also been tested in several RCTs in bipolar depression both as monotherapy and as adjunctive therapy (9 13). Although these studies were not positive and SSRIs are currently not viewed as a first-line treatment option in the treatment of bipolar depression, they may play a role in patients who are resistant to firstline therapies (14). In a previous paper, we described the results of a placebocontrolled RCT examining the efficacy and safety of the addition of lamotrigine during 8 weeks in bipolar patients with a break-through depression during treatment with lithium (15). Change in MADRS scores compared with baseline was significantly greater after the addition of lamotrigine to lithium than after the 4

74 76 Chapter 4 addition of placebo. Response rate (defined as a decrease of at least 50% of the initial MADRS score) was also significantly greater with lamotrigine compared to placebo. In this paper, we describe the results of two different treatment algorithms; treatment with lamotrigine added to lithium during the first 8 weeks (first phase) followed by further addition of paroxetine in non-responders during weeks 9 16 (second phase) vs. 8 weeks of treatment with lithium plus placebo (first phase) also followed by further addition of paroxetine in non-responders (second phase) during weeks Aims of the study The aim of this randomized controlled trial was to compare the efficacy, safety and tolerability of two treatment algorithms in patients suffering from a bipolar depression. Patients using lithium (first step) with a relapse or recurrence of a bipolar depression were randomized to the addition of lamotrigine or placebo (second step), 20 mg paroxetine was subsequently added to nonresponders of the former 2 steps (third step). Material and methods Study design This article describes the combined results of both the first and second phase of an investigator initiated (WN) study. In the first phase, 124 patients were recruited in 23 out-patient centers in the Netherlands (n = 112) and 3 centers in Spain (n = 12) between 2002 and 2005 (15). During the second phase of this study paroxetine 20 mg was open label added to non-responders at week 8. Lithium (open label) and lamotrigine or placebo (double-blind) was continued at the same dose. The study was approved by the ethical review board of the University Medical Center Utrecht, Utrecht, the Netherlands and by local institutional review boards in the Netherlands and in Spain. All patients signed informed consent prior to the start of any study procedures.

75 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine 77 Patients Patients (age 18 65) were eligible for this study if they suffered from bipolar disorder with a major depressive episode according to DSM-IV criteria confirmed by the MINI plus (16), a score of 18 on the MADRS (6) and a Clinical Global Impression for Bipolar Illness (CGI-BP) (17) severity score of 4 (moderately ill). All patients were on long-term treatment with lithium, with a lithium plasma level between 0.6 and 1.2 mm for at least 2 weeks before the start of the study. Patients with psychotic features, with a severe rapid cycling pattern (more than 10 episodes in the preceding year) with severe suicidality (score on item 10 of the MADRS 5) or with a history of alcohol or substance abuse within 1 month prior to the start of the study or dependence within 12 months, were excluded. No patients had neurological or other somatic illnesses known to affect mood. Female patients were not pregnant (negative pregnancy test prior to the study) and were required to use an effective contraceptive method. Patients were not allowed to take psychotropics other than study medications, except for benzodiazepines at a maximum of 2 mg lorazepam equivalents day. Treatments During the first phase of the study (week 0 8) patients (n = 124) received double-blind treatment with lamotrigine (up titrated to 200 mg day) or placebo added to ongoing treatment with lithium (119 patients had plasma levels between the predefined range of mm; five patients between 0.53 and 0.6 mm. Before deblinding it was decided to accept this minor protocol violation, thus including these patients in the analysis. During the second phase (week 9 16) open label paroxetine was added at a fixed dose of 20 mg day in non-responders at week 8 (CGI-BP change of 3 or more, i.e. minimally improved, unchanged or worse compared with baseline), while dosages of lithium and lamotrigine or placebo (still given double-blind) were kept stable throughout the second phase. Responders at the end of the first phase were also followed and maintained their same dosages of lithium and lamotrigine or placebo. 4

76 78 Chapter 4 Outcome measures Patient visits were scheduled at baseline (week 0) and from then onward every 2 weeks. MADRS scores and the CGI-BP Severity of Depression and Mania scores were assessed at each visit. The primary outcome for the second study phase was change from baseline on the total MADRS score at week 16. Secondary outcome measures were total MADRS score at week 16 or endpoint, CGI-BP severity of depression and of mania at week 16 or endpoint and response. Response was defined as a decrease on the MADRS of 50% and or GCI-BP change of 1 or 2 (very much or much improved) compared with baseline. An additional outcome measure was switch to (hypo) mania, defined as a score of 3 (mildly ill) on the CGI-BP severity of mania scale. Post hoc we also identified patients who had responded without a switch to (hypo) mania. At each visit adverse events were assessed and if present rated according to their severity as mild, moderate or severe. Statistics Mixed model anova for repeated measures (MMRM) was prospectively chosen as method of analysis for MADRS scores. The independent variables in this analysis were: time, treatment and their interaction. Time as categorical variable allowed estimating and testing mean differences in MADRS between the two treatments at any visit, including the primary efficacy measure for this study phase: change from baseline at week 16. For this purpose, the relevant contrasts were specified in the model for which the anova produced the estimates, SEs, confidence intervals and appropriate t-tests per visit. To make inference about an overall treatment effect over all visits in either phase of the study, time was considered a numeric trend variable in a mixed model that allowed a trend-break ( broken stick ) at the transition from the first to the second phase. In this model, a difference in time trend between the two treatment groups in either phase and changes in time trend between the two phases in either treatment group were estimated and tested. On the basis of this model predicted MADRS scores were calculated (with 95% confidence intervals) and tested between the two treatment groups at week 8 (end of phase I) and at week 16 (end of phase II). In

77 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine 79 accordance with the intent to treat principle, all randomized subjects contributed to these analyzes, missing values of MADRS scores being properly adjusted for by the restricted maximum likelihood estimation procedure used in the mixed model anova s. Clinical Global Impression for Bipolar Illness severity scores were similarly analyzed using mixed model anova with time as categorical variable. Dichotomous outcome scores (response, switch or presence of adverse effects) were compared between the two treatment groups using Fisher s exact test and last observation carried forward for missing data. Differences in the distribution of categorical nominal variables between the two treatment groups were tested using Fisher s exact test, for categorical ordinal variables the exact Chisquare trend test was used. For each test a two-sided P-value below 0.05 was considered to denote statistical significance. Results The flowchart of the study is presented in figure 1. During the first phase 64 patients received add-on lamotrigine and 60 patients received addon placebo. There were no statistical significant differences between the two groups in patients characteristics or illness severity at baseline (15) (table 1). In the lamotrigine group 52 patients (81.3%) completed the first 8 weeks of the study, 37 of them as responders and 15 as non-responders. After the first phase, nine patients (14.1%) dropped out leaving 43 patients who agreed to participate in the second phase; 34 responders and 9 non-responders. During the second phase another four patients dropped out. In the placebo group 50 patients (83.3%) completed the first 8 weeks, 28 as responders and 22 as nonresponders. Six patients (10%) dropped out, leaving 44 patients (26 responders and 18 non-responders) who continued the study. During the second phase another six patients dropped out. 4

78 80 Chapter 4 Figure 1. Flow chart of the study Assessed: 128 Randomised: 124 Not meeting inclusion criteria: 3 Not taken study medication: 1 Drop-outs: 12 Plus Lamotrigine: 64 Plus Placebo: 60 Drop-outs: 10 Phase II Phase I Responders: 37 Continued: 34 Drop-outs: 3 Drop-outs: 6 Lack of efficacy: 1 Withdrawn consent: 1 Protocol violation: 1 Completed: 52 Non-responders: 15 Completed: 50 Non-responders: 22 Plus paroxetine: 9 Plus paroxetine: 18 Non compliance: 1 Lack of efficacy: 2 Adverse events: 2 Completed: 31 Completed: 8 Completed: 14 Drop-outs: 2 Drop-outs: 4 Lack of efficacy: 1 Adverse events: 1 Responders: 28 Continued: 26 Completed: 24 Responders: 34 Responders: 26

79 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine 81 Table 1. Patient and illness characteristics Female gender (n, %) Age (years) (mean, SD) Ilness characteristics Bipolar I disorder (n, %) Rapid cycling course last 12 months MADRS (mean, SD) CGI-BP depression severity (mean, SD) CGI-BP mania severity (mean, SD) Lithium plasma level at baseline (meq) (mean, SD) Lamotrigine n=64 Placebo n=60 Total n=124 Statistic Value df P Fisher s t-test Fisher s Fisher s t-test Chi Sq. trend Fisher s t-test Previous other treatments for index depressive episode (n, %) TCA Fisher s SSRI/SNRI 15 23, Fisher s MAO-inhibitor Fisher s Antipsychotic, typical Antipsychotic, atypical Fisher s Fisher s Valproic acid Fisher s Carbamazepine Fisher s Benzodiazepine Fisher s Other Fisher s MADRS = Montgomery-Åsberg Despression Rating Scale, CGI-BP = Clinical Global Impression for Bipolar Illnes, TCA = tricyclic antidepressant, SSRI = selective serotonin re-uptake inhibitor, SNRI = serotonin-noradrenalin re-uptake inhibitor, MAO = monoamine oxidase inhibitor.

80 82 Chapter 4 Efficacy The efficacy results over both study-phases from a mixed model anova with time as categorical variable are presented in table 2. After 8 weeks, the difference between lithium plus lamotrigine vs. lithium plus placebo in the change of the MADRS scores from baseline to week 8 was significant (15); vs (P = 0.029). After 16 weeks there was no statistically significant difference in change of MADRS scores from baseline to week 16 between lithium plus lamotrigine (plus paroxetine in non-responders) vs. lithium plus placebo (plus paroxetine in non-responders); vs (P = 0.253). The difference between total MADRS scores was also no longer significant at week 16: vs (P = 0.134) (fig 2). The 95% confidence interval of the mean difference (lamotrigine minus placebo) in MADRS score at week 16 was from 7.13 to 0.97.

81 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine 83 Table 2. Efficacy Result: severity scales Lamotrigine n=64 Placebo n=60 Difference Statistic Value df P Change in MADRS score from baseline (primary outcome measure) (mean, SE) Week (0.87) (0.91) (1.26) t-test Week (1.25) (1.30) (1.81) t-test Week (1.35) (1.40) (1.95) t-test Week (1.32) (1.37) (1.90) t-test Week (1.30) (1.34) (1.87) t-test Week (1.51) (1.55) (2.16) t-test Week (1.39) (1.43) (1.99) t-test Week (1.53) (1.56) (2.19) t-test MADRS score at each visit (mean, SE) Baseline (0.76) (0.79) (1.10) t-test Week (1.23) (1.28) (1.78) t-test Week (1.42) (1.45) (2.03) t-test CGI-BP severity of depression at each visit (mean, SE) Baseline 4.56 (0.08) 4.53 (0.08) 0.03 (0.11) t-test Week (0.18) 3.08 (0.19) (0.26) t-test -1, Week (0.18) 2.56 (0.19) -0.36( 0.26) t-test CGI-BP change of depression from baseline (mean, SE) Week (0.17) 2.69 (0.17) (0.24) t-test Week (0.14) 2.15 (0.14) (0.20) t-test CGI-BP severity of mania at each visit (mean, SE) Baseline 1.03 (0.02) 1.03 (0.02) 0.00 (0.03) t-test Week (0.06) 1.14 (0.06) 0.01 (0.08) t-test Week (0.16) 1.30 (0.16) 0.07 (0.22) t-test MADRS = Montgomery-Åsberg Depression Rating Scale, CGI-BP = Clinical Global Impression for Bipolar Illness. Bold values indicate significant levels < 0.05

82 84 Chapter 4 Figure 2. Severity of depression according to the mean MADRS score at each visit for patients on placebo and lamotrigine 30 Mean MADRS score * P<0.05 * ** P<0.01 Addition of paroxetine ** ** Lamotrigine Placebo Weeks Response rates were 68.8% (44 out of 64 patients) to the lithiumlamotrigine-paroxetine algorithm, and 51.7% (31 out of 60) (P =0.066) to the lithium-placebo-paroxetine algorithm, while the response rates among completers of the study were 87% (34 out of 39 patients) and 73% (27 out of 37 patients) respectively (table 3). Other secondary outcome criteria (all CGI-BP outcomes) also did not reveal significant differences at week 16.

83 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine 85 Table 3. Efficacy results: response and switch to (hypo)mania after 8 and 16 weeks 8 weeks Lamotrigine (n = 64) Response N % N % Placebo (n = 60) Statistic P Reduction MADRS 50% 33 (51.6) 19 (31.7) Fisher CGI-BP change of depression 2 41 (64.1) 29 (49.2) * Fisher Either or both 42 (65.6) 29 (49.2) Fisher Switch to (hypo)mania ** 5 (7.8) 2 (3.3) Fisher Response and no switch to (hypo) mania *** 16 weeks 39 (60.9) 28 (46.7) Fisher Response N % N % Reduction MADRS 50% 33 (51.6) 27 (45.0) Fisher CGI-BP change of depression 2 43 (67.2) 31 (51.7) Fisher Either or both 44 (68.8) 31 (51.7) Fisher Switch to (hypo)mania ** 8 (12.5) 5 (8.3) Fisher Response and no switch to (hypo) mania *** 38 (59.4) 29 (48.3) Fisher MADRS = Montgomery-Åsberg Depression Rating Scale; CGI-BP = Clinical Global Impression for Bipolar Illness. Bold values indicate significant levels < 0.05 * 1 missing ** CGI-BP severity of mania 3 *** Reduction MADRS 50% and/or CGI-BP change of depression 2 and CGI-BP severity of mania < 3 4

84 86 Chapter 4 In table 4, the results of the piecewise linear model are presented. The estimated downward trends in time of the MADRS score which were significantly different between treatment groups in the first phase (P = 0.009), became significantly less steep in the second phase (P = in the placebo group and P < in the lamotrigine group). In the second phase, the downward trends were not significantly different anymore between the treatment groups (P = 0.19). The MADRS scores as predicted by the estimated piecewise linear model in both treatment groups was significantly different (P = 0.009) at week 8 (end of phase 1) and not significantly different (P = 0.22) anymore at week 16 (the end of phase2). Table 4. Predicted MADRS scores (95% CI) at weeks 8 and 16 (right two colums) and time trend (SE) in MADRS-score per 2 weeks (left four colums) according to a broken stick model with a trend break at week 8. The intercept (SE) equals (0.54). Phase 1: 0-8 weeks Time trend in MADRS score per 2 weeks Phase 2: 8-16 weeks Difference between phases MADRS = Montgomery-Åsberg Despression Rating Scale Predicted MADRS score P-value (test between phases) At week 8 At week 16 Placebo group (0.31) (0.31) 1.29 (0.52) (15.1 to 19.9) 11.9 (9.4 to 14.5) Lamotrigine group (0.30) (0.30) 2.99 (0.51) < (10.7 to 15.3) 9.7 (7.2 to 12.2) Difference between groups P-value (test between groups) (0.42) 0.57 (0.43) 4.5 ( 7.8 to 1.2 ) 2.2 ( 5.8 to 1.3) Eight patients (12.5%) in the lithium-lamotrigine- paroxetine group switched to (hypo) mania (five patients (7.8%) during the first 8 weeks and 3 (4.7%) after the addition of paroxetine in nonresponders). Six (9.4%) of these patients had also reached the response criterion for depression. In the lithiumplacebo-paroxetine group five patients (8.3%) switched to (hypo)mania (2 (3.3%) in the first phase and 3 (5%) in the second phase). Two of them (3.3%) had also reached the response criterion for depression (table 3).

85 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine 87 Safety Over all 16 weeks of the study, there were 15 serious adverse events (SAE) involving eight patients in the lithium-lamotrigine-paroxetine group (one patient experienced two SAEs) and five patients in the lithium-placebo-paroxetine group (also one patient with two SAEs). The four SAEs occurring during the first phase have been reported before (15). During the second phase there were11 SAEs involving nine patients. In the lithium-lamotrigine-paroxetine group two patients (no addition of paroxetine) developed a manic episode, one patient attempted suicide (no addition of paroxetine), one patient receiving paroxetine addition developed severe suicidal ideation, and one patient with high blood pressure in the first phase had an atrial flutter in the second phase (no paroxetine addition). In the lithium-placebo-paroxetine group two patients (no paroxetine addition) became severely depressed, one patient had an infection of the urinary tract (no paroxetine addition), one patient with the addition of paroxetine was admitted to a psychiatric hospital (reason unknown) with co-occurring hyperthyroidism (second phase). All patients recovered. Of the adverse events only tremor showed a difference between the lithium-lamotrigine-paroxetine and the lithium-placebo-paroxetine group (23.4% vs. 5% P = 0.042) (table 5). 4

86 88 Chapter 4 Table 5. Adverse events occurring in 5% of patients in any group (linear-by-linear association, two sided) Lamotrigine (n = 64) Bold values indicate significant levels < 0.05 Placebo (n = 60) N % N % Sex problems Pulmonary problems Blurred vision Hallucinations Throat problems Headache Obstipation Diarhea Alopecia Fatigue Nausea Flu-like symptoms Insomnia Tremor Skin problems/mild rash Dizziness Abdominal pain Rash Joint/muscle pain Back pain Coordination problems Eye problems Polyuria Agitation P

87 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine 89 Discussion Although the term treatment resistant is hardly operationalized (18) in bipolar depression, in clinical practice break-through depressions during long-term treatment including lithium, are more the rule than the exception. There is a discrepancy between the high proportion of patients receiving polypharmacy (by one estimate 80% of patients with bipolar disorder) (19, 20) and the scarceness of research data on the efficacy of polypharmacy. To our knowledge, this is the first study in bipolar disorder in which two different treatment algorithms consisting of three steps were compared. Patients who developed a depressive relapse or recurrence (break-through depression) during long-term lithium treatment (the first step) received addition of lamotrigine or placebo as a second step (first phase of the study) for 8 weeks. In the third step, non-responders after the second step were offered further addition of paroxetine during another 8 weeks (second phase of the study). At the end of the second step (week 8) there was a clinically and statistically significant advantage of change in MADRS scores from baseline (the primary outcome criterion) and on several secondary outcomes in favor of the addition of lamotrigine compared with the addition of placebo. However, by the end of the third step (week 16) there were no significant differences between the two treatment algorithms on primary or secondary outcomes. We have three possible explanations for these results. First, and in our view most likely, the addition of paroxetine in the third step was more beneficial to patients receiving addition of placebo in the second step than it was for patients who had received addition of lamotrigine, allowing the former group to catch up to at least some extent with the latter group. Whereas the lamotrigine group had plateaued already by week 8 10, the placebo group had more room for improvement. In the lamotrigine group, all patients started a potentially effective add-on treatment (lamotrigine) from baseline, while in the placebo group the nonresponders to lithium plus placebo started a potentially effective add-on treatment (paroxetine) at week 8. Indeed, a sub-analysis showed a greater effect of paroxetine in the non-responders to placebo than in the non-responders to lamotrigine (table 4). An interesting finding 4

88 90 Chapter 4 in this respect is also that the significance level between lamotrigine and placebo which was present at week 8, was even greater at week 10. Apparently, the catch up effect of paroxetine (in the placebo group) occurred after the first 2 weeks of its addition. The major objection against this explanation is that there is more evidence against rather than in favor of the efficacy of paroxetine or other SSRIs in bipolar depression (9, 12, 13). A second possible explanation for these results may be that they are a reflection of the natural waxing and waning course of bipolar disorder. However, we consider this as unlikely given the relative short period (week 12 16) during which the difference existing at week 8 (and even more strongly at week 10) receded, but it has been reported that placebo response may increase over time in bipolar disorder (21). A third possible explanation could be that an add-on study design reduces the possibility to show significant results, and this problem may even increase with multiple addon steps (21). We doubt however that this is the explanation as the significance level was at its maximum at week 10, i.e. even after most patients who dropped out after the first phase, had already left the study. In addition to the ongoing debate concerning the efficacy of antidepressants in bipolar depression (9, 12, 22, 23), it is interesting to further highlight the possible efficacy of the addition of paroxetine in non-responders to lithium plus placebo in the second phase of the study. It is remarkable that this suggestion for efficacy of paroxetine was only found in patients who had received lithium plus placebo, and not in the patients who had already received lithium plus lamotrigine. We consider it unlikely that this response to paroxetine was a placebo effect as all patients had already received placebo in the first phase. Perhaps paroxetine is effective in patients with a break-through depression during long-term treatment with lithium, but not in patients also receiving various other drugs, such as lamotrigine (as shown in our study) or treatments other than lithium, such as valproate, carbamazepine, atypical antipsychotics and or additional psychotherapy (as in the STEP-BD add-on study of the efficacy of add-on treatment with antidepressants including the SSRI sertraline) (12) To find an answer to this question, one might consider a three-arm RCT comparing the addition of both lamotrigine and paroxetine with placebo in patients with a break-through depression during treatment with lithium.

89 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine 91 Overall the tolerance of the combination of lithium with lamotrigine (or placebo) and subsequently paroxetine was very well. Of the original sample of 124 patients, 76 patients (61.3%) completed both study-phases, while seven patients (10.9%) in the lithium-lamotrigine-paroxetine group and 5 (8.3%) in the lithium-placebo-paroxetine group dropped out as a result of adverse events. Except for one patient who was lost to follow-up, all patients with SAEs recovered uneventfully. Regarding SAEs there was no statistical difference between the lithium-lamotrigineparoxetine group and the lithium-placebo-paroxetine group. It is difficult to discern whether some of the SAEs (e.g. suicidal ideation, mania) were an effect of the treatment or part of the natural course of the illness, but a relationship to drug treatment cannot be ruled out. Only tremor was reported more frequently in the lithium-lamotrigine-paroxetine group than in the lithiumplacebo-paroxetine group. A major limitation of the study was the overall relatively limited sample size, and more specifically the relative small number of patients moving from phase 1 to phase 2. With the significant difference after phase 1 but not after phase 2, we cannot conclude that the algorithm with lamotrigine is only effective at 8 weeks and not after the addition of paroxetine at 16 weeks. It remains possible that with a larger sample size, the difference would also have became significant at week 16. Moreover, as a consequence of the design with open label addition of paroxetine, the study was not capable of measuring the individual contribution of paroxetine in the treatment of bipolar depression in patients not responding to lithium plus lamotrigine or lithium plus placebo. To evaluate the efficacy of paroxetine, it would have been necessary to compare it with placebo and to re-randomize patients after phase 1. However, for such a study the sample size of our trial was much too small. Finally, the duration of this study, although long by current clinical trial standards, does not provide data how to continue further treatment in responders. This will be addressed in another paper in which we will describe the results of a 1 year follow-up in responders of both phase 1 and phase 2. In conclusion, the present results shows that in patients with a breakthrough depression during long-term treatment with lithium, the addition of 4

90 92 Chapter 4 lamotrigine is effective (first phase of this study) while further addition of paroxetine in nonresponders to lamotrigine (second phase) does not appear to provide much extra benefit. On the other hand, the finding from the second phase of an increased response to paroxetine in nonresponders to the addition of placebo supports the notion that the addition of paroxetine, and perhaps other SSRIs, may be reasonable choices for the treatment of patients with a bipolar depression despite treatment with lithium. Clearly more studies employing different treatment algorithms are needed to help discern the most optimal forms of polypharmacy for bipolar disorder.

91 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine 93 References 1. Sporn J, Sachs G. The anticonvulsant lamotrigine in treatment- resistant manicdepressive illness. J Clin Psychopharmacol 1997;17: Kusumakar V, Yatham LN. An open study of lamotrigine in refractory bipolar depression. Psychiatry Res 1997;72: Calabrese JR, Fatemi SH, Woyshville MJ. Antidepressant effects of lamotrigine in rapid cycling bipolar disorder. Am J Psychiatry 1996;153: Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960;23: Calabrese JR, Bowden CL, Sachs GS, Ascher JA, Monaghan E, Rudd GD. A doubleblind placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. Lamictal 602 Study Group. J Clin Psychiatry 1999;60: Montgomery SA, Åsberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979;134: Calabrese JR, Huffman RF, White RL et al. Lamotrigine in the acute treatment of bipolar depression: results of five double-blind, placebo-controlled clinical trials. Bipolar Disord 2008;10: Geddes J, Huffman RF, Paska W, Evoniuk G, Leadbetter R. Lamotrigine for acute treatment of bipolar depression: additional clinical trial data and a retrospective pooled analysis of response rates across all randomized trials conducted by GSK. Bipolar Disord 2007;8(Suppl. 1): Nemeroff CB, Evans DL, Gyulai L et al. Double-blind, placebo-controlled comparison of imipramine and paroxetine in the treatment of bipolar depression. Am J Psychiatry 2001;158: Shelton RC, Stahl SM. Risperidone and paroxetine given singly and in combination for bipolar depression. J Clin Psychiatry 2004;65: Vieta E, Martinez-Aran A, Goikolea JM et al. A randomized trial comparing paroxetine and venlafaxine in the treatment of bipolar depressed patients taking mood stabilizers. J Clin Psychiatry 2002;63:

92 94 Chapter Sachs GS, Nierenberg AA, Calabrese JR et al. Effectiveness of adjunctive antidepressant treatment for bipolar depression. N Engl J Med 2007;356: McElroy SL, Chang W, Nordenhem A, Paulsson B, Brecher M, Young AH. A doubleblind, placebo-controlled study with acute and continuation phase of quetiapine in adults with bipolar depression (emboldenii) rd biennial conference of the international society for bipolar disorders. Delhi, India January Goodwin GM, Anderson I, Arango C et al. ECNP consensus meeting. Bipolar depression. Nice, March Eur Neuropsychopharmacol 2008;18: van der Loos ML, Mulder PG, Hartong EG et al. Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebocontrolled trial. J Clin Psychiatry 2009;70: Sheehan DV, Lecrubier Y, Sheehan KH et al. The Mini- International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59(Suppl. 20): Spearing MK, Post RM, Leverich G, Brandt D, Nolen WA. Modification of the clinical global impressions (CGI) scale for use in bipolar illness (BP): the CGI-BP. Psychiatry Res 1997;73: Pacchiarotti I, Mazzarini L, Colom F et al. Treatmentresistant bipolar depression: towards a new definition. Acta Psychiatr Scand 2009;120: Ghaemi SN, Hsu DJ, Thase ME et al. Pharmacological treatment patterns at study entry for the first 500 STEPBD participants. Psychiatr Serv 2006;57: Simon NM, Otto MW, Weiss RD et al. Pharmacotherapy for bipolar disorder and comorbid conditions: baseline data from STEP-BD. J Clin Psychopharmacol 2004;24: Vieta E, Carne X. The use of placebo in clinical trials on bipolar disorder: a new approach for an old debate. Psychother Psychosom 2005;74: Gijsman HJ, Geddes JR, Rendell JM, Nolen WA, Goodwin GM. Antidepressants for bipolar depression: a systematic review of randomized, controlled trials. Am J Psychiatry 2004;161:

93 Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine Licht RW, Gijsman H, Nolen WA, Angst J. Are antidepressants safe in the treatment of bipolar depression? A critical evaluation of their potential risk to induce switch into mania or cycle acceleration Acta Psychiatr Scand 2008;118:

94 96

95 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebocontrolled trial with a novel design 97 Marc LM van der Loos a, Paul Mulder b, Erwin GThM Hartong c, Marc BJ Blom d, Anton C Vergouwen e, Martijn S van Noorden f, Manuela A Timmermans g, Eduard Vieta h and Willem A Nolen i for the LamLit Study Group* a Department of Psychiatry, Isala Klinieken Location Sophia, Zwolle, b Department of Biostatistics, Erasmus University Medical Center, Rotterdam, c Department of Psychiatry, Canisius- Wilhelmina Hospital, Nijmegen, d Parnassia Psychiatric Institute, The Hague, e Department of Psychiatry, Sint Lucas Andreas Hospital, Amsterdam, f Department of Psychiatry, Leiden University Medical Center, Leiden, g Praktijkondersteuning ZuidOost Brabant, Veldhoven, The Netherlands, h Bipolar Disorders Program, CIBERSAM Institute of Neuroscience, Hospital Clinic, University of Barcelona, IDIBAPS, Barcelona, Catalonia, Spain, i Department of Psychiatry, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands Bipolar Disord 2011: 13:

96 98 Chapter 5 Abstract Objective: In two previous manuscripts, we described the efficacy of lamotrigine versus placebo as add-on to lithium (followed by the addition of paroxetine in nonresponders) in the short-term treatment of bipolar depression. In this paper we describe the long-term (68 weeks) outcome of that study. Methods: A total of 124 bipolar depressed patients receiving lithium were randomized to addition of lamotrigine or placebo. After eight weeks, paroxetine was added to nonresponders for another eight weeks. Responders continued medication and were followed for up to 68 weeks or until a relapse or recurrence of a depressive or manic episode. Results: After eight weeks, the addition of lamotrigine to lithium was significantly more efficacious than addition of placebo, while after addition of paroxetine in nonresponders both groups further improved with no significant difference between groups at week 16. During follow- up the efficacy of lamotrigine was maintained: time to relapse or recurrence was longer for the lamotrigine group (median time10.0 months (confidence interval: ) versus the placebo group 3.5 months (confidence interval: )). Conclusion: In patients with bipolar depression, despite continued use of lithium, addition of lamotrigine revealed a continued benefit compared to placebo throughout the entire study.

97 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design 99 Introduction Bipolar disorders are characterized by recurrent mood episodes. While 98% of patients achieve syndromatic recovery from an acute manic or depressive episode within two years (1), 80 90% of patients suffer from new episodes thereafter (1 3). Despite the use of mood stabilizers such as lithium or valproate, especially depressive episodes and days with depressive symptoms are frequent and may prove difficult to treat (2, 4, 5). Lithium and perhaps to a lesser extent valproate have some but limited efficacy in the treatment and prevention of depressive episodes (6 8). There is a controversy regarding the use of antidepressants, both in the treatment of acute depressive episodes (9, 10) as well as in long-term treatment of bipolar disorder (11 13). Of the atypical antipsychotics, only quetiapine (14 17) and olanzapine (especially in combination with fluoxetine) (18) demonstrated efficacy in acute bipolar depression. Moreover, olanzapine (19) and quetiapine (as add-on to lithium or valproate) (20, 21) were found effective in the prevention of depressive recurrences. Another treatment option for bipolar depression is lamotrigine. The results from the acute studies in bipolar depression are contradictory. The first randomized, controlled trial (RCT) found a significant difference between lamotrigine and placebo in favour of lamotrigine, although only on secondary outcome measures (22). Four additional RCTs thereafter found no difference between lamotrigine monotherapy and placebo, probably because of a high placebo response in these studies (23, 24). Nevertheless, in a meta-analysis of all five studies, lamotrigine was more effective than placebo (25). In addition, lamotrigine has been compared with placebo and lithium in long-term treatment of bipolar disorder. In two studies (26, 27) lamotrigine was found effective in preventing depressive recurrences, while lithium was effective in preventing manic recurrences. In two previous papers, we described the short- term results of a placebocontrolled RCT investigating the effects of the addition of lamotrigine to lithium 5

98 100 Chapter 5 in patients with bipolar depression despite long-term treatment with lithium. In the first phase of that study (weeks 1 8), lamotrigine was found to be more effective than placebo (28); in the second phase (weeks 9 16), further addition of paroxetine in nonresponders to either lamotrigine or placebo (plus lithium) was associated with further improvement in both groups, while the difference between lamotrigine and placebo was no longer significant at week 16 (29). In this paper, we describe the results of the long-term outcome of that study. Methods Study design In this paper, we describe the follow-up of an investigator (WAN)- initiated, randomized, double-blind, placebo-controlled trial with three phases. The design of the study is shown in figure 1 and has been described in detail elsewhere (28, 29). During the first eight weeks of the study (phase 1), 124 patients with a bipolar I or II disorder and a depressive episode with 18 points on the Montgomery-Åsberg Depression Rating Scale (MADRS) (30) despite long-term treatment with lithium received lamotrigine or placebo in addition to ongoing treatment with lithium (28). Non-responders (n = 37) to the addition of either lamotrigine or placebo at week 8 were offered the opportunity to receive further addition of paroxetine (open label) in combination with ongoing lithium plus lamotrigine or placebo during weeks 9 16 (phase 2) (29). Responders at 8 or 16 weeks were followed until maximally week 68 (phase 3) or until they reached the primary endpoint (a depressive or manic relapse or recurrence). The medication regime (lithium, double-blind lamotrigine or placebo, and in some patients paroxetine as well) was kept stable during the entire follow-up. Patients who left the study without a relapse or recurrence were considered dropouts; reasons for dropout were recorded. The study was approved by the ethical review board of the University Medical Center Utrecht, The Netherlands, and by local institutional review boards in both countries. All patients gave written informed consent prior to initiation of any study procedure.

99 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design 101 Figure 1. Design of the study Lithium + Lamotrigine Lithium + Lamotrigine + Paroxetine in non-reponders Lithium Responders Lithium + Placebo Lithium + Placebo + Paroxetine in non-responders Phase 3 Phase 1 Phase 2 Follow-up of responders Week Treatments All patients were already receiving lithium, which was maintained at a plasma level of mmol L throughout the study. During phase 1, lamotrigine (or placebo in identical tablets) was up-titrated to 200 mg day at week 6 and maintained at this dose throughout the study. In nonresponders, at week 8 paroxetine was started at 20 mg day and also maintained at this dose throughout the study. In addition to the study medication (lithium, lamotrigine or placebo, and paroxetine) patients were not allowed to take other psychotropic drugs, with the exception of benzodiazepines at a maximum of 2 mg lorazepam equivalents day. 5 Assessments and outcome measures After baseline, patients were seen every two weeks until the end of phase 2 (week 16). During followup (phase 3), patients were assessed at week 20 and thereafter every two months. During all visits, patients were scored on the MADRS and the Clinical Global Impression Bipolar version (CGIBP) (31); adverse events

100 102 Chapter 5 and study medication compliance were also assessed. Lithium plasma levels were measured every four months. Response during phase 1 and 2 leading to follow-up was defined as a score of 1 or 2 (very much or much improved) according to the CGI-BP improvement of depression scale. Outcome of the study was the relapse or recurrence of a manic or depressive episode defined as a score 4 (at least moderate symptoms) on the CGI-BP severity of depression or mania scale. In addition, it was investigated how long the MADRS score was maintained at <50% of its baseline value after patients reached this level for the first time during phase 1 or 2. Statistics All analyses concerning efficacy during follow-up (phase 3) have to be considered in a descriptive sense without statistical testing. This is because the groups entering follow-up were selected by the treatment-dependent outcome (very) much improvement of depression on the CGI-BP and consequently had become formally incomparable at this stage. The evolution of no symptoms or only mild symptoms (both CGI-BP mania and depression severity score < 4) during the whole study was described by calculating percentages with no or only mild symptoms of the original group sizes at each visit. In addition, the sustainability of a MADRS score < 50% of baseline after the first attainment of this level was compared between the two treatment groups using a Kaplan Meier survival analysis. The frequencies of the various types of adverse events occurring in more than 5% of the patients during the continuation phase were compared between the two treatment groups using a chi-square trend test. Results Patient characteristics at baseline are presented in table 1. Treatment groups were balanced at baseline, considering the uniform distribution of p-values over the interval (0.1). The flowchart of the study, including reasons for dropping out during the three phases, is presented in figure 2.

101 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design 103 Table 1. Patient and illness characteristics at study entry Lamotrigine (n = 64) Placebo (n = 60) Total (n = 124) Statistic Value df P-value Female gender (n, %) Fisher s Age (years) (mean, SD) t-test Ilness characteristics Bipolar I disorder (n, %) Fisher s Rapid cycling course last 12 months Fisher s MADRS (mean, SD) t-test CGI-BP depression severity (mean, SD) CGI-BP mania severity (mean, SD) Lithium plasma level at baseline (meq) (mean, SD) Previous other treatments for index depressive episode (n, %) Chi Sq. trend Fisher s t-test TCA Fisher s SSRI/SNRI 15 23, Fisher s MAO-inhibitor Fisher s Antipsychotic, typical Fisher s Antipsychotic, atypical Fisher s Valproic acid Fisher s Carbamazepine Fisher s Benzodiazepine Fisher s Other Fisher s MADRS = Montgomery-Åsberg Despression Rating Scale, CGI-BP = Clinical Global Impression for Bipolar Illnes, TCA = tricyclic antidepressant, SSRI = selective serotonin re-uptake inhibitor, SNRI = serotonin-noradrenalin re-uptake inhibitor, MAO = monoamine oxidase inhibitor. 5

102 104 Chapter 5 Figure 2. Flow chart of the study Randomized: 124 Phase 2 Phase 1 Drop-outs: 6 Drop-out: 1 Drop-outs: 12 Plus lamotrigine: 64 Plus placebo: 60 Completed: 52 Completed: 50 Drop-outs: 10 Nonresponders: 15 Responders: 37 Responders: 28 Nonresponders: 22 Drop-outs: 3 Drop-outs: 2 Plus paroxetine: 9 Continued: 34 Continued: 26 Plus paroxetine: 18 Completed: 8 Drop-out: 3 Drop-outs: 2 Completed: 31 Completed: 24 Completed: 14 Drop-outs: 4 Drop-outs: 4 Responders: 5 Remained Remained Responders: 8 responders: 29 responders: 20 Drop-out: 1 Drop-out: 3 Drop-out: 1 Drop-outs: 2 Phase 3 Start follow-up: 30 Start follow-up: 25 In phase 1, a total of 124 patients were randomized to lamotrigine (n = 64) or placebo (n = 60); 102 completed this phase. After eight weeks, the difference in change of MADRS score (the primary outcome criterion during phase 1) was significant (15.37 versus 11.16, respectively p = 0.029) (28). In phase 2, nonresponders to lamotrigine (n = 9) or placebo (n = 18) received addition of paroxetine (phase 2). At the end of phase 2, both groups were further improved, while the difference between groups in change of MADRS score was no longer significant (17.91 for the lithium lamotrigine paroxetine algorithm versus (p = 0.253) for the lithium placebo paroxetine algorithm) (29).

103 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design 105 Lithium plasma levels were measured at baseline and at weeks 8, 16, 32, 48, and 64. The overall variation in mean plasma levels was mmol L for the lamotrigine group versus mmol L for the placebo group. Long-term efficacy Responders, according to the CGI-BP, during either phase 1 or phase 2 were eligible for followup. At each visit during follow-up, the number of patients who maintained responder status (CGIBP severity of depression or mania score <4) was calculated as percentage of the group who started with either lamotrigine or placebo in phase 1. These percentages maintained higher levels in the lamotrigine group than in the placebo group throughout the study (fig 3). No formal statistical test was done here because of the selected group for follow-up (see Methods). The drop in percentage between weeks 16 and 20 is caused by the treatment-dependent selection of patients entering follow-up. We also calculated time to first relapse or recurrence based on the MADRS score after having achieved responder status according to the MADRS (i.e., <50% of baseline) for the first time during phase 1 or phase 2. In this analysis, patients who left the study for any reason were censored. The time to relapse or recurrence was longer for the lamotrigine group than for the placebo group: median time 10.0 months (95% confidence interval (CI): ) versus 3.5 months (95% CI: ), respectively (see fig 4). No formal statistical test was performed. At the end of the study (68 weeks), 18 patients (28.1%) from the lamotrigine group were still in the study versus 14 patients from the placebo group (23.3%). 5

104 106 Chapter 5 Figure 3. Responders (Clinical Global Impression Bipolar version improvement of depression score of 1 or 2) as percentage of the initial group (n = 124; lamotrigine n = 64, placebo n = 60) throughout all three phases of the study Percentage responders Lamotrigine Placebo Weeks

105 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design 107 Figure 4. Probability of maintaining response (score < 50% of Montgomery-Åsberg Depression Rating Scale (MADRS) score at baseline) without recurrence (score 50% of initial MADRS) after having achieved responder status for the first time during phase 1 or phase 2. Median time to relapse or recurrence for the lamotrigine group was 10.0 months (95% confidence interval: ) versus 3.5 months (95% confidence interval: ) for the placebo group Probability of no recurrence Lamotrigine Placebo Months Safety Occurrence of severe adverse events (SAEs) and any adverse events (AEs) during the first 16 weeks of this study were described in detail previously (28, 29). During follow-up, 5 SAEs were observed: 2 in the lamotrigine group (1 patient developed an atrial flutter and 1 had high blood pressure), and 3 in the placebo group (2 patients had a severe depressive recurrence and 1 an urinary tract infection). None of the SAEs were considered to be related to study medication and all patients recovered. The AEs occurring in more than 5% of any group are listed in table 2. There was no difference between lamotrigine and placebo in the preva- lence of any AE. The total amount of AEs in both groups was remarkably low for a 68-week follow-up. 5

106 108 Chapter 5 Five patients in the lamotrigine group and 8 patients in the placebo group developed a (hypo)manic episode during follow-up. Table 2. Adverse events occuring in 5% of patients in any group (linear-by-linear association, two-sided) Lamotrigine (n = 30) Placebo (n = 25) N % N % Pulmonary problems Blurred vision Hallucinations Throat problems Joint/muscle pain Headache Nausea Flu-like symptoms Insomnia Abdominal pain Hypertension Back pain Coordination problems Eye problems P Discussion In this paper, we describe the long-term follow-up of an RCT consisting of three phases. With this quite novel design we were able to compare two different treatment algorithms, one with lamotrigine added to lithium and one with placebo added to lithium (phase 1), both followed by addition of paroxetine in nonresponders (phase 2) and a subsequent follow-up in responders after phase 1 or 2 (phase 3). We found that the algorithm with lamotrigine addition was not only more effective than the algorithm with placebo addition in the acute treatment of bipolar depression but also appeared to retain its efficacy throughout the entire study. The percentage of responders as well as the time to relapse after

107 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design 109 initial response was favourable for the lamotrigine group versus the placebo group. In our opinion, the design of our study has several advantages and disadvantages. The first advantage is that the add-on design closely resembles real life, in which many bipolar disorder patients receive multiple medications rather than monotherapy (32 34). In addition to the usual design of an RCT comparing lamotrigine or placebo added to lithium, we also could provide data on the addition of a third drug (paroxetine) to lithium and lamotrigine (or placebo) and on the long-term follow-up. Thus, our study provides useful information on the efficacy and safety of adjunctive lamotrigine compared to placebo in the acute as well as the long-term treatment of bipolar depression. A second advantage is that nonresponders after phase 1 could receive active medication (paroxetine) in phase 2, thus increasing the feasibility of the study. We hypothesize that it may also partly explain the relative low dropout rate in our study (only 38% during the first two phases and 74% over the entire study of 68 weeks), which is much lower than in the two previous long-term trials with lamotrigine (26, 27). In these studies, 98% and 85% of those who were randomized to lamotrigine or placebo, respectively, were dropouts after 68 weeks. The main disadvantage of this design is that the selection of nonresponders for receiving paroxetine after phase 1 and the selection of responders after phase 1 and 2 for continuation of the study led to an unequal number of patients leaving the study at these decision points. This eventually hampered possibilities for an analysis with statistical testing of the results of phase 3. The main limitation of our study is the relatively small sample size. This was already a problem in the analysis of phase 1, in which 124 patients participated while the study was originally planned to recruit 220 patients, and it definitely hampered statistical analysis after phase 2. Thus, our findings must be interpreted with some caution. Nevertheless, we conclude that lamotrigine was more effective than placebo as add-on treatment to lithium in patients with breakthrough bipolar 5

108 110 Chapter 5 depression and was well tolerated in combination with lithium. Moreover, lamotrigine could safely be combined not only with lithium but also with paroxetine and appeared to retain its efficacy and good tolerability throughout follow-up.

109 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design 111 References 1. Tohen M, Zarate CA Jr, Hennen J et al. The McLean- Harvard First-Episode Mania Study: prediction of recovery and first recurrence. Am J Psychiatry 2003; 160: Kupka RW, Luckenbaugh DA, Post RM et al. Comparison of rapid-cycling and nonrapid-cycling bipolar disorder based on prospective mood ratings in 539 outpatients. Am J Psychiatry 2005; 162: Goodwin FK, Jamison KR. Course and outcome. In: Manic Depressive Illness; Bipolar Disorders and Recurrent Depression, 2nd edn. New York: Oxford University Press, 2007: Judd LL, Akiskal HS, Schettler PJ et al. The long-term natural history of the weekly symptomatic status of bipolar I disorder. Arch Gen Psychiatry 2002; 59: Judd LL, Akiskal HS, Schettler PJ et al. A prospective investigation of the natural history of the long-term weekly symptomatic status of bipolar II disorder. Arch Gen Psychiatry 2003; 60: Geddes JR, Burgess S, Hawton K, Jamison K, Goodwin GM. Long-term lithium therapy for bipolar disorder: systematic review and meta-analysis of randomized con- trolled trials. Am J Psychiatry 2004; 161: Bowden CL, Calabrese JR, McElroy SL et al. A randomized, placebo-controlled 12-month trial of divalproex and lithium in treatment of outpatients with bipolar I disorder. Divalproex Maintenance Study Group. Arch Gen Psychiatry 2000; 57: Geddes JR, Goodwin GM, Rendell J et al. Lithium plus valproate combination therapy versus monotherapy for relapse prevention in bipolar I disorder (BALANCE): a randomised open-label trial. Lancet 2010; 375: Gijsman HJ, Geddes JR, Rendell JM, Nolen WA, Goodwin GM. Antidepressants for bipolar depression: a systematic review of randomized, controlled trials. Am J Psychiatry 2004; 161: Sachs GS, Nierenberg AA, Calabrese JR et al. Effectiveness of adjunctive antidepressant treatment for bipolar depression. N Engl J Med 2007; 356:

110 112 Chapter Licht RW, Gijsman H, Nolen WA, Angst J. Are antidepressants safe in the treatment of bipolar depression? A critical evaluation of their potential risk to induce switch into mania or cycle acceleration. Acta Psychiatr Scand 2008; 118: Ghaemi SN, Wingo AP, Filkowski MA, Baldessarini RJ. Long-term antidepressant treatment in bipolar disorder: meta-analyses of benefits and risks. Acta Psychiatr Scand 2008; 118: Vieta E. Antidepressants in bipolar depression. Acta Psychiatr Scand 2008; 118: Calabrese JR, Keck PE Jr, Macfadden W et al. A randomized, double-blind, placebocontrolled trial of quetiapine in the treatment of bipolar I or II depression. Am J Psychiatry 2005; 162: Thase ME, Macfadden W, Weisler RH et al. Efficacy of quetiapine monotherapy in bipolar I and II depression: a double-blind, placebo-controlled study (the BOLDER II study). J Clin Psychopharmacol 2006; 26: Young AH, McElroy SL, Bauer M et al. A double-blind, placebo-controlled study of quetiapine and lithium monotherapy in adults in the acute phase of bipolar depression (EMBOLDEN I). J Clin Psychiatry 2010; 71: McElroy SL, Weisler RH, Chang W et al. A double- blind, placebo-controlled study of quetiapine and paroxetine as monotherapy in adults with bipolar depression (EMBOLDEN II). J Clin Psychiatry 2010; 71: Tohen M, Vieta E, Calabrese J et al. Efficacy of olanzapine and olanzapine-fluoxetine combination in the treatment of bipolar I depression. Arch Gen Psychiatry 2003; 60: Tohen M, Calabrese JR, Sachs GS et al. Randomized, placebo-controlled trial of olanzapine as maintenance therapy in patients with bipolar I disorder responding to acute treatment with olanzapine. Am J Psychiatry 2006; 163: Suppes T, Vieta E, Liu S, Brecher M, Paulsson B. Maintenance treatment for patients with bipolar I disorder: results from a North American study of quetiapine in combination with lithium or divalproex (Trial 127). Am J Psychiatry 2009; 166: Vieta E, Suppes T, Eggens I, Persson I, Paulsson B, Brecher M. Efficacy and safety of quetiapine in combina- tion with lithium or divalproex for maintenance of patients with bipolar I disorder (International Trial 126). J Affect Disord 2008; 109:

111 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design Calabrese JR, Bowden CL, Sachs GS, Ascher JA, Monaghan E, Rudd GD. A doubleblind placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. Lamictal 602 Study Group. J Clin Psychiatry 1999; 60: Calabrese JR, Huffman RF, White RL et al. Lamotrigine in the acute treatment of bipolar depression: results of five double-blind, placebo-controlled clinical trials. Bipolar Disord 2008; 10: Geddes J, Huffman R, Paska W, Evoniuk G, Leadbetter R. Lamotrigine for acute treatment of bipolar depression: additional clinical trial data and a retrospective pooled analysis of response rates across all randomized trials conducted by GSK. Bipolar Disord 2006; 8 (Suppl. 1): Geddes JR, Calabrese JR, Goodwin GM. Lamotrigine for treatment of bipolar depression: independent meta-analysis and meta-regression of individual patient data from five randomised trials. Br J Psychiatry 2009; 194: Bowden CL, Calabrese JR, Sachs G et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently manic or hypomanic patients with bipolar I disorder. Arch Gen Psychiatry 2003; 60: Calabrese JR, Bowden CL, Sachs G et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently depressed patients with bipolar I disorder. J Clin Psychiatry 2003; 64: van der Loos ML, Mulder PG, Hartong EG et al. Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double- blind, placebo-controlled trial. J Clin Psychiatry 2009; 70: van der Loos ML, Mulder P, Hartong EG et al. Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine. Acta Psychiatr Scand 2010; 122: Montgomery SA, Åsberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979; 134: Spearing MK, Post RM, Leverich G, Brandt D, Nolen WA. Modification of the clinical global impressions (CGI) scale for use in bipolar illness (BP): the CGI-BP. Psychiatry Res 1997; 73: Ghaemi SN, Hsu DJ, Thase ME et al. Pharmacological treatment patterns at study entry for the first 500 STEP-BD participants. Psychiatr Serv 2006; 57:

112 114 Chapter Simon NM, Otto MW, Weiss RD et al. Pharmacotherapy for bipolar disorder and comorbid conditions: baseline data from STEP-BD. J Clin Psychopharmacol 2004; 24: Goldberg JF, Brooks JO III, Kurita K et al. Depressive illness burden associated with complex polypharmacy in patients with bipolar disorder: findings from the STEP-BD. J Clin Psychiatry 2009; 70:

113 Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design 115 5

114 116

115 117 Discussion 6

116 118 Chapter 6 Introduction Bipolar disorder is a common and debilitating disorder with a lifetime prevalence of around 2%(1) Although there is convincing evidence that patients are more often and longer depressed than manic (2-4), there was until recently no approved drug for the treatment of bipolar depression. Currently only quetiapine (both in Europe en de US) and the combination of olanzapine with fluoxetine (in the US) have been approved for the acute treatment of bipolar depression, while lamotrigine (both in Europe and in the US) has been approved for long-term treatment, specifically to prevent depressive episodes. The underpinning of existing guidelines, which also advocates other treatments including antidepressants, is confined due to the fact that research in this field is limited. As long as the highest standard of evidence in research, i.e. randomized placebo-controlled trials, depends mostly on funding from pharmaceutical companies and there was, with a few exceptions (see above), not much interest among these companies to apply for a specific indication bipolar depression, the future of new randomized controlled trials in bipolar depression is bleak (5). In addition to the add-on trial with lamotrigine as presented in this thesis and the other lamotrigine monotherapy trials, only a few other treatments have been investigated in double-blind, randomized trials with sometimes severe methodological problems. Antidepressants None of the more than 20 available antidepressants has specifically been studied and developed by pharmaceutical companies for the indication bipolar depression. As a result, there is an almost complete lack of (large) randomized placebo-controlled trials on the efficacy of antidepressants as monotherapy in bipolar depression (6;7). Next we will briefly review available evidence from the randomized placebo-controlled trials. In 1980 Mendlewicz et al. compared L-deprenil (MAO-B-inhibitor) in combination with L-5-hydroxytryptophan to placebo in 58 depressed patients

117 Discussion 119 (both unipolar and bipolar)(8). The combination outperformed placebo in clinical improvement. In 1982 Himmelhoch et al. compared tranylcypromine with placebo in 59 anergic major depressed patients(9). Although the patients were not formally diagnosed as bipolar, the authors stated that anergic major depression most typically occurs in primary bipolar illnesses. They reported that tranylcypromine showed superiority over placebo by the end of the first week, and this improvement increased in significance at each successive visit until week 6. In 1989 Cohn et al. compared fluoxetine, imipramine and placebo in a 6-week study involving 89 patients of whom 22 were already using lithium (10). Response to fluoxetine was greater than to imipramine or placebo while response to imipramine was also greater than to placebo. Due to small numbers, the differences were however not significant. In 2001 Nemeroff et al. compared the addition of paroxetine, imipramine and placebo as addition to lithium in 117 bipolar depressed patients(11). There was no significant difference between the three groups in efficacy. However, in a post-hoc analysis both antidepressants were superior in a subgroup with lithium blood serum levels below 0.8 mmol/l. In 2007 Sachs et al. published the largest and perhaps also most influential study of antidepressants in bipolar depression: the STEP-BD study with a total of 366 patients.(12). In this double-blind trial bupropion, paroxetine or placebo were added to ongoing treatment. All patients were on lithium, valproate or a combination of these two with the notion that 27% of the patients received an inadequate dose of lithium or valproate. Later, when inclusion became a problem, also other FDA approved antimanic agents (especially atypical antipsychotics such as olanzapine and quetiapine) were allowed. Other antidepressants had to be tapered off before or during the first week of the trial. Other medications could also be started during the study. Besides participating in this medication trial 69% of the patients were at the same time included in a randomized psychotherapy trial comparing cognitive behavioral therapy, interpersonal psychotherapy and family focused therapy with psychoeducation. The main finding of the study is that the antidepressants had no additional effect 6

118 120 Chapter 6 over placebo. However, due to the major limitations it is obvious that with the chosen design the widely drawn conclusion from this study that antidepressants are ineffective in bipolar depression, is not warranted. In 2010 McElroy et al. reported a 8 weeks study in which she compared 300mg. and 600 mg quetiapine to paroxetine 20 mg (n=122) and placebo (n=126) in bipolar depression(13). Both doses of quetiapine were more efficacious than placebo but there was no difference between paroxetine and placebo in response or remission. As a consequence of the (almost) lack of randomized placebo-controlled trials, and especially of well powered monotherapy trials with antidepressants, there is still much discussion about the use of antidepressants in bipolar depression. Some argue or even conclude that they are ineffective in bipolar depression(12)while others still consider them as (potential) effective, although with a small risk of switching to (hypo)mania and cycle acceleration in some antidepressants (14;15). In addition, some consider them as unsafe as they can cause a switch to (hypo)mania and cycle acceleration(15), while others claim that there is no good proof for this (16). Nevertheless, according to another STEP-BD study up to 60 % of the bipolar patients use antidepressants(12), while in yet another study the usage of antidepressants in bipolar disorder in the U.S. was found even higher: 79,33% versus 63,57% in Europe(17). Lithium, valproate and carbamazepine Although lithium has been the golden standard for treatment of bipolar disorders for more than 50 years, the proof for efficacy in the treatment of the acute bipolar depression is limited. In 1993 a review of 9 studies was published with a total of 163 patients(18). All studies were (mostly partial) cross-over designs with mostly unknown plasma levels of lithium and with various durations of the studies up to 43 days. In 8 out of 9 studies lithium was reported to be superior to placebo. However, due to methodological limitations, there is no real proof of efficacy in these studies. Subsequently, lithium has only been compared in one study with placebo and quetiapine 300 mg or 600 mg/day (19). While

119 Discussion 121 both doses of quetiapine were more efficacious than placebo, lithium (n=136, dose between mg/day) was not different from placebo (n=129). In conclusion, there is no evidence for the efficacy of lithium in the acute treatment of bipolar depression. Two meta-analyses of four small placebo-controlled trials with valproate in acute bipolar depression (total n=142) were published in 2010 (20;21). While in only one of these individual studies valproate was more effective than placebo, the pooled results showed superiority of valproate over placebo in mean depression scores as well as in response rates with an estimated number needed to treat of 5. Carbamazepine was compared to placebo in one study (n=124) with a third arm with carbamazepine in combination with a herbal medicine called Free and easy wanderer plus (FEWP) (22). Carbamazepine monotherapy had a higher clinical response rate at endpoint than placebo, but the addition of FEWP produced an even higher response rate than carbamazepine monotherapy. Atypical antipsychotics The combination of olanzapine with fluoxtine has been investigated in 2003 by Tohen(23). Olanzapine monotherapy was significant more efficacious than placebo but the combination of olanzapine with fluoxetine was superior to placebo and to olanzapine monotherapy. Based on this study the combination became the first FDA approved treatment of bipolar depression in the U.S. Subsequently also quetiapine has been tested in several studies(13;19;24;25). In all studies there was a robust difference between quetiapine and placebo in favor of quetiapine. Nowadays quetiapine has been approved for the acute treatment of bipolar depression (and also for the acute treatment of mania and for maintenance treatment) in the U.S. and in Europe. Finally, aripiprazol has been studied in bipolar depression, but surprisingly it was not found efficacious in comparison to placebo in several studies(7). 6

120 122 Chapter 6 Lamotrigine In this thesis we report a study on the efficacy of lamotrigine in the treatment of bipolar depression. When our study started the classical mood stabilizer lithium was considered the gold standard in the long-term treatment of bipolar disorder with a (presumed) stronger effect in the prevention of manic episodes than of depressive episodes(26). Moreover, the anticonvulsants valproate and carbamazepine appeared to have a similar profile. At the same time, there was some evidence that lamotrigine might be effective as monotherapy in the treatment of acute bipolar depression(27), while it also had been found effective as long-term treatment in the prevention of depressive episodes(28;29). (see also chapter 2) One other study compared lamotrigine and olanzapine/fluoxetine(30): there was a slight difference in efficacy in favor of the combination, but the drop-out rate due to adverse effects was lower for lamotrigine. This brought us to the fundamental idea of our study: the addition of lamotrigine to lithium might provide a better efficacy in the treatment of acute bipolar depression and might provide a greater stability thereafter during long-term follow-up. The study was designed as an addition strategy comparable with treatment decisions made in real life clinical practice. We selected bipolar patients who despite long-term maintenance treatment with lithium had developed a depressive episode. Two different treatment algorithms were compared; one treatment arm with addition of lamotrigine to ongoing treatment with lithium and one treatment arm with addition of placebo to lithium (phase I). After 8 weeks paroxetine was added in non-responders in both groups (phase II). Responders in both groups were followed until a relapse, a recurrence or until end of study at 68 weeks (phase III). Results Phase I (chapter 3) The main finding of our study is a greater efficacy of the addition of lamotrigine to lithium at 8 weeks than the addition of placebo to lithium

121 Discussion 123 in patients with a bipolar depression. As such, it is the first study to provide statistical significant evidence for the efficacy of lamotrigine as add on to lithium in bipolar depressed patients. After five monotherapy studies with only limited positive results, this positive trial adds prove to another treatment option in bipolar depression in addition to quetiapine, olanzapine (especially in combination with fluoxetine) and to some extend the antidepressants. The question remains why this trial had a positive outcome in contrast with the previous five monotherapy trials. The main difference between our study and the previous studies is the relative low response rate of placebo in our study (31,7%). The placebo response rate of the first study (which was positive on secondary outcome criteria) was 29% (27) but the placebo response rate of the four following studies was in the range of 40-50%.(32) About the reasons for such a high placebo response we can only hypothesize. The previous studies were all carried out in the U.S. with lots of patients without health insurance. For these patients every treatment, even an experimental treatment, may be an improvement compared to regular care. As response is often requested for further continuation in a study (and thus receiving treatment) there may also be a high pressure on patients to become a responder. As every patient in our study had an insurance, such a pressure was not the case in our patients, which may (in part?) explain the difference. Another remarkable difference is the much lower drop out rate in our study (17,7% after 8 weeks) compared to the previous studies (varying between 29% and 41%). A possible explanation for this is that almost all our investigators were also the treating physicians of the patients. This personalized approach may have added to the reliability and stability of the results. A final reason might be that all our patients were using lithium and therefore they were used to laboratory checks every few months. We hypothesize that this may have added to treatment adherence also for the other medications (lamotrigine, placebo and paroxetine) in our study. 6

122 124 Chapter 6 Results Phase II (chapter 4) In phase II we added 20 mg. paroxetine to ongoing treatment with lithium and placebo or lamotrigine in non-responders after 8 weeks. Although both groups improved further between week 8 and 16, the statistical difference between both groups disappeared. There are a few possible explanations for this effect. First the total amount of patients was further diminished during phase II. To remain statistical significant the difference in outcome between the two groups should have increased as the number of patients had decreased. Given the tendency of regression to the mean after a longer period of time, the chance for such an outcome was small. Another explanation is that it has to do with an antidepressant effect of paroxetine in these bipolar patients. As there were more non-responders in the placebo group after 8 weeks, more placebo patients received paroxetine in phase II, resulting in a catch up of the non-responders in the placebo group. However, this does not take into account that previous studies with paroxetine did not found paroxetine effective in bipolar depression(11-13). Results Phase III (chapter 5) After 16 weeks all patients who were still on lithium with addition of lamotrigine or placebo during phase I and with the partial addition of paroxetine in phase II and who had reached responder status at week 16, remained in the study and were followed until a manic or depressive relapse or until the end of the study at 68 weeks. Over all 68 weeks the difference between the treatment algorithm with lamotrigine versus the treatment algorithm with placebo appeared to remain stable. Survival without relapse or recurrence was longer for the algorithm with lamotrigine than for the algorithm with placebo. Moreover, throughout the study tolerability in general was good as reflected by a low drop-out rate due to adverse effects and the overall low amount of serious and non-serious adverse events.

123 Discussion 125 Design of the study: advantages and disadvantages The design of our study had advantages and disadvantages. The main advantage of this study is the comparability with clinical practice in which many patients with bipolar disorder receive combination therapy rather than monotherapy. Therefore, the results appear applicable to real life patients. Moreover, we consider a design with a placebo arm during the first 8 weeks (phase I) followed by the addition of an active treatment to non-responders to either the investigational drug or placebo during the following 8 weeks (phase II), an advantage for the participating patients, resulting in a greater chance to obtain informed consent from patients to participate in a trial in which they have a chance to receive placebo. Nevertheless, inclusion was a problem and the number of 124 included patients was lower than the originally planned 220 patients. This limited the power of the study. Disadvantages/limitations of the design are that we did not randomize again at week 8 (after phase I), that not all patients participated in phase II and that not all patients entered follow-up after week 16 (phase III) which seriously hampered the possibilities for formal statistical testing. Therefore, we had to use descriptive statistics for the analysis of the follow-up period. Nevertheless, we conclude that we have proved the feasibility of performing a placebo-controlled effectiveness study with multiple additions and a long follow-up. In our opinion, this opens the way to other addition trials. Clinical implications As polypharmacy is the mainstay in the treatment of bipolar disorder, we consider the results of our study clinically important as we proved that the addition of lamotrigine to lithium is indeed an option in the treatment of bipolar depressed patients: it was not only found effective as acute treatment but also appeared to sustain it s efficacy during follow up. So far the approved armory for the treatment of bipolar depression was limited to quetiapine, olanzapine in 6

124 126 Chapter 6 combination with fluoxteine and lamotrigine (for long-term treatment). Besides lithium, valproate and antidepressants might be considered as potential options. When the results of our study would be confirmed in future studies, there should definitely be a place for lamotrigine in the acute treatment of bipolar depression, at least in combination with lithium. An important advantage of lamotrigine is its benign adverse effect profile with no weight gain, sedation or metabolic syndrome. The only serious risk is a severe rash, but with careful dosing the chance for such a rash is limited and extremely rare. The long-term efficacy of drugs in the maintenance treatment of bipolar disorder In bipolar disorders the focus should always be on maintenance treatment: the prevention against new episodes using medication with a positive benefit (effectiveness)/risk (adverse effects) profile. Maybe the biggest problem in long-term studies is to choose the right design, which in practice is always a compromise between possibilities and impossibilities. There is a delicate balance between hard core randomized placebo-controlled efficacy trials with highly selected patients, and real practice effectiveness trials with a design which focuses more on generalizability. It is questionable whether the usual standard for efficacy studies (the randomized placebo-controlled monotherapy trial) is the best way to find new strategies for the long-term treatment of bipolar disorder. For registration purposes, double-blind placebo-controlled efficacy trials are requested by the authorities. The main idea behind RCT s is to make equal and comparable groups allowing a comparison between the investigated compound and placebo (or other control treatments), which is essential in efficacy studies. In acute treatment studies (of mania or depression) patients are directly randomized to start with the investigational drug or placebo (with the possibility of a third arm with a drug with proven efficacy, the so called standard treatment). For long-term trials there are however a few design options with

125 Discussion 127 advantages and disadvantages. Most long-term studies use so-called enriched designs(35). In these studies all patients start open label with the investigational drug, for instance as treatment of an acute manic or depressive episode. Thereafter, responders to treatment are randomized to continuation of the investigational drug or to switch to placebo (again with the possibility of a third arm with a drug with proven efficacy, the so called standard treatment). Good examples of this are the two long-term lamotrigine studies (28;29) in which manic or depressed patients first received open label treatment with lamotrigine. Responders to lamotrigine were randomized to continue with lamotrigine or to switch to placebo or to lithium (the standard treatment). This enriched design also has its limitations. It favors the investigational drug (in the example lamotrigine) over the standard comparison drug (in the example lithium) as only responders who also tolerated the investigational drug enter the randomized phase. An example of another design is the BALANCE study(36). In this study comparing the prophylactic effect of lithium, valproate end the combination of both, patients entered the study during an episode fo which they were treated with the combination of both drugs. Stabilized patients were randomized to continue treatment with either lithium (monotherapy), valproate (monotherapy) or to continue with the combination of both. This is also an enriched design with a selection on response and tolerance to lithium and valproate, but without favoring one of the compounds. A third example of maintenance trials is reflected by three studies by Greil et al(37), Hartong et al. (38) and Licht et al. (39) In these studies lithium was compared to carbamazepine (Greil/Hartong) or valproate (Licht). Patients entered the study during or after an acute manic or depressed episode that was treated with medications other than the investigational drugs (lithium and carbamazepine or valproate). After recovery from the episode patients were randomized to the study drugs and simultaneously the other drugs were tapered off. Subsequently patients were followed until the recurrence of a new episode or until end of follow-up. In our study we used an extension design with extended blind followup after a short-term double-blind randomized trial (phase I) in which patients 6

126 128 Chapter 6 received addition of either lamotrigine or placebo. After the acute trial (phase I and phase II) we followed the responders up to 68 weeks without new randomization after recovery, but with keeping the treatments (lamotrigine of placebo) blinded throughout follow-up. As such, our study does not prove that lamotrigine has a prophylactic effect. For that purpose we should have rerandomized all responders to lamotrigine after phase I to continue double-blind with lamotrigine or to switch to placebo (as in the first example above with the enriched design). Besides that we did not aim for that goal, we also had too few patients for such a study. For randomizing twice during one study, many more patients should have been required. Moreover, it can lead to an overpowered first part of the study or an underpowered second part of the study (and not to mention the extra problems with the addition of paroxetine in non-responders after 8 weeks). Therefore, we choose for a naturalistic, blind follow-up in responders. Limitation of the add-on design A limitation of our study is that we only gathered information on adding new medication(s). We have no information on how to continue treatment. Would it have been possible to taper off lithium or lamotrigine, and what to do with paroxetine? This problem is until now an almost unmet need. Most studies so far are designed to gather information on short -and long-term efficacy and tolerability of starting new compounds and combinations, but hardly on how they might be stopped. This is probably partly related to the fact that sponsoring of a trial investigating stopping of medication is not interesting for pharmaceutical companies. Moreover, the long-term maintenance studies (with enriched designs) are set up to proof efficacy (and tolerability) but not to find out which patients are in need of long-term treatment and which patients are not. Medications are often added to the ineffective (or only partially effective) treatment, without knowing whether it would also have been an option to switch from the (partially) ineffective treatment to another drug. I.e. would it also be possible to stop the first drug (in our study lithium) instead of adding new medication to the first drug.

127 Discussion 129 Patients can be reluctant to stop medication or combinations of medications. At least some patients want to continue medication if they believe that these drugs kept them stable over a long time. This is underpinned by an intended study in which bipolar depressed responders to the addition of an antidepressant (to ongoing treatment with mood stabilizers) (40) were asked to consent to be randomized to double-blind continuing or stopping of the added antidepressant. It proved almost impossible to find patients who accepted to be randomized after a successfully treated depressive episode. Some patients only wanted to keep their medication, but there were also patients who were reluctant to continue their medication, for instance because of adverse effects, or because they were ill-affected to take (too many) drugs. Finally the study was completed by comparing two non-randomized groups of patients who themselves had decided to stop the antidepressant or to continue the antidepressant with questionable results. The only randomized discontinuation trial with antidepressants in bipolar depression (41) showed no significant benefit from continuing antidepressants in the long-term treatment of bipolar patients on the primary outcome (mean change on the depressive subscale of the CMF). Only time to a first new depressive episode was significantly longer for the group that continued with the antidepressant. However, also these results are questionable, giver the fact that this was a follow-up study of the STEP-BD trial with all it s methodological limitations described above. Conclusion In our study the addition of lamotrigine to ongoing treatment with lithium was more effective than the addition of placebo in the treatment of acute bipolar depression. Moreover, we found after further continuation of treatment, that lamotrigine appeared to maintain its effect. Due to the fact that we did not rerandomize patients after the first 8 (or 16) weeks of treatment, the results of the follow-up phase cannot be considered as evidence for the long-term effectiveness of lamotrigine. However, there is such evidence from two other studies, in which 6

128 130 Chapter 6 responders to lamotrigine were randomized to continue with lamotrigine or to switch to placebo (or lithium)(28;29). Further research on the treatment of bipolar depression should be directed to find (other) long-term treatment(s) with maximum effectiveness and with minimal adverse effects. In addition, future research should not only be directed to the effectiveness and safety of starting medication(s) but also to the risks and benefits of stopping medication(s). Moreover, the optimal balance between efficacy and adverse effects is different for each individual and therefore a challenge to the cooperation between doctor and patient. Psychoeducation is not a matter of convincing patients to accept adverse effects, it is more a quest to find the right medication or combination of medications fit for the individual patient with optimal functioning in all aspects of human life.

129 Discussion 131 References 1. Regeer EJ, ten HM, Rosso ML, Hakkaart-van RL, Vollebergh W, Nolen WA. Prevalence of bipolar disorder in the general population: a Reappraisal Study of the Netherlands Mental Health Survey and Incidence Study. Acta Psychiatr Scand 2004 Nov;110(5): Judd LL, Akiskal HS, Schettler PJ, Coryell W, Endicott J, Maser JD, et al. A prospective investigation of the natural history of the long-term weekly symptomatic status of bipolar II disorder. Arch Gen Psychiatry 2003 Mar;60(3): Judd LL, Akiskal HS, Schettler PJ, Endicott J, Maser J, Solomon DA, et al. The longterm natural history of the weekly symptomatic status of bipolar I disorder. Arch Gen Psychiatry 2002 Jun;59(6): Kupka RW, Altshuler LL, Nolen WA, Suppes T, Luckenbaugh DA, Leverich GS, et al. Three times more depression than mania in both bipolar I and bipolar II disorder. Bipolar disorders 2007; 9: Vieta E. Antidepressants in bipolar depression. Acta Psychiatr Scand 2008 Nov;118(5): Gijsman HJ, Geddes JR, Rendell JM, Nolen WA, Goodwin GM. Antidepressants for bipolar depression: a systematic review of randomized, controlled trials. Am J Psychiatry 2004 Sep;161(9): Vieta E, Locklear J, Gunther O, Ekman M, Miltenburger C, Chatterton ML, et al. Treatment options for bipolar depression: a systematic review of randomized, controlled trials. J Clin Psychopharmacol 2010 Oct;30(5): Mendlewicz J, Youdim MB. Antidepressant potentiation of 5-hydroxytryptophan by L-deprenil in affective illness. J Affect Disord 1980 Jun;2(2): Himmelhoch JM, Fuchs CZ, Symons BJ. A double-blind study of tranylcypromine treatment of major anergic depression. J Nerv Ment Dis 1982 Oct;170(10): Cohn JB, Collins G, Ashbrook E, Wernicke JF. A comparison of fluoxetine imipramine and placebo in patients with bipolar depressive disorder. Int Clin Psychopharmacol 1989 Oct;4(4): Nemeroff CB, Evans DL, Gyulai L, Sachs GS, Bowden CL, Gergel IP, et al. Double- 6

130 132 Chapter 6 blind, placebo-controlled comparison of imipramine and paroxetine in the treatment of bipolar depression. Am J Psychiatry 2001 Jun;158(6): Sachs GS, Nierenberg AA, Calabrese JR, Marangell LB, Wisniewski SR, Gyulai L, et al. Effectiveness of adjunctive antidepressant treatment for bipolar depression. N Engl J Med 2007 Apr 26;356(17): McElroy SL, Weisler RH, Chang W, Olausson B, Paulsson B, Brecher M, et al. A double-blind, placebo-controlled study of quetiapine and paroxetine as monotherapy in adults with bipolar depression (EMBOLDEN II). J Clin Psychiatry 2010 Feb;71(2): Post RM, Altshuler LL, Leverich GS, Frye MA, Nolen WA, Kupka RW, et al. Mood switch in bipolar depression: comparison of adjunctive venlafaxine, bupropion and sertraline. Br J Psychiatry 2006 Aug;189: Leverich GS, Altshuler LL, Frye MA, Suppes T, McElroy SL, Keck PE, Jr., et al. Risk of switch in mood polarity to hypomania or mania in patients with bipolar depression during acute and continuation trials of venlafaxine, sertraline, and bupropion as adjuncts to mood stabilizers. Am J Psychiatry 2006 Feb;163(2): Licht RW, Gijsman H, Nolen WA, Angst J. Are antidepressants safe in the treatment of bipolar depression? A critical evaluation of their potential risk to induce switch into mania or cycle acceleration. Acta Psychiatr Scand 2008 Nov;118(5): Post RM, Leverich GS, Altshuler LL, Frye MA, Suppes T, Keck PE, et al. Differential clinical characteristics, medication usage, and treatment response of bipolar disorder in the US versus The Netherlands and Germany. Int Clin Psychopharmacol 2011 Mar;26(2): Zornberg GL, Pope HG, Jr. Treatment of depression in bipolar disorder: new directions for research. J Clin Psychopharmacol 1993 Dec;13(6): Young AH, McElroy SL, Bauer M, Philips N, Chang W, Olausson B, et al. A doubleblind, placebo-controlled study of quetiapine and lithium monotherapy in adults in the acute phase of bipolar depression (EMBOLDEN I). J Clin Psychiatry 2010 Feb;71(2): Bond DJ, Lam RW, Yatham LN. Divalproex sodium versus placebo in the treatment of acute bipolar depression: a systematic review and meta-analysis. J Affect Disord 2010 Aug;124(3):

131 Discussion Smith LA, Cornelius VR, Azorin JM, Perugi G, Vieta E, Young AH, et al. Valproate for the treatment of acute bipolar depression: systematic review and meta-analysis. J Affect Disord 2010 Apr;122(1-2): Zhang ZJ, Kang WH, Tan QR, Li Q, Gao CG, Zhang FG, et al. Adjunctive herbal medicine with carbamazepine for bipolar disorders: A double-blind, randomized, placebo-controlled study. J Psychiatr Res 2007 Apr;41(3-4): Tohen M, Vieta E, Calabrese J, Ketter TA, Sachs G, Bowden C, et al. Efficacy of olanzapine and olanzapine-fluoxetine combination in the treatment of bipolar I depression. Arch Gen Psychiatry 2003 Nov;60(11): Calabrese JR, Keck PE, Jr., Macfadden W, Minkwitz M, Ketter TA, Weisler RH, et al. A Randomized, Double-Blind, Placebo-Controlled Trial of Quetiapine in the Treatment of Bipolar I or II Depression. Am J Psychiatry 2005 Jul;162(7): Thase ME, Macfadden W, Weisler RH, Chang W, Paulsson B, Khan A, et al. Efficacy of quetiapine monotherapy in bipolar I and II depression: a double-blind, placebocontrolled study (the BOLDER II study). J Clin Psychopharmacol 2006 Dec;26(6): Geddes JR, Burgess S, Hawton K, Jamison K, Goodwin GM. Long-term lithium therapy for bipolar disorder: systematic review and meta-analysis of randomized controlled trials. Am J Psychiatry 2004 Feb;161(2): Calabrese JR, Bowden CL, Sachs GS, Ascher JA, Monaghan E, Rudd GD. A doubleblind placebo-controlled study of lamotrigine monotherapy in outpatients with bipolar I depression. Lamictal 602 Study Group. J Clin Psychiatry 1999 Feb;60(2): Bowden CL, Calabrese JR, Sachs G, Yatham LN, Asghar SA, Hompland M, et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently manic or hypomanic patients with bipolar I disorder. Arch Gen Psychiatry 2003 Apr;60(4): Calabrese JR, Bowden CL, Sachs G, Yatham LN, Behnke K, Mehtonen OP, et al. A placebo-controlled 18-month trial of lamotrigine and lithium maintenance treatment in recently depressed patients with bipolar I disorder. J Clin Psychiatry 2003 Sep;64(9): Brown E., Keck P.E., Deldar A., Adams D.H., Williamson D.J. Olanzapine/Fluoxetine Combination versus Lamotrigine in the treatment of Bipolar I Depression

132 134 Chapter van der Loos ML, Mulder PG, Hartong EG, Blom MB, Vergouwen AC, de Keyzer HJ, et al. Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar depression: a multicenter, double-blind, placebo-controlled trial. J Clin Psychiatry 2009 Feb;70(2): Calabrese JR, Huffman RF, White RL, Edwards S, Thompson TR, AscherJA, et al. Lamotrigine in the acute treatment of bipolar depression: results of five double-blind, placebo-controlled clinical trials. Bipolar Disord 2008 Mar;10(2): van der Loos ML, Mulder P, Hartong EG, Blom MB, Vergouwen AC, van Noorden MS, et al. Efficacy and safety of two treatment algorithms in bipolar depression consisting of a combination of lithium, lamotrigine or placebo and paroxetine. Acta Psychiatr Scand 2010 Sep;122(3): van der Loos ML, Mulder P, Hartong EG, Blom MB, Vergouwen AC, van Noorden MS, et al. Long-term outcome of bipolar depressed patients receiving lamotrigine as add-on to lithium with the possibility of the addition of paroxetine in nonresponders: a randomized, placebo-controlled trial with a novel design. Bipolar Disord 2011 Feb;13(1): Zimmerman M, Posternak MA, Ruggero CJ. Impact of study design on the results of continuation studies of antidepressants. J Clin Psychopharmacol 2007 Apr;27(2): Geddes JR, Goodwin GM, Rendell J, Azorin JM, Cipriani A, Ostacher MJ, et al. Lithium plus valproate combination therapy versus monotherapy for relapse prevention in bipolar I disorder (BALANCE): a randomised open-label trial. Lancet 2010 Jan 30;375(9712): Greil W, Ludwig-Mayerhofer W, Erazo N, Schochlin C, Schmidt S, Engel RR, et al. Lithium versus carbamazepine in the maintenance treatment of bipolar disorders--a randomised study. J Affect Disord 1997 Apr;43(2): Hartong EG, Moleman P, Hoogduin CA, Broekman TG, Nolen WA. Prophylactic efficacy of lithium versus carbamazepine in treatment-naive bipolar patients. J Clin Psychiatry 2003 Feb;64(2):

133 Discussion Licht RW, Nielsen JN, Gram LF, Vestergaard P, Bendz H. Lamotrigine versus lithium as maintenance treatment in bipolar I disorder: an open, randomized effectiveness study mimicking clinical practice. The 6 trial of the Danish University Antidepressant Group (DUAG-6). Bipolar Disord 2010 Aug;12(5): Post RM, Leverich GS, Nolen WA, Kupka RW, Altshuler LL, Frye MA, et al. A reevaluation of the role of antidepressants in the treatment of bipolar depression: data from the Stanley Foundation Bipolar Network. Bipolar Disord 2003 Dec;5(6): Ghaemi SN, Ostacher MM, El-Mallakh RS, Borrelli D, Baldassano CF, Kelley ME, et al. Antidepressant discontinuation in bipolar depression: a Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) randomized clinical trial of longterm effectiveness and safety. J Clin Psychiatry 2010 Apr;71(4):

134 136

135 Summary 137

136 138 Design of the study In this thesis we describe an investigator (W.A. Nolen ) initiated double-blind randomized placebo-controlled trial (RCT) comparing two different treatment algorithms with a follow-up until a maximum of 68 weeks. In bipolar outpatients (men or women aged at least 18 years), with a depression despite long-term treatment with lithium (at serum levels of mmol/l) lamotrigine or placebo was added to ongoing treatment with lithium. In non-responders after 8 weeks paroxetine was open-label added to ongoing treatment with lithium plus lamotrigine or placebo. Responders after 8 or 16 weeks were followed until a relapse or recurrence or until the maximum duration of the study; 68 weeks (fig 1). Figure 1. Design of the study Lithium + Lamotrigine Lithium + Lamotrigine + Paroxetine in non-reponders Lithium Responders Lithium + Placebo Lithium + Placebo + Paroxetine in non-responders Phase 3 Phase 1 Phase 2 Follow-up of responders Week The study was approved by the ethical review board of the University Medical Center Utrecht, the Netherlands and by local institutional review boards in the Netherlands and in Spain. All patients gave written informed consent.

137 Summary 139 Results Figure 2. Flow chart of the study Randomized: 124 Phase 2 Phase 1 Drop-outs: 6 Drop-out: 1 Drop-outs: 12 Plus lamotrigine: 64 Plus placebo: 60 Completed: 52 Completed: 50 Drop-outs: 10 Nonresponders: 15 Responders: 37 Responders: 28 Nonresponders: 22 Drop-outs: 3 Drop-outs: 2 Plus paroxetine: 9 Continued: 34 Continued: 26 Plus paroxetine: 18 Completed: 8 Drop-out: 3 Drop-outs: 2 Completed: 31 Completed: 24 Completed: 14 Drop-outs: 4 Drop-outs: 4 Responders: 5 Remained Remained Responders: 8 responders: 29 responders: 20 Drop-out: 1 Drop-out: 3 Drop-out: 1 Drop-outs: 2 Phase 3 Start follow-up: 30 Start follow-up: Patients were randomised, 112 in the Netherlands and 12 in Spain. Phase 1: week 1-8 (chapter 3) Of the 64 patients who started in the lamotrigine group (fig 2), 52 patients (81%) completed the first 8 weeks versus 50 (83%) of the 60 patients in the placebo group. On the primary outcome measure (change in MADRS score from baseline to week 8; vs respectively; P=0.024) as well as on some secondary outcome measures lamotrigine was significantly more efficacious than placebo (fig 3).

138 140 Figure 3. Severity of depression according to the mean MADRS score at each visit for patients on placebo and lamotrigine MADRS Score, mean * ** 5 0 Lamotrigine Placebo Baseline Wk 2 Wk 4 Wk 6 Wk 8 *p =.031 **p =.006 Phase 2: week 9-16 (chapter 4) After 8 weeks responders to the lithium-lamotrigine and the lithiumplacebo combination continued the study with lithium and (still double-blind) lamotrigine or placebo, while open-label paroxetine was added in nonresponders. Although both groups improved further, the difference after 16 weeks between lithium plus lamotrigine (plus paroxetine in non-responders) versus lithium plus placebo (plus paroxetine in non responders) was not statistically significant anymore (change in MADRS score from baseline to week versus respectively; p=0.253). Other secondary outcome criteria also did not reveal significant differences at week 16 (fig 4).

139 Summary 141 Figure 4. Severity of depression according to the mean MADRS score at each visit for patients on placebo and lamotrigine (plus addition of paroxetine in non-responders at week 8) 30 Mean MADRS score * P<0.05 * ** P<0.01 Addition of paroxetine ** ** Lamotrigine Placebo Weeks Phase 3: week (chapter 5) During follow-up responders after 8 or 16 weeks were maintained on study medications until a manic or depressive episode or until the maximum duration of the study, i.e. until week 68. Descriptive statistics showed that the difference (as percentage of responders of the initial group) between the lamotrigine group versus the placebo group remained stable throughout the follow-up period (fig 5 and 6).

140 142 Figure 5. Responders (CGI-BP improvement of depression score of 1 or 2) as percentage of the initial group (n = 124, lamotrigine n = 64, placebo n = 60) throughout all three phases of the study Percentage responders Lamotrigine Placebo Weeks Also time from first response (score <50% of MADRS score at baseline) to recurrence (score 50% of initial MADRS) showed a similar pattern in favour of the lamotrigine group. The median survival time for responders of the lamotrigine group was 10.0 months (95% CI: ) versus 3.5 months (CI: ) for the responders in the placebo group.

141 Summary 143 Figure 6. Probability of maintaining response (score <50% of MADRS score at baseline; n = 79; lamotrigine: n = 41; placebo: n = 38) without recurrence (score 50% of initial MADRS) since having achieved responder status during phase 1 or phase Probability of no recurrence Lamotrigine Placebo Months Safety and Adverse Events During the 68 weeks of the study there were 20 Serious Adverse Events (SAE), 15 during the first 16 weeks and 5 thereafter. 10 patients in the lamotrigine group (2 patients had 2 SAEs) group versus 8 patients in the placebo group experienced a SAE. Most SAEs concerned serious mood swings as could be expected in a trial with 124 bipolar patients. Statistically there were no differences between lamotrigine and placebo. The only statistical difference between lamotrigine and placebo in adverse events was tremor at 16 weeks which occurred more often in the lamotrigine group versus the placebo group. So, we can conclude that the addition of lamotrigine to lithium (with further addition of paroxetine in non-responders) was safe and had an adverse events profile comparable to the addition of placebo to lithium.

142 144 Conclusions After this study several conclusions considering the treatment and future research of patients with a bipolar depression can be drawn. 1. The addition of lamotrigine to lithium in patients with bipolar depression is efficacious and overall well tolerated. 2. After the further addition of paroxetine in non-responders to the combination of lithium plus lamotrigine or placebo the statistical significance between the treatment arm with lamotrigine group and the treatment arm with placebo group disappeared with ongoing improvement in both groups. However, the treatment arm with lamotrigine remained numerically more effective, also during follow-up up to one year. 3. These findings indicate that lamotrigine - in addition to its use in the longterm prevention of depressive episodes - also deserves a place in the acute treatment of bipolar depression. 4. The gap between scientific research with randomized placebo-controlled trials and clinical practice can be bridged by designing studies comparing combinations of medications.

143 Summary 145

144 146

145 Samenvatting 147

146 148 Design van de studie In dit proefschrift beschrijven wij een door W.A. Nolen geinitialiseerd dubbel blind placebo gecontroleerd onderzoek waarin twee behandeling algoritmes vergeleken werden met een vervolg tot een maximum van 68 weken. Bij bipolaire depressieve patiënten (man of vrouw, ouder dan 18 jaar) met een bipolaire depressie ondanks langdurige behandeling met lithium (bloedspiegels mmol/l werd lamotrigine of placebo toegevoegd. Na acht weken kregen non-responders in beide groepen (lithium +lamotrigine of lithium +placebo) open label paroxetine (20mg) extra toegevoegd. Responders na 8 of 16 weken werden gevolgd tot een terugval in een depressieve of (hypo)manische episode of tot het eind van de studie na 68 weken. Figuur 1. Ontwerp van de studie Lithium + Lamotrigine Lithium + Lamotrigine + Paroxetine in non-reponders Lithium Responders Lithium + Placebo Lithium + Placebo + Paroxetine in non-responders Fase 3 Fase 1 Fase 2 Follow-up of responders Week Voorafgaand aan de studie was toestemming verleend door de medisch ethische toetsingscommissie van het UMC Utrecht en door de

147 Samenvatting 149 lokale toetsingscommissies in Nederland en in Spanje. Alle patiënten gaven voorafgaand aan de studie informed consent. Results Figuur 2. Flow Chart van de studie Randomized: 124 Phase 2 Phase 1 Drop-outs: 6 Drop-out: 1 Drop-outs: 12 Plus lamotrigine: 64 Plus placebo: 60 Completed: 52 Completed: 50 Drop-outs: 10 Nonresponders: 15 Responders: 37 Responders: 28 Nonresponders: 22 Drop-outs: 3 Drop-outs: 2 Plus paroxetine: 9 Continued: 34 Continued: 26 Plus paroxetine: 18 Completed: 8 Drop-out: 3 Drop-outs: 2 Completed: 31 Completed: 24 Completed: 14 Drop-outs: 4 Drop-outs: 4 Responders: 5 Remained Remained Responders: 8 responders: 29 responders: 20 Drop-out: 1 Drop-out: 3 Drop-out: 1 Drop-outs: 2 Phase 3 Start follow-up: 30 Start follow-up: patiënten werden gerandomiseerd, 112 in Nederland en 12 in Spanje. Fase I: week 1-8 (hoofdstuk 3) 52 van de 64 patiënten(81%) in de lamotrigine groep die in de studie startten voltooiden de eerste 8 weken van het onderzoek versus 50 van de 60 patiënten (83%) in de placebo groep (fig 2). Op de primaire uitkomstmaat; de

148 150 verandering gemeten van baseline tot en met week 8 op de Montgomery- Åsberg Depression Rating Scale (MADRS) (en op sommige secundaire uitkomstmaten) bleek de toevoeging van lamotrigine aan lithium statistisch significant (-15,38 versus p=0.024) effectiever dan de toevoeging van placebo aan lithium (fig 3). Figuur 3. Ernst van de depressie gemeten als gemiddelde MADRS score tijdens elke visite voor patiënten op lamotrigine en placebo 30 MADRS Score, gemiddeld * ** 5 0 Lamotrigine Placebo Baseline Wk 2 Wk 4 Wk 6 Wk 8 *p =.031 **p =.006 Fase II: week 9-16 (hoofdstuk 4) Na 8 weken vervolgden responders op zowel de lithium lamotrigine combinatie als de responders op de lithium placebo combinatie de studie met dezelfde (nog steeds dubbel blinde) medicatie. Bij non-responders in beide groepen werd open label 20 mg paroxetine toegevoegd. Beide groepen verbeterden verder, maar het verschil tussen de groep met lithium + lamotrigine + paroxetine (vanaf 8 weken in non-responders) versus de groep met lithium + placebo + paroxetine (vanaf 8 weken in non-responders) was niet langer statistisch significant ( versus p=0.253) (fig 4). Ook andere secundaire uitkomstmaten waren niet statistisch significant.

149 Samenvatting 151 Figuur 4. Ernst van de depressie volgens de MADRS op elke bezoek voor patiënten met lamotrigine en placebo toevoeging aan lithium (plus additie van paroxetine na 8 weken bij non-respons) 30 Gemiddelde MADRS score * P<0.05 * ** ** P<0.01 Plus additie van paroxetine ** Lamotrigine Placebo Weken Fase III: week (hoofdstuk 5) Responders na 8 of 16 weken werden verder gevolgd met behoud van dezelfde (deels dubbelblinde) medicatie tot een nieuwe depressieve of (hypo) manische episode of tot het eind van de studie na 68 weken. Uit beschrijvende statistiek bleek dat het verschil tussen de lamotrigine groep en de placebo groep (gemeten als percentage responders van de 2 oorspronkelijke groepen) stabiel bleef gedurende de follow-up (fig 5 en 6).

150 152 Figuur 5. Responders (volgens de CGI-BP verbetering depressie score 1 of 2) als percentage van de oorspronkelijke groep (n = 124, n = 64 voor lamotrigine en n = 60 voor placebo) gedurende alle drie de fases van de studie Percentage responders Lamotrigine Placebo Weken Ook de tijd vanaf de eerste keer dat een patiënt responders was (MADRS score < 50% van de initiële score) tot terugval (MADRS score 50% van de initiële score) liet hetzelfde patroon zien. De gemiddelde tijd als responder voor de lamotrigine groep was 10.0 maanden (95% CI: ) versus 3.5 maanden (95% CI ) voor de responders in de placebo groep.

151 Samenvatting 153 Figuur 6. Kans om responder te blijven (score <50% of MADRS score op baseline; n = 79; lamotrigine: n = 41; placebo: n = 38) zonder terugval (score 50% van de oorspronkelijke MADRS) nadat men responder geworden is gedurende fase 1 of fase Waarschijnlijkheid van terugval Lamotrigine Placebo Maanden Veiligheid en bijwerkingen Gedurende de 68 weken van deze studie waren er 20 SAE s (Serious Adverse Events), 15 gedurende de eerste 16 weken en 5 daarna. 10 patiënten in de lamotrigine groep (2 patiënten hadden 2 SAE s) versus 8 patiënten in de placebo groep kregen een SAE. De meeste SAE s waren heftige stemmingswisselingen. Dit was niet onverwacht in een groep van 124 bipolaire patiënten. Over de gehele periode was er alleen na 16 weken voor wat betreft. tremor een statistisch verschil tussen lamotrigine en placebo ten gunste van placebo. Voor de rest was er geen enkel statistisch significant verschil tussen lamotrigine en placebo over de gehele studie. Concluderend kunnen we stellen dat de toevoeging van lamotrigine aan lithium (met daarna de toevoeging van paroxetine bij non-respons) veilig was en

152 154 dat de bijwerkingen van de toevoeging van lamotrigine aan lithium vergelijkbaar waren met de toevoeging van placebo aan lithium. Conclusies Na deze studie kunnen we een aantal conclusies trekken over de behandeling en toekomstig onderzoek bij patiënten met een bipolaire stoornis gedurende een depressieve fase. 1. De toevoeging van lamotrigine aan lithium bij patiënten met een bipolaire stoornis gedurende een depressieve fase is effectief en veilig. 2. Na de toevoeging van paroxetine op 8 weken bij non-responders verdween het statistisch significante verschil tussen de behandeling met lamotrigine en lithium versus de behandeling met placebo en lithium. Beide groepen verbeterden nog wel verder tussen de 8 en 16 weken. Ook tijdens het vervolg tot 68 weken bleef de behandel arm met lamotrigine in aantal effectiever dan de behandel arm met placebo. 3. Deze resultaten laten zien dat lamotrigine - in toevoeging aan het gebruik voor de preventie op de lange termijn van nieuwe depressieve episoden - ook een plaats verdient in de acute behandeling van de bipolaire depressie. 4. De kloof tussen wetenschappelijk onderzoek en de dagelijkse behandel praktijk kan overbrugd worden door het ontwerpen van studies die kijken naar combinaties van medicatie.

153 Samenvatting 155

154 156

155 Curriculum Vitae M.L.M. van der Loos 157