When First-Line Treatments Fail in Bipolar Disorder

Mark A. Frye, MDProfessor and ChairDepartment of Psychiatry and PsychologyDirector, Mayo Clinic Depression CenterRochester, Minnesota

Faculty Disclosure

Dr. Frye: Grant Support—Assurex Health, Mayo Foundation, Medibio; Consultant (Mayo)—Actify Neurotherapies, Allergan, Intra-Cellular Therapies Inc., Janssen, Myriad, Neuralstem, Inc., Takeda, Teva Pharmaceuticals; CME/Travel/Honoraria—American Physician Institute, CME Outfitters, Global Academy for Medical Education.

Mayo Clinic has a financial interest in AssureRx, OneOme, and the technology referenced in this presentation.

Disclosure

• The faculty have been informed of their responsibility to disclose to the audience if they will be discussing off-label or investigational use(s) of drugs, products, and/or devices (any use not approved by the US Food and Drug Administration).– The off-label use of antidepressants in bipolar depression will be discussed.

• Applicable CME staff have no relationships to disclose relating to the subject matter of this activity.

• This activity has been independently reviewed for balance.

Learning Objectives

• Discuss when to declare treatment failure

• Describe how to approach treatment of mania after first failed treatment

• Describe how to approach treatment for bipolar depression after first failed treatment

First-Line Treatments Fail• Mania

– FDA approved – Have they been optimized?– Adjunctive atypical antipsychotic – Clozapine– ECT

• Depression– FDA approved – Have they been optimized?– Antidepressants– Modafinil / Armodafinil– Ketamine– Bright light therapy– Partial wake therapy– ECT

• Maintenance– Subsyndromal symptoms

ECT = electroconvulsive therapy.

Mania MattersEpisodes associated with neuroanatomic change?

Frye MA, et al. Psychiatry Res. 2007;154(3):259-265. Tsai G, et al. Prog Neurobiol. 1995;46(5):531-540. Altshuler LL. Biol Psychiatry. 1993;33(8-9):563-565.

T1-weighted sagittal MRI anterior cingulate/medial prefrontal cortex PRESS 1H-MRS (TR/TE = 3s/30ms voxel size 3x3x3 cm3)

*Aripiprazole, asenapine, olanzapine, quetiapine, risperidone monotherapy and adjunct to lithium or divalproex and with/without psychosis.

FDA Approved Bipolar Disorder Treatments*Agent Manic Mixed Depression Maintenance

Aripiprazole + + – +

Asenapine + + – –

Cariprazine + + + –

Lurasidone – – + –

Olanzapine + + – +

Olanzapine/Fluoxetine – – + –

Quetiapine/XR + + + +

Risperidone (Oral / IM) + + – + (IM)

Ziprasidone + + – +

Chlorpromazine + – – –

Carbamazepine ER + + – –

Divalproex DR/ER + + – –

Lamotrigine – – – +

Lithium + – – +

Double-Blind Comparison of Clonazepam vs Lorazepam in Acute Mania

Bradwejn J, et al. J Clin Psychopharmacol. 1990;10(6):403-408.

Cipriani A, et al. Lancet. 2011;378(9799):1306-1315.

• 68 randomized trials of acute mania (n=16,073)

• Acceptability = any-cause early discontinuation

• Multiple treatments meta-analysis

• Accounts for direct and indirect comparisons

Cariprazine for Acute Mania Associated with Bipolar I Disorder

Randomized, double-blind, placebo-controlled trial (2010–2011); cariprazine 3 to 6 mg/day vs cariprazine 6 to 12 mg/day vs placebo over 3 weeks; 497 patients with BD-I manic or mixed episodes; primary endpoint: change YMRS total score; secondary endpoints: response, remission

BD-I = bipolar I disorder; YMRS = Young Mania Rating Scale.Calabrese JR, et al. J Clin Psychiatry. 2015;76(3):284-292.

Atypical Antipsychotics in Acute Mania

Pros – As a class, effective in acute mania and mixed episodes– Rapid control of acute mania/mixed, rapid cycling, psychosis/no

psychosis– Sustained improvement of symptoms

Cons – Tardive dyskinesia, neuroleptic malignant syndrome– Weight gain, related dysmetabolic effects

Tarr GP, et al. J Affect Disord. 2011;134(1-3):14-19. Yildiz A, et al. Neuropsychopharmacology. 2011;36(2):375-389.

Typical Antipsychotics in Acute ManiaPros

– Efficacious for acute mania– Haloperidol may be more rapidly efficacious than olanzapine,

quetiapine, ziprasidone

Cons/adverse effects– Acute EPS, tardive dyskinesia, akathisia, neuroleptic malignant

syndrome

Negative impact on course of illness– ↑ post-mania depressive symptom severity– ↑ frequency of major depressive episodes

Vietta, et al. 2010. Muralidharan K, et al. J Affect Disord. 2013;150(2):408-414. Goikolea JM, et al. Eur Neuropsychopharmacol. 2013;23(4):305-316. Kane JM. J Clin Psychiatry. 1999;60 Suppl 5:43-49.

Clozapine in Refractory Bipolar Disorder

• Open trials suggest mood stabilization and polypharmacy reduction

• Faster antimanic response for clozapine (n=15, mean dose ~ 300 mg) vs chlorpromazine (n=12, mean dose 166 mg) in a 3-week open randomized trial of lithium resistant mania

• Adjunctive clozapine, in comparison to treatment as usual– Significant reduction in polypharmacy– Significant improvement in symptoms of mania, positive and

negative symptoms in a group of treatment-resistant bipolar outpatients (n=38)

Frye MA, et al. J Affect Disord. 1998;48(2-3):91-104. Barbini B, et al. Int Clin Psychopharmacol. 1997;12(2):109-112. Suppes T, et al. Am J Psychiatry. 1999;156(8):1164-1169.

Clozapine Reduced Recurrent Suicidal Behavior

Meltzer HY, et al. Arch Gen Psychiatry. 2003;60(1):82-91.

Lithium in Acute Mania

• Gold standard – benchmark• Lithium non-response differs from other

mood stabilizers• Clinical predictors account for < 50% of

variance, suggesting genetic factors• Prophylactic response familial• Numerous side effects, narrow therapeutic

index• Believed to reduce suicide rates via

unknown mechanism

Frye MA, et al. J Clin Psychopharmacol. 1998;18(6):461-464. Goodwin FK. JAMA. 1990;264(8):950. American Psychiatric Association. Practice Guideline. 2002. Bowden CL, et al. JAMA. 1994;271(12):918-924.

Advertisement from Harper’s New Monthly Magazine, 1892, from the author’s collection

Variable Lithium Response Rate

Frye MA, et al. J Affect Disord. 1998 Mar;48(2-3):91-104.

RapidCycling

NonrapidCycling

MixedMania

EuphoricMania

SubstanceAbuse

NoSubstance

Abuse

(-) FamilyHistory

(+) FamilyHistory

>3Episodes

Few LifetimeEpisodes

DMIPattern

MDIPattern

D

D

M

M

PoorResponse30%

GoodResponse70%

Based on Bipolar Subtype

DMI = Depression mania euthymic ininterval. MDI = Mania depression euthymic interval

Lithium and Suicidality

• Meta-analysis 48 RCTs– 23 lithium vs placebo– 13 lithium vs active comparator

• N=6674 (12 studies unipolar patients,19 bipolar patients)• Lithium reduced risk of suicide (OR 0.13, 95% CI 0.03–0.66) • Lithium reduced all cause mortality (OR 0.38, 95% CI 0.15–0.95)• No effect on non suicidal self injury (OR 0.6, 95% CI 0.27–1.32)• Mood dependent vs independent (impulsivity)

Cipriani A, et al. BMJ. 2013;346:f3646. Mitka M. JAMA. 2013;310(4):360.

Maintenance Treatment of Bipolar DisorderDifferential Response to Lithium and Carbamazepine

BP I = bipolar I disorder; BP II = bipolar II disorder; BP NOS = bipolar disorder not otherwise specified.Greil W, et al. J Clin Psychopharmacol. 1998;18(6):455-460.

Carbamazepine Levels Correlation with Improvement

• Anticonvulsant serum levels (4–12 mcg/mL)• Mood stabilization serum levels unclear

– plasma carbamazepine (n=10, r=.21, ns) – plasma-10, 11 epoxide (n=10, r=.62, P<.06)– CSF carbamazepine (n=10, r=.23, ns)– CSF-10, 11 epoxide (n=10, r=.67, P<.01)

• Induction of CYP450 3A3/4– Decreases serum concentrations of many medications– Autoinduction 3 to 5 weeks (ie, after hospital discharge) with

need to adjust dose

Centorrino F, et al. Bipolar Disord. 2003;5(5):370-374. Bowden CL. J Clin Psychiatry. 1996;57 Suppl 13:4-9. Review. Post RM, et al. Arch Gen Psychiatry. 1983;40(6):673-676.

Valproate for ManiaDose-Response Effect

• Prospective study of 374 patients with acute mania stratified into 6 groups based on valproate serum level ranges (lowest level < 55.0 mcg/mL)

Allen MH, et al. Am J Psychiatry. 2006;163(2):272-275.

RESULTS• Linear relationship between

valproate serum level and therapeutic response

• Efficacy significantly > placebo beginning at 71.4–85.0 mcg/mL

• Efficacy was associated with highest valproate serum levels (> 94 mcg/mL)

Divalproex and Carbamazepine in Acute ManiaPros

– Effective in manic and mixed episodes– Effective in alcohol withdrawal and relapse prevention– Several effective in migraine prevention

Cons– Ineffective in acute mania (LTG, TPX, GBP)– P450 3A/4 hetero-induction– Weight gain and endocrine disturbances (VAL)– Teratogenicity (VAL, CBZ)– Rash risk

CBZ = carbamazepine; GBP = gabapentin; LTG = lamotrigine; TPX = topiramate; VAL = valproate.Novick, et al. 2009. Goodwin, et al. 2010. Frye, et al. 2006. Harden, et al. 2009. Goodwin, et al. 2009. Jiang, et al. 2009.

Other Anticonvulsant Drugs

• Oxcarbazepine– 1 negative randomized, double-blind, placebo-controlled trial– No placebo-controlled studies in adults

• Lamotrigine– 2 unpublished negative trials

• Gabapentin – Negative placebo-controlled add-on study (LI, VPA)

• Topiramate– 4 negative placebo-controlled trials

Wagner KD, et al. Am J Psychiatry. 2006;163(7):1179-1186. Rosa AR, et al. CNS Neurosci Ther. 2011;17(3):167-177. Pande AC, et al. Bipolar Disord. 2000;2(3 Pt 2):249-255. Kushner SF, et al. Bipolar Disord. 2006;8(1):15-27.

ECT for Acute Mania

• ECT is a mood stabilizer

• 2 controlled studies of acute mania• ECT vs lithium• ECT vs lithium + haloperidol

• ECT reported significant benefits for acute mania

Mukherjee S, et al. Convuls Ther. 1988;4(1):74-80. Small JG, et al. Arch Gen Psychiatry. 1988;45(8):727-732.

Target Dose Range for Acute Mania

Mood Stabilizer: Safety and Tolerability Concerns

In: Ketter TA (ed). Advances in the Treatment of Bipolar Disorder. 2005. Physician’s Desk Reference. 2008.

Antipsychotic: Safety and Tolerability Concerns

In: Ketter TA (ed). Advances in the Treatment of Bipolar Disorder. 2005. Physician’s Desk Reference. 2008.

Frye MA. N Engl J Med. 2011;364(1):51-59.

Bipolar Depression: Best Practices

• FDA approved– Olanzapine fluoxetine (OFC)– Quetiapine monotherapy– Lurasidone mono/adjunct– Cariprazine

• Maximize the mood stabilizer• Antidepressants FDA off-label

– Do they work? Are they safe?• Psychotherapy• Novel treatment

FDA off-label: Antidepressants are not indicated for treatment of bipolar depression.

The Old Guitarist Pablo Picasso 1903 The Blue Period

Response Rates of Atypical Antipsychotics in Bipolar Depression

OFC = olanzapine/fluoxetine combination. *P<.05; †P<.001 vs placebo. Calabrese JR, et al. Am J Psychiatry. 2005;162(7):1351-1360. Thase ME, et al. J Clin Psychopharmacol. 2006;26(6):600-609. Tohen M, et al. Arch Gen Psychiatry. 2003;60(11):1079-1088. Thase ME, et al. Psychopharmacol. 2009;29(1):38. Sachs GS, et al. J Clin Psychiatry. 2011;72(10):1413-1422.

Res

po

ns

e (%

)

0

10

20

30

40

50

60

70

OLZ/OFC Quetiapine Aripiprazole Ziprasidone

Placebo Active Active

†

*

†† * *

10 mg / 7.5 mg40mg 300 and 600 mg ~17 mg ~90 mg

PREVAIL 2: Results

Loebel A, et al. Am J Psychiatry. 2014;171(2):160-168.

Cariprazine vs Placebo in Bipolar I Depression

aMixed-effects model for repeated measures, intent-to-treat population; P values were not adjusted for multiple comparisons. Cariprazine0.75 mg/day compared with placebo: *P<.05; **P<.01; ***P<.001. Cariprazine 1.5 mg/day compared with placebo: †P<.05; ††P<.01; †††P<.001. Cariprazine 3.0 mg/day compared with placebo: #P<.05; ##P<.01; ###P<.001. Durgam S, et al. Am J Psychiatry. 2016;173(3):271-281.

Meta-Analysis Lamotrigine in Acute Bipolar Depression

Geddes JR, et al. Br J Psychiatry. 2009;194(1):4-9. van der Loos ML, et al. J Clin Psychiatry. 2009;70(2):223-231.

Favors DrugFavors Placebo

0.371223 Risk Ratio 2.6938

1.26 (1.10,1.44)Overall (95% CI)

8.81.63 (1.05,2.53)LAMLIT

20.71.26 (0.95,1.67)SCA10022

19.91.24 (0.91,1.70)SCA30924

21.71.09 (0.81,1.48)SCA40910

20.6 1.11 (0.83,1.48)SCAA2010

8.31.71 (1.08,2.69)SCAB2001

Weight(%)

Risk Ratio(95% CI)Study

Meta-Analysis Divalproex in Acute Bipolar Depression

Muzina DJ, et al. J Clin Psychiatry. 2011;72(6):813-819. Davis LL, et al. J Affect Disord. 2005;85(3):259-266. Ghaemi SN, et al. J Clin Psychiatry. 2007;68(12):1840-1844.

Relative risk of remission in patients treated with divalproex vs placebo

Maximize the Mood Stabilizer Lithium and Bipolar Depression

IMI = impramine; Li = lithium; PAR = paroxetineNemeroff CB, et al. Am J Psychiatry. 2001;158(6):906-912.

*

n=19 n=19

n=22

n=14n=17

n=21

0

-2

-4

-6

-8

-10

-12

Li + PARn=33

Li + IMIn=36

Li Onlyn=43

Me

an

Ch

an

ge

in H

AM

-D S

co

re Li+ < 0.8 mEq/L Li+ 0.8 mEq/L

Antidepressants Most Common Initial Treatment for Patients with Bipolar Disorder in the United States in 2002–2003

Baldessarini RJ, et al. Psychiatr Serv. 2007;58(1):85-91.

Antidepressants Not Effective for Bipolar Depression

• Meta-analysis 16 studies acute antidepressant Rx vs placebo or active comparator in BD-I / BD-II depressed patients (n=3113)

• The pooled treatment estimates– Clinical response (RR=1.17, 95% CI, 0.88–1.57; P=.28)– Clinical remission (RR=1.14, 95% CI, 0.90–1.45; P=.28)

• Pooled treatment estimates for 1000 patients– No increase risk of switch

• In smaller analysis– 43% TCA, 15% venlafaxine, 7% SSRI, 5% bupropion

Sidor MM, et al. J Clin Psychiatry. 2011;72(2):156-167. Sidor MM, et al. Curr Psychiatry Rep. 2012;14(6):696-704.

Depressive Episode Relapse with Antidepressant Discontinuation

Altshuler L, et al. Am J Psychiatry. 2003;160(7):1252-1262.

Risk Factors for Switch

• Mixed depression• TCA vs venlafaxine• History of antidepressant-induced mania (AIM)• Absence of antimanic mood stabilizer

– First 3 months associated with greatest liability• Low thyroid stimulating hormone (with TCAs)• Polymorphism (s/s or s/l) at 5-HTTLPR• Hyperthymic temperament• Comorbid alcoholism• Female gender and comorbid anxiety disorder• Age (peripubertal > adolescents)• BD-I > BD-II

Viktorin A, et al. Am J Psychiatry. 2014;171(10):1067-1073. Frye MA, et al. Am J Psychiatry. 2009;166(2):164-172.

Baseline Mixed Depression Associated with Treatment Emergent Mania

• Prior to Antidepressant Treatment• 3 YMRS items significantly higher in

TEM– motor-energy– speech– thought content

• Factor analysis to identify clusters of YMRS items that covaried and analysis of variance only identified motor/verbal activation (F(2,169)=3.99, P=.02)

TEM = Treatment Emergent Mania; YMRS = Young Mania Rating Scale.Frye MA, et al. Am J Psychiatry. 2009;166(2):164-172.

DSM-5 Mixed Specifier

Akiskal HS, et al. J Affect Disord. 2000;59 Suppl 1:S5-S30.

Mayo Clinic Individualized MedicineBiobank for Bipolar Disorder

SLC6A4 Polymorphism and Antidepressant Induced Mania

SLC6A4 S Allele and AIM: Meta-Analysis Results

Frye MA, et al. J Clin Psychiatry. 2015;76(2):174-180.

Meta-analysis marginally significant evidence of association between S allele and AIM+ (P=.059)

Pharmacogenomic Haplotype Analysis L-A-10 Protective

Frye MA, et al. J Clin Psychiatry. 2015;76(2):174-180.

Haplotype analysis suggests an association between AIM and haplotypes composed of the 5HTTLPR, rs25531, and the intron 2 VNTR in the SLC6A4 gene, with the L-A-10 haplotype being associated with reduced risk of AIM

6-Week, Randomized Placebo-Controlled Evaluation of Adjunctive Modafinil for Bipolar Depression

Frye MA, et al. Am J Psychiatry. 2007;164(8):1242-1249.

N = 85 Bipolar I/II depression Inadequate response to mood stabilizers ± AD Rx

8-Week Randomized Double-Blind Adjunctive Armodafinil in Acute Bipolar I Depression: Results

aResponse: ≥ 50% IDS-C30 decrease.Calabrese JR, et al. J Clin Psychiatry. 2014;75(10):1054-1061.

Meta-Analysis of Modafinil/Armodafinil in Bipolar Depression

Nunez NA, et al. Accepted for Presentation 2019 New Research American Society of Clinical Pharmacology (ASCP).

Ketamine and Other NMDA Antagonists: Early Clinical Trials and Possible Mechanisms in Depression

The A) top plot shows results one day after initiation of ketamine (heterogeneity: χ2=4.27, df=4, P=.51, I2=0%). The B) bottom plot shows results 1 week after initiation of ketamine (heterogeneity: χ2=1.14, df=5, P=.95, I2=0%).Newport DJ, et al. Am J Psychiatry. 2015;172(10):950-966.

Ketamine for Treatment-Resistant Bipolar Depression – Replication

• Ketamine noncompetitive NMDA antagonist

• FDA approved as a general anesthetic

• 0.5 mg/kg over 40 minutes vs 1 infusion of saline placebo

• Almost immediate reductions in depression rating scores

Zarate CA Jr, et al. Biol Psychiatry. 2012;71(11):939-946.

Adjunctive Pramipexole in Acute Bipolar Depression(Pooled) 6-week Randomized Double-Blind

Response: ≥ 50% HDRS/MADRS decrease.Goldberg JF, et al. Am J Psychiatry. 2004;161(3):564-566. Zarate CA Jr, et al. Biol Psychiatry. 2004;56(1):54-60.

Sit DK, et al. Am J Psychiatry. 2018;175(2):131-139.

Midday bright light therapy 7000-lux, 4000K, 60 min/day (23 enrolled/completed)

ECT Bipolar Depression

• 6-week, 6-site, randomized trial of 3×/week RUL ECT vs algorithm based pharmacologic treatment (n=73)– Response rate 74% (17/23) vs 35% (7/20, P<.01)

• Bitemporal generally acknowledge to have greater efficacy and side effects

Schoeyen HK, et al. Am J Psychiatry. 2015;172(1):41-51. Tohen M, et al. Am J Psychiatry. 2015;172(1):3-5. Kotzalidis GD, et al. Am J Psychiatry. 2015;172(3):294.

TMS Meta-Analysis

34 StudiesN=1383 patients

TMS superior to sham for MDD

Effect size 0.55 (0.38 to 0.72) (P<.001)

Slotema CW, et al. J Clin Psychiatry. 2010;71(7):873-884.

TMS in Bipolar Depression

• Meta-analysis of 19 TMS studies in bipolar depression (N=181)– Stimulation Targets: Left, Right, Bilateral DLPFC– High vs low or sequential stimulation frequency– Response: TMS 44% (47/106) vs Sham 25% (19/75, P0.01)

• Bilateral sequential (1 Hz Rdlpfc → 10 Hz Ldlpfc) vs sham rTMS for 4 weeks (N=49)– No significant difference in baseline to end point change, response or

remission rates• Substantial clinical trial design heterogeneity

– Stimulation target– Laterality– High (10 Hz) vs low (1 Hz) stimulation

McGirr A, et al. World Psychiatry. 2016;15(1):85-86. Fitzgerald, et al. 2016.

Significant Reduction in Bipolar Depressive Symptoms with dTMS

Tavares DF, et al. Neuropsychopharmacology. 2017;42(13):2593-2601.

TSH and with Depressive Relapse in Lithium Maintained Patients with Bipolar

Frye MA, et al. Acta Psychiatr Scand. 2009;120(1):10-13.

Free T4 and Depressive Severity inLithium Maintenance

P<.01; Beck Depression Inventory 10–16 = mild depressionFrye MA, et al. Am J Psychiatry. 1999;156(12):1909-1914.

Adjunctive Levothyroxine in Bipolar Depression

*P<.05 vs placebo (ITT; LOCF). Adjunctive levothyroxine (300 µg/day) or placebo in patients with bipolar I or II disorderHAM-D = Hamilton Rating Scale for Depression.Stamm TJ, et al. J Clin Psychiatry. 2014;75(2):162-168.

Intensive Psychotherapies Improve Bipolar Depression

• N=293 bipolar depressed outpatients• Protocol medications + 9 months:

– FFT (family-focused therapy)– IPSRT (interpersonal and social rhythm

therapy)– CBT (cognitive-behavioral therapy)– CC (collaborative care)

• Intensive psychotherapies – Higher recovery rate– Shorter time to recovery– 1.6× more likely to be clinically well

during any study month

Miklowitz DJ, et al. Arch Gen Psychiatry. 2007;64(4):419-426.

Time to Recovery (Days)

0.0

0.2

0.4

0.6

0.8

0 100 200 300 400

CCCBT

FFTIPSRT

Maintenance of Antidepressant Response after Group IPSRT Group for Bipolar Disorder

*P<.05, N=6. YMRS = Young Mania Rating Scale; IDS-C = Inventory of Depressive Symptomatology-Clinician Rated; BDI-II = Beck Depression Inventory-II.Hoberg AA, et al. Perspect Psychiatr Care. 2013;49(4):226-234.

6

3533

2

1820

3

21

15

0

5

10

15

20

25

30

35

40

YMRS IDS-C BDI-II

Baseline

Post (2 wk)

12-week

* * **

Conclusions• Optimize first treatment before defining failure• Assess residual symptoms and augment switch accordingly

– Psychosis – AAP clozapine– Euphoric – complimentary mood stabilizer– Mixed specifier – anticonvulsant, AAP, reduce AD Rx– Comorbidity (migraine, smoking cessation, Etoh)

• Meta-analysis can guide next steps– Lamotrigine and divalproex acute bipolar depression– Lithium suicidality– Armodafinil / modafinil bipolar depression– Ketamine bipolar depression– Judicious use of antidepressants

• Evidence base does not support monotherapy use• Switch rate is not 0%

Thank You

Q&A