Sequential Parallel Comparison Design for Trials with High Placebo Response

Transcription

1 Sequential Parallel Comparison Design for Trials with High Placebo Response Anastasia Ivanova Department of Biostatistics UNC at Chapel Hill

2 Placebo response 2

3 Increased placebo response over time Data from 86 major depressive disorder trials submitted to FDA during 1986 to 2008 (Khin et al., 2011) + US trials - Non US trials 3

4 Diminished treatment effect over time. + US trials - Non US trials 4

5 Rising placebo response RCTs and failed trials Undurraga and Baldessarini 2012

6 Placebo response In many studies in depression and schizophrenia considered by FDA over 12 year period, 1987 through 1999, in which investigational drug could not be distinguished from placebo also included an active standard drug that could not be distinguish from placebo (Laughren, 2001). 6

7 Placebo response Yellow pills make the most effective antidepressants, like little doses of pharmaceutical sunshine Red pills can give you a more stimulating kick More is better Placebos taken four times a day deliver greater relief than those taken twice daily. More $ is better Waber, RL, Shiv, B, Carmon, Z, Ariely, D (2008). Commercial 7 features of placebo and therapeutic efficacy. JAMA 299:

8 Partial List of Illnesses Cited as Having Significant Placebo Responses Organized by FDA Center for Drug Evaluation and Research Review Divisions Anesthesia, Analgesic and Rheumatology Arthritis Pain Psoriatic arthritis Rheumatic diseases Anti-viral Herpes simplex Cardiovascular and Renal Heart failure, congestive Hypertension Ear, Nose and Throat Tinnitus Gastroenterology Crohn s disease Dyspepsia and gastric motility Gastric and duodenal ulcers GERD Irritable bowel syndrome Gastroenterology (cont.) Nausea Obesity Reflux esophagitis Ulcerative colitis Neurology Alzheimer s Disease Autism Epilepsy Headache Migraine Multiple Sclerosis Parkinson s disease Restless leg syndrome Psychiatry ADHD Anxiety disorders Binge eating disorder Bipolar mania Psychiatry (cont.) Chronic Fatigue syndrome Depression Insomnia Panic disorders Schizophrenia Social Phobia Pulmonary and Allergy Allergies Asthma Cough Cystic Fibrosis Food allergy Reproductive and Urologic Benign prostatic enlargement Erectile dysfunction Premenstrual dysphoric disorder Sexual dysfunction, women Vulvar vestibulitis Urinary incontinence 8

9 Design options to lower placebo response Option 1. Parallel single stage design R A N D O M I Z E PLACEBO DRUG 9

10 Option 2: The Placebo Lead-in PLACEBO No Response R A N D O M I Z E PLACEBO DRUG Use placebo lead-in (run-in) to eliminate placebo responders 10

11 Advantages of Placebo Lead-In Advantages Possible increase in power from larger effect size in placebo non-responders 11

12 Placebo response Impact of the placebo lead-in duration on the placebo response in 86 major depressive disorder trials 30 trials without placebo lead-in had an average HAMD total and SD 1.87, but 56 MDD trials with placebo lead-in had an average HAMD total -7.6 and SD

13 Disadvantages of Placebo Lead-In Disadvantages Longer trial duration Might fail to eliminate placebo responders Analysis has shown that placebo lead-in periods (at least those that are single blind) rarely deliver their theoretical benefit. Trivedi and Rush (1994) reported that meta-analyses of 101 antidepressant studies reveal that a placebo lead-in does NOT (1) lower the placebo response rate, (2) increase the drug-placebo difference, or (3) affect the drug response rate post-randomization. The number of patients recruited is larger than the n of the trial 13

14 Option 3: The Sequential Parallel Comparison Design (SPCD) R A N D O M I Z E PLACEBO DRUG No Response R A N D O M I Z E PLACEBO DRUG 14

15 Option 3: The Sequential Parallel Comparison Design Sequential Parallel Comparison Design (SPCD) is a clinical trial methodology developed in 2003 in Massachusetts General Hospital (Fava et al., 2003) after parallel first stage, re-study the patients who do not respond to placebo Unlike a placebo lead-in, all patients are utilized and, some patients are utilized twice Massachusetts General Hospital holds a portfolio of patents related to SPCD, licensed by PPD. 15

16 The Sequential Parallel Comparison Design SPCD is sometimes referred to as Sequential Parallel Design (SPD) Sequential Parallel Design with Re- Randomization or SPD - ReR (Chen et al., 2011) Doubly Randomized Delayed-Start Design (Liu et al., 2012) 16

17 SPCD: defining the outcome R A N D O M I Z E q 1 PLACEBO DRUG p 1 No Response R A N D O M I Z E q 2 PLACEBO DRUG p 2 Let q 1 and p 1 be placebo and drug response rates in Stage 1 Let q 2 and p 2 be placebo and drug response rates in Stage 2, that is, among placebo non-responders 17

18 SPCD: Hypothesis testing Parallel single stage trial: population = all comers H 0 : p 1 = q 1 Placebo lead-in: population = placebo non-responders H 0 : p 2 = q 2 SPCD all comers placebo non-responders H 0 : p 1 = q 1 p 2 = q 2 H 1 : p 1 > q 1 OR p 2 > q 2 18

19 Advantages Advantages of the SPCD Unlike placebo lead-in, no eligible patients are recruited and then not used Increase in power for any given sample size From potentially larger effect size in placebo non-responders From reuse of patients More responses are observed compared to parallel design or placebo lead-in For any given power, overall trial duration is typically shorter because sample size is smaller 19

20 Disadvantages Disadvantages of the SPCD Longer trial duration for individual subjects compared to the parallel design and placebo leadin (because lead-in phase is usually shorter than full follow-up for response) Some controversy regarding hypothesis being tested. However, - Use of placebo non-responders (as in placebo lead-in) is well accepted for Phase 3 by FDA - 9 completed SPCD MDD trials and 9 ongoing trials, including completed and ongoing pivotal trials

21 SPCD data analysis Linear combination test (Fava et al., 2003) ˆ θ = pˆ qˆ ˆ θ = pˆ qˆ T = w ˆ θ ˆ 1 + (1 w) θ2 w Var( ˆ ) (1 w) Var( ˆ ) 2 2 θ1 + θ1 Since cov( ˆ θ, ˆ θ ) 1 2 asymptotically, T~N(0,1) 21

22 SPCD data analysis 1. Linear combination test (Fava et al., 2003) test parameter w ˆ θ + (1 w) ˆ θ 2. Score test (Ivanova, Qaqish, Shoenfeld, 2011), binary outcome T = 1 2 w Var( ˆ ) (1 w) Var( ˆ ) 2 2 θ1 + θ1 ( ) / ( ) r = p q p q test parameter 3. Weighted combination of Z-scores (Liu et al., 2012) T = wz + 1 wz 1 2 test parameter where Z 1 is the test statistic for stage 1 and Z 2 is for stage 2

23 Ziprasidone (max 160 mg/day) vs. placebo for depression n = 120 Actual Response Rate Drug Placebo Difference Stage % 31.9% 12.9% Stage % 28.0% -4.2% * Conventional Design p-value = 0.19 SPCD p-value = 0.42 (response rates of Stage 1, using 50:50 randomization) Note: All p-values are two-sided. SPCD p-values are obtained using the score test with an r = 1 (Ivanova et al., 2011). Single stage p-values are obtained using the Fisher s exact test.

24 L-methylfolate (7.5 mg/day) vs. placebo for SSRI-resistant major depression n = 148 Funded by Pamlab, Inc. Actual Response Rate Drug Placebo Difference Stage % 28.5% -9.1% Stage % 9.1% 8.0% * Conventional Design p-value = 0.12 * SPCD p-value = 0.96 (response rates of Stage 1, using 50:50 randomization) 24

25 ADAPT-A: Aripiprazole Augmentation of SSRIs for major depression with inadequate antidepressant therapy response n = 221 Funded by Bristol-Myers Squibb Actual Response Rate Drug Placebo Difference Stage % 17.4% 1.1% Stage % 7.9% 10.1% 9X Conventional Design p-value = 0.86 SPCD p-value = 0.19 (response rates of Stage 1, using 50:50 randomization) 25

26 L-Methylfolate (15 mg/day) Augmentation of SSRIs vs. placebo for SSRI-resistant major depression n = 75 Funded by Pamlab, Inc. Actual Response Rate Drug Placebo Difference Stage % 19.6% 17.2% Stage % 9.5% 18.2% 1X Conventional Design p-value = 0.12 SPCD p-value = 0.03 Stage 1 p-value = 0.21 Stage 2 p-value = 0.22 (response rates of Stage 1, using 50:50 randomization) n = 19 (25%) : n = 56 (75%) Drug : Placebo 26

27 Example of a Dose-Finding Study with SPCD: Alkermes Phase 2 Study of ALKS 5461 in Depression ALKS 5461, a novel opioid modulator, in patients with major depressive disorder and inadequate response to standard therapies Primary Endpoint - Hamilton Depression Rating Scale (HAM-D17) Secondary Endpoints - Montgomery-Åsberg Depression Rating Scale (MADRS) - Clinical Global Impression (CGI-S) 27

28 Alkermes Phase 2 Study of ALKS 5461 in Depression R A N D O M I Z E n = 142 9:2:2 n = 99 PLACEBO n = 22 LOW DOSE n = 21 HIGH DOSE No Response R A N D O M I Z E n = 20 PLACEBO n = 23 LOW DOSE n = 22 HIGH DOSE 4 wk trt 1 wk taper 4 wk trt 1 wk taper 28

29 Alkermes Phase 2 Study of ALKS 5461 in Depression Response rate Low+High dose Placebo Difference SPCD Stage % 25.3% 16.6% SPCD Stage % 5.0% 32.8% 2X Stage 1 p-value = Stage 2 p-value = SPCD p-value =

30 A Two-way Enriched Design (TED) (Ivanova and Tamura, 2011) R A N D O M I Z E PLACEBO DRUG No Response Response R A N D O M I Z E R A N D O M I Z E PLACEBO DRUG PLACEBO DRUG 30

31 Sequential Enriched Design (SED) (Chen and Tamura, 2014) PLACEBO No Response R A N D O M I Z E PLACEBO DRUG Response R A N D O M I Z E PLACEBO DRUG 31

32 References Chen Y.F., Yang Y., Hung H., Wang S. Evaluation of performance of some enrichment designs dealing with high placebo response in psychiatric clinical trials. Contemporary Clinical Trials 2011; 32: Doros G, Pencina M, Rybin D, Meisner A, Fava M. A repeated measures model for analysis of continuous outcomes in sequential parallel comparison design studies. Statistics in Medicine in press. Fava, M, Schoenfeld, D. U.S. provisional Pat. App. No. 60/459,517, filed March 31, 2003; U.S. Pat. App. No. 10/814,852, filed March 31, Corresponding U.S. Patents include U.S. Pat. No. 7,647,235, issued January 12, 2010; U.S. Pat. No. 7,840,419, issued November 23, 2010; and U.S. Pat. Nos. 7,983,936; 8,145,504; 8,145,505; and 8,219,419. Fava M., Evins A, Dorer D., Schoenfeld D. The problem of the placebo response in clinical trials for psychiatric disorders: culprits, possible remedies and a novel study design approach. Psychotherapy and Psychosomatics 2003; 72: Fava, M, Mischoulon D, Iosifescu D, Witte J, Pencina M et al. A double-blind, placebocontrolled study of Aripiprazole adjunctive to antidepressant therapy among depressed outpatients with inadequate response to prior antidepressant therapy (ADAPT-A Study). Psychotherapy and Psychosomatics 2012; 81, Huang X., Tamura R. Comparison of test statistics for the sequential parallel design. Statistics in Biopharmaceutical Research 2010; 2:

33 References Ivanova, A., Qaqish, B., Schoenfeld D. Sample size and power calculations for the sequential parallel design. Statistics in Medicine 2011; 30: Ivanova A, Tamura RN. A two-way enriched clinical trial design: combining advantages of placebo lead-in and randomized withdrawal. Statistical Methods in Medical Research 2011; available in Early View. Laughren TP. The scientific and ethical basis for placebo-controlled trials in depression and schizophrenia: an FDA perspective. European Psychiatry 2001; 16: Liu Q, Lim P, Singh J, Lewin D, Schwab B, Kent J. Doubly randomized delayed-start design for enrichment studies with responders or nonresponders. Journal of Biopharmaceutical Statistics. 2012; 22(4): Papakostas GI, Shelton RC, Zajecka JM, Etemad B, Rickels K, Clain A, et al. L- methylfolate as adjunctive therapy for SSRI-resistant major depression: results of two randomized, double-blind, parallel-sequential trials. Am J Psychiatry 2012; 169, Silberman, S Placebos are getting more effective. Drugmakers are desperate to know why. Wired Magazine 2009, Tamura RN, Huang X. An examination of the efficiency of the sequential parallel design in psychiatric clinical trials. Clinical Trials: Journal of the Society of Clinical Trials 2007; 4: Trivedi MH, Rush AJ. Does a placebo run-in or placebo treatment cells affect the 33 efficacy of antidepressant medications? Neuropsychopharmacology 1994; 11:33-43.