Open-label surgical trials for Parkinson disease: Time for reconsideration

POINT OF VIEW

Open-Label Surgical Trials for Parkinson Disease: Time for Reconsideration

I

nsanity: doing the same thing over and over again and expecting different results. —Albert Einstein

Six consecutive times in the past decade, a surgically administered experimental therapy for Parkinson disease (PD) that had shown great promise in the laboratory and positive results in open-label clinical trials1–7 failed to demonstrate a statistically significant clinical benefit greater than placebo when subsequently tested in a randomized, double-blind, sham surgery-controlled trial.8–13 These include trials testing transplantation of human embryonic mesencephalon,1,2,8,9 transplantation of embryonic porcine mesencephalon,3,10 transplantation of human retinal pigmented epithelial cells mounted on gelatin microcarriers,6,12 intraputaminal infusion of glialderived neurotrophic factor (GDNF),4,5,11 and intraputaminal gene delivery of the trophic factor neurturin using an adeno-associated viral vector (AAV2-NTN).7,13 When the first of these double-blind trials was proposed in the 1990s, the ethics of conducting clinical trials with sham surgery controls was hotly debated.14–16 Today, the importance of evaluating experimental surgical therapies for PD with double-blind controlled trials is more widely accepted17,18 because: (1) PD patients who participate in surgical trials are particularly susceptible to placebo response19,20; (2) dopamine appears to play a central role in the placebo response21–23; (3) double-blind sham-controlled clinical trials have repeatedly contradicted the findings of open-label studies; (4) to date no patient has been seriously harmed by undergoing a sham procedure for PD24; and (5) patients, institutional review boards, and clinical researchers understand and accept the need for sham surgery controls in trials of surgical interventions for PD.17,25 In truth, had these sham surgery-controlled trials not been conducted, several of the surgical therapies for PD listed above might now be widely employed based solely on unsubstantiated open-label observations. The cost of these failed double-blind trials to patients, investigators, study sponsors, and society are substantial. Sham surgery-controlled trials involve relatively large numbers of patients, and are time consuming and expensive. They impose significant opportunity costs on researchers and funding sources by diverting scarce

resources and investigator effort from other possibly more fruitful research. Moreover, patients with PD who participate in these trials must delay or forego other established therapies (eg, deep brain stimulation); and patients receiving the active treatment are exposed to both the known risks of an intracranial procedure and the unknown risks of an experimental intervention that may have no true therapeutic value. Finally, these highly publicized repeated failures might negatively impact upon patient recruitment for future trials of more promising agents. These 6 negative results not only demonstrate the value of double-blind sham surgery-controlled trials for testing new surgical therapies for PD; they also suggest that the methodologies currently employed for the initial evaluation of such therapies are flawed, as each of these initial trials was ultimately shown to have yielded a falsepositive result (type I error). False-negative results (type II error) in the double-blind studies are much less likely to have occurred, as those studies employed similar entry criteria and protocols, and included larger sample sizes. Although it is true that the primary role of an early or phase I trial is to assess the safety and tolerability of a new intervention rather than its efficacy, it is also true that clinical results derived from such trials are routinely used both to determine whether it is worthwhile proceeding with larger controlled trials and to calculate the necessary sample size required to achieve a statistically significant result. Therefore, inaccurate measures of clinical response during an initial open-label trial can have serious consequences, including the needless performance of a costly, risky, and underpowered double-blind trial if the response is overestimated and the abandonment of a truly valuable therapy if the response is underestimated. In an attempt to understand this recent pattern of failure better, we collected the clinical results from both the initial open-label and subsequent sham surgery-controlled, blinded trials of each of these therapies (Table). The data were extracted either from the published literature (11 studies) or through the cooperation of the study investigators (2 studies) (P. Kaplan, personal communication; R. Gross and C. W. Olanow, personal communication). Whenever possible, we recorded the motor C 2011 American Neurological Association V 5

6

12 Cere-120

F/U ¼ follow-up; GDNF, glial-derived neurotrophic factor; H&Y, Hoehn and Yahr scale;; Hu ¼ human; Imp ¼ improved; N/A, not available; UPDRS, Unified Parkinson Disease Rating Scale.

2 deaths (unrelated to surgery or therapy) 17% 20% 12 UPDRS-III (off ) 38 20 36% 12

22% 21% 12 UPDRS-III (off ) 35 36 48% 12 UPDRS-III (off ) 6 Spheramine

45% 12 UPDRS-III (off ) 10

UPDRS-III (off )

6 deaths; 1 off-state dyskinesia

25% serious device-related complications 4.5% 10% 6 UPDRS-III (off ) 17 17 57% UPDRS-III (off ) 5 Intraputaminal GDNF

24

Unknown 31% 32% 18 UPDRS-III (off ) 9 9 15% 12 12 Porcine fetal

UPDRS-III (off )

Uncontrolled Dyskinesia in 56.5% 7.3% UPDRS-III (off ) 11 11 30% 20.5 UPDRS-III (off ) 6 Hu fetal 2

N/A H&Y (off ) 7 Hu fetal 1

15% with 90% confidence. In our opinion, this is the minimum standard that is acceptable for physicians and patients to determine if the potential risks of an experimental procedure warrant proceeding with a larger controlled trial. Of note is that more recent preliminary surgical trials for PD have included this minimum number of patients. • Staggered enrollment. We recommend that patients be operated on in groups of 2 to 4, with sufficient time intervals between stages to provide the investigators and the data safety monitoring board an

July 2011

•

•

•

•

opportunity to detect serious risks before exposing the next group of patients to the therapy. Study duration of at least 1 year. An experimental surgical therapy for PD may require a relatively long period of time before beneficial or adverse effects can be accurately determined. Studies lasting a year or more will allow for the resolution of transient perioperative complications, and permit a better assessment than short-term studies as to whether the potential benefits and risks associated with a given surgical therapy justify performing further studies. Again, this time frame is already standard for surgical trials of novel PD therapies. Randomization. Although it may be difficult to justify sham surgery controls in initial trials of novel interventions, patients who meet inclusion criteria for the study could be randomized to active treatment or a control group that receives best medical therapy. Blinded evaluations. Both placebo effect and observer bias have the potential to inflate estimates of clinical response and underestimate side effects in open-label trials of experimental PD therapies. Controlling for observer bias can be accomplished easily and inexpensively by using a single-blind, doubleobserver study design. Here, a separate, blinded investigator who is independent of the surgeon and the treating investigator performs all evaluations related to safety and efficacy. Subjects can be gowned and capped to mask any evidence of a surgical procedure. Alternatively, subjects can undergo standardized videotape examinations, which can then be randomly distributed for blinded evaluation by independent raters. A comparison to historical controls is simply not sufficient. Full ascertainment. It is important to ensure full ascertainment of the study cohort so as not to lose track of patients who might fail to return for follow-up because they are experiencing a bad outcome. Even if patients wish to withdraw from the trial, every effort should be made to continue to follow each subject to the study’s conclusion, including making visits to the patient’s place of residence if they cannot or are unwilling to return to the medical center. Failure to achieve complete ascertainment could lead to an underestimation of adverse events or overestimation of clinical benefits associated with a given intervention.

We believe that these modest recommendations will permit a more accurate initial evaluation of experimental surgical therapies for PD and may also be applicable to other surgical fields. Applying these standards should reduce the number of patients who are unnecessarily subjected to the risks of an experimental therapy without a 7

ANNALS

of Neurology

reasonable expectation of benefit, and increase the likelihood that those therapies that are evaluated in doubleblind controlled trials will ultimately prove efficacious.

Acknowledgment We thank Dr K. Kieburtz for his constructive criticisms and Dr J. Godbold for his expert advice.

type 2-neurturin) to patients with idiopathic Parkinson’s disease: an open-label, phase I trial. Lancet Neurol 2008;7:400–408. 8.

Freed CR, Greene PE, Breeze RE, et al. Transplantation of embryonic dopamine neurons for severe Parkinson’s disease. N Engl J Med 2001;344:710–719.

9.

Olanow CW, Goetz CG, Kordower JH, et al. A double-blind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease. Ann Neurol 2003;54:403–414.

10.

GenVec. Genzyme and Diacrin report preliminary results from phase 2 trial of NeuroCell(TM)-PD. Available at: http://www.genvec.com/go.cfm?do¼Press.View&prid¼141. Accessed on January 14, 2010.

11.

Lang AE, Gill S, Patel NK, et al. Randomized controlled trial of intraputaminal glial cell line-derived neurotrophic factor infusion in Parkinson’s disease. Ann Neurol 2006;59:459–466.

12.

R initial phase IIb Titan Pharmaceuticals announces SpheramineV results. Available at: http://medicalnewstoday.com/articles/ 113908.php. Accessed on January 14, 2010.

13.

Marks WJ Jr, Bartus RT, Siffert J, et al. Gene delivery of AAV2neurturin for Parkinson’s disease: a double-blind, randomised, controlled trial. Lancet Neurol 2010;9:1142–1143.

14.

Freeman TB, Vawter DE, Leaverton PE, et al. Use of placebo surgery in controlled trials of a cellular-based therapy for Parkinson’s disease. N Engl J Med 1999;341:988–992.

15.

Macklin R. The ethical problems with sham surgery in clinical research. N Engl J Med 1999;341:992–996.

16.

Dekkers W, Boer G. Sham neurosurgery in patients with Parkinson’s disease: is it morally acceptable? J Med Ethics 2001;27:151–156.

17.

Kim SYH, Frank S, Holloway R, et al. Science and ethics of sham surgery: a survey of Parkinson disease clinical researchers. Arch Neurol 2005;62:1357–1360.

18.

Olanow CW. Double-blind, placebo-controlled trials for surgical interventions in Parkinson disease. Arch Neurol 2005;62: 1343–1344.

19.

Goetz CG, Wuu J, McDermott MP, et al. Placebo response in Parkinson’s disease: comparisons among 11 trials covering medical and surgical interventions. Mov Disord 2008;23:690–699.

20.

Kaptchuk TJ, Stason WB, Davis RB, et al. Sham device v. inert pill: randomized controlled trial of two placebo treatments. BMJ 2006; 332:391–397.

21.

de la Fuente-Ferna´ndez R, Ruth TJ, Sossi V, et al. Expectation and dopamine release: mechanism of the placebo effect in Parkinson’s disease. Science 2001;293:1164–1166.

Potential Conflicts of Interest R.L.A.: consultancy, Ceregene. M.T.: consultancy, St. Jude/ ANS, Ipsen; expert testimony, Bendin, Sumrall & Ladner, LLC, Paul, Weiss, Rifkind, Wharton, and Garrison, LLP; grants/grants pending, Ceregene, Solvay/Abbott, Allergan, St.Jude/ANS, Medtronic; speaking fees, Medtronic, Allergan, Glaxo, Novartis, Boehringer Ingelheim; payment for development of educational presentations, Medtronic. C.W.O.: consultancy, Ceregene, Schering, Novartis/Orion, Teva/Lundbeck, Abbott/Solvay; expert testimony, Welding Industry; stock/stock options, Ceregene, Clintrex.

Ron L. Alterman, MD Department of Neurosurgery, Mount Sinai School of Medicine, New York, NY

Michele Tagliati, MD Department of Neurology, Cedars Sinai Medical Center, Los Angeles, California

C. Warren Olanow, MD Departments of Neurology and Neuroscience, Mount Sinai School of Medicine, New York, NY

References 1.

Freed CR, Breeze RE, Rosenberg NL, et al. Survival of implanted fetal dopamine cells and neurologic improvement 12 to 46 months after transplantation for Parkinson’s disease. New Engl J Med 1992;327:1549–1555.

2.

Hauser RA, Freeman TB, Snow BJ, et al. Long-term evaluation of bilateral fetal nigral transplantation in Parkinson’s disease. Arch Neurol 1999;56:179–187.

22.

Pollo A, Benedetti F. The placebo response: neurobiological and clinical issues of neurological relevance. Prog Brain Res 2009;175: 283–294.

3.

Schumacher JM, Ellias SA, Palmer EP, et al. Transplantation of embryonic porcine mesencephalic tissue in patients with PD. Neurology 2000;54:1042–1050.

23.

Zubieta JK, Stohler CS. Neurobiological mechanisms of placebo response. Ann N Y Acad Sci; 2009;1156:198–210.

24.

4.

Patel NK, Bunnage M, Plaha P, et al. Intraputamenal infusion of glial cell line-derived neurotrophic factor in PD: a two-year outcome study. Ann Neurol 2005;57:298–302.

Frank S, Kieburtz K, Holloway R, Kim SY. What is the risk of sham surgery in Parkinson disease clinical trials? A review of published reports. Neurology 2005;65:1101–1103.

25.

5.

Slevin JT, Gash DM, Smith CD, et al. Unilateral intraputaminal glial cell-line derived neurotrophic factor in patients with Parkinson’s disease: response to 1 year of treatment and 1 year of withdrawal. J Neurosurg 2007;106:614–620.

Frank SA, Wilson R, Holloway RG, et al. Ethics of sham surgery: perspective of patients. Mov Disord 2008;23:63–68.

26.

Stover NP, Bakay RAE, Subramanian T, et al. Intrastriatal implantation of human retinal pigment epithelial cells attached to microcarriers in advanced Parkinson’s disease. Arch Neurol 2005;62:1833–1837.

Olanow CW, Gracies J-M, Goetz CC, et al. Clinical pattern and risk factors for dyskinesias following fetal nigral transplantation in Parkinson’s disease: a double-blind video-based analysis. Mov Disord 2009;24:336–343.

27.

Patel NK, Gill SS. GDNF delivery for Parkinson’s disease. Acta Neurochir Suppl 2007;97(pt 2):135–154.

Marks WJ, Ostrem JL, Verhagen L, et al. Safety and tolerability of intraputaminal delivery of CERE-120 (adeno-associated virus sero-

DOI: 10.1002/ana.22453

6.

7.

8

Volume 70, No. 1