DELAYEDSTART DESIGN WITH DOPAMINERGIC THERAPIES - PowerPoint PPT Presentation

Title: DELAYEDSTART DESIGN WITH DOPAMINERGIC THERAPIES


1
DELAYED-START DESIGNWITH DOPAMINERGIC THERAPIES

• Stanley Fahn, MD
• Columbia University
• New York, NY
• AAPS Workshop29 April 2008

2
All Neuroprotective Trials Have Failed

• I want to challenge this statement.

3
Neuroprotective Trials in PDVarious Agents all
unsuccessful
4
Neuroprotective Trials in PDMAO-B inhibitors
all successful
5
Neuroprotective Trials in PDOther Agents
6
DATATOP RESULTS
9 mo
P lt 0.001
PSG, 1993
7
RESULTS FROM DATATOP(Secondary Outcome)

• Average annual rate of decline in UPDRS scores,
  Mean S.D.
• Treatment Total UPDRS
• Placebo 14.02 12.32
• Tocopherol 15.16 16.12
• Selegiline 7.00 10.76
• Tocopherol Selegiline 7.28 11.11
• P Value lt 0.001

8
Change in Total UPDRS in the Blind-Date Study

• Duration Placebo Selegiline
  Difference
• 1 month 0.50 7.73 -1.52 7.54 2.02
• 3 months 1.57 9.41 -0.85 9.42 2.42
• 9 months 4.18 10.12 1.63 10.61 2.55
• 15 months 5.63 10.73 0.46 10.88 5.17
• 21 months 7.06 12.70 1.51 10.36 5.55
• ? L-dopa p 0.0002
• mg/d 181 246 106 205 p 0.003

9
Placebo
P 0.0003
Deprenyl
Shoulson et al., 2002
10
Total UPDRS On Scores following
the addition of levodopa
Placebo
Selegiline
P 0.003
11
The ELLDOPA Study Change in total UPDRS from
baseline
P lt0.0001
Placebo
150mg
300mg
600mg
Medications withdrawn
Baseline
12
All Neuroprotective Trials Have Failed

• What failed was that the results could not
  eliminate the possibility that a symptomatic
  effect could account for the benefit. Therefore,
  neuroprotection could not be proven.

They were not delayed-start designs.
13
DOPAMINERGIC THERAPIES

• EFFICACY IN IMPROVING SYMPTOMS
• levodopa gt dopamine agonists gt amantadine gtgt
  MAOIs

14
PRIMARY OUTCOMES THAT HAVE BEEN USED IN CLINICAL
TRIALS TO INVESTIGATE NEUROPROTECTION IN PD

• De novo subjects
• - Need for dopaminergic therapy
• - Change in UPDRS scores
• - UPDRS scores after withdrawal of symptomatic
  drug
• - UPDRS scores after delayed- vs early-start
• 2. Subjects on medication
• - Change in UPDRS scores
• - Change in DAT neuroimaging (DAT or other
  ligand scans)
• - Change in clinical features

15
PRIMARY OUTCOMES THAT HAVE BEEN USED IN CLINICAL
TRIALS TO INVESTIGATE NEUROPROTECTION

• De novo subjects
• - Need for dopaminergic therapy
• Selegiline (DATATOP)
• Tocopherol (DATATOP)
• Riluzole
• CGP 3466 (TCH 346)
• CEP-1347 (PRECEPT)

16
PRIMARY OUTCOMES THAT HAVE BEEN USED IN CLINICAL
TRIALS TO INVESTIGATE NEUROPROTECTION

• De novo subjects
• - Change in UPDRS scores
• Coenzyme Q10 (QE2, QE3)

17
PRIMARY OUTCOMES THAT HAVE BEEN USED IN CLINICAL
TRIALS TO INVESTIGATE NEUROPROTECTION

• De novo subjects
• - UPDRS scores after withdrawal of symptomatic
  drug
• Levodopa (ELLDOPA trial)

18
STUDY DESIGNS THAT HAVE BEEN USED IN CLINICAL
TRIALS TO INVESTIGATE NEUROPROTECTION

• De novo subjects
• - UPDRS scores after delayed- vs early-start
• Rasagiline (TEMPO, ADAGIO)
• Pramipexole

19
STUDY DESIGNS THAT HAVE BEEN USED IN CLINICAL
TRIALS TO INVESTIGATE NEUROPROTECTION

• 2. Subjects on medication
• - Change in UPDRS scores
• Selegiline
• - BLIND-DATE
• - Swedish selegiline trial

20
STUDY DESIGNS THAT HAVE BEEN USED IN CLINICAL
TRIALS TO INVESTIGATE NEUROPROTECTION

• 2. Subjects on medication
• - Change in DAT SPECT scans
• Neuroimmunophilin (GPI 1485)

21
STUDY DESIGNS THAT HAVE BEEN USED IN CLINICAL
TRIALS TO INVESTIGATE NEUROPROTECTION

• 2. Subjects on medication
• - Change in clinical features
• GDNF intraputamenal infusion
• Creatine (a NET-PD trial)

22
CLINICAL TRIAL DESIGNS WITHDRAWAL vs. DELAYED
START

• Saga of the ELLDOPA Trial

23
EARLIER VS. LATER LEVODOPA IN PARKINSON DISEASE
(The ELLDOPA Study)
A Multi-Center, Controlled Investigation by the
Parkinson Study Group  Supported by NIH and
DOD Carbidopa/levodopa and matching placebos
provided gratis by Teva Pharmaceuticals
24
2004351 (Dec 9)2498-2508
25
GOAL AND PURPOSE

• To determine if levodopa alters the natural
  history of PD. Does it hasten it, slow it, or
  have no effect?
• Should levodopa be started earlier or later?

26
Origin of dopamine toxic hypothesis

• Cohen G. The pathobiology of Parkinson's
  disease biochemical aspects of dopamine neuron
  senescence. J Neural Transm 1983(Suppl
  19)89-103.
• Dopamine forms hydrogen peroxide when
  metabolized by MAO. This leads to the formation
  of oxyradicals causing oxidative stress and
  neuronal death.

27
DESIGN PLAN

• Enroll subjects with newly diagnosed PD, so that
  a placebo control group can be included.
• Randomize the subjects.
• After a treatment period, withdraw the treatment
  and determine if the actively treated group is
  better or worse than the placebo-treated group.

28
(No Transcript)
29
Problems With a Withdrawal Design noted by Paul
Leber

• A fundamental problem is that it is unattractive
  to patients and their families. It compels a
  patient who is doing well to take a placebo.
• Worsening of the disease by the time of the
  withdrawal phase makes it difficult to conduct
  the drug withdrawal phase if the drug has a
  symptomatic benefit.
• The washout period needs to be long enough to
  establish a return of symptoms. Not stated the
  posibility of a prolonged symptomatic benefit.

30
POTENTIAL PROBLEMS TO OVERCOME - 1

• Symptomatic benefit from levodopa could unblind
  the subject and the investigator.
• Solution
• Utilize a Treating Investigator and a Primary
  Rater.
• The Primary Rater sees the subject only twice at
  baseline and after the washout period, and has no
  contact with the subject during the trial.
• The Primary Outcome is based on the Primary
  Raters evaluations.

31
POTENTIAL PROBLEMS TO OVERCOME 2

• Worsening of PD in the placebo group could
  require symptomatic Rx, especially if the
  treatment period is long.
• Solutions
• 1) Enroll newly diagnosed patients.
• 2) Keep the duration relatively short. 40 weeks
  (9 months) of Rx.

32
POTENTIAL PROBLEMS TO OVERCOME - 3

• A short duration study loses statistical power to
  detect progression of disease.
• Solution Use a large sample size and a
  dose-response design to enhance power.

33
POTENTIAL PROBLEMS TO OVERCOME - 4

• Duration of the washout period needs to be of
  adequate duration to allow PD symptoms to return
  to their natural untreated state.
• Solution A 2-week washout period appears to be
  adequate.

34
POTENTIAL PROBLEMS TO OVERCOME - 5

• Worsening of symptoms during washout period could
  result in a subject terminating from the study.
• Solutions
• 1) Enroll only patients with early, mild
  Parkinsons disease.
• 2) Examine the subject at both Weeks-1 and -2
  during the washout period.

35
N Engl J Med 1967276374-379.
36
(No Transcript)
37
(No Transcript)
38
They evaluated 15 patients taking
carbidopa/levodopa (411.7 25.5 mg/d) at days 1,
8, and 15 following discontinuation of their
medication. Total UPDRS increased 3.9 1.3 from
day 1 to day 8 (p 0.01), and 1.8 0.9 from day
8 to day 15 (p 0.08)
39
(No Transcript)
40
Hauser and Holford,22 analyzed the withdrawal of
levodopa in 20 patients and reported that the
mean half-life of loss of clinical benefit was
7.9 days (95 percent confidence interval, 2.2 to
30.4 days). They suggest that a washout period of
32 days (4 half-lives) may be required to
eliminate 90 percent of the symptomatic effects
of levodopa.
41
ELLDOPA STUDY DESIGN (N361)
Placebo
150 mg/day
300 mg/day
Screening
600 mg/day
Begin washout
Baseline
0
3
24
41
9
40
42
Weeks
1st ßCIT SPECT
2nd ßCIT SPECT
Final Exam
42
ASSESSMENTS

• Clinical assessment
• (UPDRS)
• Neuroimaging assessment
• (ß-CIT SPECT)

43
The ELLDOPA Study Change in total UPDRS from
baseline
P lt0.0001
Placebo
150mg
300mg
600mg
Medications withdrawn
Baseline
44
SPECT SCAN RESULTS IN ELLDOPA
Subjects with Low Putamen lt 3.25 (n116)
-1.4
-6.0
-4.0
-7.2
P (vs. Placebo)
0.17
0.4
0.015
P (dose-response) 0.036
45
INTERPRETATIONS

• THE CLINICAL COMPONENT
• The clinical component of this study failed to
  find any evidence that levodopa treatment is
  harmful or hastens progression of PD. On the
  contrary, it suggests neuroprotection.
• But was the washout period of 2 weeks long
  enough? Because of this uncertainty, we cannot
  conclude that levodopa has been proven to be
  protective.

46
INTERPRETATIONS

• THE IMAGING COMPONENT
• The interpretation of the results of the ß-CIT
  SPECT component of the study is also uncertain
  because we cannot exclude the possibility that
  levodopa has a pharmacologic effect on DAT
  binding (i.e., reduces DAT activity), which could
  explain the decreased ß-CIT uptake.

47
Discusses randomized withdrawal, randomized start
and neuroimaging as a surrogate marker.
48
CONCLUSIONWhat study design should a clinical
trial use to prove neuroprotection if the
experimental drug has symptomatic benefit, such
as the dopaminergic drugs?

• The only design that can test an effect on
  natural history from dopaminergic drugs would be
  the delayed-start design.

49
THANK YOU