Transcript Phil Skolnick, Ph.D.

DEVELOPING MEDICATIONS TO TREAT SUBSTANCE
USE DISORDERS: PITFALLS AND PROMISES
Phil Skolnick, Ph.D., D.Sc. (hon.)
Director, Division of Pharmacotherapies and
Medical Consequences of Drug Abuse
The Gains In Medications To Treat SUDs
Are Incremental
Why Haven’t We Been More Successful At Developing
Medications To Treat Addictions ?
It’s Not For Lack Of Putative Targets Identified In
Preclinical Studies
-Kappa opioid receptor antagonists
-Dopamine (D3) receptor antagonists
-5HT2A receptor antagonists
-5HT2C receptor agonists
-mGluR5 receptor antagonists
-CRH1 antagonists
-FAAH inhibitors
-Orexin antagonists
Compounds acting at these targets may be effective against multiple
drugs of abuse
Why Haven’t We Been More Successful At Developing
Medications To Treat Addictions ?
• Historically, big pharma has not embraced
developing medications to treat addictions*
(*with the exception of nicotine)
Some of the factors often cited for this indifference:
-Cost of developing an NCE (current estimates: a bit south of $2BB
including cost of capital)
-Perceived small market size (translates to return on investment)
Suboxone sales (2009) of >$750,000,000 belie this perception
-Difficulties in executing a clinical trial campaign using patients with
SUDs (often many comorbid conditions + lifestyle issues)
-Stigma associated with addiction to illegal substances
High Cost of Developing Drugs To Treat
Neuropsychiatric Disorders
Science, 329: 502-504,2010
The Drug Development Cycle Time Has
Increased Dramatically
Time from First Pharmacological Testing to New
Drug Approval, 1963 - 1997
Years
1960s
1970s
1980s
1990s
3.8
1.9
2.4
4.4
4.3
5.2
5.6
5.6
0.0
1st Pharm Testing to IND File
2.1
2.8
6.7
5.0
IND File to NDA Subm
2.0
10.0
15.0
NDA Subm to Approval
Source: Parexel’s Pharmaceutical R&D Statistical Sourcebook, 2002/2003.
6
Why Haven’t We Been More Successful At Developing
Medications To Treat Addictions ?
-Recent mergers (e.g. Pfizer/Wyeth, Schering-Plough/Merck) have resulted
in a consolidation (read: net loss) of programs, including CNS.
-Several big pharmas (GSK, A/Z) have dropped discovery medicine
programs in psychiatry .
-Both a reduction in the number of potential partners AND lack of
enthusiasm for drugs to treat psychiatric disorders dampens enthusiasm
for investment in CNS based startups/biotechs, often the drivers of
innovation.
The Current Probability of Success in Developing Drugs to Treat CNS
Disorders Is Dismal
Lead
C.C.
Ph. I
Ph . II Ph. III
NDA
3%  8%  17%  25%  59%  85%
Pharma Retrenchment Will Affect The SUD Pipeline
-Most molecules that we explore are “repurposed” – with the
cutback in psychiatry R&D, many promising early –mid stage
molecules will not be developed.
-The upside: there may be a bumper crop of mid-stage
molecules in the near-intermediate term.
-The trend away psychiatry research in big pharma will have a
negative, trickle-down effect on CNS focused biotech/small
pharma, since there are fewer “end game” (licensing) partners.
-Our new champions may be small/mid-size pharma: e.g.
Alkermes, Titan, Catalyst, R/B
But will these companies have the resources and
infrastructure to develop truly innovative medicines?
THERAPEUTIC STRATEGIES TO TREAT
SUDs
• Replacement/Substitution (e.g. methadone,
buprenorphine, nicotine)
• Magic Bullets and Arrows (e.g. cocaine vaccine;
nicotine vaccine; mAbs to methamphetamine;
engineered cocaine esterases)
• New Age [targets developed because of a
better understanding of the neuroscience of
addiction] (e.g. mGluR5 antagonists, 5HT2C
agonists, D3 antagonists)
Therapeutic Horizon For SUDs
Near Term (≤3 y)
• NicVax
• Probuphine
• Vivitrol (for relapse prevention to opiates) <APPROVED 2010>
• Bupropion
Mid Term (4-7 y)
• Vigabatrin
• 5HT2c agonist (lorcarserin)
• Buspirone (D3 antagonist +)
• Bupropion/Naltrexone (Contrave)
Therapeutic Horizon for SUDs
Long Term (≥7 y)
• D3 antagonist (NCE, not buspirone)
• mGluR5 antagonist (fenobam)
• Vaccines (cocaine and others)
• mAbs (stimulants, opiates)
• κ antagonist (JDTic)
• PPAR-γ agonists (glitazones)
Probuphine® : An Implantable SR
Formulation of Buprenorphine
 Subcutaneous implant that delivers low,
continuous, steady-state levels of
buprenorphine for 6 months
 Strengths/Advantages
• Reduces risk of diversion
• Improves compliance – one good decision
every 6 months
• Reduces side-effects associated with
fluctuating drug levels
• Provides efficacious therapy while reducing
overall drug exposure to the patient
12
Study 1: Difference in Mean % Negative
Urines (Δ from placebo)
Mean Treatment Difference (%)
20
P = .0006
18
16
14
P = .0253
12
28% - 9%
10
8
40% - 29%
4
2
0
Weeks 1-16
13
Weeks 17-24
Study 1: Secondary Efficacy Outcomes
Outcome Measure (24 weeks)
Probuphine > Placebo
Treatment Retention
p < .0001
Patient-Rated Opioid Withdrawal
p = .0039
Clinician-rated Opioid Withdrawal
p = .0004
Global Improvement of Opioid
Addiction (Clinician-Rated)*
p < .0003
Global Severity of Opioid Addiction
(Clinician-Rated)*
p = .0086
Opioid Craving Visual Analog
Scale (VAS)
p = .0009
* Responder Analysis
14
Probuphine
• Follow on Ph III trial initiated in Q2’10
• Trial compares the effects of probuphine (100
patients), suboxone (100 patients) and
placebo (50 patients)
• Enrollment completed Q’3 2010, results
expected in Q2’11
NO ATTEMPT TO “OVERSMOKE” THE EFFECT OF NICVAX
STATUS OF NICVAX CLINICAL PROGRAM
• Multiple Phase III studies (each with ~1,000
patients) initiated under an “SPA” with the
FDA. Primary endpoint is long term
abstinence from smoking.
• First trial initiated Q4’09 and enrollment
completed Q3’10. Final data expected Q4’11.
• Second trial initiated Q1’10 and enrollment is
expected to be complete by Q1’12.
Bupropion
• A dual (DA/NE) uptake inhibitor
• Approved for the treatment of depression and
smoking cessation
• No significant difference from placebo in a
large trial treating methamphetamine
dependence (Elkasheff, et al., 2007). Primary
outcome measure: the proportion of
participants having a methamphetamine-free
week.
Bupropion Increases Abstinence In Methamphetamine
Users (Success defined as ≥ 2 weeks of methamphetamine
free urine lasting through the end of study)
McCann and Li (2010) – reanalysis of Elkasheff, et al.,2008
Bupropion vs. Methamphetamine
Total Sample
Abstinence During
Last 3 Weeks
Placebo
Bupropion
Failures
Successes
68
(94.4%)
4
(5.56%)
65
(82.3%)
14
(17.7%)
3.2 x
Reanalysis of data from Elkasheff, et al., 2008
P = 0.02 (Chi-square test)
Bupropion Is Most Effective In Producing Abstinence In
“Light-to-Moderate” Methamphentamine Users (success
defined as ≥ 2 weeks of methamphetamine free urine
lasting through the end of study)
No measure of medication
compliance was included in this
study. Based on compliance
data in other studies, these
data may underestimate the
efficacy of bupropion in
producing ≥ 2 weeks of EOSA
McCann and Li (2010) – reanalysis of Elkasheff, et al.,2008
Bupropion vs. Methamphetamine
“Low” Baseline Use (≤ 18 days)
Abstinence During
Last 3 Weeks
Placebo
Bupropion
Failures
Successes
34
(97.1%)
1
(2.86%)
27
(75.0%)
9
(25.0%)
8.7 x
Reanalysis of data from Elkasheff, et al., 2008
P = 0.007 (Chi-square test)
Is Bupropion Effective For The Treatment of
Methamphetamine Abuse?
•Based on reanalysis of the Elkasheff, et
al. (2007) report, a second trial was
initiated.
•Multisite, double blind study comparing
bupropion (150 mg b.i.d.) and control in
methamphetamine. Last patient
randomized January, 2011. Data expected
Q2’11.
Therapeutic Horizon For SUDs
Near Term (≤3 y)
• NicVax
• Probuphine
• Vivitrol (for opiate addiction)
• Bupropion
Mid Term (4-6 y)
• 5HT2c agonist (lorcarserin)
• Buspirone (D3 antagonist +)
• Bupropion/Naltrexone (Contrave)
• Vigabatrin
Lorcaserin, A 5HT2C Agonist, Is In Late Stage
Development
-Arena has completed phase III clinical testing of Lorcaserin for the primary
indications of obesity and weight management. An NDA was submitted in
December, 2009.
Lorcaserin is currently in phase III for the treatment of obese and overweight
patients with type 2 diabetes. The results will be filed as a supplement to the
NDA.
An advisory committee recommended against approval and in Q4’10 the FDA
did not approve lorcarersin to treat obesity. The sponsor is currently conducting
additional studies to address the FDA’s concerns.
Why Lorcaserin?
o Preclinical data suggest 5-HT2C agonists may be
effective for the treatment of nicotine, cocaine and other
drugs of abuse.
o Lorcaserin is appears to be safe and well tolerated in
clinical trials. The most common side effect: headache
o Lack of serious adverse effects in clinical studies:
• No increased risk of depression or suicidal ideation
• No cardiac valvular side effects
5-HT2C receptor agonists: preclinical studies
predict efficacy in multiple SUDs
Nicotine
Drug Discrimination
Cocaine
Self-Administration
Cue-induced Reinstatement of
Nicotine
Self-Administration
Cocaine Seeking
Cunningham et al., 2011, in prep.
Zaniewska et al.,
2007
Ro 60-0175
Grottick et al., 2001
Reinstatement models
Responses
Initiation
Reinstatement
- Cocaine priming
Maintenance
- Footshock stress
Extinction
- Conditioned cue
Cocaine
Saline
Session
Adapted from Erb, Shaham and Stewart, 1996
5-HT2C receptor agonists: preclinical studies
predict efficacy in multiple SUDs
Cue-induced Reinstatement of Meth
Seeking
Lorcaserin
Drug-induced Reinstatement of Meth
Seeking
Lorcaserin
Why Haven’t We Been More Successful At Developing
Medications To Treat SUDs?
Clinical Trial Conduct/Execution
•There are multiple reasons why clinical trials fail;
while most often blamed, the failure of a mechanism/
validity of our models is only one potential source of
failure.
•The three most important aspects of a successful
clinical trial: COMPLIANCE, COMPLIANCE, AND
COMPLIANCE
Medication Compliance Is Required For Adequate
Testing Of A Hypothesis In The Clinic
Noncompliance May Have Spoiled Cocaine Dependence
Drug Trial
• SAN FRANCISCO (EGMN) - A trial of the anticonvulsant
vigabatrin to treat cocaine dependence may have failed
because the patients weren't taking it and not because the
drug didn't work, an analysis of the study results suggests.
• A subsequent analysis of urine samples retained from the
study showed that fewer than 40 percent of 53 patients in the
vigabatrin arm who completed the 12-week study had urine
drug levels that would indicate adherence to the medication
regimen. The subsequent urinalyses suggested that at five of
the 11 study sites, fewer than half of the patients had taken
the medication as prescribed.
• Treatment adherence was rated at 85% using pill counts and
patient self-reports to measure treatment adherence.
(headline from HEALTH DAILY NEWS)
VACSP/NIDA Study # 1026
Phase 2, Double-Blind, Placebo-Controlled Trial of
Modafinil for Methamphetamine Dependence
Placebo
68
Modafinil 200 mg
72
Modafinil 400 mg
70
An ICC of 0.21 indicates a very low agreement between
measures
The Conduct Of Clinical Trials Is Unlikely To Change
in The Short/Mid Term: How do we tackle the
compliance problem?
•Electronic monitoring (e.g. MEMS) [easily gamed]
•Measurement of drug/metabolite in a biological fluid
should be incorporated into every SUD trial protocol. Use
of “compliant” subjects in statistical evaluation.
•Recognition by regulatory authorities that a “snapshot of
compliance” may be the only practical surrogate in a real
world setting.
•Incorporation of riboflavin into formulations has been
used as a means of detecting compliance in both placebo
and drug groups. Alternatively, the use of “homeopathic”
amounts of drug in the placebo, sufficient for detection in
blood/urine, prevents subjects from “gaming” the trial.
Why Haven’t We Been More Successful At Developing
Medications To Treat Addictions ?
Clinical Trial Issues (cont’d).: Are current outcome measures
appropriate?
-Can we make a compelling argument that medicationinduced:
1) Reductions in drug use during the course of an 8 or 12
week trial
2) Changes in an instrument (or subscale) such as the ASI
Provide significant, long term benefit to the patient---Not only regulatory bodies (e.g. FDA) but 3rd party payers
(insurance companies, formulary committees) must be
convinced for a medication to be approved and
commercially successful.