Business of Drug Discovery

1
Business of Drug Discovery
David RidleyJanuary 2009
2
David Ridley

• Assistant Professor at Duke University
• Teaching health economics for MBA students
• Research co-author of priority review voucher
  proposal

3
Bill Gates described priority review voucher at
Davos last year

• But some of the highest-leverage work that
  government can do is to set policy and disburse
  funds in ways that create market incentives for
  business activity that improves the lives of the
  poor. Under a law signed by President Bush last
  year, any drug company that develops a new
  treatment for a neglected disease like malaria or
  TB can get priority review from the Food and Drug
  Administration for another product they've made.
  If you develop a new drug for malaria, your
  profitable cholesterol-lowering drug could go on
  the market a year earlier. This priority review
  could be worth hundreds of millions of dollars.

4
Drug DiscoveryWho, What, When, Where, How

• When? Drug development timeline
• How much money?
• By whom? License in R. Contract out D
• Where? Clinical trial location
• What? Orphan and neglected diseases

5
RD Is Risky Costly
Compound Success Rates by Stage
Years
Discovery (2-10 years)
0
Preclinical laboratory animal tests
2
5,00010,000Screened
4
Phase I 20-80 healthy volunteers to determine
safety dosage
Phase II 100-300 volunteers to look for efficacy
side effects
6
250Enter Preclinical Testing
8
Phase III 1000-5000 volunteers to monitor
adverse reactions to long-term use
10
5Enter Clinical Testing
12
FDA Review Approval
14
Additional post-market testing
16
1Approved by the FDA
Net Cost 802 million invested over 15 years
Source DiMasi et al. 2003, Tufts
6
Time
7
Vary by disorder
Source Chris Adams Van Brantner (FTC), Health
Affairs, 2006 Probabilities and duration
estimated using PharmaProjects data Did not
observe actual costs
8
Tuberculosis
Probability of anti-TB by 2010 lt 5 Source
Glickman et al., Science, 2006 But
anti-infectives already used now in phase III so
might be higher
9
Patent 20 yearsEffective patent lt 20 years

• Patent life is 20 years
• Effective patent life is considerably shorter
• Clinical trials
• FDA time
• Patent restoration
• 1/2 RD and all FDA time
• lt5 years extra
• lt 14 years total
• But time value of so 1 year delay for
  blockbuster costs 300 million (Ridley et al.
  2006)

10
Drug DiscoveryWho, What, When, Where, How

• When? Drug development timeline
• How much money?
• By whom? License in R. Contract out D
• Where? Clinical trial location
• What? Orphan and neglected diseases

11
802 million

• Cost per new drug averages 802 million (year
  2000 dollars) (Source DiMasi, Hansen, Grabowski,
  2003)
• The estimate has been criticized, even used as
  the title for a critical book
• Is it really just 200 million?
• Cut in ½ for opportunity cost of capital
• Cut in ½ again for share of dry wells

12
Cost of RD

• We often hear 800 million but with post-approval
  costs its 900 million and with inflation
  adjustments its gt 1300 million
• Accountants might say, however, 200 million (
  inflation)

13
Sarafem different?

• How costly are line extensions?
• Manufacturer sought a line extension for Prozac
• Branded Sarafem as treatment for premenstrual
  dysphoric disorder (PMDD)
• Quick search of clinical studies. Appears that 3
  clinical trials testing effectiveness of Sarafem
  for PMDD. Trials had 260, 42, and 19 patients. At
  10,000 per patient cost is 3.2 million
• Caveat havent seen records, this is just back
  of envelope

14
Torcetrapib different?

• Average 130 million (in 2000 dollars)
• DiMasi et al. (2003)
• other costs cost of capital dry wells
  800m
• Torcetrapib 800 million (?!)
• Pfizer announced that it had pulled the plug on
  the medicine entirely, turning the companys
  nearly 1 billion investment in it into a total
  loss. NYT 2006
• Pfizer had spent close to US1 billion in the
  development of Torcetrapib which included US 800
  million in the clinical trials itself - case
• About 90 million for plant in Ireland
• The numbers are correct. They are based on a
  simple sum of outlays related to Torcetrapib RD
  over the years. They do not include
  cost-of-capital or NPV. Pfizer friend
• Why Torcetrapib more costly?
• More people 7500 in treatment, 7500 in control
  in Illuminate 10,000 others

15
Devices diagnostics different?

• Clinical trials often arent required for
  approval
• For devices, 510(k) submissions demonstrate
  substantial equivalence to approved
• But
• For more innovative, clinical trials necessary
  for Premarket Approval (PMA)
• For convincing providers and payers
• For the future. Critical GAO report encouraged
  FDA to subject devices to more rigorous review

16
Biotech different?

• Faster
• More nimble and creative (different corporate
  culture)
• Fewer safety issues (more relevant to early
  biotech era development)
• Slower
• Less experience in clinical development and
  interacting with regulators
• More costly manufacturing process RD and
  production of clinical supplies

Source DiMasi and Grabowski 2007
17
Pre-Approval Capitalized Cost per Approved New
Molecule

• Based on a 5-year shift and prior growth rates
  for the preclinical and clinical periods
• New pharmaceutical RD is not 802 or 899 but
  1318

Source DiMasi and Grabowski 2007
18
Pre-Approval Cash Outlays per Approved New
Molecule
All RD costs (basic research and preclinical
development) prior to initiation of clinical
testing
Based on a 5-year shift and prior growth rates
for the preclinical and clinical periods
Source DiMasi and Grabowski 2007
19
Clinical Trial Size Increases with

• Smaller treatment difference relative to noise
  (variation) (?)
• Want smaller false positive rate (a)
• Want higher power (1- ?)
• Intervention group size

20
Pivotal Trials

• Often 2 pivotal trials, but 1 for orphan
• The Food, Drug and Cosmetic Act calls for
  adequate and well-controlled clinical trials to
  support approval of a product for a specific
  indication. In traditional drug development
  terms, this language translates into at least two
  pivotal trials to support registration. The
  purpose of these two clinical trials is to have
  one trial confirm the results of the other,
  thereby reducing the probability of outcomes
  associated with false positives (Brenner
  Ellis-Grosse 2006).
• For example, all 4 of the following groups
• Duke treatment control (placebo)
• Quintiles (CRO) treatment control (placebo)
• But orphan drugs are often approved based on one
  pivotal Phase III clinical trial (or a combined
  Phase II/III trial).

21
RD by function, PhRMA, 2006
Does not account for time value of . Source
PhRMA Industry Profile 2008
22
Postapproval Drug Safety

• Ridley, Kramer, Tilson, Grabowski, Schulman
  (2006)
• 11 companies participated
• Accounting for 71 of 2003 sales by top 20
• For 2003
• Mean pharma sales 17 billion
• Mean safety spend per company 56 million (0.3
  of sales)
• Estimated spending by top 20 drug manufacturers
  800 million
• Context
• Total budget for FDAs Office of Drug Safety in
  2003 22.1 million
• Industry spend on RD 2003 15.6 of sales

23
(No Transcript)
24
Drug DiscoveryWho, What, When, Where, How

• When? Drug development timeline
• How much money?
• By whom? License in R. Contract out D
• Where? Clinical trial location
• What? Orphan and neglected diseases

25
Who innovates? Big pharma or others?2006
Domestic RD by PhRMA Members
17 of dollars, but probably gt 17 of
products Source PhRMA Industry Profile 2008
26
Who conducts trials? Contract research is an
option

• For contract research, how does the price for
  government and foundations compare to the
  industry price?
• we provide the US government and Bill and
  Melinda Gates grantees (and other non for profit
  entities working in the neglected diseases space)
  with preferential terms of business.
• One way is rather simple - discounted rates off
  our fees/hourly rates. US discount and non-US
  discounts are different, with the more aggressive
  reduction being provided for services in non
  US-locations.
• Another way is to provide rebates at the end of a
  calendar year based on revenues (tranche system).
  I would love to chat further if you are
  interested. .
• - Vaila Clements, VP Public Health and Government
  Services, Quintiles Transnational, January 2009

27
Drug DiscoveryWho, What, When, Where, How

• When? Drug development timeline
• How much money?
• By whom? License in R. Contract out D
• Where? Clinical trial location
• What? Orphan and neglected diseases

28
T / F

• True or false clinical trials are moving to
  India and China

29
Regional Distribution of Biopharmaceutical
Clinical Trial Sites

• Emerging (green) grew from 8 in 2002 to 20 in
  2006
• Largest losses in Western Europe between 2002 and
  2004, while decline in North America between 2004
  and 2006
• Growth is 5-10 so not necessarily leaving, just
  not growing as fast

Western Europe and North America declining but
large
Source Berndt, Cockburn, Thiers 2008
30
Shares of Biopharmaceutical Clinical Trial Sites

• Top 5 (US, Germany, France, Canada, and Spain)
  host 66
• Emerging now host 17 of sites

Source Berndt, Cockburn, Thiers 2008
31
Average Relative Annual Growth Rates in Shares

• Decreasing shares in North America, Europe,
  Venezuela
• Increasing shares in China and other emerging
  markets

Source Berndt, Cockburn, Thiers 2008
32
Drug DiscoveryWho, What, When, Where, How

• When? Drug development timeline
• How much money?
• By whom? License in R. Contract out D
• Where? Clinical trial location
• What? Orphan and neglected diseases

33
Problem Little private financial incentive to
develop treatments for
34

• How can we induce companies to develop treatments
  for orphan diseases, neglected diseases, and
  bioterrorist attacks?

35
Solution Push Pull Mechanisms

• Pull Mechanisms fund outputs (drugs, vaccines)
• Advance Markets
• Guaranteed price creating 3 billion market
• Malaria, TB, HIV
• Transferable voucher for extended patent life
• Prize for treatment for neglected diseases or
  bioterrorism
• Extra patent life for a different drug

• Push Pull
• Orphan Drug Act
• Tax credits grants (push)
• Marketing exclusivity (pull)
• Priority Review Voucher
• Priority review for other drug
• Dengue, Chagas,

36
1983 U.S. Orphan Drug Act

• Eligibility
• Disease affects few people in U.S. (lt200,000)
• Or no reasonable expectation that profitable
• Low expected returns sufficient not necessary
• Push
• RD tax credits ½ clinical, even if not approved
• Clinical research grant programs
• Pull
• FDA counseling and priority review
• Guaranteed market exclusivity of 7 years

37
U.S. Definition of Disease Prevalence

• From www.fda.gov/orphan/
• (i) The disease or condition for which the drug
  is intended affects fewer than 200,000 people in
  the United States or, if the drug is a vaccine,
  diagnostic drug, or preventive drug, the persons
  to whom the drug will be administered in the
  United States are fewer than 200,000 per year as
  specified in Sec. 316.21(b), or
• (ii) For a drug intended for diseases or
  conditions affecting 200,000 or more people, or
  for a vaccine, diagnostic drug, or preventive
  drug to be administered to 200,000 or more
  persons per year in the United States, there is
  no reasonable expectation that costs of research
  and development of the drug for the indication
  can be recovered by sales of the drug in the
  United States as specified in Sec. 316.21(c).

38
Orphan Drug Laws
39
Orphan Drugs LawsMore Details on US EU
40
Most Important IncentiveMarket Exclusivity

• Market exclusivity was most sought after
  incentive in Orphan Drug Act, said FDA
• Particularly important for
• biologicals that had uncertain patents
• older chemical entities with useful orphan drug
  indications

41
Orphan DrugsLower Costs

• Orphan drugs have smaller and shorter clinical
  trials
• The 7 orphan drugs in 1999 were approved with an
  average of 588 patients
• By contrast, non-orphan approvals in the late
  1990s had over 5,000 patients on average

42
Orphan DrugsLower Lifetime Sales Profile
Orphan
Non-orphan
Orphan
Source Grabowski and Vernon 2003 1990-1994 Sales
of New Drug Introductions
43
Orphan Drugs Can Be Lucrative
Source WSJ 2005 Update Cerezyme 300K/yr,
1.1b in 2007 (NYT March 2008)
44
Approvals of Drugs for Rare Diseases
1983 Orphan Drug Act
Source WSJ 2005
Source FDA
45
Orphan Drug Act Conclusions

• Orphan Drug Act has been a success in encouraging
  many new drug approvals for rare diseases
• But few U.S. orphan drug approvals for neglected
  diseases. Malaria vaccine would qualify as
  orphan, but with so few high-paying customers,
  exclusivity does little good
• Need additional incentive mechanisms for
  neglected diseases

46
?Harm
0.1 1 10
100
Annual Global Burden (Millions of DALYs)
0 0.1
1 10
Developed Countries Share of DALY Burden
?Disparity
Source Ridley, Grabowski, Moe
DALYDisability-adjusted life years
47
Priority Review Voucher

• Senator Brownback wrote After reading their
  proposal in Health Affairs, I met with Ridley and
  colleagues to discuss the idea further, and I
  subsequently drafted an amendmentIndeed, their
  idea is the heart of my Elimination of Neglected
  Diseases (END) amendment.
• Became law in 2007

48
Priority Review Voucher
3. Voucher user fee
Priority review at FDA (6 vs. 18) orphan credits
Treatment for neglected disease that is approved
licensed as generic
Developer of treatment for neglected disease
Manufacturer of potential blockbuster
49
Priority Review Voucher Creates Win-Win

• Benefits people suffering from neglected diseases
• Benefits patients in the U.S. who receive earlier
  access to the priority drug
• U.S. patients also get earlier access to the
  generic because drugs with faster FDA review
  typically have earlier patent expirations
• Concept could also be applied to treatments for
  bioterrorism

50
SEC. 1102. PRIORITY REVIEW TO ENCOURAGE
TREATMENTS FOR TROPICAL DISEASES.Subchapter A of
chapter V of the Federal Food, Drug, and Cosmetic
Act (21 U.S.C. 351 et seq.) is amended by adding
at the end the followingSEC. 524. PRIORITY
REVIEW TO ENCOURAGE TREATMENTS FOR TROPICAL
DISEASES.(a) Definitions- In this section(1)
PRIORITY REVIEW- The term priority review', with
respect to a human drug application as defined in
section 735(1), means review and action by the
Secretary on such application not later than 6
months after receipt by the Secretary of such
application, as described in the Manual of
Policies and Procedures of the Food and Drug
Administration and goals identified in the
letters described in section 101(c) of the Food
and Drug Administration Amendments Act of
2007.(2) PRIORITY REVIEW VOUCHER- The term
priority review voucher' means a voucher issued
by the Secretary to the sponsor of a tropical
disease product application that entitles the
holder of such voucher to priority review of a
single human drug application submitted under
section 505(b)(1) or section 351 of the Public
Health Service Act after the date of approval of
the tropical disease product application.(3)
TROPICAL DISEASE- The term tropical disease'
means any of the following
On September 27, 2007, President Bush signed H.R.
3580, "Food and Drug Administration Amendments
Act of 2007" into Law
51
Conclusion

• When? Each of 3 phases of trials take couple of
  years
• How much? 800 million
• ½ is cost of capital, ½ is dry wells
• Update 1300
• So if 325 million, youre average
• Big trials costly but need bigger if
• Smaller treatment difference, Higher noise
  (variation),
• Smaller false positive rate (alpha), Higher
  power (1-beta)
• By whom?
• License in 17 of for in-license
• Contract out Quintiles lower rate for Gates
• Where?
• Emerging markets now host 17 of sites
• What?
• Incentives for orphan, neglected, biodefense