Transcript Document
Measuring Trends in the Development of
New Drugs: Time, Costs, Risks and
Returns
Joseph A. DiMasi, Ph.D.
Director of Economic Analysis
Tufts Center for the Study of Drug Development
Tufts University
SLA Pharmaceutical & Health Technology
Division Spring Meeting
Boston, MA, March 19, 2007
Agenda
New Drug development times
Risks in new drug development
R&D costs and returns for new drugs
Pace of competitive development
Impact of improvements to the R&D
process
Trends in new drug pipelines
New Drug Development Times
Mean U.S. Approval and Clinical Phases for
U.S. New Drug Approvals, 1963-2004
12
10
Total Phase
Years
8
6
IND Phase
4
Approval Phase
2
0
19
64
19
68
19
72
19
76
19
80
19
84
19
88
Year of NDA Approval
Source: Tufts CSDD, 2006
Points are 3-year moving averages
19
92
19
96
20
00
20
04
Clinical and Approval Times Vary
Across Therapeutic Classes, 2002-04
10.4
Neuropharmacologic
1.7
9.0
Antineoplastic
0.8
6.6
Antiinfectives
Endocrine
6.3
AIDS Antivirals
6.3
1.9
1.3
0.6
5.6
Gastrointestinal
4.8
Cardiovascular
0
7.5
8.0
6.3
14
Years
Clinical Phase
Source: Tufts CSDD, 2006
8.5
6.9
2.8
1.5
9.8
7.6
1.9
5.2
Anesthetic/Analgesic
12.1
Approval Phase
New Drug Development Risk
Approval Success Rates for NCEs
Also Vary by Therapeutic Class
40.4%
Antiinfective
27.2%
Oncology/Immunology
19.9%
Respiratory
15.2%
Cardiovascular
14.4%
CNS
10.9%
GI/Metabolism
0
Approval Success Rate
Source: Tufts CSDD Impact Report, 8(3): May/June 2006
45
Pharmaceutical R&D Productivity
New Drug Approvals Are Not Keeping
Pace with Rising R&D Spending
60
40
R&D Expenditures
30
20
New Drug Approvals
15
0
1963
0
1968
1973
1978
1983
1988
R&D expenditures are adjusted for inflation
Source: Tufts CSDD Approved NCE Database, PhRMA, 2005
1993
1998
2003
R&D Expenditures
(Billions of 2004$)
NCE Approvals
45
Recent Productivity Decline in the Drug
Industry: Is this a Unique Phenomenon?
“In 1960 the trade press of the U.S. drug
industry began to refer to the last few
years as constituting a “research gap,”
commenting that the flow of important new
drug discoveries has for some inexplicable
reason diminished.”
Source: U.S. Senate, Report of the Subcommittee on Antitrust and Monopoly, 87th Congress,
1st Session, “Study of Administered Prices in the Drug Industry,” June 27, 1961, p.136
Pharmaceutical R&D Costs and Returns
Opportunity Cost for Investments
Consider two investment projects, A and B
Both projects require the same out-of-
pocket expenditure (say, $400 million)
However, returns to A are realized
immediately, but investors must wait 10
years before returns to B are realized
Rational investors would conclude that B
is effectively much costlier than A
Out-of-Pocket and Capitalized
Costs per Approved Drug
Millions of 2000 $
900
802
800
700
600
500
400
300
200
100
0
466
336
403
282
121
Preclinical
Clinical
Out-of-Pocket
Total
Capitalized
Source: DiMasi et al., J Health Economics 2003;22(2):151-185
Pre-approval and Post-approval
R&D Costs per Approved Drug
Millions of 2000 $
1200
897
1000
802
800
543
600
403
400
200
0
140
95
Out-of-Pocket
Capitalized
Post-approval
Pre-approval
Source: DiMasi et al., J Health Economics 2003;22(2):151-185
Total
Annual Growth Rates
for Out-of-Pocket R&D Costs
11.8%
7.8%
7.0% 7.6%
6.1%
2.3%
Preclinical
Clinical
1970s to 1980s approvals
Total
1980s to 1990s approvals
Source: DiMasi et al., J Health Economics 2003;22(2):151-185
Mean Number of Subjects in NDAs
for NMEs
Subjects
5,507
3,233
1,576
5,621
3,567
1,321
1977-80 1981-84 1985-88 1990-92 1994-95 1998-01
Approval Period
Sources: Boston Consulting Group, 1993; Peck,
Food and Drug Law J, 1997; PAREXEL, 2002
Clinical Trial Complexity Index (Phases I-III)
150
140
130
120
110
100
90
1992
1993
1994
Source: DataEdge, 2002
1995
1996
1997
1998
1999
2000
Summary for R&D Costs
R&D costs have grown substantially, even in
inflation-adjusted terms
The growth rate for discovery and preclinical
development costs has decreased substantially
Conversely, clinical costs have grown at a much
more rapid rate
New discovery and development technologies
(e.g., genomics) hold the promise of lower costs in
the long-run (but perhaps higher costs in the
short-run)
Summary for R&D Costs (cont.)
Evidence and conjectures regarding factors
affecting growth in clinical costs
More clinical trial subjects
Increased complexity: more procedures per
patient
Patient recruitment and retention
Treatments associated with chronic and
degenerative diseases
Testing against comparator drugs
Returns to New Drug Development
Present Values of Net Sales and R&D Cost
for New Drugs by Sales Decile (millions of 2000 $)
After-tax net sales
3000
2500
2000
1500
1000
After-tax average R&D Cost
500
0
1
2
3
4
5
6
7
8
Deciles
Source: Grabowski et al., PharmacoEconomics 2002; 20(Suppl 3):11-29
9
10
Biopharmaceutical R&D Costs
Transition Probability
Transition Probabilities for Clinical Phases
83.7%
71.0%
68.5%
64.2%
56.3%
44.2%
30.2%
Phase I-II
Phase II-III
Biotech
Phase IIIApproval
Pharma
Source: DiMasi and Grabowski, Managerial and Dec Econ 2007, in press
21.5%
Phase I Approval
Clinical Development and Approval Times
19.5
Biotech
Pharma
12.3
29.3
26.0
32.9
33.8
16
18.2
97.7
90.3
0
120
Months
Phase I
Phase II
Phase III
RR
Source: DiMasi and Grabowski, Managerial and Dec Econ 2007, in press
Pre-Approval Out-of-Pocket (cash outlay) and Time
Costs per Approved New Biopharmaceutical*
Millions (2005$)
1,241
682
626
615
559
417
361
265
198
Preclinical**
Clinical
Out-of-pocket
Time
Total
Capitalized
* Based on a 30.2% clinical approval success rate
** All R&D costs (basic research and preclinical development) prior to initiation of clinical testing
Source: DiMasi and Grabowski, Managerial and Dec Econ 2007, in press
Why Might Biopharma Cost Differ?
Biotech firms may be more nimble and creative
(different corporate culture)
Replacement therapies may confront fewer safety
issues (more relevant to early biotech era
development)
However, biotech firms have less experience in
clinical development and in interacting with
regulatory authorities
Manufacturing process R&D and production of
clinical supplies much more expensive for
biopharmaceuticals
Biopharmaceutical and Pharma
R&D Costs Compared
Pre-Approval Out-of-Pocket Cost
per Approved New Molecule
Millions (2005$)
672
522
559
452
361
198
136
316
150
Preclinical*
Biotech
Clinical
Pharma
Total
Pharma (time-adjusted)**
* All R&D costs (basic research and preclinical development) prior to initiation of clinical testing
** Based on a 5-year shift and prior growth rates for the preclinical and clinical periods
Source: DiMasi and Grabowski, Managerial and Dec Econ 2007, in press
Pre-Approval Capitalized Cost
per Approved New Molecule
1,318
Millions (2005$)
1,241
899
879
626
615
376
523
439
Preclinical*
Biotech
Clinical
Pharma
Total
Pharma (time-adjusted)**
* All R&D 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, Managerial and Dec Econ 2007, in press
The Pace of Competitive Development
Period of First-in-Class Approval
Market Exclusivity for First-in-Class
has Declined: Mean Time to First
Follow-on Approval
8.2
1970s
5.9
1980-84
5.1
1985-89
2.8
1990-94
1.8
1995-98
0
9
Years
Source: DiMasi and Paquette, PharmacoEconomics 2004;22(Suppl 2):1-14
Percent of Follow-on Drugs Reaching
Clinical Milestone Prior to First-inClass Drug Reaching Same Milestone
50
42%
35%
36%
27%
14%
10%
0
1985-1989
1990-1994
Period of First-in-Class Approval
IND Filed
1st Human Test
Source: DiMasi, Paquette, PharmacoEconomics 2004;22(Suppl 2):1-14
1995-1998
Follow-on Approvals Create Competition
Resulting in Price Discounts
8
Number of New Drugs
7
6
5
5
5
5
3
2
0
2
0
-10% to -3%
-3% to3%
3% to 15%
15% to 40%
>40%
Relative Price Discount
Mean Price for Existing Drugs
Price Leader
* Analysis based on FYs 1995-1999.
Source: DiMasi, 2000 [http://aspe.hhs.gov/health/reports/drugpapers/dimassi/dimasi-final.htm]
Impact of Improvements in Drug
Development Productivity
Cost reduction
Cost Reductions from Higher Clinical
Success Rates
35%
30%
25%
20%
15%
10%
5%
0%
34
33
32
31
30
29
28
27
26
25
24
23
22
21
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
%
%
%
%
%
%
%
%
%
%
%
%
%
%
Success Rate
Average phase cost
Phase cost adjusted for cost of failures
Source: DiMasi, PharmacoEconomics 2002; 20(Suppl 3):1-10
Cost Reductions from Simultaneous
Percentage Decreases in All Phase Lengths
Cost reduction
30%
25%
20%
15%
10%
5%
0%
0%
5%
10% 15% 20% 25% 30% 35% 40% 45% 50%
Phase time reduction
Clinical cost
Total cost
Source: DiMasi, PharmacoEconomics 2002; 20(Suppl 3):1-10
Trends in Drug Development Pipelines
Clinical Testing Pipelines for Large Pharmaceutical Firms*
Have Grown in Recent Years (Phase I Starts per year)
100
80
60
40
20
0
1993-1997
1998-2002
* Ten largest pharmaceutical firms
Source: Tufts CSDD Impact Report, 8(3): May/June 2006
2003-2005
Trends in New Drug Development
Pipelines* by Therapeutic Class
7.9%
11.6%
13.7%
Antiinfective
8.7%
9.1%
Cardiovascular
14.3%
20.6%
19.6%
21.9%
CNS
GI/Metabolism
4.3%
8.3%
10.7%
20.2%
Oncology/Immunologic
20.5%
6.5%
4.8%
Respiratory
0%
27.2%
9.1%
Percent of Phase I Pipeline
1993-97
1998-02
* Ten largest pharmaceutical firms
Source: Tufts CSDD Impact Report, 8(3): May/June 2006
2003-05
30%
Large Pharmaceutical Firms* are
Increasingly Licensing-in New Drugs
100
77.8
Percent
80
70.3 71.8
60
40
23.6 25.0
14.8
20
7.4 6.2
3.2
0
Self-originated
Licensed-in
1993-97
1998-02
Licensed-out
2003-05
* Ten largest pharmaceutical firms
Source: Tufts CSDD Impact Report, 8(3): May/June 2006
Conclusions
Drug development has been and still is costly,
risky, and lengthy
Periods of market exclusivity have shrunk for
first-in-class drugs
The potential payoffs for improvements in the
development process are substantial
After a period of decline, more new drugs are
now entering clinical testing pipelines