Transcript Ani Bhattacharya - Pharma

Portfolio Management and
R & D Productivity
Anirban Bhattacharya, PhD
8th Annual Project and Portfolio Management Forum
Is Pharma “over-spending” in R&D ?
Pharma Industry has one of the most expensive
R&D efforts
Company
EXXON MOBIL
CHEVRON
CONOCOPHILLIPS
GENERAL ELECTRIC
GENERAL MOTORS
FORD
HEWLETT-PACKARD
MCKESSON
IBM
PROCTER & GAMBLE
PFIZER
APPLE
BOEING
MICROSOFT
ARCHER-DANIELS-MIDLAND
JOHNSON & JOHNSON
DELL
UNITED TECHNOLOGIES CORP.
DOW CHEMICAL
KRAFT FOODS
2010 REVENUE 2010 R&D SPEND 2010 R&D SPEND OPTIMAL Difference
($ IN MILLIONS) ($ IN MILLIONS)
(%of sales)
R&D SPEND
$341,578
$189,607
$175,752
$149,060
$135,592
$128,954
$126,033
$112,084
$99,871
$78,938
$67,791
$65,225
$64,306
$62,484
$61,682
$61,587
$61,494
$54,326
$53,674
$49,207
$1,012
$526
$230
$3,939
$6,962
$5,000
$2,959
$407
$5,720
$1,950
$9,538
$1,782
$4,121
$8,714
$56
$6,844
$661
$1,746
$1,660
$583
*Top 20 firms traded on US Exchanges
Optimal R&D Spend calculated on the basis of RQ (Research Quotient)
Article by A.M.Knott on Harvard Review (May 2012): “The trillion-dollar R&D Fix”
0.30%
0.28%
0.13%
2.64%
5.13%
3.88%
2.35%
0.36%
5.73%
2.47%
14.07%
2.73%
6.41%
13.95%
0.09%
11.11%
1.07%
3.21%
3.09%
1.18%
$136,486
$56,690
$100,350
$19,947
$15,570
$14,405
$43,907
$111,598
$10,359
$7,816
$6,304
$9,468
$8,142
$9,210
$29,947
$5,371
$16,218
$6,192
$9,356
$7,254
$135,474
$56,163
$100,119
$16,008
$8,608
$9,405
$40,948
$111,190
$4,639
$5,866
($3,234)
$7,686
$4,021
$496
$29,891
($1,472)
$15,557
$4,446
$7,695
$6,671
73% drop in R&D productivity !
From Abundance to Scarcity
(in quantity AND quality!)
Golden Period:
1996-2004
NMEs approved per year (average) = 3
Average 5th year sales per NME ($ MM) = 515
5th year sales per year in total ($ Bn) = 18.3
R&D spend per year ($Bn) = 65
5th year sales per 1$Bn R&D spend ($MM) =
275
“Drought” Period:
NMEs approved per year (average) = 22
Average 5th year sales per NME ($ MM) = 430
5th year sales per year in total ($ Bn) = 9.4
R&D spend per year ($Bn) = 125
5th year sales per 1$Bn R&D spend ($MM) = 75
The 90s remained the most fruitful period in the history of Pharma
PDUFA + “good” FDA behavior (e.g. HIV)
Targeting chronic diseases with new MoAs
Development of many “fast-followers”
Establishment of “surrogate markers”
Creation of new diseases (OAB, RLS, etc.)
2005-2010
Is this a
“rebound
effect” or an
“R&D
productivity
crisis”?
What is R&D productivity?
Defining R&D Productivity
Required Investment
R&D Productivity is an aggregate representation of:
R&D efficiency: ability to translate inputs (ideas, resources, money, etc.)
into defined outputs (usually approvals and launches), over a defined period
of time; it is simply measured by a “cost per launch”
R&D effectiveness: ability to produce outputs with certain intended and
desired qualities/outcomes (value to patients, physicians and/or payers; and
substantial commercial value); it is simply measured by a “value per launch”
S.M..Paul et al. “How to improve R&D productivity”
Nature Reviews/Drug Discovery vol.9; March 2010; p.203-214
Created Value
R&D Productivity Report Card
The good ……
The bad ……
Annual number of NME/NBEs approved by the
FDA stable over the last 60 years
60
50
40
30
20
10
10
20
02
98
94
90
86
82
78
74
70
66
62
58
54
06
20
20
19
19
19
19
19
19
19
19
19
19
19
19
19
50
0
- The number of new drugs approved
per US$ 1 Bn spent in R&D has halved
every 9 years since 1950, falling around
80-fold
- The cost of developing one NME
raised 38-fold, from $50M in the 50s to
$1.8 Bn in the 2000s
FDA approvals of NME/NBEs
The ugly ……
NPV for NME is -$65
IRR is 7.5% (less than cost of capital at 10%)
Negative trend: IRR was 12% in 1997-2001
Better financials for biologics: IRR at 13% and NPV at $1.26 billion
R&D Efficiency Assessment
Developmental Cost
Risk
Time
Increased Development Costs
 The first randomized controlled trial, published in 1948, recruited 109 patients and
randomized 107 of them
 Between 1987 and 2001, the number of patients per pivotal trial for anti-hypertensive agents
rose from 200 to 450
 Between 1993 and 2006, the average number of patients across the pivotal trials in diabetes
rose from 900 to over 4,000
 The first long-acting insulin analogue, glargine, was approved in 1999 following 3 pivotal
trials; the newest long-acting insulin analogue , degludec, was filed in 2011 following 12
pivotal trials
 The first pivotal trial for Merck’s simvastatin, published in 1994, recruited 4,400 pts; a pivotal
trial for Merck’s Anacetrapib is recruiting more than 30,000 pts
According to a study (E. David et al.), between 1997 and
2010, the cost of development increased by 8%
Declining R&D Success Rates
(adapted from Bain drug economics model, 2003* and from KMR 2007-2011**)
1995-2000*
2000-2003*
2003-2007**
2007-2011**
Preclinical: 7
Phase I: 6
Phase II: 3
Phase III: 1.5
Regulatory: 1.1
Launch:
1.0
Preclinical: 13
Phase I:
9
Phase II: 5
Phase III: 1.6
Regulatory: 1.1
Launch:
1.0
Preclinical: 24
Phase I: 15
Phase II: 7
Phase III: 1.8
Regulatory: 1.2
Launch:
1.0
Preclinical: 30
Phase I: 19
Phase II: 9
Phase III: 1.9
Regulatory: 1.2
Launch:
1.0
Cum. Success Rate: 14 %
Cum. Success Rate: 8 %
Cum. Success Rate: 4 %
Cum. Success Rate: 3 %
2% for Small molecules and 11% for Biologics
Small Molecules 43.6 (61%) 26.6 (42%) 11.1 (18%) 2.0 (60%) 1.2 (85%) 1
Biologics
8.8 (75%) 6.6 (56%) 3.7 (44%) 1.6 (79%) 1.3 (79%) 1
PC
Ph.I
Ph.II
Ph.III
Reg
Launch
• There are other reports with different numbers (e.g. according to E.David et al., between
1997 and 2010, the cumulative PoS lost 5 points in %), however the negative trend remains a constant !
Increased time for ClinDev
 While the time spent in Discovery has remained stable over time at
4.5 years, the time spent in development has increased significantly
 According to E.David et al. , between 1997 and 2010, clinical development time
has increased by 15 months
 According to KMR, between 1998 and 2011, clinical development time has
increased by 2 years (from 11.5 to 13.7 years)
 The total time for development currently averages 8 years, with high
variability by TA (adapted from KMR report)
 Phase I => 2 year
 Phase II => 3 year
 Phase III => 3 year
LO, Ph.II & Ph.III as main cost drivers
Even though Phase III has by far the highest average cost per project ($150M), the higher number of
projects and the capital cost over time make Phase II and Lead Optimization average costs higher!
Early commercial involvement will help to make the right choices
to optimize resource allocation in LO and Early Development !
S.M.Paul et al. “How to improve R&D productivity”
Nature Reviews/DrugDiscovery vol.9; March 2010; p.203-214
Decrease of PoS for Ph.II and Phase III are the two most important cost
drivers: the critical role of VoI
Impact on the average cost of drug discovery & development ($1,778M) of the ten most important cost drivers
S.M.Paul et al. “How to improve R&D productivity”
Nature Reviews/DrugDiscovery vol.9; March 2010; p.203-214
Optimizing R&D Efficiency: the cost of an NME can be cut
by 50%
 Reduce cost and time of development




Trial size, sites/investigators, CRO management, low-cost countries, partnerships
Reduction of ph.III from 2.5 to 2 years will reduce the cost by $100M
Adaptive/seamless ph.II/III trial designs will save time and cost
Time/cost of development is “disease-specific” (e.g.CV worse than ID)
 Optimize PoS
 Reduce attrition in ph.II/III with early PoC studies (reliable biomarkers)
 More validated/druggable targets & greater use of translational phenotypic assays
 Sufficient number of projects by phase to ensure 1 launch/year:
 If PoS in ph.III increases from 70% to 90% , the number of products entering ph.I can
decrease from 9 to 7
 Redirecting resources from drugs with low PoS: e.g. 1 ph.III has same cost of 10 Ph.I
 Moving from FIPCo (Fully Integrated Pharma Co) to FIPNet (Fully Integrated Pharma
Network)
S.M.Paul et al. (Nature Reviews/Drug Discovery, vol 9; March 2010; p.203-214)
R&D Effectiveness Assessment
Pharma needs to create new medicines able to
surpass an ever-improving SoC, being the
victim of its own success.
No more low-hanging fruits!
Decreasing Sales for New Products
716M
556M
1992-1996
1997-2001
482M
2002-2006
408M
2007-2011
Average Peak Sales for New Products ($ M)
Adapted from Accenture Research Report, based on data from various sources
Variable Return from New products
New drugs launched in 2000-2006 showed an average IRR
of 7.5% ; they can be grouped in quartiles based on
revenues generated:
 1st quartile: 2% of them with IRR of 28%
 2nd quartile: 4% of them with IRR of 12%
 3rd quartile: 40% of them with IRR of 8%
 4th quartile: 54% of them with IRR of 6%
E David et.al., Nature Reviews / Drug Discovery vol.8, Aug.2009 p.509-510)
The questionable value of new medicines: BIC or
me-toos?
 From its inception in 2004, Germany’s IQWIG has classified 70% of
drugs reviewed as drugs with “unproven benefit”; from Jan.2011,
Value Dossier required with NDA
 According to the French HTA system, in recent years, only 12% of
new medicines are bringing significant clinical benefits over SoC;
nearly 60% of new medicine had no additional value
 Between 1998 and 2008, the UK’s NICE granted restricted or no
market access to almost 60% of the drugs launched by the top ten
pharma companies
 In the US, only one third of the new medicines achieve a
formulary listing that allows unrestricted use, with
reimbursement
Optimizing R&D Effectiveness: Focused portfolio
on Core TAs
 Several challenges (cheaper generics, more aggressive payors, more difficult
science, etc.) have increased the competitive requirements
 Companies are realizing they cannot compete effectively in every TA
 70% of the BBs launched in 1970-2000 were in TAs where the marketer had significant
presence
 Category leaders completed 2 times more deals, 70% higher success rate in ClinDev
and had 5 times more revenues as compared to other competing firms
 More attention to be paid to specialization of capabilities and integration of focus
areas from research through to commercialization
 Changing mindset: from “playing everywhere” to “play to win”
Optimizing R&D Effectiveness: Darwinian approach to
decision making
 Objective evaluation of projects and elimination of decision-making biases,
allowing only the best program to survive
 Changing mindset from “win with any innovation” to “raise the bar for
innovation”
 From “targeting the broadest population in which the new drug has statistically
significant (though often clinically marginal!) benefits” to “targeting patients with the
greatest benefit”
 Recently launched Pfizer’s crizotinib targets only 5% of lung cancer patients with ALK
oncogene, but has very high efficacy: with a target population WW of 50,000 this
product will surpass $500MM by 2015
 Incentives need to give less weight to milestone accomplishments and more to
measures of quality and strategic intensity
 Ability to re-allocate resources across different franchises to better invest R&D
money
Optimizing R&D Effectiveness: Regain trust of all
customers
 From PoC to PoC&EB (Proof of Concept and Economic Benefit): providing
value to all customers
 Instead of searching for a gap in the market where to sell a product in development,
design a product to fill a well-identified market gap
 Gap defined by the needs of the patients, the physicians, the regulators, the HTAs
and the payors
 From artificial patients identification with surrogates to patient segmentation to
maximize drug value (=>personalized medicine!)
 Identify and define, before starting clinical development, what outcomes matter
to all customers, including what evidence is required
 From treating payors as a problem to solving their problems
Assessing the Balance between
Efficiency and Effectiveness
The Role of Portfolio Management
R&D Productivity - Implications
 R&D Productivity grows with:
Number of projects
PoS
Value of projects
 R&D Productivity decreases with:
Time to complete projects
Cost of projects
But these 5 elements are inter-connected and Portfolio
Management, if started early, can help optimize them!
DOP – Disease Opportunity Profile
 Available to R&D at Target Identification
 Defines the “opportunity” and the “challenges” in the marketplace,
clarifying KSFs (Key Success Factors)
 The DOP is continuously updated and shared with the critical players
in R&D
 The Assessment of the disease focuses on:






Definition of Target Population
Evaluation of the Level unmet medical need
Identification of key differentiators from a ‘gold standard”
Assessment of the Competitive landscape (incl. LOEs)
Understanding of key P&R requirements
SWOT analysis
 DOP scoring and threshold
Disease Target Assessment quantifies the opportunity
in each of the key attributes of medical need
Opportunity
Level
Achieved by
Gold Standard
Efficacy
Safety /
Tolerability
Convenience
Mortality
Morbidity
Cost
Measuring Unmet need using published objective
clinical trial data
symptom relief, slowing of progression, restoring lost
Efficacy: function, pharmacokinetics
Side effects: frequency and severity of each
Convenience: mode and frequency of dosing
Mortality: age-adjusted excess risk of mortality
pain, disability, hospitalization, quality of life,
Morbidity: complications
Costs: direct (drug and non-drug) and indirect
Impact of
Impact of efficacy on mortality, morbidity, and cost
Efficacy:
Assign a score to
each component of
unmet need:
0
1
No unmet need
2
3
4
5
Substantial unmet need
TCP – Target Candidate Profile
 Available to R&D at Lead Optimization
 It states the minimal attributes for the new product to be
commercially viable
 It identifies, among all product attributes, the key “value drivers”
 It focuses on the key differentiators from SoC
 It helps the performance of a more effective “Lead Optimization”
phase while offering guidance for GnG decisions
 It includes a “bucket” forecast with possible upsides
 It is always complemented by the relevant DOP
 TCP scoring and prioritization (facilitating the “early kill”)
Profile Alpha is highly innovative vs. current gold
standard
Sources of Difference in Unmet Need between Alpha and Current Gold Standard
Unmet Need Score
2.50
2.45
5.5%
2.40
-0.8%
2.35
2.30
2.5%
0.6%
2.48
1.2%
2.25
0.1%
0.4%
2.20
2.25
2.15
2.10
Current Gold
Standard
Efficacy
Side Effects Convenience
Mortality
Morbidity
Direct Cost
Indirect Cost
Unmet need scores range from 0 (no unmet need) to 5 (substantial unmet need)
Sum of % differences equals overall relative improvement in unmet need: 9.6%
Note: Bars may not sum to overall % due to rounding
Alpha
The greater the reduction in medical need, the
larger the peak share achieved
100%
Viagra
Peak Patient Share
90%
Aricept
80%
70%
60%
Lipitor
50%
Vfend
Advair
40%
30%
Singulair
20%
Boniva
Lescol
0%
-15%
-10%
-5%
0%
5%
10%
15%
Percent Reduction in Unmet Need
Confidential Equinox Group Information
20%
25%
30%
Strategic intent
DOP & TCP to support Go / No-go decisions while
achieving efficient ROI
Disease Opportunity Profile
Target Candidate Profile
Unmet Need
Competitive
Landscape
Market
Overview
Payer Pressure
Regional
Contribution
Decision Criteria
• Commercial Viability
• Identification of
Non-negotiable
Attributes
• Additional value
drivers for potential
economic upside
# Killer Experiments
# Go / No-go
VOI (Value of Information) to Assess early
experiments
70%
yes
30%
Positive
Result
Revenue
Negative
Result
$0
Positive
Result
Clin.
Trial
Cost
$100M
30%
Negative
Result
Clin.
Trial
no
Revenue
$0
$500M
yes
$75M
Cost ?
70%
Test
$50M
no
no
Clin.
Trial
$0
30%
Positive
Result
Revenue
$500M
Negative
Result
Revenue
$0
yes
$100M
70%
Adapted from N.Rosati, Expert Rev.Pharmacoeconomics Outcomes Res. 2(2), 2002
Prioritizing the Portfolio
1.
2.
3.
4.
5.
6.
7.
Strategic Fit
Core competencies (clinical and commercial)
Technical feasibility and complexity
Criticality of launch timing
PoS (including tractability, target validation, etc.)
Clinical cost to launch
Commercial opportunity:
1.
2.
3.
4.
5.
8.
Level and prevalence of unmet medical needs
Competitive pressure
Product Differentiation and Value proposition
Payors’ pressure and P&R risks
Expected Peak sales
Financials (eNPV, ROI, IRR, payback period, etc.)
Different weights by Phase and BIC vs FIC
George W. Merck, 1950
“ We try never to forget that medicine is for the people.
It is not for the profits. The profits follow, and if we
have remembered that, they have never failed to
appear. The better we have remembered it, the
larger they have been.”