RiskAssessmentCBIFalcon25June2012

Risk Assessment and
Management for New Product
Planning
Robert J. Falcon
Endo Pharmaceuticals

Planning is Critical for Success
“How you run the race - your
planning, preparation, practice, and
performance - counts for
everything. Winning or losing is a
by-product, an aftereffect, of that
effort.”
John Wooden
©2012 Endo Pharmaceuticals, Inc. All rights reserved. 2

Overview
 Why Managing New Product Risk is Important
 What problems are we having and why?
 Forecasts and the scientific method
 Importance of the Target Product Profile (TPP)
 Sample TPP & PPOs – Kestrel Pharmaceuticals
 Use Decision Analysis to capture & quantify
development options
 Keep it simple – Manage forecast complexity

Concern Over Early Stage New Product Development
– Commercial potential [value] is
being assessed at the beginning of
every development program1
– “Sales estimates” identified as the
parameter management most
wishes to be more reliable in
biotech valuations2
– 80% of products launched in past
decade did not live up to
expectations3
– # of FDA drugs approved per each
$1B of R&D spend halved every 9
years since 1950 4
– $4B-$12B = average R&D spend
for every drug approved (1997 –
2011)5
1- John LaMattina, former Pfizer VP R&D, LifescienceLeader.com, Dec 2011
2 – Dec 2011 Linked in Survey by Avance (www.avance.ch)
3 – Campbell Alliance, PA Bio, March 29, 2012 “What Decisions Impact Value the Most”
4 – Nature Reviews, Drug Discovery, March 2012 p. 191
5 – Forbes, March 12, 2012 p. 38

Average Spend per New Drug Approved is Soaring
Research Spending Per New Drug
Company Ticker
Number
of drugs
approved
R&D
Spending
Per Drug
($Mil)
Total R&D
Spending
1997-2011
($Mil)
AstraZeneca AZN 5 11,790.93 58,955
GlaxoSmithK
line GSK 10 8,170.81 81,708
Sanofi SNY 8 7,909.26 63,274
Roche
Holding AG RHHBY 11 7,803.77 85,841
Pfizer Inc. PFE 14 7,727.03 108,178
Johnson &
Johnson JNJ 15 5,885.65 88,285
Eli Lilly &
Co. LLY 11 4,577.04 50,347
Abbott
Laboratories ABT 8 4,496.21 35,970
Merck & Co
Inc MRK 16 4,209.99 67,360
Bristol-
Myers
Squibb Co. BMY 11 4,152.26 45,675
Novartis AG NVS 21 3,983.13 83,646
Amgen Inc. AMGN 9 3,692.14 33,229
Sources: InnoThink Center For Research In Biomedical Innovation; Thomson
Reuters Fundamentals via FactSet Research Systems
Source: http://www.forbes.com/sites/matthewherper/2012/02/10/the-truly-staggering-cost-of-inventing-new-drugs/

Company Ticker
Number
of drugs
approved
R&D
Spending
Per Drug
($Mil)
Total R&D
Spending
1997-2011
($Mil)
GlaxoSmithK
line GSK 10 8,170.81 81,708
Sanofi SNY 8 7,909.26 63,274
Roche
Pfizer Inc. PFE 14 7,727.03 108,178
Johnson &
Johnson JNJ 15 5,885.65 88,285
Eli Lilly &
Co. LLY 11 4,577.04 50,347
Abbott
Merck & Co
Inc MRK 16 4,209.99 67,360
Bristol-
Myers
Squibb Co. BMY 11 4,152.26 45,675
Novartis AG NVS 21 3,983.13 83,646
Amgen Inc. AMGN 9 3,692.14 33,229

Company Ticker
Number
of drugs
approved
R&D
Spending
Per Drug
($Mil)
Total R&D
Spending
1997-2011
($Mil)
GlaxoSmithK
line GSK 10 8,170.81 81,708
Sanofi SNY 8 7,909.26 63,274
Roche
Pfizer Inc. PFE 14 7,727.03 108,178
Johnson &
Johnson JNJ 15 5,885.65 88,285
Eli Lilly &
Co. LLY 11 4,577.04 50,347
Abbott
Merck & Co
Inc MRK 16 4,209.99 67,360
Bristol-
Myers
Squibb Co. BMY 11 4,152.26 45,675
Novartis AG NVS 21 3,983.13 83,646
Amgen Inc. AMGN 9 3,692.14 33,229
Pharma Only

Benchmarking Opportunity
What are Lilly, BMS, Merck, and Novartis
doing differently than the others?

Higher Cost per Approval Driven By Failures
Source: NATURE REV. DRUG DISC. Sept 29 2011 p. 527
But why are Phase II failures not increasing at Phase I & III rate?

Are Phase II Trial Hurdles set too low?
– Approved drugs enrolled median 171 people in each Ph II, while
drugs failing Ph. III or later enrolled median 69 people (01/00-3/11). 1
– Failures in Ph III due to inadequately defined endpoints & no efficacy 2
– New mechanisms of action have higher failure rates, but were not tested
thoroughly enough in Ph II
– Ph II endpoints were not objective or were patient reported
– Why? Wishful thinking by project teams and incentives tied to targets
– Lilly vs Pfizer POS in their Phase III Alzheimer trials
1- Nature Medicine, Vol 18 No 4, April 2012 (http://lesusacanada.org/docs/presentations/deloitterecap-lhp_les-nj_mar-22_2012.pdf?sfvrsn=2)
2 – “Why Products Fail in Ph III”, In Vivo, April 2006 (Trials from 1990-2002)
Source: NATURE REV. DRUG DISC.
Sept 29 2011 p. 527

Why Should Forecasting be Concerned??
 Success or failure of compound affects your job
 Sharing your concerns in an appropriate manner
may open some eyes
 History may not repeat itself, but the patterns
(such as weak Ph II trials) speak for themselves

Early vs Late Stage New Product Development
Source: "A More Rational Approach to New-Product Development."
Harvard Business Review, March 2008, p. 99

Challenge
 Build business case like you are going to trial
 Science/Evidence Based approach – Document!
 Provide “realistic” forecasts and valuations
 Base Case = 50% Case
 “Gretzky” vision
 Model the market of tomorrow, not today
 Multiple market scenarios likely needed
 Resist “Irrational Exuberance”
 Risk associated with inflated forecasts
 Consider all “options,” but let the data lead!

Forecasts and the Scientific Method
“Scientific method refers to a body of techniques
for…acquiring new knowledge, or correcting and integrating
previous knowledge” 1
“[A] basic expectation is to document, archive and share all
data and methodology so they are available for careful
scrutiny by other scientists, giving them the opportunity to
verify results by attempting to reproduce them.” 2
“The goal of a scientific inquiry is to obtain knowledge in the
form of testable explanations that can predict the results of
future experiments.” 3
1 – Wikipedia http://en.wikipedia.org/wiki/Scientific_method
2 - Ibid
3 - Ibid

“Take Issue with the Assumptions!!!”
 Sales forecast is never wrong, but….
 Did you forecast the appropriate future market?
 Is the target indication appropriate?
 Is your assumption source(s) valid?
 Should you use an Incidence, prevalence, or
patient flow model for your chronic therapy?
 Oncology – Start with deaths for met drug?
 Will the diagnostic be ready?
 How will BD activities affect pipeline forecast?
 Must measure incremental value!

Forecast is Dependent Upon Assumptions of Other Groups
Pricing &
Reimb
Health
Outcomes
Marketing
Medical
Clinical
Market
Research
Regulatory
R&D
Forecasting
 Need to get all contributors on
the same page
 Forecaster must communicate
early the impact of decisions made by
each contributor.
 e.g., Clinical Dev Options
 The vehicle through which all
components/inputs of the forecast first
put side by side is the Target Product
Profile (TPP)

Target Product Profile
 Forecast with no TPP? You’re making “stuff” up!!
 Forecaster must add his/her input to TPP
 TPP must reflect future world(s), not today
 Step out of your comfort zone, and push others do
the same
 How does each TPP contributor impact your
forecast?
 Challenge teammates tactfully
 Facilitate the discussion to assure everybody is on
same page
 Start with the end in mind – TPP paints the picture

Example – Regulatory vs Reimbursement
 “Growing tension” between data requirements
 Regulatory – Absolute efficacy data
 Reimbursement (Payers) – Effectiveness data
 What is the economic value of this drug??
 Comparative effectiveness (US)
 Value-based pricing (EU)
 Indirect comparison or Head-to-Head required?
 Does comparator differ by market?
 Preliminary P & R strategy S/B in place by Phase
II (ideally) or entering Phase III (required)
 Forecaster needs to be matchmaker!
Source: “Addressing Reimbursement during Development,”
PharmaVoice, May 2012, P 20-22

TPP Input - Be Like Columbo (Peter Falk)
– Disheveled and disarming Lt.
Columbo
– Consistently underestimated
by his suspects
– Formidable eye for detail and
meticulously dedicated
approach
– Ask a lot of unassuming
open-ended questions
– Restrain any personal
opinions
– Let the evidence lead
analysis

Sample TPP – 1 of 3
Kestrel Pharmaceuticals - KP101 Target Product Profile
(Note: Company name and “Talon Inhibitor” are fictitious and for presentation only)
Base PPO High PPO Low PPO
Indication(s) and
Usage
NSCLC: KP101, used in
combination with carboplatin, is
indicated for 1st
Line treatment of
patients overexpressing Talon.
NSCLC: KP101, used alone
(monotherapy), is indicated for 1st
Line treatment of patients
overexpressing Talon.
NSCLC: KP101, used in combination
with carboplatin, is indicated for 1st
Line treatment of patients regardless
of Talon status.
Route of
Administration
KP101 is an oral tab KP101 is an oral tab KP101 is an oral tab
Dosage Form(s) 2.5 mg tab 4.0 mg tab 2.5 mg tab
Dosage Regimen One 2.5mg tab QD. Patients
continue to receive KP101 until
unacceptable toxicity or disease
progression.
One 4.0mg tab QD. Patients
continue to receive KP101 until
unacceptable toxicity or disease
progression.
One 2.5mg tab QD. Patients continue
to receive KP101 until unacceptable
toxicity or disease progression.
Efficacy vs.
Competition
KP101 in combination with
carboplatin demonstrates 50%
improvement in Talon+ patients
over Combo X, the current Standard
of Care.
ORR= 30%
TPP= 9.0mos
OS= 22mos
KP101, used alone, demonstrates
100% improvement in Talon+
patients over Combo x, the current
Standard of Care.
ORR= 40%
TPP= 12.0mos
OS= 30 mos
KP101/Carbo combination in all
patients (regardless of Talon status)
demonstrates 20% improvement in
OS, but equal ORR and TTP
compared to Combo X, the Current
Standard of Care.
ORR= 20%
TPP= 6.0mo
OS= 18 mos

Side Effects vs.
Competition
10% increase in gr 3 and 4
neutropenia, similar leucopenia,
mucositis, and peripheral
neuropathy compared Combo X
but a 50% reduction in mucositis
and peripheral neuropathy
compared to Combo x
mucositis, and peripheral neuropathy
compared Combo X
Pharmacoeconomics KP101 must show significant
improvement in survival over
standard of care to justify cost of
therapy and increase in AE profile.
The Talon Screen will identify
patients most likely to respond.
Price of TALON screen must be
cost effective to justify use vs the
cost of empirically treating all
patients.
KP101 must show significant
The Talon Screen will identify
patients most likely to respond.
Price of TALON screen must be
cost effective to justify use vs the
cost of empirically treating all
patients.
KP101 must show significant
Although patients will not incur cost
of the Talon Screen, efficacy will be
hampered relative patients who
targeted through use of the TALON
screen.
Positioning Thesis KP101 in combination with
Carboplatin will be the SOC 1st
Line therapy for all TALON+
patients with metastatic NSCLC.
KP101, used alone, offers superior
efficacy and eliminates platinum-
related side effects and will be the
SOC 1st
Line therapy for all
TALON+ patients with metastatic
NSCLC.
KP101 in combination with
Carboplatin offers superior OS
relative to current Standard of Care
for 1st
Line patients with metastatic
NSCLC.

Competitive
Assessment
Taxane-Platinum doublets are
currently used for 1st
Line NSCLC
patients. The competitive pipeline is
full of agents that are selective for a
particular molecular target, but none
are selective for Talon.
Consideration should be given to
combining KP101 with other agents
broadly used in NSCLC
Line NSCLC
are selective for Talon.
Line NSCLC
are selective for Talon. Consideration
should be given to combining KP101
with other agents broadly used in
NSCLC
Pricing and
Market Access
Strategy
Preliminary pricing research
indicates potential for premium-
pricing to SOC and parity pricing to
other recently launched selective
therapies. Assumes proper clinical
trial health outcomes support the
TPP
Preliminary pricing research
strongly supports potential for
premium-pricing to SOC and parity
pricing to other recently launched
selective therapies. Assumes proper
clinical trial health outcomes
support the TPP
Preliminary pricing research indicates
likelihood of parity or slight premium
pricing to SOC but less pricing of
other recently launched selective
therapies. Assumes proper clinical
trial health outcomes support the TPP
PPO Probability 60% 25% 15%

TPP Drives Decision Analysis
 Decision Analysis provides a means for dialog
between the forecaster and project team
 This dialog enables uncertainties, concerns,
expectations, and assumptions to be clarified
 Forecaster must quantify many qualitative product
characteristics to impact the forecast
 Better side effect profile
 Oral introduced into an IV dominated market
 ER tab launched into an IR market
 Will our product’s characteristic differentiation
generate any traction in the market?

What is Decision Analysis?
 Iterative process
 Involves uncovering specific key factors that affect the
forecast and their magnitude
 Allows you to concentrate on what is important
 Uses subjective probability assessments from subject
matter experts to obtain the likelihood of future events
 Provides the means to make an informed rather than
purely intuitive or “gut” decision
Source: Skinner, Intro to Decision Analysis, 2nd ed

Apply Decision Analysis to Target Profile
 Prepare forecast model for each of the Target
Product Profiles (or PPOs)
 Early stage = high level
 Medical/Clinical need to estimate probabilities
 Utilize ranges to estimate typical uncertainties
such as share, price, for each scenario
 Appropriate colleagues need to contribute
 Utilize monte-carlo simulation*
 Correlate variables where appropriate
 Sensitivity Analysis (tornado) will expose areas to
target market research
* Make sure senior management is on board

Share Decision Tree Hypothetical Example
Start
Monotherapy
or Combo?
Improvement in
Efficacy over SOC?
Monotherapy
Combo Therapy
Side Effects
Same, Better,
Worse
Side Effects
Same, Better,
Worse
67% - 100%
33% - 67%
Side Effects
Same, Better,
Worse
10% - 33%
Improvement in
Efficacy over SOC?
Side Effects
Same, Better,
Worse
33% - 67% Side Effects
Same, Better,
Worse
10% - 33%
Side Effects,
Same, Better,
Worse
67% - 100%
Other Factors that will determine share:
1. What other new Products will launch? How do they compare?
2. What currently marketed products will go generic?
3. What new or emerging alternative therapies might make this
flowchart obsolete?
Decision Tree will have dozens of branches. Pick the most likely
(50% case), and at least one upside and downside.
What will the impact of your chosen efficacy, side effect, &
competitive set have on share? Price? Treatment Rate?

Decision Trees
 Can help to quantify a range of assumptions
 E.g., Share based on product attributes, competitive product
launches, launch of disruptive therapies, etc
 Utilize primary market research to quantify impact
 Can be used to quantify sales, NPV, RANPV, etc
 Once all assumptions have been mapped
 Practical use
 Decision Trees must be kept to a high level
 Leverage monte-carlo simulation to cover the ranges within
each variable

Tornado Analysis
– Shows sensitivity of forecast
model to changes in variables.
– Holds all but one variable
constant, then measures impact of
remaining variable.
– Shows where the big uncertainties
are in the forecast
– Helps to target those assumptions
where reduction of uncertainty can
“tighten-up” the forecast.

Forecast Complexity – Double Edged Sword
 Forecast complexity is not an issue to forecaster
 Complexity may confuse your audience
 Forecaster use of any modeling technique that
management is not open to may be a CLM
 Tornado Analysis is normally fine
 Explaining Monte-carlo simulation to a VP unfamiliar or
uncomfortable with the term may be counterproductive
 Keep it as simple as you can, add the complexity if
you must!!

Takeaways
 Forecasting must actively drive the New Product
Planning process (especially the TPP)
 Step (and think) out of your comfort zone, and
encourage your colleagues to do the same
 Make sure the dialog with SMEs happens, but
challenge teammates tactfully (Columbo)
 Keep the early analysis to a high level – make the
experts on your team assign probabilities to
various unknowns
 If you just take assumptions that are given you
and drop them in forecast, you are expendable!!

Questions??

RiskAssessmentCBIFalcon25June2012

RiskAssessmentCBIFalcon25June2012

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to RiskAssessmentCBIFalcon25June2012

Similar to RiskAssessmentCBIFalcon25June2012 (20)

RiskAssessmentCBIFalcon25June2012

Editor's Notes