Personalized Medicine – future impact

Pharma industry perspective

Alison Ayers
Vice President, Oncology Worldwide Commercial Development,
Oncology Business Unit
Pfizer Inc

Confidential and Proprietary – Not for further distribution

What Are We Trying to Accomplish?

Right Drug Right Patient Improved Survival

New treatment
1.0 Comparator

Overall Survival Probability

0.8

0.6

0.4

0.2

0
0 6 12 18 24 30 36
Drug targeted to specific Patient identified through
oncogene or aberrant molecular profiling of Significant improvement
pathway driving the their tumor in survival
specific tumor

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
What Are the Issues in “Personalized medicine”
Development and Commercialization?

Drug • Identification of the best targets to pursue

Development
• Identification of right biomarker prior to Ph3 initiation
• Complexities of Clinical Trial design for co-development of
drug and diagnostic
• Tissue availability

Regulatory • Path for simultaneous approval of Dx and Tx

• Regulatory pathway for “do-overs” and next generation Dx
• Labeling requirements for companion diagnostics

Commercial • Commercial viability of drug development in small patient

populations
• Payer willingness to reimburse
• Cost and access to companion diagnostic
• Logistics
DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
CDRH View of Optimal Drug/
Diagnostic Co-Development Paradigm

Clinical test validation of a new diagnostic for use in selecting drug therapy or avoiding
drug therapy should be characterized by studying the test in relation to the intended
clinical outcome in patient subgroups with and without the analyte of interest:

SOC + Drug
Mutant +
SOC + Placebo
Biomarker data
All Subjects
available
SOC + Drug
Mutant ‒
Issues SOC + Placebo

• Complex randomization + extensive molecular profiling means large, expensive,

long study
• IRB and patient concern
– 50% of patients with biomarker WILL NOT get targeted drug
– 50% of patients WITHOUT biomarker WILL get targeted drug with less likelihood of benefit
– Cross-over resolves some ethical issues, but impairs Overall Survival end-point

FDA Drug/Diagnostic Co-Development Paper – April 2005

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
Futur State: Drug and Diagnostic Co-Development
Paradigm

IPASS trial experience: ability to gain molecular data in clinical trial setting
Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma; Mok et al, NEJM Sept 09

261
132
1217 437 Mutant +
129
Biomarker data
All Subjects
available
91
Mutant ‒
Issues 85
176

Future State:
Streamlined Ph 3 clinical trials focusing on target patient population
Adaptive and cross-over designs; acceptance of PFS as endpoint
Simultaneous approval of companion diagnostic

FDA Drug/Diagnostic Co-Development Paper – April 2005

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
The Challenge of Developing Personalized Medicine
for Small Patient Populations

Number of
Tumor Type Molecular Event Rate Patients in US

Sarcoma ALK Translocation ~90% ~75-150 patients

ALK Mutation or
Neuroblastoma ~20-25% ~200 patients
Amplification

ALCL NPM-ALK Translocation ~70-80% ~300-1,200 patients3

NSCLC EM4-ALK Translocation ~3-7% ~7,600 patients

TOTAL ~8,800 patients

Future State: Development and Regulatory Path Based on the

Molecular Defect Regardless of Histology

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
Case Study: Gefitinib – First EGFR Inhibitor to Gain
(and Lose, and Gain) Regulatory Approval
May 2003 Dec 2003 Sept 2008 April 2009
“All Comers” No New Patients Phenotypic Genotypic

• FDA grants • 2 Ph3 trials in • IPASS study in • EMEA approves

accelerated approval unselected patients selected patients Iressa in patients with
in NSCLC on the show no survival (Asian, never/light NSCLC with
basis of response benefit smokers) reports activating EGFR
mutations
rate in 2Ph 2 trials positive data
(216 patients) • FDA restricts use • August 2009 AZ
of Iressa • Iressa superior to announces deal with
• Survival data from – No new patients standard chemo for DxS for companion
ongoing Ph3 trials Progression Free diagnostic for EGFR
required for full • MAA withdrawn in Survival (HR=0.74) mutation
approval Europe
• Iressa only marketed
in Japan

Market Potential Market Potential Market Potential Market Potential

US, EU US, EU US, EU US, EU

• ~580,000 • Zero • 130,000 • 60-90,000

patients p.a.
• never/light smokers • EGFR mutation +ve
with NSCLC (assuming US
approval can be
gained)
DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
Gefitinib IPASS study – significant clinical benefit in
mutation positive : negative impact in mutation negative
Overall: Phenotypically Selected Population EGFR-Mutation – Positive

Progression-Free Survival
1.0
Progression-Free Survival

1.0
Hazard ratio, 0.74 (95% CI, 0.65-0.85) Hazard ratio, 0.48 (95% CI, 0.36-0.64)
p<0.001 p<0.001
0.8 0.8

Probability of
Probability of

Events: geftinib, 453 (74.4%); Events: geftinib, 97 (73.5%);

carboplatin + paclitaxel, 497 (81.7%) 0.6 carboplatin + paclitaxel, 111 (86.0%)
0.6

0.4 0.4

Carboplatin Gefitinib Carboplatin Gefitinib

0.2 0.2
+ paclitaxel + paclitaxel
0.0 0.0
0 4 8 12 16 20 24 0 4 8 12 16 20 24
Months Since Randomization Months Since Randomization

Gefitinib
Gefitinib Superior
Superior to
to Chemo
Chemo in
in EGFR
EGFR
Mutant
Mutant Population
Population
EGFR-Mutation – Negative
Progression-Free Survival

1.0
Hazard ratio, 2.85 (95% CI, 2.05-3.98)
0.8
p<0.001 Future State:
Probability of

Events: geftinib, 88 (96.7%);

0.6 carboplatin + paclitaxel, 70 (82.4%)
• Knowledge of the “right”
0.4 Carboplatin
+ paclitaxel biomarker before initiation of Ph 3
0.2
Gefitinib
0.0 • Genomically guided treatment for
0 4 8 12 16 20 24
lung cancer with extended survival
Months Since Randomization

Chemo
Chemo Superior
Superior to
to Gefitinib
Gefitinib in
in Mutation
Mutation
Negative Population
Negative Population

Mok et al. NEJM 361 (10): 947, Figure 1. September 3, 2009

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
Potential Treatment Alogrithm for Advanced-Stage Non-
Small-Cell Lung Cancer: (Good Performance Status): 2012

Potential
Potential Treatment
Treatment Algorithm
Algorithm

ALK
ALK Fusion
Fusion ++ ve
ve EGFR
EGFR Mutation
Mutation Molecular ERCC1
ERCC1 Low
Low Molecular ERCC1
ERCC1 High
High

Molecular
Chemotherapy
Chemotherapy
ALK
ALK Inhibitor
Inhibitor EGFR
EGFR inhibitor
inhibitor TS
TS High
High TS
TS Low
Low (Nonplatinum)
(Nonplatinum)

Molecular

RRM1
RRM1 Low
Low RRM1
RRM1 High
High

Platinum/
Platinum/ Platinum/
Platinum/
Gemcitabine
Gemcitabine Platinum/Other
Platinum/Other ±± Pemetrexed
Pemetrexed ±±
(or
(or Other)
Other) ±± Bevacizumab
Bevacizumab or
or Bevacizumab
Bevacizumab or
or
Bevacizumab
Bevacizumab or or Cetuximab
Cetuximab Cetuximab
Cetuximab
Cetuximab
Cetuximab

Adapted from: Gandara DR et al. Clin Lung Cancer. 2009:10:6; 392-394.

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
The Issue of the Tissue

Potential Genomic Tests for Lung

Cancer Treatment Decisions
EGFR Amplification IHC
EGFR Mutation PCR
EML4-ALK FISH
Fine Needle Aspirate ERCC1 IHC
RRM1 IHC

• Limited availability of tissue via fine needle aspirate

• Varying tissue acquisition sources
• Cost of conducting multiple molecular diagnostic tests

Future State:
Multi-plex tests to optimize data from limited tissue
Next generation and non-invasive tests
Clear regulatory path for approval

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
Impact of Personalized Medicine on Adjuvant therapy
for Breast Cancer : the impact of Her2 Mab
HERA Trial
100%

80%

60%
DFS (%)

40%

20%

0%
0 6 12 18 24 30 36
Months from Randomization
Trastuzumab + Chemotherapy
Chemotherapy alone (Control)

Events 3-Yr DFS HR 95% CI p Value

Trastuzumab 218 80.6 0.63 0.53, 0.75 <0.0001
Control 316 74.0

Smith IE. Lancet. 2007.

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
Impact of Personalized Medicine on Adjuvant therapy
for Breast Cancer : the impact of Her2 Mab

HERA Trial Future State: Biomarker and

100%
Treatment Research focus
Resistant
80% Effective

60%
DFS (%)

Future State: Identify and exclude Unnecessary

40% from clinical trials to “enrich” for No improvement
high risk population vs chemo alone
20%
Martine Piccart-Gebhart

0% Optimizing Anti-HER2
treatment for Early BC
0 6 12 18 24 30 36
ECCO15/ESMO34 Sept 09
Months from Randomization
Trastuzumab + Chemotherapy
Chemotherapy alone

Events 3-Yr DFS HR 95% CI p Value

Trastuzumab 218 80.6 0.63 0.53, 0.75 <0.0001
Control 316 74.0

Smith IE. Lancet. 2007.

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
Personalized Medicine Development:
Positive and Negative drivers

Pros Cons

• May improve probability of • Label limited to specific population

success for drug discovery and • May need to pay for Diagnostic
PHARMA

development development
• May reduce Ph3 study size and • Cost and scope of molecular profiling
costs if end points achieved more
rapidly/more robust difference • May select wrong biomarker
e.g., EGFR vs k-ras
• Potential for greater clinical
benefit in targeted population • Many logistical issues e.g., tissue
availability, regulatory path

Pros Cons

• More compelling “cost per QALY” • Cost of Diagnostics

PAYERS

• Enables payer to predict • Ability to change clinical practice based

budget impact on predictive/prognostic diagnostic test
• Easier to demonstrate higher • Dx may become form of “prior auth”
value for target populations –Dependent on accuracy of Dx test

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
Future opportunity for accelerated Drug Development
timelines?

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
Future opportunity to reduce ineffective healthcare
expenditure

• Simon-Kucher and Partners Spring 2009 Health Insights

Publications, SK&P reported
“…in discussions with Simon-Kucher and Partners, most payers
have overwhelmingly favored paying for biomarker tests,
particularly in oncology. For these payers, the biomarker test is an
investment and the goal is to reap the savings of not using the
high cost therapies for patients unlikely to respond to treatment”

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only
Potential New Drug Discovery and Development
Paradigm in the Era of Personalized Medicine

Molecular Guided Development

Targeted Research Right Drug, Right Patient
Plan

• Target oncogenes correlated
with malignant cell growth and
metastasis
• “Roadmap” for development
teams for biomarkers and
patient selection
• Molecular guided strategy for • Routine use of Molecular
clinical trial patient selection Diagnostics to Guide
– Trials in enriched populations treatment decisions
to accelerate POC and target • Optimized patient outcomes
higher response/survival and improved survival
• Focus on poor prognosis • Robust Value Proposition
patients not well served by
current treatments
• Co-development of Companion
Diagnostic

DRAFT – Subject to Further Review – Company Confidential and Proprietary – Internal Use Only