1. IRESSA A Case Study in Personalised Medicine Dr Rose McCormack
Personalised Healthcare
AstraZeneca
A Presentation for the Translational Genomics Meeting
Royal College of Physicians, London
15th January 2013

2. An Overview The drug The biomarkers and clinical trials
Implications for Diagnostics
Our experience and learnings
An Overview

3. The drug The drug The biomarkers and clinical trials
Implications for Diagnostics
Our experience and learnings
The drug

4. Gefitinib (IRESSA™): A brief overview
Gefitinib is a once-daily 250mg oral medication that targets and blocks the activity of the EGFR-TK
Gefitinib was the first EGFR-TK inhibitor to be approved for use in non-small cell lung cancer
Gefitinib has demonstrated longer progression-free survival, better tolerability and quality of life compared with doublet chemotherapy (carboplatin/paclitaxel) in first-line treatment for EGFR mutation-positive advanced NSCLC.
On the 1st July 2009 the European Commission granted marketing authorisation for gefitinib for the treatment of adults with locally advanced or metastatic NSCLC with activating mutations of EGFR-TK across all lines of therapy
Gefitinib is currently approved for the treatment of 1st line EGFR M+ advanced NSCLC patients in 85 countries worldwide
EGFR-TK, an enzyme that regulates intracellular signalling pathways implicated in cancer cell proliferation and survival.
Mutation of the EGFR gene occurs in about 10-15% of non-small cell lung cancers (NSCLC) in Europe and around 30-40% in Asia
First approved in Japan in 2002

5. The biomarkers and clinical trials
The drug
The biomarkers and clinical trials
Implications for Diagnostics
Our experience and learnings
The biomarkers and clinical trials

6. Lessons in Biomarker Analysis
Chromosome
DNA
mRNA
Protein
Gene
Copy Number
(FISH)
DNA Mutation Analysis
(e.g. ARMS)
Expression
Analysis (e.g. Array, RT-PCR etc)
Protein Expression Analysis (e.g. IHC)
PI3K
Grb-2
Ras
SOS
PTEN
Akt
Raf
MEK
mTOR
STAT 3/5
This diagram shows where the potential predictors of response to EGFR TKIs (EGFR mutations, gene copy number and protein expression) lie in the process of translating chromosome information to protein expression, and the downstream EGFR signalling pathways.
Reference
Pao & Miller. J Clin Oncol 2005; 23:
MAPK
Tumour cell
survival
Tumour cell
proliferation
Pao & Miller 2005

7. The Trials: A Brief History
ISEL, INTEREST: Unselected trials in pre-treated setting
IRESSA registration Japan
ISEL
INTEREST
2002
2003
2004
2005
2006
2007
2008
2009
EGFR protein expression
EGFR gene copy number

8. EGFR mutations: First observed in 2004
Lynch et al 2004 (New Eng J Med 350: )
Activating Mutations in the Epidermal Growth Factor Receptor Underlying Responsiveness of Non-Small-Cell Lung Cancer to Gefitinib
Reprinted with permission. © 2005 American Society of Clinical Oncology. All rights reserved.
Paez et al 2004 (Science 304: )
EGFR Mutations in Lung Cancer: Correlation with Clinical Response to Gefitinib Therapy
Mitsudomi T et al: J Clin Oncol 23 (11), 2005:
Mutations of the epidermal growth factor receptor gene predict prolonged survival after gefitinib treatment in patients with non-small-cell lung cancer with postoperative recurrence.
The presence of EGFR mutations was first described in 2004.
The frequency of these EGFR mutations was found to vary between different clinical sub-populations.
Multiple examples of the response to gefitinib in representative four patients with recurrent non–small-cell lung cancer

9. The Trials: A Brief History
ISEL, INTEREST: Unselected trials in pre-treated setting
IPASS: Clinically selected trial in first line setting
IRESSA registration Japan
ISEL
INTEREST
IPASS
2002
2003
2004
2005
2006
2007
2008
2009
EGFR protein expression
EGFR gene copy number
EGFR mutations

10. Docetaxel 75 mg/m2 every 3 weeks
INTEREST: Phase III study of gefitinib vs docetaxel in pre-treated NSCLC
Endpoints
Patients
Progressive or recurrent disease following CT
Considered candidates for further CT with docetaxel
1 or 2 CT regimens (≥1 platinum)
PS 0-2
Primary
Overall survival
(co-primary analyses of non-inferiority in all patients and superiority in patients with high EGFR gene copy number)
Secondary
Progression-free survival
Objective response rate
Quality of life
Disease related symptoms
Safety and tolerability
Exploratory
Biomarkers
EGFR mutation
EGFR protein expression
EGFR gene copy number
K-Ras mutation
Gefitinib 250 mg/day
Docetaxel 75 mg/m2 every 3 weeks
1:1 randomization
Reference
Kim ES et al. Lancet 2008; 372:
1466 patients
Kim 2008 10

11. Probability of progression-free survival Probability of survival
INTEREST Results
OS: NI margin 1.154, PP population
HR (96% CI) =1.020 (0.905, 1.150) n=1433, deaths=1169 Median survival: Gefitinib 7.6m, Docetaxel 8.0m
PFS: EFR population
HR (95% CI) =1.04 (0.93, 1.18), p=0.466
n=1316, progressions=1137
Median PFS: Gefitinib 2.2m, Docetaxel 2.7m
1.0
1.0
Gefitinib Docetaxel
Gefitinib Docetaxel
0.8
0.8
0.6
0.6
Probability of progression-free survival
Probability of survival
0.4
0.4
Reference
Kim ES et al. Lancet 2008; 372:
0.2
0.2
0.0
0.0 4 8 12 16 20 24 28 32 36 40 4 8 12 16 20 24 28 32 36 40
Months
Months
Kim 2008 11

12. INTEREST: Summary of key subgroup analyses
Overall Survival
ORR (%)
Gefitinib v. Docetaxel
Progression-free Survival
Overall
Ever smoker
Never smoker
Asian
Non-Asian
EGFR FISH+
EGFR FISH-
EGFR Mutation+
EGFR Mutation-
9.1 v Overall
Ever smoker
Never smoker
19.7 v. 8.7 Asian
6.2 v. 7.3 Non-Asian
13.0 v. 7.4 EGFR FISH+
7.5 v EGFR FISH-
42.1 v EGFR Mutation+
6.6 v. 9.8 EGFR Mutation-
Overall
Ever smoker
Never smoker
Asian
Non-Asian
EGFR FISH+
EGFR FISH-
EGFR Mutation+
EGFR Mutation-
References
Kim ES et al. Lancet 2008; 372:
Douillard et al. J Clin Oncol; 26 (May 20 Suppl.): Abstract 8001.
0.5
1.0
1.5
2.0
0.5
1.0
1.5
2.0
2.5
HR (Gefitinib vs docetaxel) and 95% CI
HR (Gefitinib vs docetaxel) and 95% CI
Kim 2008; Douillard 2010 12

13. INTEREST: Overlap of biomarkers
EGFR FISH +
n=117
249 patients evaluable for EGFR expression, FISH and mutations
n=16
n=84
+++ n=24
EGFR expression +
n=189 4
Reference
Douillard et al. J Clin Oncol 2009, in press.
EGFR mutation +
n=39 3
n=8
Douillard 2010
--- n=37 13

14. Carboplatin AUC 5 or 6 and Paclitaxel 200mg/m2 3 wkly
IPASS: Phase III study of gefitinib versus doublet chemotherapy in first line NSCLC
Gefitinib 250 mg/day
Carboplatin AUC 5 or 6 and Paclitaxel 200mg/m2 3 wkly
1:1 randomization
Patients
Adenocarcinoma histology
Never smokers or light ex-smokers*
PS 0-2
Provision of tumour sample for biomarker analysis strongly encouraged
Primary
Progression free survival (non-inferiority)
Secondary
Objective response rate
Quality of life
Disease related symptoms
Overall survival
Safety and tolerability
Exploratory
Biomarkers
EGFR mutation
EGFR gene copy number
EGFR protein expression
Endpoints
1217 patients from East Asian countries
Reference
Mok TS et al. N Engl J Med 2009; 361:
*Never smokers:<100 cigarettes in lifetime; light ex-smokers: stopped 15 years ago and smoked  10 pack yrs
Carboplatin/paclitaxel was offered to IRESSA patients upon progression
PS, performance status; EGFR, epidermal growth factor receptor
Mok 2009 14

15. IPASS: Progression Free Survival
Objective response rate 43% vs 32% p=0.0001
Probability of PFS
1.0
Carboplatin /
paclitaxel
Gefitinib N
Events
609
453 (74.4%)
608
497 (81.7%)
0.8
HR (95% CI) = (0.651, 0.845) p<0.0001
0.6
Median PFS (months) 4 months progression-free 6 months progression-free 12 months progression-free
5.7 61% 48% 25%
5.8 74% 48% 7%
0.4
Primary objective exceeded: Gefitinib demonstrated superiority relative to carboplatin / paclitaxel in terms of PFS
0.2
Reference
Mok TS et al. N Engl J Med 2009; 361:
0.0 4 8 12 16 20 24
Months
At risk :
IRESSA
609
363
212 76 24 5
Carboplatin / paclitaxel
608
412
118 22 3 1
Mok 2009 15 15 15

16. Time from randomisation (months)
IPASS: EGFR mutation is a strong predictor for differential PFS benefit between gefitinib and doublet chemotherapy
Gefitinib EGFR M+ (n=132) Gefitinib EGFR M- (n=91) Carboplatin / paclitaxel EGFR M+ (n=129)
Carboplatin / paclitaxel EGFR M- (n=85)
1.0
EGFR M+ HR=0.48, 95% CI 0.36, 0.64 p<0.0001
EGFR M-
HR=2.85, 95% CI 2.05, 3.98 p<0.0001
0.8
Probability of PFS
0.6
Treatment by subgroup interaction test, p<0.0001
0.4
Reference
Mok TS et al. N Engl J Med 2009; 361:
0.2
0.0 4 8 12 16 20 24
Time from randomisation (months)
Mok 2009 16 16 16

17. IPASS: PFS by Biomarkers (ITT)
Treatment-by-subgroup interaction test p-value
p= for EGFR gene copy number
p= for EGFR expression
p< for EGFR mutation
Favours gefitinib
Favours carboplatin / paclitaxel
Known mutation status
EGFR Mutation+
EGFR Mutation-
Known expression status
EGFR+
EGFR-
Known FISH status
EGFR FISH+
EGFR FISH-
0.25
0.5
1.0
2.0
4.0
Hazard Ratio (Gefitinib : Carboplatin / Paclitaxel) and 95% CI
Fukuoka JCO 2011

18. IPASS: Overlap of biomarkers
EGFR FISH+ n=249
EGFR expression +
n=266 25 51 13
+++ = 132 28
--- = 31 34
EGFR Mutation+ n=261 15
N=329 with known biomarker status for all 3 biomarkers
Fukuoka JCO 2011

19. EGFR Mutation Analysis
Lessons in Biomarker Analysis
Chromosome
DNA
mRNA
Protein
EGFR Gene
Copy Number
EGFR Mutation Analysis
EGFR Expression
Analysis
EGFR Protein Expression Analysis
What? Choose your biomarker(s) carefully
How?
The tool that you use to measure your biomarker must be robust and reliable
The cut-offs used to define biomarker positive, negative, and unknown subgroups must be appropriate
Samples must be available
PI3K
Grb-2
Ras
SOS
PTEN
Akt
Raf
MEK
mTOR
STAT 3/5
This diagram shows where the potential predictors of response to EGFR TKIs (EGFR mutations, gene copy number and protein expression) lie in the process of translating chromosome information to protein expression, and the downstream EGFR signalling pathways.
Reference
Pao & Miller. J Clin Oncol 2005; 23:
MAPK
Tumour cell
survival
Tumour cell
proliferation
Pao & Miller 2005

20. IPASS: Attrition factors in biomarker analysis
1217 randomised patients (100%)
Reasons for samples not evaluable: Sample not available, insufficient quantity to send, cytology only, sample at another site
1038 biomarker consent
(85%)
683 provided samples
(56%)
565 histology
118 cytology
Evaluable for:
EGFR mutation: 437 (36%)
EGFR gene copy number: 406 (33%)
EGFR expression: 365 (30%)
Reference
Mok TS et al. N Engl J Med 2009; 361:
Mok 2009, Fukuoka 2009 20 20 20

21. Implications for diagnostics
The drug
The biomarkers and clinical trials
Implications for Diagnostics
Our experience and learnings
Implications for diagnostics

22. IPASS reports September 2008, partway through the European MAA review of INTEREST (Phase III study of Gefitinib vs docetaxel in pre-treated NSCLC)
Gefitinib is indicated for the treatment of adult patients with locally advanced or metastatic non‑small cell lung cancer (NSCLC) with activating mutations of EGFR‑TK.

23. Getting the right treatment to the right patient
All NSCLC patients
Affected by.....
Test Availability and Ease of Test Order
Patients tested
Easily accessible testing
High quality testing (high detection rates)
Access to suitable samples
Reasonable Test Turnaround Time
Needs.....
Test sensitivity
Test ease of use
Interpretation of results
Test Turnaround time
Test result
Positive
Treated

24. EGFR mutation testing: diversity in 2009
Japan UK
Centralised lab network
Testing reimbursed
High level of patients tested
Tests used:
CLAMP
INVADER
Sequencing
Cycleave
De- centralised lab network
Testing not reimbursed
Low level of patients tested
Tests used:
TheraScreen
Sequencing
Pyrosequencing

25. Understanding and Supporting the Dx Environment
Patient
Physician
(respiratory, surgeon, oncologist)
Lab service provider/Molecular Biologist
Sample
Test result
Diagnostic company/ies
Platform provider
Pathologist
egfr-mutation.com
egfr-test.com

26. Challenges associated with a PHC approach
Physicians adopting a PHC approach
Access to testing
Diversity of testing methods /capabilities across countries
Multiple individuals involved from decision to test to getting results
Challenges
Testing reimbursement
Availability of suitable samples for testing
Time taken to generate test results

27. Our experience and learnings
The drug
The biomarkers and clinical trials
Implications for Diagnostics
Our experience and learnings
Our experience and learnings

28. The Biomarker Journey
Ideally biomarker to indication in biomarker population is a straightforward, well planned process
In reality this is a challenging process in which your direction changes e.g. molecular disease segment, clinical characteristics, different biomarkers, techniques, cut-offs etc.
Patient/ Disease
Biomarker/Dx Tool
What patients do you intend to treat?
What are you measuring?
Do you need a diagnostic to identify those patients?
How are you measuring it?
Is there an existing assay available to identify the patients?
How do you define your cut-offs?
Do you need to develop a diagnostic test suitable for selecting patients eligible for therapy?
Can you develop an appropriate tool that can be used to measure in a robust and reliable way?

29. Personalised Healthcare development today and in the future
Gefitinib experience
Gefitinib had standard drug development
Predictive biomarker for gefitinib discovered by external collaborator ~7 years after start of clinical trials
Retrospective investigation was required to show the significant clinical benefit for those patients identified by diagnostic test
Ultimately identified patients most likely to benefit offers an alternative treatment option to doublet chemotherapy in newly diagnosed advanced/metastatic NSCLC
Future Therapies
Personalised Healthcare research discovers predictive biomarker in preclinical models before start of clinical development
Clinical programme prospectively selects biomarker eligible patients, targeted to patients most likely to respond
Early engagement with health authorities and payers
Co-development of drug and diagnostic
Drug launched globally, linked to diagnostic established within the diagnostic environment

30. Summary
Gefitinib is approved in Europe for a biomarker targeted population
Gefitinib was developed during a time of rapid progress in the understanding of molecular mechanisms of cancer, therefore the route to approval was not straightforward
In future, pharmaceutical companies are unlikely to be able or willing to follow a similar development path for new agents
There are several useful learnings for future biomarker targeted products
Understand the science
Assess efficacy in a biomarker defined, targetted population, as early as possible
Maximise tissue samples, no sample means no biomarker result
Diagnostic test adoption is as important as the drug
Pharmaceutical companies, Physicians, Pathologists, Academics and Regulators are learning about this together
Engage early
Considerable challenges on both sides
Opportunity for collaboration

31. Confidentiality Notice
This file is private and may contain confidential and proprietary information. If you have received this file in error, please notify us and remove it from your system and note that you must not copy, distribute or take any action in reliance on it. Any unauthorized use or disclosure of the contents of this file is not permitted and may be unlawful. AstraZeneca PLC, 2 Kingdom Street, London, W2 6BD, UK, T: +44(0) , F: +44 (0) ,
Author | 00 Month Year
Set area descriptor | Sub level 1