Transcript IRESSA: A journey of experience from broad to biomarker

IRESSA: A journey of experience from broad to
biomarker populations
Claire Watkins
Global Product Statistician, AstraZeneca
EFSPI meeting on Oncology
Basel, 24th June 2010
Outline
• A brief history of IRESSA (gefitinib)
• Lessons learned
• Looking to the future of biomarker targeted
drug development
What is IRESSA and how does it work?
http://www.egfr-info.com/EGFR-lung-cancer/
European Indication – approved June 2009
IRESSA 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.
The ideal
Biomarker targeted drug
Studies
Indicated for Biomarker+
The reality
Broad
population?
Biomarker targeted drug
Studies
Clinical
characteristics?
Biomarker(s)?
Which
biomarker?
What cut-off?
Indicated for Biomarker+
IRESSA - May 2001
“Dramatic” Tumour shrinkage in patient with metastatic NSCLC
IDEAL 1&2 – NSCLC Phase II non-comparative - 2002
Response rate, %
30
25
20
15
18
19
250 mg
500 mg
12
9
10
5
0
250 mg
500 mg
IDEAL 2 – USA
IDEAL 1 – Japan and Europe
Vertical bars represent 95% CI.
Kris 2003, Fukuoka 2003
Japan and US approvals
•
Japan – full approval granted July 2002
Indication: Inoperable or recurrent non small cell lung cancer.
Precautions related to Indication 1. Efficacy and safety of IRESSA in patients without
previous chemotherapy regimens have not been established. 2. Efficacy and
safety of IRESSA in post-operative adjuvant therapy have not been established.
•
US – accelerated approval granted May 2003:
IRESSA is indicated as monotherapy for the treatment of patients with locally
advanced or metastatic non-small cell lung cancer after failure of both platinumbased and docetaxel chemotherapies.
The effectiveness of IRESSA is based on objective response rates. There are no
controlled trials demonstrating a clinical benefit, such as improvement in diseaserelated symptoms or increased survival.
•
US – Phase III post approval pre-treated commitment studies including:
•
•
•
•
ISEL – OS superiority vs placebo
INTEREST – OS non-inferiority vs docetaxel
IBREESE – Symptom improvement superiority vs placebo
Question – what is needed from these studies to lift the conditional approval?
ISEL – reports December 2004
OS
TTF
HR (95% CI) =0.82 (0.73, 0.93) p=0.0006
n=1316, progressions=1137
1.0
1.0
0.8
0.8
Proportion without
treatment failure
Proportion surviving
HR (95% CI) =0.89 (0.77, 1.02) p= 0.0871
by primary stratified log rank test
n=1692, deaths=976
[Adjusted Cox analysis HR 0.86 (0.76-0.99)
p=0.0299]
0.6
0.4
IRESSA
Placebo
0.2
0.6
0.4
0.2
0.0
0.0
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16
Months
Months
Objective Response Rate
8.0% vs 1.3%, p<0.0001
Thatcher 2005
ISEL OS subgroups by smoking status and histology
Proportion surviving
Treatment by smoking interaction test p=0.047
Never smoked (n=375)
1.0
Ever smoked (n=1317)
HR 0.67; 95% CI 0.49, 0.92;
p=0.012
0.8
HR 0.92; 95% CI
0.79, 1.06; p=0.242
IRESSA
Placebo
0.6
0.4
0.2
0.0
0
2
4
6
8 10 12 14 16
0
2
4
6
8 10 12 14 16
Proportion surviving
Treatment by race interaction test p=0.043
Asian origin (n=342)
1.0
HR 0.66; 95% CI
0.48, 0.91; p=0.010
0.8
Non-Asian origin (n=1350)
HR 0.92; 95% CI 0.80, 1.07;
p=0.294
0.6
0.4
0.2
0.0
0
2
4
6
8
10 12 14 16 0 2 4
Cox regression analysis Time (months)
6
8 10 12 14 16
Thatcher 2005, Chang 2006
Regulatory reactions
•
•
•
•
•
MHLW open public mtg 17th Jan 05
FDA Advisory committee 4th March 05
MHLW open public mtg (2) 10th March 05
MHLW open public mtg (3) 17th March 05
MHLW open public mtg (4) 24th March 05
• FDA restricts labelling
IRESSA is indicated as monotherapy for the continued treatment of
patients with locally advanced or metastatic non-small cell lung
cancer after failure of both platinum-based and docetaxel
chemotherapies who are benefiting or have benefited from
IRESSA
• Japan – no change to labelling
EGFR biomarkers
ISEL, INTEREST:
Unselected trials in
pre-treated setting
ISEL
IRESSA
registration
Japan
INTEREST
IPASS
2002
2005
2007
2009
EGFR protein expression
EGFR gene copy number
EGFR mutations
IPASS: Clinically
selected trial in first
line setting
ISEL: OS by EGFR gene copy number
Treatment by gene copy number interaction test p=0.047
High (+)
Low (-)
N=114, E=68
Cox HR=0.61 (0.36, 1.04)
p=0.07
N=256, E=157
Cox HR=1.16 (0.81, 1.64)
p=0.42
Percent 1.0
surviving
0.8
1.0
IRESSA
Placebo
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0.0
0.0
0
2
4
6 8 10 12 14
Time (months)
IRESSA
Placebo
16
0
2
4
6 8 10 12 14
Time (months)
16
OS could not be analysed by EGFR mutation status as there were
only 5 mutation positive patients on placebo. The ORR was 38%
in the 21 mutation positive patients treated with IRESSA
Hirsch 2006
INTEREST: Phase III study of IRESSA vs
docetaxel in pre-treated NSCLC
Endpoints
Primary
Patients
• Progressive
or
recurrent disease
following CT
IRESSA
250 mg/day
• Considered
candidates for
further CT with
docetaxel
•1
or 2 CT regimens
(≥1 platinum)
• PS
0-2
1:1 randomization
Docetaxel
75 mg/m2 every
3 weeks
• Overall survival
•(co-primary analysesa 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
• 1466 patients
amodified
Hochberg procedure applied to control for multiple testing
CT, chemotherapy; PS, performance status;
EGFR, epidermal growth factor receptor
• Biomarkers
•EGFR mutation
•EGFR protein expression
•EGFR gene copy number
•K-Ras mutation
Kim 2008
INTEREST: OS and PFS and ORR
OS: NI margin 1.154, PP population
PFS: EFR population
HR (96% CI) =1.020 (0.905, 1.150)
n=1433, deaths=1169
Median survival: IRESSA 7.6m,
Docetaxel 8.0m
HR (95% CI) =1.04 (0.93, 1.18), p=0.466
n=1316, progressions=1137
Median PFS: IRESSA 2.2m,
Docetaxel 2.7m
1.0
Probability of progressionfree survival
Probability of survival
1.0
IRESSA
Docetaxel
0.8
0.6
0.4
0.2
0.0
IRESSA
Docetaxel
0.8
0.6
0.4
0.2
0.0
0
4
8
12
16
20
Months
24
28
32
36
40
0
4
8
12
16
20
24
28
32
36
Months
ORR [EFR population]: 9.1% IRESSA, 7.6% Docetaxel; p=0.3257
Kim 2008
40
INTEREST: Summary of key subgroup analyses
INTEREST
Overall Survival
Overall
ORR (%)
IRESSA v. Docetaxel
9.1 v. 7.6
Ever smoker
Never smoker
Asian
Non-Asian
Progression-free Survival
Overall
Overall
Ever smoker
Ever smoker
Never smoker
Never smoker
19.7 v. 8.7 Asian
Asian
6.2 v. 7.3
Non-Asian
Non-Asian
EGFR FISH+
13.0 v. 7.4 EGFR FISH+
EGFR FISH+
EGFR FISH-
7.5 v. 10.1 EGFR FISH-
EGFR FISH-
EGFR Mutation+
42.1 v. 21.1 EGFR Mutation+
EGFR Mutation+
EGFR Mutation-
6.6 v. 9.8
EGFR Mutation-
0
0.5
1.0
1.5
2.0
HR (IRESSA vs docetaxel) and 95% CI
Unadjusted
analysis
PP population
for clinical factors
ITT population for
biomarker factors
EGFR Mutation-
0
0.5
1.0
1.5
2.0
2.5
HR (IRESSA vs docetaxel) and 95% CI
EFR
population
Adjusted
EFR
analysis
population
Kim 2008; Douillard 2010
IPASS: Phase III study of IRESSA versus doublet
chemotherapy in first line NSCLC
Endpoints
Patients
• Adenocarcinoma
histology
• Never
smokers or
light ex-smokers*
• PS
0-2
Primary
IRESSA
250 mg/day
Secondary
1:1 randomization
• Provision
of tumour
sample for
biomarker analysis
strongly
encouraged
• Progression free survival
(non-inferiority)
Carboplatin AUC 5
or 6 and Paclitaxel
200mg/m2 3 wkly
• 1217 patients from East Asian countries
• 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
*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
18
IPASS reports September 2008,
partway through the European MAA
review of INTEREST
IPASS: Exceeded primary objective and demonstrated
superior PFS for IRESSA versus doublet chemotherapy
IRESSA
Carboplatin /
paclitaxel
N
Events
609
453 (74.4%)
608
497 (81.7%)
HR (95% CI) = 0.741 (0.651, 0.845) p<0.0001
IRESSA demonstrated superiority relative
to carboplatin / paclitaxel in terms of PFS
Mok 2009
Primary Cox analysis with covariates; ITT population
HR <1 implies a lower risk of progression on IRESSA
HR, hazard ratio; CI, confidence interval; PFS, progression-free survival
20
IPASS: Superior PFS and ORR with IRESSA vs doublet
chemotherapy; PFS effect not constant over time
Probability 1.0
of PFS
Carboplatin /
IRESSA
N
Events
0.8
609
453 (74.4%)
paclitaxel
608
497 (81.7%)
HR (95% CI) = 0.741 (0.651, 0.845) p<0.0001
0.6
5.8
74%
48%
7%
Median PFS (months)
5.7
4 months progression-free
61%
6 months progression-free
48%
12 months progression-free 25%
0.4
Primary objective exceeded: IRESSA
demonstrated superiority relative to carboplatin /
paclitaxel in terms of PFS
0.2
0.0
At risk :
IRESSA
Carboplatin /
paclitaxel
0
4
8
12
16
20
24 Months
609
608
363
412
212
118
76
22
24
3
5
1
0
0
Objective response rate 43% vs 32% p=0.0001
Mok 2009
Primary Cox analysis and logistic regression with covariates; ITT population
HR <1 implies a lower risk of progression on IRESSA
21
IPASS: Superior progression-free survival and response
rate for IRESSA in EGFR mutation positive patients
IRESSA EGFR M+ (n=132)
Probability
of PFS
1.0
Carboplatin / paclitaxel EGFR M+ (n=129)
0.8
EGFR M+
HR=0.48, 95% CI 0.36, 0.64
p<0.0001
0.6
0.4
Objective response rate
71.2% vs 47.3%
0.2
p=0.0001
0.0
0
4
8
12
16
20
24
Time from randomisation (months)
Mok 2009
M+, mutation positive
22
IPASS: Superior progression-free survival and response rate for
doublet chemotherapy in EGFR mutation negative patients
Probability
of PFS
IRESSA EGFR M- (n=91)
1.0
Carboplatin / paclitaxel EGFR M- (n=85)
0.8
0.6
EGFR M-
HR=2.85, 95% CI 2.05, 3.98
p<0.0001
0.4
Objective response rate
1.1% vs 23.5%
0.2
p=0.0013
0.0
0
4
8
12
16
20
24
Time from randomisation (months)
Mok 2009
M-, mutation negative
23
IPASS: EGFR mutation is a strong predictor for differential
PFS benefit between IRESSA and doublet chemotherapy
Probability
of PFS
IRESSA EGFR M+ (n=132)
IRESSA EGFR M- (n=91)
Carboplatin / paclitaxel EGFR M+ (n=129)
Carboplatin / paclitaxel EGFR M- (n=85)
1.0
0.8
Treatment
by
subgroup
interaction
test,
p<0.0001
EGFR M+
HR=0.48, 95% CI 0.36, 0.64
p<0.0001
0.6
EGFR M-
HR=2.85, 95% CI 2.05, 3.98
p<0.0001
0.4
0.2
0.0
0
4
8
12
16
20
24
Time from randomisation (months)
Mok 2009
M+, mutation positive; M-, mutation negative
24
European Indication – approved June 2009
IRESSA 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.
Lessons learned
• Understand the biology
• Make friends with your translational scientists
• Determine whether to go down the targeted biomarker route as
early as possible
• “The tissue is the issue” – collect as many samples as you can
• No sample = no biomarker
• Pathologists are key
• Conflict between push for faster studies and push for targeted
healthcare
• Fast recruiters are not often the most experienced at sample collection
• A targeted drug is useless without a diagnostic
• Co-development has its own unique challenges
• Ensure an understanding of prognostic vs predictive
• A predictive factor cannot be identified from a single arm study
• A poor prognostic factor can be a good predictive factor for a new
agent
Prognostic vs Predictive
Not predictive
Not prognostic
Prognostic
12
12
10
10
8
8
Olaparib
6
Comparator
Olaparib
6
Comparator
4
4
2
2
0
HRD+
+
HRD-
0
-
12
HRD+
HRD-
+
-
12
10
10
Predictive
8
8
Olaparib
6
Comparator
Olaparib
6
Comparator
4
4
2
2
0
HRD+
HRD-
+
-
0
HRD+
HRD-
+
-
Blue=Experimental, Purple=comparator
Lessons learned
• It matters
• What you measure
• How you measure it
• How you define positive (cut-off)
Tissue
sample
Diagnostic
test
MAGIC ALGORITHM!
Biomarker
status
Positive or
negative
It matters what you measure
Protein
expression
Gene
copy
number
EGFR
Gene
mutation
It matters how you measure it
FISH
Fluorescence
Gene
copy
number
IHC
Protein
expression
CISH
EGFR
Gene
mutation
Sequencing
ARMs
PNA-LNA
PCR clamp
It matters how you define positive (cut-off)
Staining
intensity
Staining
percentage
# of
copies
FISH
CISH
Fluorescence
Gene
copy
number
IHC
Protein
expression
New diagnostics
may use more than
one biomarker to
define positivity
Pattern
of
copies
EGFR
Gene
mutation
Sequencing
ARMs
PNA-LNA
PCR clamp
Type of
mutation
INTEREST: Overlap of biomarkers (EGFR gene copy
number by FISH, EGFR expression by IHC, EGFR mutation)
EGFR
expression +
n=189
EGFR FISH +
n=117
n=73
n=16
n=84
+++ n=24
4
EGFR mutation +
n=39
3
n=8
--- n=37
249 patients
evaluable for
EGFR
expression,
FISH and
mutations
Douillard 2010
32
Lessons learned
• It matters
• What you measure
• How you measure it
• How you define positive (cut-off)
Tissue
sample
Diagnostic
test
MAGIC ALGORITHM!
Biomarker
status
Positive or
negative
• Consider if there is a surrogate for the biomarker e.g. clinical
characteristics, another marker
INTEREST: EGFR mutation appeared to be
associated with some clinical characteristics
% of
samples
EGFR
mutation
positive
60
50
40
30
20
10
0
Overall EGFR mutation positive rate 14.8% (44/297)
Douillard 2010
K-Ras and EGFR mutations rarely co-exist
in the same tumour
5 incidences across 19 studies totalling around 3300 patients
Study
AstraZeneca studies
INTEREST
ISEL
INVITE
Literature
Wu et al 2008
Yamamoto et al 2008
Zhu et al 2008
Do et al 2008
Sasaki et al 2008
Na et al 2007
Massarelli et al 2007
Bae et al 2007
Hirsch et al 2006
van Zandwijk et al 2007
Yokoyama et al 2006
Suzuki et al 2006
Tam et al 2006
Tomizawa et al 2005
Shigematsu et al 2005
K-Ras mutations
N evaluable
N (%) K-Ras+
EGFR mutations
N evaluable
N (%) M+
Number
K-Ras+/M+
275
152
90
49 (17.8)
12 (7.9)
24 (26.7)
297
215
65
44 (14.8)
26 (12.1)
6 (9.2)
1
0
1
237
86
206
200
190
133
70
115
152
9 (3.8)
26 (30.2)
30 (14.6)
25 (12.5)
21 (11.1)
17 (12.8)
16 (22.9)
6 (5.2)
12 (7.9)
349
150
215
120
617
21 (6.0)
6 (4.0)
21 (9.8)
4 (3.3)
50 (8.1)
235
86
204
200
195
133
71
115
215
41
349
150
241
120
519
96 (40.9)
10 (11.6)
34 (16.7)
73 (36.5)
82 (42.1)
32 (24.1)
7 (9.9)
20 (17.4)
26 (12.1)
13 (31.7)
102 (29.2)
38 (25.3)
116 (48.1)
29 (24.2)
120 (23.1)
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
35
Lessons learned
• Engage with regulators early
• Everyone is learning as they go along
• FDA in particular has stated positions that may not be practical in
all cases
• >90% evaluable samples
• Prove don’t work in –ve
IPASS: Attrition factors in biomarker analysis
1217
randomised
patients
(100%)
1038
biomarker
consent
(85%)
Reasons for samples not
evaluable: Sample not available,
insufficient quantity to send,
cytology only, sample at another
site
683
provided
samples
(56%)
•565 histology
• 118 cytology
Evaluable for:
EGFR mutation: 437 (36%)
EGFR gene copy number:
406 (33%)
EGFR expression:
365 (30%)
Mok 2009, Fukuoka 2009
37
Lessons learned
• Engage with regulators early
• Everyone is learning as they go along
• FDA in particular has stated positions that may not be practical in
all cases
• >90% evaluable samples
• Prove don’t work in –ve
• Don’t want to do a repeat of Phase IIIs
• Issues of generating a strong signal in a small early study
• Payers are key stakeholders
• Randomised Phase IIs
• Keep an eye to the future
• New or revised tests, markers, tissue types
• Flexible consent
• Be aware that science will move on as your study is ongoing
Personalised Healthcare development
today and in the future
Today
2013
• Predictive biomarker for IRESSA
discovered by external collaborator ~7
years after start of clinical trials
• Took ~4.5 further years retrospective
research to show significant increase in
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
§ Personalised Healthcare research
discovers predictive biomarker in
preclinical models before start of
clinical development
§ Early engagment with payers and
health authorities ensures that drug is
targeted to patients likely to respond
§ Clinical programme prospectively
tailored for responders, used for codevelopment of drug and diagnostic
§ Drug launched globally, linked to
diagnostic
Summary
• IRESSA is approved in Europe for a biomarker targeted
population
• But it took a long time to get there
• 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
• Maximise tissue samples
• Diagnostic is as important as the drug
• Pharmaceutical companies and regulators are learning about
this together
• Engage early
• Considerable challenges on both sides
• Opportunity for collaboration
References
•
•
•
•
•
•
•
•
•
•
Kris MG, Natale RB, Herbst RS, et al: Efficacy of gefitinib, an inhibitor of the epidermal growth factor
receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: A randomized trial. JAMA
290:2149-2158, 2003
Fukuoka M, Yano S, Giaccone G, et al: Multi-institutional randomized phase II trial of gefitinib for previously
treated patients with advanced non-small-cell lung cancer. J Clin Oncol 21:2237-2246, 2003
Thatcher N, Chang A, Parikh P, et al. Gefitinib plus best supportive care in previously treated patients with
refractory advanced non-small-cell lung cancer: results from a randomised, placebo-controlled, multicentre
study (Iressa Survival Evaluation in Lung Cancer). Lancet 366: 1527–37, 2005
Chang A, Parikh P, Thongprasert S, et al: Gefitinib (IRESSA) in Patients of Asian Origin with Refractory
Advanced Non-small Cell Lung Cancer: Subset Analysis from the ISEL Study. J Thoracic Oncol 1: 8: 847855, 2006
Hirsch FR, Varella-Garcia M, Bunn PA, et al. Molecular predictors of outcome with gefitinib in a Phase III
placebo-controlled study in advanced non-small-cell lung cancer. J Clin Oncol 24: 5034-5042, 2006
Kim ES, Hirsch V, Mok T, et al: Gefitinib versus docetaxel in previously treated non-small-cell lung cancer
(INTEREST): a randomised phase III trial. Lancet 372:1809-1818, 2008
Douillard JY, Hirsch V, Mok T, et al: Molecular analyses from a phase III trial comparing gefitinib with
docetaxel in previously treated non-small-cell lung cancer (INTEREST). J Clin Oncol 26 (May 20 Suppl):
Abstract 8001, 2008
Mok T, Wu Y, Thongprasert S, et al: Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma NEJM
361: 947-957, 2009
Douillard J, Hirsh V, Mok T, et al: Molecular predictors of outcome with gefitinib and docetaxel in previously
treated non-small-cell lung cancer: data from the randomised phase III INTEREST trial J Clin Oncol 5:744752, 2010
Fukuoka M, Wu Y, Thongprasert S, et al. Biomarker analyses from a phase III, randomized, open-label, firstline study of gefitinib (G) versus carboplatin/paclitaxel (C/P) in clinically selected patients (pts) with
advanced non-small cell lung cancer (NSCLC) in Asia (IPASS). J Clin Oncol 27 (15s suppl): Abstract 8006,
2009