Transcript Malcolm Moore - Lebanese Society of Medical Oncology (LSMO)

Malcolm Moore
Professor of Medicine and Pharmacology,
Princess Margaret Hospital, University of Toronto, Toronto, Canada
 Director of the Bras Family New Drug
Development Program, Princess Margaret
Hospital, University of Toronto, Canada
 Head of Division of Medical Oncology
and Hematology, Princess Margaret
Hospital and University of Toronto
 Senior Scientist in the Division of
Experimental Therapeutics at the Ontario
Cancer Institute, Canada
 He is the author of more than 160 peer-
reviewed publications and has given over
140 invited lectures worldwide
Princess Margaret Hospital
Charting a new landscape in advanced
pancreatic cancer
Malcolm Moore
Princess Margaret Hospital
University of Toronto
Toronto, Canada
Pancreatic cancer:
a major therapeutic challenge
 Major health burden
– fourth leading cause of cancer death in the USA
– fifth leading cause of cancer death in the EU
 Fatal disease with 98% mortality rate1,2
– OS rate is among the shortest of any solid tumour
 Poor prognosis
– usually locally advanced or metastatic at diagnosis
– most patients unsuitable for surgery
– current treatment options are limited
1Parkin
OS = overall survival
DM, et al. CA Cancer J Clin 2005;55:74–108
2Michaud DS. Minerva Chir 2004;59:99–111
Gemcitabine: a standard of care in
pancreatic cancer
Patients surviving (%)
100
Gemcitabine
(n=63)
5-FU
(n=63)
p value
5.65
4.41
0.0025
Median survival
(months)
80
60
40
Gemcitabine
5-FU
20
0
0
2
5-FU = 5-fluorouracil
4
6
8
10 12
Months
14
16
18
20
Burris H, et al. J Clin Oncol 1997;15:2403–13
Lack of progress in therapy for
advanced pancreatic cancer
 Treatment options remained limited over last decade
despite promising phase II results with gemcitabine
plus various different agents
 More than 10 phase III trials of new drug versus
gemcitabine or in combination with gemcitabine
– none demonstrated a survival benefit
 Increased toxicity with several combination regimens
Randomised phase III trials in pancreatic
cancer (median OS in months)
Gem
Gem + X
p value
Gem ± marimastat (Bramhall, Br J Cancer 2002)
Gem ± tipifarnib (Van Cutsem, J Clin Oncol 2004)
Gem ± exatecan (Abou-Alfa, J Clin Oncol 2006)
Gem ± CPT-11 (Rocha-Lima, J Clin Oncol 2006)
Gem ± pemetrexed (Oettle, Ann Oncol 2006)
5.5
6.0
6.2
6.6
6.3
5.5
6.4
6.7
6.3
6.2
NS
NS
NS
NS
NS
Gem ± 5-FU bolus (Berlin, J Clin Oncol 2002)
Gem ± capecitabine (Herrmann, J Clin Oncol 2007)
Gem ± 5-FU/LV (Riess, J Clin Oncol 2005)
Gem ± capecitabine (Cunningham, ECCO 2005)
(preliminary results)
Gem ± cisplatin (Heinemann, J Clin Oncol 2006)
Gem ± oxaliplatin (Louvet, J Clin Oncol 2005)
Gem ± oxaliplatin (Poplin, ASCO 2006)
5.4
7.3
6.2
6.0
6.7
8.4
5.9
7.4
NS
NS
NS
0.026
6.0
7.1
4.9
7.5
9.0
5.9
NS
NS
NS
Gem ± erlotinib (Moore, J Clin Oncol 2007)
Gem ± bevacizumab (Kindler, ASCO 2007)
Gem ± cetuximab (Philip, ASCO 2007)
5.9
6.1
5.9
6.4
5.8
6.4
0.011
NS
NS
Gem = gemcitabine; Gem + X = combination regimen being investigated
LV = leucovorin; NS = not significant
Gemcitabine/cetuximab versus gemcitabine
alone: phase III study (n=735)
Gemcitabine +
cetuximab
Gemcitabine
monotherapy
6.5
6.0
Median OS (months)
HR=1.09
(95% CI: 0.93–1.27)
p=0.14
Progression-free survival
(PFS) (months)
3.5
Response rate (RR) (%)*
12
Patients experiencing ≥1
grade 4 toxicity (%)
14
HR=1.13
(95% CI: 0.97–1.30)
p=0.058
3.0
14
11
*Includes unconfirmed responses
HR = hazard ratio; CI = confidence interval
 Phase III data did not confirm promising phase II results
Philip PA, et al. J Clin Oncol 2007;25(Suppl. 18 Pt I):s199 (Abstract 4509)
Gemcitabine/bevacizumab versus
gemcitabine/placebo: phase III CALGB study
Gemcitabine +
bevacizumab (n=302)
Gemcitabine +
placebo (n=300)
Median OS (months)
5.7
6.0
PFS (months)
4.8
4.3
1.9
11.2
40.7
3.0
8.3
35.7
RR (%)*
Complete response
Partial response
Stable disease
*Includes unconfirmed responses
 Phase III data did not confirm promising phase II results
Kindler HL, et al. J Clin Oncol 2007;25(Suppl. 18 Pt.I):s199 (Abstract 4508)
Rationale for targeting EGFR in
anticancer therapy
Epidermal growth
factor receptor (EGFR)
Extracellular
Intracellular
P
PI3K
Akt
P
MAPK
JNK
 Proliferation
 Invasion
 Metastasis
 Angiogenesis
 Inhibition of apoptosis
Woodburn J. Pharmacol Ther 1999;82:241–50; Lynch TJ, et al. N Engl J Med 2004;350:2129–39
Knowlden JM, et al. Endocrinology 2003;144:1032–44
Chakravarti A, et al. Cancer Res 2002;62:4307–15
Rationale for targeting EGFR in
pancreatic cancer
 EGFR overexpression is common (30–95%)1,2
 Elevated EGFR and EGF is thought to be
associated with3–5
– more aggressive disease
– increased tumour size
– late clinical stage
– poor prognosis
– reduced sensitivity to chemotherapy
1Tobita
K, et al. Int J Mol Med 2003;11:305–9
A, et al. Hum Pathol 2001;32:1184–9
3Ueda S, et al. Pancreas 2004;29:1–8
4Nicholson R, et al. Eur J Cancer 2001;37:S9–S15
5Xiong H, et al. Semin Oncol 2002;29:31–7
2Srivastava
Data overview: PA.3 study
PA.3: study schema
Stratified by
 Centre
 ECOG PS (0/1 vs 2)
 Stage of disease
(locally advanced/
distant metastases)
(n=569)
 Primary endpoint = overall
R
A
N
D
O
M
I
S
E*
Gemcitabine 1,000mg/m2 i.v.
weekly
+
erlotinib 100/150mg/day p.o.
(n=285)
Gemcitabine 1,000mg/m2 i.v.
weekly
+
placebo 100/150mg/day p.o.
(n=284)
survival
 Secondary endpoints include progression-free survival (PFS), quality of
life (in selected countries), response rate (RR), toxicity, biomarkers
*1:1 randomisation
ECOG = Eastern Cooperative Oncology Group;
PS = performance status; i.v. = intravenous; p.o. = oral
Moore M, et al. J Clin
Oncol 2007;25:1960–6
PA.3: key eligibility criteria
 Locally advanced or metastatic adenocarcinoma of
the pancreas
 Age 18 years
 PS 0–2
 Measurable or non-measurable disease
 Prior radiotherapy for local disease allowed
 No prior chemotherapy, except for 5-FU or
gemcitabine as a radiosensitiser
 EGFR-positive status not required
Moore M, et al. J Clin Oncol 2007;25:1960–6
PA.3: patient characteristics*
 569 patients randomised
– 521 patients at 100mg/placebo
– 48 patients at 150mg/placebo
Gemcitabine +
erlotinib (n=285)
Gemcitabine +
placebo (n=284)
64
64
52/48
43/57
30/51/19
30/52/18
Locally advanced/metastatic (%)
24/76
25/75
Pain score 20/>20/unknown (%)
46/51/3
45/53/2
94
92
Median age (years)
Female/male (%)
PS 0/1/2 (%)
Measurable disease (%)
*Baseline characteristics of the 100mg cohort did not differ substantially from the whole
population
Moore M, et al. J Clin Oncol 2007;25:1960–6
Data on file, OSI Pharmaceuticals Inc.
PA.3: OS in the total
population* (ITT)
Survival probability (%)
100
80
Median survival
(months)
1-year
survival
Gemcitabine + erlotinib
6.24
23
Gemcitabine + placebo
5.91
17
60
p=0.038
HR=0.82 (95% CI: 0.69–0.99)
40
20
0
0
6
*Includes patients with locally advanced and
metastatic disease at both 100mg/day and
150mg/day doses of erlotinib
ITT = intent to treat; HR = hazard ratio; CI =
confidence interval
12
Months
18
24
Moore M, et al. J Clin Oncol 2007;25:1960–6
PA.3: overall survival in
the 100mg cohort
Median survival 1-year
(months)
survival
Survival probability (%)
100
80
Gemcitabine + erlotinib (n=261)
6.4
23.8
Gemcitabine + placebo (n=260)
6.0
19.4
p<0.028
HR=0.81 (95% CI: 0.68–0.97)
60
40
20
0
0
6
12
Months
18
24
Moore M, et al. J Clin Oncol 2007;25:1960–6
PA.3: PFS in
the 100mg cohort
Erlotinib + gemcitabine (n=261)
Placebo + gemcitabine (n=260)
Survival probability (%)
100
75
p=0.006
HR=0.76* (95% CI: 0.64–0.92)
50
 30% increase in PFS
25
0
0
*HR adjusted for PS and
extent of disease at baseline
6
12
Months
18
24
Data on file, OSI Pharmaceuticals Inc.
PA.3: all subgroups benefit from treatment
with erlotinib plus gemcitabine
Factors
n
HR
95% CI
Erlotinib: placebo*
521
0.81
0.7–1.0
PS 0–1
PS 2
432
89
0.87
0.70
0.7–1.1
0.5–1.1
Locally advanced
Distant metastases
124
397
0.93
0.80
0.6–1.3
0.7–1.0
Pain intensity 20
Pain intensity >20
238
268
0.72
1.00
0.6–0.9
0.8–1.3
EGFR positive
EGFR negative
EGFR unmeasured
70
66
385
0.82
0.75
0.86
0.5–1.3
0.5–1.2
0.7–1.1
Male
Female
273
240
0.74
1.00
0.6–0.9
0.8–1.3
Age <65 years
Age 65 years
274
247
0.78
0.94
0.6–1.0
0.7–1.2
Caucasian
Black
Asian
Prior radiosensitising
chemotherapy**
No prior radiosensitising
chemotherapy**
456
13
34
0.88
0.67
0.61
0.7–1.1
0.2–2.2
0.3–1.3
42
0.62
0.3–1.2
479
0.86
0.7–1.0
*Stratified by PS and extent of disease
**No prior chemotherapy
allowed, other than as a radiosensitiser
0
0.50
1.00
HR
1.50
2.00
2.50
Moore M, et al. J Clin Oncol 2005;23(Suppl. 16 Pt I):s1 (Abstract 1)
Genentech: Tarceva® full prescribing information
Importance of stable disease
Median duration of SD >5 months
Gemcitabine
+ erlotinib
SD
Gemcitabine
+ placebo
SD
0
10
20
CR/PR
CR/PR
30
40
50
Patients (%)
 Disease control rate (CR + PR + SD): gemcitabine plus erlotinib 59%;
gemcitabine plus placebo 49.4%; p=0.036
CR = complete response
PR = partial response; SD = stable disease
60
Data on file, OSI Pharmaceuticals Inc.
Acceptable safety profile with addition
of erlotinib: no synergistic toxicities
Grade 3/4 AEs
20
Gemcitabine + erlotinib
Gemcitabine + placebo
Patients (%)
15
10
5
0
Rash

Diarrhoea
Infections
Weight loss
Serious adverse drug reactions are infrequent
AE = adverse event
Tarceva® Summary of Product Characteristics
Significance of PA.3
 First randomised trial to show significantly prolonged
survival with any regimen versus single-agent
gemcitabine in advanced pancreatic cancer
– first demonstration of OS benefit with EGFR
inhibitor in combination with chemotherapy
 Results led to
– FDA approval of the combination in patients with
locally advanced and metastatic pancreatic cancer
– EMEA approval of the combination in patients with
metastatic pancreatic cancer
EGFR = epidermal growth factor receptor
FDA = Food and Drug Administration; EMEA = European Medicines Agency
Erlotinib in the routine clinical
setting: a case study
Case study: patient history
 20-year-old university student
 History of abdominal pain for 6 months and gradually
increasing abdominal distension
 Felt a lump in upper abdomen
 Past medical health was good but there was a strong
family history of cancer
 CT scan was performed in December 2005
CT = computerised tomography
Case study: imaging at baseline
 Liver biopsy found well-differentiated adenocarcinoma of
pancreatic or biliary tract origin
Case study: initial treatment
 Patient commenced treatment with gemcitabine plus
erlotinib 100mg/day
 Two days after starting erlotinib, the patient
developed pustular rash that increased in intensity
– stopped erlotinib and came to clinic
 On examination, confirmed as grade 3 rash
Erlotinib-related rash: NCI-CTC grade 3
Definition
 Severe, generalised
erythroderma (generalised
reddening and scaling of skin)
 Macular, papular or vesicular
eruption (small fluid-filled
blisters)
 Desquamation covering >50%
of BSA
– OR <50% of BSA, but very
symptomatic with vesicular
eruption
NCI-CTC = National Cancer Institute-Common
Toxicity Criteria; BSA = body surface area
Management recommendations for
symptomatic rash
Mild
 No treatment
OR
Moderate
Severe
 Hydrocortisone
2.5% cream
AND
OR
 Hydrocortisone
1% or 2.5% cream
 AND/OR
 Clindamycin 1% gel
 Treat as moderate
 Clindamycin 1% gel
 Systemic steroids†
AND
OR
 Pimecrolimus 1%
cream
 Consider dose
interruption
AND
 Synthetic tetracycline*
2-weekly assessments
*Doxycycline or minocycline 100mg b.i.d.
†Prednisone, methylprednisolone 30mg once
daily (tapered over 30 days)
b.i.d. = twice daily
Lynch TJ, et al. Oncologist 2007;12:610–21
Case study: rash management
 Erlotinib stopped for 1 week while rash treated with
oral minocycline and topical steroids
 Restarted erlotinib at 100mg/day
 Grade 3 rash recurred by week 7 at which point
erlotinib was temporarily halted, then restarted at a
reduced dose of 50mg/day
Case study: continuing status
 Repeat CT scan after 8 weeks showed stable
pancreatic mass and reduction in liver metastases
 Patient continued on gemcitabine plus
50mg/day erlotinib
 After 6 months of treatment, the patient had chronic
grade 1 rash and diarrhoea
– CT scan showed further improvement in
liver lesions
Case study: imaging at 6 months
Case study: current status
 September 2006: patient underwent resection of
pancreatic tail mass and liver lesions
– confirmed well-differentiated
cystic adenocarcinoma
 Patient remained well until May 2007 when two new
liver lesions were observed
– treated with radiofrequency ablation
 December 2007
– three new liver lesions observed
– restarted gemcitabine plus erlotinib
Future developments
Identifying patients most likely to benefit
from gemcitabine plus erlotinib
 Prior to treatment
– molecular markers
– demographic/clinical characteristics (e.g. age,
gender, histology, PS)
 During treatment
– RR; disease control rate
– other surrogate markers (e.g. presence of rash)
PA.3: significant improvement
in OS in PS 2 patients
Survival probability (%)
100
Gemcitabine + erlotinib (n=53)
Median: 4.16 months
95% CI: (4.37–6.34)
Gemcitabine + placebo (n=52)
Median: 3.20 months
95% CI: (2.63–4.11)
80
60
p≤0.001
HR=0.47 (95% CI: 0.31–0.71)
40
20
0
0
6
12
Months
18
24
Data on file, OSI Pharmaceuticals Inc.
PA.3: skin rash associated with erlotinib
 72% of patients developed skin rash
 Significantly higher likelihood of skin rash in
– patients younger than 65 years (p=0.01)
– patients with a good PS (p=0.03)
 Presence of rash was associated with
– higher rate of disease control (p=0.05)
– longer survival (HR=0.74, p=0.037)
 Generally develops within 7–10 days of starting therapy
and may spontaneously resolve and reappear
 No correlation with duration of therapy
 Reversible following drug discontinuation
PA.3: degree of rash correlates
with OS (100mg cohort)
Survival probability (%)
100
Grade 0 Grade 1 Grade 2
(n=79) (n=108) (n=103)
Median survival
(months)
80
5.29
5.75
10.51
16
11
43
1-year survival (%)
60
40
Grade 0
Grade 1
Grade 2
20
p<0.0001, HR=0.71
0
0
5
10
Months
15
20
Data on file, OSI Pharmaceuticals Inc.
PA.3: observed skin rash
 Potential reasons
– variability of drug absorption
– variability of metabolism
– ability of rash to predict a more
immunocompetent individual
– pharmacogenomic basis
 Further study in this area is necessary to understand the
value of rash in predicting survival in individual patients
– some patients with no rash do obtain clinical benefit
from erlotinib
– some patients who develop grade 2 rash had grade 1
rash initially
Phase II trial: assessing potential predictive markers
and dose escalation according to skin reactions
Advanced
pancreatic
cancer:
gemcitabine +
erlotinib 100mg
(n~400)
No rash
or grade 1
rash
4 weeks
Grade 2
rash
Gemcitabine +
erlotinib dose
escalation to grade 2
rash or DLT
PD
Gemcitabine + erlotinib
100mg
PD
Gemcitabine +
erlotinib 100mg
PD
 Primary endpoint: OS
 Mandatory tissue collection for biomarker testing
 Secondary endpoints: PFS, disease control, safety, correlation of
EGFR-related biomarkers with outcome (EGFR, EGF, TGF-α, KRAS,
pAkt, pMAPK)
DLT = dose-limiting toxicity; PD = progressive disease
EGFR: a predictive factor?
 In NSCLC, EGFR IHC and EGFR FISH status have not yet
been demonstrated to correlate with benefit from EGFR
inhibitors in a prospective, randomised trial
 In NSCLC, activating mutations of EGFR (exons 18–21)
may be correlated with increased benefit from EGFR TKIs
such as gefitinib or erlotinib
 These appear to be the most promising candidate
biomarkers for EGFR-targeted therapy, e.g. erlotinib
 HOWEVER, incidence of activating mutations in pancreatic
cancer = 1.5%
 Results from one tumour type can not be extrapolated to
another tumour type
NSCLC = non-small cell lung cancer; IHC = immunohistochemistry
FISH = fluorescence in-situ hybridisation; TKIs = tyrosine-kinase inhibitors
PA.3: survival by EGFR IHC status
(100mg cohort)
EGFR IHC+ (n=70)
1.00
Erlotinib (n=34)
Placebo (n=32)
0.75
HR=0.75
0.50
95% CI: 0.46–1.23
(p=NS)
0.25
0
Survival probability
Survival probability
EGFR IHC– (n=66)
1.00
Erlotinib (n=41)
Placebo (n=29)
0.75
HR=0.82
0.50
95% CI: 0.50–1.32
(p=NS)
0.25
0
0
6
12
Months
18
24
0
6
12
18
24
Months
Data on file, OSI Pharmaceuticals Inc.
PA.3: implications of EGFR gene copy
number for survival
EGFR FISH+
(n=50)
EGFR FISH–
(n=57)
HR
p
value
Number of partial responses
5/48
4/53
1.36
0.68
Median OS (months)
5.3
7.8
1.24
0.32
Median PFS (months)
3.6
4.2
1.85
0.004
N.B. results are irrespective of treatment received
Gemcitabine Gemcitabine
+ erlotinib
+ placebo
HR
HR +
p value for
p value interaction
Survival (months)
EGFR FISH+
5.2
5.2
0.90
0.73
EGFR FISH–
8.4
6.7
0.60
0.08
1.41
(p=0.31)
Moore M, et al. J Clin Oncol 2007(Suppl. 18 Pt I):s203(Abstract 4521)
KRAS: a predictive factor?
 KRAS point mutations in codon 12
 KRAS mutations associated with resistance to
cetuximab in CRC
 KRAS mutations associated with shorter survival in
pancreatic cancer compared with wild-type KRAS
CRC = colorectal cancer
Lee J, et al. Cancer 2007;109:1561–9
PA.3: implications of KRAS mutations
for survival
KRAS mutant
(n=92)
KRAS WT
(n=25)
HR
p
value
Number of partial responses
9/90
0/21
–
0.20
Median OS (months)
6.7
5.4
0.63
0.37
Median PFS (months)
3.8
3.7
0.86
0.53
N.B. results are irrespective of treatment received
Gemcitabine Gemcitabine
+ erlotinib
+ placebo
HR
HR +
p value for
p value interaction
Survival (months)
KRAS mutant
6.0
7.4
1.07
0.78
KRAS WT
6.1
4.5
0.66
0.34
WT = wild-type
1.71
(p=0.29)
Moore M, et al. J Clin Oncol 2007(Suppl. 18 Pt I):s203:(Abstract 4521)
Limitations of PA.3 biomarker data
 Data are from retrospective analyses and include only
small numbers of patients
– no data as yet from prospective, randomised study
designed to investigate biomarkers for predicting
clinical benefit with erlotinib plus gemcitabine
– studies are planned but are in early stages
No studies or guidelines to date suggest use of a
specific biomarker to select patients to receive
erlotinib plus gemcitabine
Randomised phase II biomarker study
Advanced
pancreatic
cancer failing
previous
chemotherapy
or unsuitable for
first-line
chemotherapy
PS 0–2, n=200
Erlotinib 150mg
Placebo + BSC
PD
Subsequent
treatment at
investigators’
discretion
PD
Erlotinib 150mg
or other active
treatment
 Primary endpoint: PFS
– biomarkers: EGFR IHC, EGFR FISH, EGF, TGF-α,
amphiregulin, KRAS mutations, EGFR mutations, RANTES
 Secondary endpoints: response, disease control, OS, CA19-9, safety
BSC = best supportive care
Challenges of finding predictive
biomarkers in pancreatic cancer
 Practical difficulties of tissue acquisition in pancreatic cancer
– numbers of patients with samples often small
 Current status of translational research
– EGFR FISH positivity (only in 5–10% of patients)
– EGFR mutations (no activating mutations)
– KRAS mutation status (90% tumours positive, simulations suggest
benefit in PA.3 highly unlikely to be driven by wild-type carriers only)
 Recent analyses inconclusive
– NCIC
• EGFR FISH
• KRAS
 False results have major implications
– may deny patient access to life-extending treatment
– patients may receive a therapy they are unlikely to benefit from
Randomised phase III trial: adding
bevacizumab to erlotinib plus gemcitabine
Previously
untreated
metastatic
pancreatic cancer
(n=600)
Erlotinib (100mg) +
gemcitabine +
bevacizumab (5mg/kg
every 2 weeks)
PD
Standard regimen used
in PA.3., i.e. erlotinib
(100mg) +
gemcitabine
+ placebo
PD
 Primary endpoint of OS not met
 Secondary endpoints were met, however, demonstrating that the
combination has some clinical activity
 Confirmed efficacy of gemcitabine plus elotininib control arm, in line
with PA.3
 Data will be presented at an upcoming congress in 2008
Bevacizumab plus gemcitabine plus
erlotinib or cetuximab
Previously
untreated
advanced
pancreatic cancer
(measurable
disease), PS 0–2
Bevacizumab +
gemcitabine +
cetuximab
R
Bevacizumab +
gemcitabine +
erlotinib
BGC
(n=27)
BGE
(n=24)
RR
19%
21%
SD
59%
67%
Median PFS (months)
3.6
3.6
6-month survival rate
41%
38%
Preliminary results
(n=51)
Kindler HL, et al. J Clin Oncol 2006;24(Suppl. 18 Pt I):s188 (Abstract 4040)
Randomised phase III trial: erlotinib plus
concurrent/sequential gemcitabine/capecitabine
Advanced
pancreatic
cancer
Erlotinib +
capecitabine
PD
Gemcitabine
PD
Erlotinib +
gemcitabine
PD
Capecitabine
PD
R
 Primary endpoint = time to second progression
 n=132/250
PI: Prof V Heinemann (Germany)
Erlotinib plus gemcitabine: conclusions
 Erlotinib plus gemcitabine = a new treatment option
for patients with advanced pancreatic cancer
 Further research underway to establish if patients
who will obtain greatest benefit from this regimen can
be identified
– also investigating relationship between rash and
clinical benefit
 Erlotinib and gemcitabine now provides a backbone
of therapy for pancreatic cancer
– future trials will evaluate addition of other novel
agents, e.g. bevacizumab, to further extend survival