Transcript Safety & tolerability

Safety & Tolerability of Biologics
Dubai, United Arab Emirates
January 19th, 2009
Prof. Joachim R. Kalden
Director emeritus
Department of Internal Medicine III
Div. of Molecular Immunology
Niklolaus-Fiebiger-Center
University of Erlangen-Nuremberg
Overview
Monitoring of safety: registries
Safety: Tuberculosis
Safety: Serious infectious events
Safety: Malignancy
Safety: Lymphoma
European registries
Post approval commitment to EMEA that Wyeth
would monitor safety
Professional Societies independently supported
establishment of registries to monitor safety
Established in a number of countries
independently but with different approaches,
with or without
 Reference groups
 Efficacy endpoints
Why important?
Only source of comparisons with competitors
Providing rich stream of data for differentiation
Currently under analysed – registries focusing
on limited number of questions
Potential for pooled analyses to give even
greater power
Why are registries important to
Wyeth
Meet post approval commitment
Large cohorts of Enbrel patients treated in
clinical practice
Source of comparative data with MAbs
Resource for evaluation of safety and efficacy
Potential to combine data to provide increased
power to undertake analyses of rare events
Original registries
Sweden
 ARTIS – nationwide but organised on a regional basis
–
–


STURE – Stockholm registry
SSAGT – Southern Sweden
Both ARTIS and the regional registries publish results
No concurrently recruited controls but use other RA cohorts for reference
UK

BSRBR - national registry powered to detect 2 fold increase in lymphoma in
comparison to a DMARD treated cohort
Germany
 RABBIT – national registry to describe the long term effectiveness
–
–
–
–

treatment continuation
clinical outcomes
long term hazards
direct and indirect costs
Comparison with conventional DMARD treatment from national database
These 3 registers meet together annually with companies
Use standard report which is sent companies twice a year for inclusion in safety
reports to regulators
Estimates of patient numbers
Country
Started
Date of
Estimate
Total
TNFi
Controls
Sweden
1999
2008
14000
National DB
UK
2001
2008
12000
3300
Germany
2001
2008
5300
National DB
Spain
2000
2007
8320
N/A
Norway
2000
2006
1500
DMARDs
Denmark
2000
2006
1021
None
Czech
2001
2007
1040
Switzerland
2000
2005
1600
France
Total
44781
SCQM
Monitoring of safety: problems
Controlled trials
 Relatively few pts, not the same patient population as
in clinical practice, seldom long-term, randomisation
may create well-balanced comparator
Spontaneous reporting
 Very low reporting rates; only certain adverse drug
reactions (with attribution)
Long-term observational studies
 Difficult to obtain enough compliance, need for
comparator data
Advantages and disadvantages of
registries
Advantages
 Usually much larger than
clinical trials
 Greater power than RCTs to
detect rare events
 Enrolment reflects clinical
practice
 Potential for studying
numerous outcomes
 Suited to long term follow-up
 Can examine complex
situations not suited to RCTs
 Results can usually be
generalised
Disadvantages
 Non randomised, subject to
bias
 Confounding by indication
 Missing data
 Potential for confounding
 Channelling bias
 Choice of reference group
Conditions Where Mechanism of Action
Differences May Affect Safety Profile
TB
Serious Infectious Events (SIEs)
Malignancy
Tuberculosis (TB)
FDA MedWatch spontaneous
reporting system (AERS): 2001
TB associated with
infliximab
70 reported cases of TB
after treatment with
infliximab for a median of
12 weeks
Normal
 40/70 had extra-
pulmonary disease
Post-infliximab
Keane J. et al. NEJM 2001;345:1098-1104.
Inhibition of IFN gamma
Effect of drugs on IFN production in whole blood cultures stimulated with
M tb culture filtrate. Median and interquartile ranges n=15
Saliu et al. J Infect Dis 2006 .
TB associated with clinical trials:
TB events despite screening
Anti-TNF Agent
Etanercept
Required
Screening for TB
(soluble receptor)
Not mandated
Pooled analysis of all clinical trials across
indications (Europe / N. America)4
(N=11,390; 2 cases TB)
Adalimumab
(mAb)
TB Event Rate
(per 1000 pt-yr)
0.11
Yes
Pooled analysis Europe1
3.3
Pooled analysis N. America1
0.8
REACT Trial7
5.0
Infliximab
(mAb)
Yes
Pooled analysis
Not available
START Trial2
13.1
Infliximab vs abatacept trial3
12.1
Certolizumab
Pooled analysis
(mAb)
Yes
Not available
RAPID I Trial4
6.9
RAPID II Trial5
12.5
Enbrel and TB: Portugal
TB events associated with anti-TNF agents were compared for 960 pts treated
between 1999 – 2005 in Portugal*:
Percentage of Treated Patients Developing TB
2.5
2.3%
(4/171)
2
1.5
1.5%
Etanercept
Infliximab
(8/456)
% TB
1
Adalimumab
0.5
0.3%
(1/333)
0
Infliximab
Adalimumab
Etanercept
* 13 events total: 9 RA (n=639); 3 AS (n=200); 1 PsA (n=101)
.
Fonseca JE et al. Acta Reumatol Port. 2006;31:247-53.
BIOBADASER: Risk and incidence
of TB in Spain
Patient-years
of exposure
to TNF
antagonists
Cases
<March 2002
8,671
41
472 (384–642)
19 (11–32)
5,8 (2,5–15,4)
>March 2002
8,717
15
172 (103–285)
7 (3–13)
2,4 (0,8–7,2)
4,576
2
43 (11–175)
1,8 (0,28–7,1)
Undetermined
4,170
13
311 (181–636)
13 (6–25)
4,8 (1,04–44,3)
Treatments starts
 100%
compliance
 <100%
compliance
TB rate per
100,000 p/y
RR vs. general
population
(IC 95%)
RR vs. EMECAR
(IC 95%)
Annual incidence rate / 100,000 p. y. General population 25; EMECAR (RA pre-biologic era)
Carmona et al. Arthritis Rheum 2005; 52(6):1766-72; Gómez-Reino et al. Arthritis Care & Research 2007.
BSRBR: Anti-TNFs and risk of TB
191
200
180
160
136
Rate/100,000 yrs
140
120
100
80
50
60
40
20
Dixon WG et al. THU0134. EULAR 2008
ab
In
fl i
xi
m
ab
da
l im
um
A
Et
an
er
ce
pt
0
RATIO: Factors predictive of TB
Use of specific biologics is predictive of TB in anti-TNFtreated patients (n=67)
•
•
•
•
•
Last anti-TNF
received
No. of cases
HR (95% CI) vs
etanercept
P-value
Adalimumab
27
10.05 (1.92-52.61)
0.006
Etanercept
35
Infliximab
5
8.63 (1.38-53.78)
0.02
Incidence TBC 39.2/100,000 pt/year
ETA: 6.0/100,000 pt/year
INF or ADA: 71.5/100,000
General Population: 8.7/100,000 pt/year
Two thirds (30/45) of the patients who developed TB had negative skin tests
Tubach F et al. OP-0014. EULAR 2008
Serious Infectious Events
(SIEs)
Etanercept and serious infectious events
Pooled analysis of randomised clinical trials for Enbrel in RA
Serious infectious events
3.5
3
2.5
Events per 2
100 pt-yr 1.5
1
0.5
0
Placebo/control
Etanercept
Etanercept open label
extensions
No difference vs. placebo
Hamza S et al. EULAR 2007 ARD 2007;66(2): THU0153 Abstract
Etanercept and opportunistic infections
Pooled analysis of randomised clinical trials for Enbrel in RA
Opportunistic infections
0.35
0.3
0.25
Events per 0.2
100 pt-yr 0.15
0.1
0.05
0
Placebo/control
Etanercept
Etanercept open label
extensions
No difference vs. placebo
Hamza S et al. EULAR 2007 ARD 2007;66(2): THU0153 Abstract
BSRBR: Etanercept and mAbs vs.
DMARDS
SIE rates relative to DMARDS in
first 90 days of therapy
6
5
4
Adjusted
Incidence 3
Rate Ratio
2
1
0
ETN
Dixon WG et al. A&R 2007;56(9):2896-904
INFLIX
ADAL
RABBIT: Etanercept and SIEs Herpes virus infections
RABBIT Registry (Germany)*
Evaluated RA patients for reactivation
or first infection of Herpes virus
infections (Varicella Zoster Virus,
Herpes simplex Virus)
“Our data suggest a different mode of
action of TNF antibodies and the
soluble TNF receptor fusion protein
etanercept in respect to the
reactivation of a latent herpes
infection.”
Reactivation of Herpes virus
infections suggest a loss of cellmediated immunity
Risk of Infection vs. Control
Hazard
Ratio
P value
Etanercept
1.2
0.53
Infliximab
2.1
0.01
Adalimumab
2.0
0.01
*Strangfeld A. et al. Oral Presentation Abstract OP0214 Friday June 15, 2007 EULAR 2007
ARTIS: Hospitalisation risk for
infection – all anti-TNFs
Relative risk (9.5% CI)
1.5
1
0.5
0
Year 1
Year 2
Year 3
Time since treatment start
Askling J, et al. Ann Rheum Dis. 2007 66:1339–44
ARTIS: Serious infection rate in patients
treated with a 2nd TNF antagonist
Patients hospitalized prior to
treatment with TNF inhibitor
(n=2,692)
All TNF inhibitor-treated
patients
Infections leading to
hospitalisation/100
patient-years
Infections leading to
hospitalisation/100
patient-years
(n=4,167)
16
7.0
12
8
4.5
4
0
First TNF inhibitor
Second TNF inhibitor
10.0
16
12
8
5.4
4
0
First TNF inhibitor
Crude Incidence
Adapted from Askling J, et al. Ann Rheum Dis. 2007 66:1339–44
Second TNF inhibitor
Malignancies
Malignancy
Malignancy Risk: ENBREL vs. control,
derived from a large patient database*
Analyses
Number of Patients
Point Estimate
95% CI
Lymphoma only
19591
0.7(OR)
0.3-1.6
All malignancies
13001
1.0 (OR)
0.8-1.3
Skin cancer
13001
1.2 (OR)
1.0-1.5
Lower Risk
*Wolfe F and Michaud K. A&R 2007;56:1433-1439; 2886-2895.
1.0
Higher Risk
ARTIS: Anti-TNF and solid cancers
Swedish national registry ARTIS
 Data compared with Swedish nationwide cancer & census
registers
Prevalent RA
n = 53067
Incident RA
n = 3703
RA on anti-TNF
n = 4160
n
SIR
(95% IC)
n
SIR
(95% IC)
n
SIR
(95% IC)
All cancers
3379
1.05 (1.01-1.08)
138
1.1 (0.9-1.3)
67
0.9 (0.7-1.2)
Melanoma
120
1.19 (0.99-1.42)
4
0.9 (0.2-2.2)
1
0.3 (0.0-1.8)
Askling et al. Ann Rheum Dis 2005;64:1414–1420
ARTIS: Anti-TNF and solid cancers
Swedish national registry ARTIS
 Data compared with Swedish nationwide cancer & census
registers
Prevalent RA
n = 53067
Incident RA
n = 3703
RA on anti-TNF
n = 4160
n
SIR
(95% IC)
n
SIR
(95% IC)
n
SIR
(95% IC)
All cancers
3379
1.05 (1.01-1.08)
138
1.1 (0.9-1.3)
67
0.9 (0.7-1.2)
Melanoma
120
1.19 (0.99-1.42)
4
0.9 (0.2-2.2)
1
0.3 (0.0-1.8)
Skin (nonmelanoma)
374
1.66 (1.50-1.84)
5
0.7 (0.2-1.6)
11
3.6 (1.8-6.5)
Lung
330
1.48 (1.33-1.65)
23
2.4 (1.5-3.6)
10
1.8 (0.9-3.3)
Askling et al. Ann Rheum Dis 2005;64:1414–1420
ARTIS (2008): No increase in cancer
following anti-TNF therapy
No increased cancer risk with anti-TNF therapy
No. primary
cancers
Anti-TNF
treatment
Askling J et al. OP-0013. EULAR 2008
169
Relative risk (95% CI)
Compared to
national RA cohort
Compared to
general population
0.94 (0.80-1.12)
1.04 (0.89-1.21)
RA and cancer
National Data Base for Rheumatic Diseases (NDBRD)
 13,001 RA patients (49,000 p/y) of observation (1998/2005)
– At least 1 year of follow-up
– 49% of whom exposed to biological therapy
 Cancer rates compared with population rates US National Cancer
Institute database
 Incidence of new cancers in patients with anti-TNF/ without anti-
TNF
 ORs as estimates RR adjusted for 6 confounders: age, sex,
education level, smoking history, RA severity and prednisone
use
Wolfe , ACR 2006 and Arthritis and Rheum 2007
Lymphoma
Lymphoma and rheumatic disease
Several studies suggested an increased
lymphoma risk in patients with rheumatic
disease
Risk may be tied to degree of disease severity
and inflammation
What is the impact of biologics on this?
Lymphoma risk and RA disease
activity: Pre-biologics
Patients: Swedish in-patient registry with RA
1964–1995, who developed lymphoma > 1 year
after discharge (RA and lymphoma diagnosis
validated)
Controls: Individually matched RA patients
without lymphoma from same source (378 cases
and controls)
All records retrospectively reviewed to assess
disease activity and DMARD therapy
Baecklund E, et al. Arthritis Rheum. 2006;54:692-701.
Outlook
TNFalpha antagonists might improve or prevent
important comorbidities

Cardiovascular diseases

Lymphomas

Decreasing inflammation

Decreasing activation of endothelial cells

Normalizing pathologic lipid profiles

Normalizing insulin resistance
By:
Lymphoma risk and rheumatoid
arthritis disease activity
71.3
Inflammatory Activity
High
7.7
Medium
Low
0.1
1
10
Unadjusted Odds Ratio (95% CI)
Baecklund E, et al. Arthritis Rheum. 2006;54:692-701.
100
No increased risk of lymphoma in RA
patients upon treatment with anti-TNF
Swedish population-based cohort study of RA pts:
 one prevalent cohort (n = 53067)
 one incident cohort (n = 3703)
 one TNFi -treated cohort 1999 through 2003 (n = 4160)
Prevalent and incident RA pts have an increased risk of lymphoma
(SIR = 1.9 and 2.0, respectively) and leukaemia (SIR = 2.1 and 2.2,
respectively)
RA pts treated with TNF antagonists had a tripled lymphoma risk
(SIR = 2.9)
However, after adjustment (sex, age, and disease duration) the risk
was not higher than in the other RA cohorts
.
Askling et al. Ann Rheum Dis 2005;64:1414–1420
Further safety issues
1.
Infections
2.
Pregnancies
3.
Non-tb pulmonary disease
4.
Heart failures
5.
Surgical issues
6.
Vaccination
7.
Haematological chances
8.
Demylating diseases
9.
Allergic reactions
Summary: Areas in which data suggest a
difference between mAbs and Etanercept
Tuberculosis
Risk of developing TB appears to be greater with mAbs than with Etanercept
based upon:
 Pooled analyses of randomized controlled trials
 Multiple national registries
 MOA
Malignancy
Possible risk for development of lymphomas or other malignancies in
patients treated with a TNF-antagonist, including Etanercept, cannot be
excluded
 Further analyses pending
Serious Infections
Differences in risk for infections may exist between Etanercept and mAbs;
however, data are inconclusive
 RCTs suggest a difference
 Registries suggest no difference