Transcript FDA Evaluation of Hepatotoxicity Related to

FDA Evaluation of Hepatotoxicity
Related to Tyrosine Kinase Inhibitors
J Chang,1 M Rand,2 G Blumenthal,1 P Cortazar,1 C Kulick,2
E Hausman,2 S Chang,2 R Pratt,2 B Habtemariam,3 Y Chen,3
J Bullock,3 RC Orbach,3 I Zineh,3 R Justice,1 R Pazdur4
1Division
of Drug Oncology Products; 2Office of Surveillance and Epidemiology;
3Office of Clinical Pharmacology; 4Office of Oncology Drug Products
Background
Tyrosine kinase inhibitors (TKIs) interfere with several critical
signal transduction pathways in cancer
As of March 2010, FDA had approved 9 TKIs (imatinib,
dasatinib, nilotinib, erlotinib, gefitinib, lapatinib, sorafenib,
sunitinib, and pazopanib) to treat a variety of malignancies
All TKIs are associated with severe liver injury (SLI) and have
different categories for liver injury, ranging from boxed
warnings for fatal hepatotoxicity to adverse reactions for
elevated LFTs
LFTs=liver function tests.
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Methods

Analyses included a literature review and an evaluation of postmarketing
reports of SLI associated with TKIs from FDA’s Adverse Event Reporting
System* (AERS) from various dates of marketing approval until March 15,
2010. SLI is defined as acute liver injury (elevated LFTs, bilirubinemia, or
jaundice) in combination with 1 of the following events: death, liver
transplantation or placement on liver transplant list, hepatic encepholopathy,
coagulopathy, or renal impairment

Drug utilization data were reviewed for comparison across TKIs. The
estimated number of nationally projected dispensed prescriptions from mail
order/specialty pharmacy and outpatient retail pharmacy settings was
obtained from Wolters Kluwer Source PHAST Prescription Monthly™ from
2006 through 2009

For a mechanistic explanation, laboratory and exposure data were reviewed
from 8 new drug application (NDA) submissions. Gefitinib data are excluded
because it is no longer marketed in the United States. Exposure-response
properties of TKI-induced liver enzyme and bilirubin elevations were
evaluated
* Limitations of AERS data include: underreporting, stimulated reporting, biased reporting, variable quality and
content, and no certainty that the reported event is due to the product.
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Results
Literature search: As of April 2010, there were 19 case reports
of liver injury with imatinib, 7 with erlotinib, 5 with sorafenib,
4 with sunitinib, and 2 with gefitinib. 15 of these reports
were captured in AERS
Drug utilization results: Total oral TKI market increased
from approximately 313,000 prescriptions in 2006 to 348,000
prescriptions in 2009. Imatinib had the largest proportion
of dispensed prescriptions
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Results (cont’d)
Total Number of Nationally Projected Dispensed Prescriptions for
TKIs in US Mail-Order and Outpatient Retail Pharmacies, 2006-2009
FDA Approval Year
Total Prescriptions
2006-2009, N
Market Share
2006-2009, %
Imatinib
2001
652,997
45.1
Erlotinib
2004
479,114
33.1
Sunitinib
2006
148,864
10.3
Lapatinib
2007
66,008
4.6
Dasatinib
2006
47,353
3.3
Gefitinib
2003
17,325
1.2
Sorafenib
2005
25,726
1.8
Nilotinib
2007
9,671
0.7
Pazopanib
2009
105
0
1,447,163
100
TKI
TOTAL
Source: Wolters Kluwer SOURCE PHAST Prescription, 2005-2009.
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Postmarketing Cases
Of the 497 cases evaluated for SLI, 55 were categorized as
possibly or likely associated with a drug
The majority of SLI cases were associated with sunitinib (22)
and imatinib (13)
Most common event terms were hepatic failure, hepatic
necrosis, hepatic coma or encephalopathy, and
hepatorenal failure
67% of postmarketing cases were from foreign sources, and
45% of cases involved patients enrolled in clinical trials
No dasatinib cases were classified as possibly or likely
associated with a drug
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Postmarketing TKI Cases of SLI*
Sunitinib
n=22
Imatinib
n=13
Gefitinib†
n=7
Erlotinib
n=6
Sorafenib
n=2
Nilotinib
n=2
Lapatinib
n=1
Pazopanib
n=2
Age (years)
Median
Range
64
45-83
50
4-78
72
48-88
67
53-77
56
56
38
22-53
54
54
29
21-37
Gender
Male
Female
Unknown
14
8
1
10
3
4
3
3
3
1
2
0
0
1
0
2
RCC 16
GIST 2
Other 4
CML 8
GIST 3
Other 2
CML 1
GIST 1
Breast 1
Synovial
sarcoma 2
Demographics
Cancer type
1
NSCLC 6
Pancreatic 1
NSCLC 5
Other 1
Thyroid 1
RCC 1
* Limitations of AERS data include: underreporting, stimulated reporting, biased reporting, variable quality and content, and no
certainty that the reported event is due to the product.
† Gefitinib is no longer marketed in the United States.
CML=chronic myelogenous leukemia; GIST=gastrointestinal stromal tumor; NSCLC=non–small cell lung cancer;
RCC=renal cell carcinoma.
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Postmarketing TKI Cases of SLI* (cont’d)
Sunitinib
n=22
Imatinib
n=13
Gefitinib†
n=7
Erlotinib
n=6
Sorafenib
n=2
Nilotinib
n=2
Lapatinib
n=1
Pazopanib
n=2
Reported Adverse Event and Event Characteristics
16
2
2
10‡
1
0
4
0
1
5
0
0
1
0
1
2
0
0
1
0
0
2
0
0
0
2
1
1
0
2
1
0
0
0
0
0
0
0
0
0
Dose (mg)
Median
Range
50
25-50
400
200-400
250
None
150
100-150
800
None
800
None
1500
None
800
None
Time to event (days)
Median
Range
27
7-314
150
8-312
22
4-180
28
12-71
68
56-80
63
6-120
126
126
45
34-55
Outcome
Death
Hospitalization
Life-threatening
Other
16
4
2
0
8
5
0
8
6
1
0
0
5
1
0
0
1
0
0
1
2
0
0
0
0
1
0
0
2
0
0
0
(+) dechallenge
(+) rechallenge
5
0
1
0
0
2
1
0
1
0
0
0
1
0
0
0
Hepatic failure
Hepatic necrosis
Hepatic coma or
encephalopathy
Hepatorenal failure
Other§
* Limitations of AERS data include: underreporting, stimulated reporting, biased reporting, variable quality and
content, and no certainty that the reported event is due to the product.
† Gefitinib is no longer marketed in the United States.
‡ 2 Patients required liver transplant.
§ Other includes: hepatitis fulminant (2), liver disorder (1), hepatitis (1), hepatic cirrhosis (1).
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Pharmacogenomics
This is being explored as a potential tool to understand the
mechanistic basis of drug-induced liver injury (DILI) and
manage its clinical risks
Susceptibility to idiosycratic DILI is currently attributed to the
complex interaction of multiple factors such as genetic,
environmental, and drug-related factors
Genetic susceptibility to DILI is mostly associated with
variations in immune response and/or drug metabolism and
transport genes
Lapatinib-induced hepatotoxicity has recently been linked to a
HLA allele
HLA=human leukocyte antigen.
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Examples of Genotypic Associations for DILI
Drug
Therapeutic Class
HLA Allele
Reference (PMID)
Flucloxacillin
Antibiotic
HLA-B*5701
Daly et al. 2009
[19483685]
Ximelagatran
Thrombin inhibitor
HLA-DRB1*07
and HLA-DQA1*02
Kindmark et al. 2008
[17505501]
Lumiracoxib
NSAID
HLA-DQA1*0102
Singer et al. 2010
[20639878]
Antineoplastic
HLA-DQA1*0201
Spraggs et al. 2011
[21245432]
Lapatinib
NSAID=nonsteroidal anti-inflammatory drug; PMID=PubMed identifier.
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Clinical Pharmacology

Preliminary review of the data from NDA submissions suggests that TKIs inhibit bilirubin
clearance. As shown below, sorafenib (sor), nilotinib (Nil), and pazopanib (pazo) exhibit
concentration-dependent inhibition of the hepatic UGT1A1 enzyme in vitro. Inhibition of
UGT1A1 is expected to increase total bilirubin in vivo.
100
% Inhibition
80
sor:estradiol
sor:SN-38
Nil:bilirubin
Nil:estradiol
pazo:HFC
60
40
20
0
0.1
0.5
1.0
5.0
10.0
50.0 100.0
Drug Concentration (M)
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Clinical Pharmacology (cont’d)
Probability of ALT >3 x ULN With
Increasing Pazopanib Concentrations
Probability of bilirubin >2 x ULN With
Increasing Pazopanib Concentrations
100
80
Logistic regression
Observed proportion (95% CI)
80
Probability of
ALT >3 x ULN (%)
Probability of
ALT >3 x ULN (%)
Logistic regression
Observed proportion (95% CI)
60
40
20
60
40
20
4/63
3/63
1/64
17/63
2/63
16/63
11/63
0
8/64
0
20
40
60
80
Steady State Trough
Concentration (g/mL)
With increasing trough concentrations of pazopanib,
the probability of ALT >3 x ULN increases with
increasing pazopanib plasma concentrations
0
0
20
40
60
80
Steady State Trough
Concentration (g/mL)
No concentration-dependent increase in total
bilirubin >2 x ULN is observed
ALT=alanine aminotransferese; CI=confidence interval; ULN=upper limit of normal.
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
Conclusions

The pharmacogenomics data for lapatinib thus far suggest a genetic
association for DILI. More research into the predictive value,
sensitivity, and specificity of these biomarkers is needed to
understand their clinical utility for predicting patients who are
susceptible

These associations may be used in the future to identify individuals at
greater risk of developing liver injury

Consideration of both genetic and nongenetic factors in DILI may help
to identify these patients and improve therapy management.
Investigations of the exposure response analysis for other TKIs are
ongoing

Review of postmarketing and clinical pharmacology data suggests a
role for TKIs in causing hepatotoxicity. However, review of
postmarketing AERS data* precludes class labeling for SLI for all TKIs

Greater numbers of cases for imatinib and sunitinib may relate to
greater national market share of TKIs
* Limitations of AERS data include: underreporting, stimulated reporting, biased reporting, variable quality and
content, and no certainty that the reported event is due to the product.
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.
TKI Labeling for Hepatotoxicity
Drug
Boxed
Warning
Imatinib*
Dasatinib
Nilotinib
Erlotinib
Gefitinib†
Lapatinib
Sunitinib
Warnings and
Precautions
Fatal Hepatotoxicity, Severe
Liver Injury (requiring liver
transplant),
Acute Liver Failure,
Elevated LFTs
Elevated LFTs
Hepatic Failure and Hepatorenal
Syndrome (including fatalities)
Elevated LFTs
(Precautions section)
Fatal Hepatotoxicity
Fatal Hepatotoxicity
Fatal Hepatotoxicity
Fatal Hepatotoxicity
Sorafenib*
Pazopanib
Fatal Hepatotoxicity
Fatal Hepatotoxicity
Adverse
Reactions
Additional Data From
Clinical Trials or
Postmarketing Experience
Elevated LFTs
Hepatitis, Hepatic Failure,
and Hepatic Necrosis
(including some fatalities)
Elevated LFTs
Elevated LFTs
Hepatitis
Hepatitis, Hepatotoxicity
Elevated LFTs
Hepatotoxicity
Hepatotoxicity
Drug-induced hepatitis,
Liver Dysfunction including reports of hepatic
failure and death
Elevated LFTs
Updated changes are in bold and italics
* Imatinib and sorafenib prescribing information updated in italics.
† Gefitinib is no longer marketed in the United States.
Adapted from Chang J et al. Presented at: ASCO Annual Meeting; June 3-7, 2011; Chicago, IL.