Risk and Outcomes of DILI in Patients with Underlying Liver Disease

INEDSYS Hepatology Club Meeting

April 2022

Drug Induced Liver Injury in Patients with

Underlying Liver Disease

Naga Chalasani, MD

Indiana University School of Medicine

@nagachalasani

-- 1 of 35 --

Disclosures

• Consulting

• Madrigal
• Zydus
• GSK
• Merck
• Altimmune
• Foresite

• Research support

• DSM
• Exact Sciences

• Speaking

• None

-- 2 of 35 --

Outline

• Background

• Changes in drug disposition in NASH and cirrhosis

• Is NASH a risk factor for DILI?

• Is DILI worse in patients w/CLD and cirrhosis?

• Can a medication be selectively hepatotoxic in certain types of

liver disease?

-- 3 of 35 --

Drug Induced Liver Injury

• ~3-10% of acute liver injury in the US

• Single, major cause of acute liver failure

• Common cause for a medication to be

abandoned during development

• Common cause for withdrawal or restriction of

use of an approved medication

 Troglitazone, bromfenac, trovafloxacin, bosentan,

telithromycin, lumiracoxib, ximelagatran, tolvaptan,

obeticholic acid.

April 2014

-- 4 of 35 --

Direct Hepatotoxicity

 Expected outcome, dose related

 Acetaminophen

 Aspirin

 Niacin

 Many antineoplastic agents

 Serum enzyme elevations

 Acute hepatic necrosis

 Sinusoidal obstruction syndrome

 Lactic acidosis, steatosis and hepatic dysfunction

 Nodular regenerative hyperplasia

April 2014

-- 5 of 35 --

Idiosyncratic Hepatotoxicity

 Unexpected outcome, rare, and generally not-dose related

 Isoniazid

 Troglitazone

 Amoxicillin/Clavulanic acid

 Minocycline

 Diclofenac

 Idiosyncrasy: immunologic or metabolic

 Etiology, generally unknown

 Challenge to clinical and basic research

-- 6 of 35 --

Drug Induced Liver Injury: Phenotypes

Hepatitis: hepatocellular, cholestatic or mixed

Serum enzyme elevations only

Acute hepatic necrosis (direct hepatotoxicity)

Bland cholestasis (estrogens, androgens)

Chronic hepatitis, cirrhosis

Nonalcoholic steatohepatitis (NASH)

Lactic acidosis, steatosis & hepatic dysfunction

Nodular regeneration, sinusoidal obstruction

-- 7 of 35 --

Cholestatic vs Hepatocellular

Drug induced liver injury is usually classified as either

Hepatocellular (acute viral hepatitis like)

Cholestatic (similar to biliary obstruction)

R ratio: ALT/Alk P [expressed as x ULN]

HC >5, Cholestatic <2, Mixed 2-5

In general population, HC is more dangerous than CS, but

in cirrhosis even cholestatic liver injury becomes

important

April 2014

-- 8 of 35 --

Hy’s law & Temple Corollary

Hy’s Law Observation made by late Hyman Zimmerman suggesting a 1 in 10

mortality risk of DILI if the following 3 criteria are met:

1. Serum ALT or AST > 3 x ULN

2. Serum total bilirubin elevated to >2xULN, without initial

findings of cholestasis (elevated serum alkaline phosphatase)

3. No other reason can be found to explain the combination of

increased aminotransferases and bilirubin, such as viral

hepatitis A, B, C or other preexisting or acute liver disease

Temple’s Corollary An imbalance in the frequency of ALT > 3 x ULN between active

treatment and control arms in a randomized controlled trial. This

is used to assess for hepatotoxic potential of a drug from

premarketing clinical trials.

ALT >10 ULN from the

drug (Senior Rule)

For every 10 cases of ALT > 10 ULN, there would be 1 Hy’s law

episode

-- 9 of 35 --

eDISH (Evaluation of Drug Induced Serious Hepatotoxicity)

Studies are ongoing

(a) to modify eDISH for patients with elevated baseline liver tests

(b) to plot Peak Tbili vs Peal AlkP for cholestatic DILI

-- 10 of 35 --

Idiosyncratic DILI is not entirely

dose-independent

Lammert C, et al, Chalasani N. H epatolog y 2008

compounds to have

caused

< 10 mg

(n=54)

10-50 mg

(n=83)

>50 mg

(n=97)

P-value

ALT > 3 ULN (%) 19 27 30 0.14

Jaundice (%) 33 40 44 0.19

Liver Failure (%) 17 12 32 0.009

Death (%) 11 11 28 0.004

Transplant (%) 0 2 13 <0.001

Median # prescriptions in

2005

4,746,500 4,938,000 3,733,000 0.3

• Numerous reports of DILI precipitating when there is dose escalation

(e.g., azathioprine, statins, duloxetine)

-- 11 of 35 --

Drugs which are hepatically

metabolized are more hepatotoxic

Compounds Proportion of compounds causing

ALT > 3

ULN

(n=61)

Jaundice

(n=93)

Liver Failure

(n=49)

Liver

Transplant

(n=15)

Fatal DILI

(n=39)

Extensive

hepatic

metabolism

34% 43% 28% 9% 23%

Without

extensive

hepatic

metabolism

10% 34% 9% 1% 4%

P-value 0.007 0.2 0.001 0.45 0.0003

Lammert C, et al. H epatolog y 2009

-- 12 of 35 --

Combined effect of Dose and Metabolism

Non extensive hepatic

metabolism (n=64)

Extensive hepatic metabolism

(n=162)

Average Daily Dose Average Daily Dose

≤ 10 mg

(n=10)

11-49 mg

(n=22)

≥50mg

(n=32)

≤ 10 mg

(n=43)

11-49 mg

(n=61)

≥50mg†

(n=58)

ALT>3x ULN

(n=61)

2

(20%)

4

(18%)

5

(14%)

8

(19%)

15

(24.5%)

27*

(46.5%)

Jaundice

(n=93)

4

(40%)

10

(45%)

13

(40%)

14

(32.5%)

20

(33%)

32

(55%)

Liver failure

(n=49) 0 2

(9%)

4

(12.5%)

9

(21%)

8

(13%)

26*

(45%)

Liver

Transplant

(n=15)

0 0 2

(6.2%) 0 2

(3.3%)

11**

(20%)

Fatal DILI

(n=39) 0 2

(9%)

2

(6.2%)

6

(14%)

7

(11.4%)

22*

(34%)

Lammert C, et al, Chalasani N. Hepatology 2009

-- 13 of 35 --

DILI Score & Rules of Two & Three

• Rule of Two: Daily dose ≥ 100 mg & lipophilicity

(logP≥3)

• Rule of Three: Daily dose ≥ 100 mg + lipophilicity

(logP≥3)+ Reactive metabolite formation

Chen et al. Hepatology 2013; 58:201-213

Chen et al. Hepatology 2016; 64: 931-940

-- 14 of 35 --

Mechanisms behind altered pharmacokinetics in cirrhosis

Mechanism Comments

1 Reduced Intrinsic Capacity

• Decreased enzymes
• Decreased activity

There may be etiology specific variability in

different CYPs. Alcohol and infection can

affect some CYPs selectively

2 Reduction in blood flow

3 Shunts Intrahepatic shunts in the fibrous bands –

may account for up to 65% of hepatic

blood flow

Spontaneous portasystemic shunts and

TIPS can significantly modify

4 Changes in protein binding

5 Reduced delivery of oxygen Oxygen is important for CYP activity

6 Altered transporter expression and

function

Hepatocyte uptake as well as efflux into

biliary canaliculi can be affected

7 Gut CYP activity TIPS significantly reduces small bowel

CYP3A activity 15

-- 15 of 35 --

Hepatic DME and transporters in NASH

Marie S, et al, Cherrington NJ. Acta Pharmaceutica Sinica B. 2023; 12:1-28

-- 16 of 35 --

Clinically important alterations in pharmacogene expression in

association with severity in NAFLD

Powell NR, et al, Chalasani N. Nature Communications 2023, March 7 Online

-- 17 of 35 --

Underlying chronic liver disease & DILI

-- 18 of 35 --

DILI in patients with chronic liver disease

• Are individuals with chronic liver disease are more

susceptible to idiosyncratic DILI?

• Is DILI associated with worse outcomes in individuals with

chronic liver disease?

• Causality assessment for DILI in individuals with chronic

liver disease

-- 19 of 35 --

Drugs reported (or predicted) to have an increased risk of

hepatotoxicity in liver disease with some evidence

Anti-tuberculosis drug (e.g., INH, rifampin, pyrazinamide)

HAART (nevirapine)

Methimazole

Methotrexate (chronic DILI)

Nefazodone

Propoxyphene

Tamoxifen (steatohepatitis)

Valproate

Vitamin A

Modified from Lewis & Stine. Aliment Pharmarcol Ther 2013; 37: 1132-1156

-- 20 of 35 --

Underlying liver disease & risk for chronic DILI

Tamoxifen Obesity and underlying steatosis may increase the risk for tamoxifen-related

steatohepatitis

Amiodarone Obesity is considered a risk factor of amiodarone related chronic liver disease.

Methotrexate Obesity, hepatitis C, and heavy alcohol consumption are considered as risk

factors

Irinotecan Obesity is a risk factor for irinotecan induced steatohepatitis

Stavudine There may be worsening of underlying steatosis/steatohepatitis Massart J, et al. J Clinical and Translational Research 2017

• The risk of drug induced hepatic steatosis due to a novel glucagon-receptor

antagonist as well as due to basal insulin peglispro was significantly higher in

individuals carrying PNPLA3 risk alleles.

Guzman CB, et al. Hepatol Commun 2018; 23: 561-570

Pillai S, et al. Pharmacogenomics J, 2018; 18: 487-493

-- 21 of 35 --

Normal ALT +

statins

Abnormal

ALT *+

statins

Liver dz

Controls

Statin duration (yr) 0.48 ± 0.08 0.48 ± 0.08

Statin discontinue 10.7% 11.1%

↑ AST or ALT 1-10 x ULN 1.7% 4.7% 6.4%

p=0.002 p=0.2

↑ AST or ALT >10 x ULN 0.2% 0.6% 0.4%

p=0.6 p=0.6

Individuals with underlying liver disease are not at higher risk for statin

hepatotoxicity

Chalasani N et al. Gastroenterology 2004; 126; 1287-1292

-- 22 of 35 --

Underlying liver disease & increased risk for all-cause DILI

Hypothesis

Individuals with chronic liver disease are not at increased risk for DILI due to common

hepatotoxic agents.

Methods

Using the Indiana Health Information Exchange, we compared the frequency of

suspected DILI between individuals with likely chronic liver disease (CLD cohort) and two

control groups with no biochemical evidence for liver disease over a 10 year period.

• CLD cohort had serum ALT > 45 U/L on at least two occasions occurring 6-24 months

apart in the absence of positive anti-HCV antibody, hepatitis B surface antigen, heavy

alcohol consumption, or hypotension (N=25,499).

• Control group 1 had ALT ≤ 45 U/L (N=276,897) and Control group 2 had ALT ≤ 31 U/L in

men and ALT ≤ 19 U/L in women (N=212,809) on at least two occasions occurring 6-24

months apart.

• Suspected DILI was defined as ALT > 200 U/L and/or AP > 250 U/L and/or total bilirubin

> 2.5 mg/dl on at least two consecutive occasions within 3 months after receiving a

prescription for one of 10 candidate prescription medications, in the absence of

positive anti-HCV antibody, hepatitis B surface antigen, heavy alcohol consumption, or

hypotension.

Lammert C & Chalasani N. Clinical Gastro Hepatol 2019

-- 23 of 35 --

Frequency of suspected DILI in 3 study groups

Control group 1 CLD Cohort Control group 2

Exposed

(n = 62,936)

Exposed

(n = 6,334)

Exposed

(n = 48,928)

Weight (Kg, mean ± SD) 87 ± 28 91 ± 33 86 ± 28

BMI (Kg/m2) 31 ± 9 31 ± 9 30 ± 8

AST at baseline (IU/L) 30 ± 14 76 ± 151 20 ± 14

ALT at baseline (IU/L) 19 ± 8 83 ± 122 15 ± 5

AP at baseline (IU/L) 77 ± 43 150 ± 173 75 ± 38

T. Bili at baseline (mg/dL) 0.6 ± 0.5 1 ± 2.4 0.6 ± 0.4

Type 2 diabetes (%) 24 29 24

Hypertension (%) 50 48 51

Suspected DILI (%) 221 (0.4%) 136 (2.1%) 160 (0.3%)

P<0.001 P <0.001

Lammert C & Chalasani N. CGH 2019

Frequency of suspected DILI in CLD was ~ 6

fold higher than control group 1 (OR 6.2, 95%

CI: 5-7.7) and control group 2 (OR 6.7, 95% CI:

5-7.7).

The frequency of death or liver transplant

within 6 months of suspected DILI was 20.6%

in CLD cohort and is higher than control group

1 (14.5%, p=0.056) and control group 2 (8.8%,

p=0.003).

-- 24 of 35 --

Other data also suggest an increased risk for

DILI in NAFLD

• A South Korean population-based study suggested the risk for DILI

doubled after a NAFLD diagnosis ( IRR 3.81 within 90 days prior to

NAFLD diagnosis: IRR 7.72 after NAFLD diagnosis.

• An Italian prospective study, NAFLD was significantly associated DILI

(OR 3.95, 95% CI: 1.35-11.48, p=002)

Hwang S, Won S, Lee L. Clin Gastro Hepatol 2021 (Letter)

Tarantino G, et al. Hepatology Research 2007; 37; 410-415

-- 25 of 35 --

Saroglitazar: reports of DILI in PBC but not in

NAFLD/NASH

• Saroglitazar, a combined PPARα/γ agonist, is approved for

dyslipidemia and NASH treatment in India

• No reports of bona fide DILI from saroglitazar in NAFLD or

dyslipidemia

• In Phase2 PBC clinical trial, saroglitazar was associated with DILI

• 3 out of 13 patients on 4 mg dose had DILI – 1 highly likely, 2 probable

• 1 out of 14 patients on 2mg dose had DILI – probable causality

• No jaundice

• Rapidly reversed w/discontinuation

Vuppalanchi R, et al Chalasani N. Journal of Hepatology 2022; Vol 76: 75-85

-- 26 of 35 --

DILI from Saroglitazar in PBC

-- 27 of 35 --

DILI in pre-existing liver disease is associated with worse outcomes

• In the DILIN prospective study, 10% had pre-existing liver disease

Known pre-existing

liver disease

(n=89)

No pre-existing

liver disease

(n=810)

P-value

6-month Outcomes (%)

• All-cause mortality
• Liver-related mortality
• Transplant

16

9.1

3.4

5.2

2.4

4.1

<0.001

0.04

1.0

Chalasani N, et al for the US DILIN. Gastroenterology 2015

Liver disease

n=53

No Liver Ds

n=790

P-value

Liver related death

due to DILI

4 (7.5%) 14 (1.8%) 0.02

Liver

Transplantation

0 13 (1.6%) 0.6

Spanish Registry – Long term outcomes data. J Hepatology 2021

-- 28 of 35 --

Relationship between Charlson Comorbidity Index & six-month

mortality in suspected DILI

Ghabril M, et al, Chalasani N. Gastroenterology 2019; 157: 1245-1252

-- 29 of 35 --

Cholestatic DILI is equally bad as hepatocellular

DILI in patients w/chronic liver disease

• In general, cholestatic DILI has better prognosis and risk of liver

related mortality is lower

• However, in patients with advanced chronic liver disease, especially

cirrhosis, cholestatic liver disease is equally deleterious

Cholestatic

(N=28)

Mixed

(N=13)

Hepatocellular

(N=48)

P-value

All deaths 25% 15.4% 10.4% 0.22

Liver Deaths 10.7% 7.7% 8.3% 0.88

Liver transplants 0 0 6.3% 0.4

Chronic DILI 24% 0 11.9% 0.175

Chalasani N, et al for the US DILIN. Gastroenterology 2015

-- 30 of 35 --

Common Differential Diagnoses

• Viral hepatitis A, B, C, E and CMV/EBV

• Gall bladder & other biliary tract pathology

• Infiltrative & obstructive processes

• Ischemia/sepsis

• Heavy alcohol consumption (with acetaminophen)

• Other medications

• Idiopathic autoimmune hepatitis

• Natural history of the disease

-- 31 of 35 --

A different definition for DILI in cirrhosis?

ALT Baseline Visit 7 Visit 13 Visit 20 Visit 26 Visit 27 EOS

Abnormal levels, n (%) 29 (54) 36 (68) 35 (65) 28 (52) 29 (54) 27 (50) 26 (48)

>3 ULN, n (%) 4 (7) 2 (4) 4 (7) 3 (6) 3 (6) 2 (4) 2 (4)

>5 ULN, n (%) 2 (4) 2 (4) 1 (2) 2 (4) 2 (4) 2 (4) 1 (2)

Alk Phos Baseline Visit 7 Visit 13 Visit 20 Visit 26 Visit 27 EOS

Abnormal levels, n (%) 19 (35) 18 (33) 19 (35) 19 (35) 20 (37) 17 (32) 15 (28)

>1.5 ULN, n (%) 3 (6) 5 (9) 3 (6) 3 (6) 3 (6) 4 (7) 5 (9)

>2 ULN, n (%) 0 (0) 1 (2) 1 (2) 1 (2) 0 (0) 0 (0) 1 (2)

Total Bilirubin Baseline Visit 7 Visit 13 Visit 20 Visit 26 Visit 27 EOS

Abnormal levels, n (%) 7 (13) 10 (19) 10 (19) 10 (19) 7 (13) 9 (17) 10 (19)

>1.5 ULN, n (%) 3 (6) 4 (7) 3 (6) 5 (9) 3 (6) 4 (7) 7 (13)

>2 ULN, n (%) 2 (4) 0 (0) 1 (2) 1 (2) 1 (2) 3 (6) 2 (4)

Natural history of liver biochemistries in NASH Cirrhosis: Placebo arm of

an interventional trial

Shamseddin H, et al. Drug Safety 2020

-- 32 of 35 --

Attempts to avoid DILI are critical in patients with

advanced liver disease

• Patient and provider education

• Avoiding medications & Dietary Supplements with

high DILI potential

Chalasani N, et al for the US DILIN. Gastroenterology 2015

Top 10 therapeutic

classes and individual

agents to cause DILI in

the USA (N=899)

-- 33 of 35 --

Take home messages

• There is a suggestion that NASH is a predisposition for DILI

• DILI outcomes are worse in patients with underlying chronic liver

disease & cirrhosis

• Both hepatocellular & cholestatic forms of DILI may lead to liver

deterioration

• Thorough work up for competing etiologies is essential

• Avoiding potential hepatotoxic medications and herbal agents is

critical

-- 34 of 35 --

@ nagachalasani

-- 35 of 35 --