Who Gets Who Gets AAlcoholic lcoholic LLiver iver

DDisease?isease?

Chris DayChris Day

Newcastle University Newcastle University

Fatty hepatocytes Fatty hepatocytes

Intracellular fat

deposition

Intracellular fat

deposition

Alcoholic Fatty Liver (Steatosis) Alcoholic Fatty Liver (Steatosis)

Fibrosis with necrosis Fibrosis with necrosis

Inflammation Inflammation

Fat deposits Fat deposits

Alcoholic SteatoHepatitis (ASH) Alcoholic SteatoHepatitis (ASH)

Cirrhosis Cirrhosis

Fibrosis Fibrosis

Nodules surrounded by fibrosis

Nodules surrounded by fibrosis

Regenerative nodule Regenerative nodule

CirrhosisCirrhosis

Fatty Liver

(60-100%)

Normal

(0-30%)

Cirrhosis

(<10%)

Steatohepatitis

(20-30%)

Prevalence of disease

stages in unselected

heavy drinkers

WHY??

Dose & pattern of alcohol intake Dose & pattern of alcohol intake

Dietary factorsDietary factors –– Specific nutrients?Specific nutrients?

–– General excess (obesity)General excess (obesity)

Smoking Smoking Becker 2002, Becker 2002, KlatskyKlatsky 20062006

Coffee drinkingCoffee drinking KlatskyKlatsky 20062006

Gut bacteria (the “microbiome”) ? Gut bacteria (the “microbiome”) ?

NonNon--genetic (inherited) factorsgenetic (inherited) factors

Leon et alLeon et al Lancet 2006Lancet 2006

ALD & Dose: population levelALD & Dose: population level

Cirrhosis death rate correlates

with per capita alcohol intake

ALD & Dose: individual levelALD & Dose: individual level

0

1

2

3

4

5

6

7

8

0 0.1-3 3-6 6-9 9-12 >12

Cirrhosis

Any liverdisease

% of

category

Drinks per day

Bellentani et al 1997Bellentani et al 1997

ALD risk and ALD risk and patternpattern of intakeof intake

Risk of ALD increased by:Risk of ALD increased by:

Drinking outside mealDrinking outside meal--timestimes

RiskRisk : 3.4[1.7: 3.4[1.7--6.6] 6.6] BellentaniBellentani 19971997

Drinking beer/spirits rather than wineDrinking beer/spirits rather than wine

RiskRisk : 2.5[1.7: 2.5[1.7--3.3] 3.3] Becker 2002Becker 2002

Daily versus weekend drinking Daily versus weekend drinking

RiskRisk : 2.5 [1.8: 2.5 [1.8--4.6] 4.6]

StokkelandStokkeland 20082008

Johansen, Friis, Skovenborg, Gronbaek BMJ 2006

Is wine Is wine vsvs beer effect due to beer effect due to

confounding “lifestyle” factors?confounding “lifestyle” factors?

Diet and ALD: the role of Diet and ALD: the role of

obesityobesity

0

5

10

15

20

25

30

35

Steatosis Hepatitis Cirrhosis

% o

ve

rw

eig

ht

With lesion

Withoutlesion

1604 heavy drinkers1604 heavy drinkers

Overweight = minimum Overweight = minimum

BMI>25 women, >27 men BMI>25 women, >27 men

in previous 10yrin previous 10yr

In multivariate analysis In multivariate analysis

obesity/glucose best obesity/glucose best

predictors of cirrhosis predictors of cirrhosis

(Risk (Risk > x 2> x 2 for obesity) for obesity)

Naveau et al 1997Naveau et al 1997

Raynard et al 2002Raynard et al 2002

Rates of liver cirrhosis in 1,230,662 UK women

(mean age 56yrs) followed up for mean 6.2 years

Liu Br Med J 2010

Relative risks of contributions of BMI and alcohol to liver

disease mortality (adjusted for all risk factors).

Hart Br Med J 2010

9,559 Scottish men followed up for mean 29 years

Obesity & ALD: ?explanation for Obesity & ALD: ?explanation for

associationassociation

Mechanisms of ALD & obesity related Mechanisms of ALD & obesity related

liver disease (NAFLD) very similar:liver disease (NAFLD) very similar: –– Metabolic overloadMetabolic overload

–– Oxidative, ER and cytokine mediated stressOxidative, ER and cytokine mediated stress

–– The The microbiomemicrobiome//endotoxinendotoxin

Direct Direct fibrogenicfibrogenic (scar producing) (scar producing)

effects of obesityeffects of obesity –– High Insulin/glucoseHigh Insulin/glucose

–– ““AdipokineAdipokine” profile” profile

Risk of ALD and coffee drinkingRisk of ALD and coffee drinking

Coffee intakeCoffee intake Relative Risk of ALDRelative Risk of ALD

Never/seldomNever/seldom 1.01.0

<1<1 0.7 [0.40.7 [0.4--1.1]1.1]

11--33 0.6[0.40.6[0.4--0.8], p<0.0010.8], p<0.001

≥4≥4 0.2[0.10.2[0.1--0.4], p<0.0010.4], p<0.001

Per cup/dayPer cup/day 0.8[0.70.8[0.7--0.9], p<0.0010.9], p<0.001

Adjusted for age, weight, smoking, gender, ethnicity

Klatsky 2006

Gut bacteria Gut bacteria -- the “microbiome” the “microbiome”

-- and ALD susceptibility?and ALD susceptibility?

Abundant evidence that the gut Abundant evidence that the gut

microbiomemicrobiome may be involved in ALD may be involved in ALD

pathogenesis pathogenesis PurohitPurohit 20082008

NonNon--alcoholic steatohepatitis (NASH) as alcoholic steatohepatitis (NASH) as

a transmissible disease ? a transmissible disease ? HenaoHenao--Mejia Nature 2012Mejia Nature 2012

What about ALD?What about ALD?

–– Faeces from ALD patients induce ALD in Faeces from ALD patients induce ALD in

alcohol fed mice alcohol fed mice LlopisLlopis 20132013

Tilg & Day Nat Clin Practice 2007

Gender and ALDGender and ALD

Females develop ALD at “lower” intakeFemales develop ALD at “lower” intake

Explanation: Explanation: blood alcohol levels for same blood alcohol levels for same dose per body weightdose per body weight VolVol distribution distribution Marshall 1983Marshall 1983

EstrogenEstrogen sensitises the liver (sensitises the liver (KupfferKupffer cells) to cells) to bacterial (bacterial (endoendo) toxins) toxins –– gut permeability to gut permeability to endotoxinendotoxin

–– Endotoxin receptors (CD14) & release of Endotoxin receptors (CD14) & release of inflammatory mediators from liver inflammatory inflammatory mediators from liver inflammatory cells (cells (egeg. . TNFTNFaa)) Thurman et al 1999Thurman et al 1999

Tilg & Day Nat Clin Practice 2007

Are genes importantAre genes important??

Concordance rates (%) among 15 924 male twin

pairs age 67-77. Alcoholism and cirrhosis ↑ 25%.

Alcoholism genetic

ALD genetic

Risk “over-and-above” the risk of

alcoholism

Hrubec 1981

MZ DZ

Alcoholism 26.3 11.9

Cirrhosis 14.6 5.4

MZ DZ

Alcoholism 26.7 12.2

Cirrhosis 16.9 5.3

1981 1996

Alcoholism genetic

ALD risk now shared with alcoholism

No longer a significant independent

genetic risk for ALD

Reed 1996

Ethnic variation in susceptibility Ethnic variation in susceptibility

Caetano & Clark 1998, Stinson 2001, Caetano & Clark 1998, Stinson 2001,

KlatskyKlatsky 2006 2006

Finding the genesFinding the genes

HypothesisHypothesis--drivendriven

–– Candidate gene caseCandidate gene case--control studiescontrol studies All studies in ALD thus farAll studies in ALD thus far

Almost all underpowered (too small)Almost all underpowered (too small)

HypothesisHypothesis--free (generating)free (generating)

–– “Whole” genome scanning “Whole” genome scanning None in ALD as yet (underway) None in ALD as yet (underway) –– BUT:BUT:

Clues from studies in NAFLDClues from studies in NAFLD

• USA Study population: (Black n=1032, White n=636, Hispanic n=383).

• 9229 non-synonymous exonic SNPs.

• Steatosis defined non-invasively by 1H-MRS.

Figure 2.

Association between a sequence variant in PNPLA3 (I148M) and hepatic triglyceride (TG)

content. (a) PNPLA3 is a 481 residue protein that contains a patatin-like domain at the N

terminus (residues 10-179, UniProt http://pir.uniprot.org/uniprot/ ) with consensus sequences

for a Ser-Asp catalytic dyad (Gly-X-Ser-X-Gly and Asp-X-Gly/Ala)14. The I148M substitution

(rs738409) is located between the consensus sequences for the catalytic dyad and is highly

conserved. (b) Median hepatic TG content, body mass index (BMI), homeostatic model

assessment of insulin resistance (HOMA-IR) and plasma TG levels in the Dallas Heart Study.

Values for hepatic fat content were compared using ANOVA. Age, gender, BMI, diabetes

status, ethanol use and global ancestry were included as covariates in the model. ( c) Median

Romeo et al. Page 9

Nat Genet. Author manuscript; available in PMC 2009 June 1.

NIH

-PA

Auth

or M

anu

scrip

tN

IH-P

A A

uth

or M

an

uscrip

tN

IH-P

A A

uth

or M

an

uscrip

t

Figure 2.

Association between a sequence variant in PNPLA3 (I148M) and hepatic triglyceride (TG)

content. (a) PNPLA3 is a 481 residue protein that contains a patatin-like domain at the N

terminus (residues 10-179, UniProt http://pir.uniprot.org/uniprot/ ) with consensus sequences

for a Ser-Asp catalytic dyad (Gly-X-Ser-X-Gly and Asp-X-Gly/Ala)14. The I148M substitution

(rs738409) is located between the consensus sequences for the catalytic dyad and is highly

conserved. (b) Median hepatic TG content, body mass index (BMI), homeostatic model

assessment of insulin resistance (HOMA-IR) and plasma TG levels in the Dallas Heart Study.

Values for hepatic fat content were compared using ANOVA. Age, gender, BMI, diabetes

status, ethanol use and global ancestry were included as covariates in the model. ( c) Median

Romeo et al. Page 9

Nat Genet. Author manuscript; available in PMC 2009 June 1.

NIH

-PA

Auth

or M

anu

scrip

tN

IH-P

A A

uth

or M

an

uscrip

tN

IH-P

A A

uth

or M

an

uscrip

t

PNPLA3 (I148M) – Carriage of G allele associated with increased steatosis (p=5.9 x 10-10)

Presence of NASH

OR 1.5 (1.12-2.04)

Fibrosis >F1

OR 1.5 (1.09-2.12)

N=591

PNPLA3 PNPLA3 -- AdiponutrinAdiponutrin

Anstee et al, 2013

He et al, 2010 Huang et al, 2010 Huang et al, 2011 Pirazzi et al, 2012

But why the link with NASH & scarring?

Case control studies of PNPLA3 Case control studies of PNPLA3

in ALD in ALD

Study Numbers Ethnicity Associated

phenotype

Tian, 2010 1,221 Mexican-

European

Risk of ALD and

cirrhosis

Seth, 2010 548 UK Risk of ALD and

cirrhosis

Stickel , 2010 1,043/376 German Risk of ALD and

cirrhosis

Trepo, 2011 658 Belgian Risk of ALD and

cirrhosis

Trepo 2012 571 Belgian and

French

Risk of HCC

Nischalke 2011 160 Germany Risk of HCC

TM6SF2 (E167K) – Carriage of minor allele associated with increased steatosis (p=5.7 x 10-8)

See also: Holmen et al, Nature Genetics, 2014

– Carriage of major allele increased cholesterol and cardiovascular disease

Multivariate analysis (additive model) incorporating Age, Gender, BMI, T2DM and PNPLA3 rs738409 genotype.

i.e. OR per copy of the minor allele carried

Nature Communications, 2014; 10.1038/ncomms5309.

↑ IHTG

↑Steatosis

↑ Fibrosis

? ↑ NASH ↑ Cardiovascular Disease

↓ LDL-C ↓ TG

T-allele Lys167 (E167K)

↓ Hepatic VLDL Excretion

C-allele Glu167 (E167K)

↑ Total Chol ↑ LDL ↑ TG

↑ Hepatic VLDL Excretion

↑ Intrahepatic Fatty Acid Flux

Obesity Alcohol T2DM

TM6SF2 rs58542926

Does TM6SF2 genotype determine the CAUSE of death in

NAFLD/ALD?

TM6SF2 gene variant determines TM6SF2 gene variant determines

risk of liver and heart disease risk of liver and heart disease

Dongiovanni 2014

ConclusionsConclusions

ALD is a typical “complex” disease with susceptibility ALD is a typical “complex” disease with susceptibility likely due to a combination of genetic/environmental likely due to a combination of genetic/environmental factorsfactors

Several modifiable environmental risk factors such Several modifiable environmental risk factors such as diet/coffee drinking identifiedas diet/coffee drinking identified

Role of the Role of the microbiomemicrobiome as a modifiable and as a modifiable and ?transmissible risk factor worthy of further study?transmissible risk factor worthy of further study

PNPLA3 is the only robust genetic association PNPLA3 is the only robust genetic association reported thus far and is also associated with HCCreported thus far and is also associated with HCC

Genetic factors may explain CAUSE of death in Genetic factors may explain CAUSE of death in heavy drinkersheavy drinkers

GGenome wide scans are awaited with interestenome wide scans are awaited with interest