“Portal hypertension – Update 2018”

Jaime Bosch

Liver Unit, Hospital Clinic-IDIBAPS, University of Barcelona and CIBERehd, Barcelona, Spain

Hepatology, University Clinic for Visceral Surgery and Medicine, Inselspital, Bern University, Switzerland

Disclosures Consultant/Advisory Board Conatus, Exalenz, Actelion,

BioVie, Brudy, BLB

Symposium Gastroenterology / Hepatology

Sanct Gallen 2018 Science And Novel Clinical Tools in GAstroenteroLogy, Liver diseases and ENdoscopy

Klinik für Gastroenterologie/Hepatologie November 15, 2018

Outline

- Relevance - Stages of cirrhosis

- Rationale basis of therapy

- Treatment of compensated patients

- Preventing first decompensation

- Preventing first bleeding and other complications

- Treatment in decompensated patients

- Preventing further decompensation and death

- Treatment of acute variceal bleeding

Cirrhosis burden in Europe • 29 million subjects with liver diseases in EU (alcohol, HCV, HBV, metabolic syndrome) • Cirrhosis: 170,000 deaths per year • HCC: 47,000 deaths per year • 5,500 liver transplants/year

Higher than breast cancer Blachier et al. J Hepatol 2013

Variceal bleeding

G-E varices Ascites

Portal Hypertension

Bosch et al, Nature Rev Gastro Hepatol 2009

normal range subclinical PH

Clinically significant portal

hypertension CSPH

ascites

encephalopathy

varices

SBP

10

0

mmHg HVPG

bleeding

5

The Portal Hypertension Fever

Concept (1) The complications of portal

hypertension can be prevented or reversed by preventing the increase

or by decreasing the HVPG

Concept (2) Treatments aimed at correcting

portal hypertension have a greater potential of changing the natural

history and improving survival than local treatments

ARCHITECTURAL LIVER DISRUPTION IS THE MAIN MECHANISM THAT LEADS TO AN INCREASED INTRAHEPATIC RESISTANCE

Clinically-significant portal hypertension is responsible for complications of cirrhosis that define decompensation

Cirrhotic Liver

Clinically significant PORTAL HYPERTENSION

Varices Variceal hemorrhage

Ascites Encephalopathy

pressure

Portal vein

(-1.8 mmHg)

Decrease in HVPG at SVR-24 in 226 patients with compensated cirrhosis treated with DAA

HVPG (mmHg) % above threshold values

P< 0.05

• Clinically significant portal hypertension may persist despite elimination of the cause of cirrhosis

• There is no etiological therapy for NASH

Effects of an intensive lifestyle intervention program on HVPG in patients with cirrhosis and obesity

The Sport-Diet study

Hepatology Volume 65, Issue 4, pages 1293-1305, 6 MAR 2017 DOI: 10.1002/hep.28992 http://onlinelibrary.wiley.com/doi/10.1002/hep.28992/full#hep28992-fig-0002

HVPG (mmHg)

Berzigotti et al, Hepatology 2017 Effects mediated by inhibition of pro-inflammatory cytokines and of bacterial translocation

P<0.0001

Baseline HVPG

16-wk HVPG

D= - 1.7 mmHg; - 10.7%

0

10

20

30

40

24

1816

32

46

%

+

Primary and Secondary Prophylaxis of Variceal Bleeding Current Guidelines

Prevention of First Variceal Bleeding

NSBB (non-selective beta-blocker)*

or

Endoscopic band ligation (EBL)

Prevention of Recurrent Variceal Bleeding

NSBB (non-selective beta-blocker)

plus

Endoscopic band ligation (EBL)

De Franchis et al Baveno VI, J Hepatol 2015

*Only recommended therapy for high-risk small varices

“There is no indication for treating patients without high-risk varices”

Pathophysiology of Portal Hypertension

Role of hemodynamic factors

Increased Blood Flow

(splanchnic vasodilation)

Increased Portal Pressure

Resistance (dynamic / structural)

Increased

D Pressure = Resistance x Blood Flow

Ascites Bleeding PSE

Formation of Collaterals & Varices

HVPG > 10 mmHg

Initial factor Late factor

Current treatments

First reports on the use of NSBB to prevent variceal bleeding in cirrhosis

Still valid in 2018 !

Carvedilol, a new NSBB with anti-a1 adrenergic activity has a greater effect decreasing HVPG

Sinagra et al, AP & T 2014

- Overall “good HVPG response”: 55-75% (Vs 30-50% with NSBBs)

- Easier to adjust dose - Better tolerated

Recommended dose: 6.25 – 12.5 mg/day

New!

New attempt at early therapy: The PREDESCI Study

PREventing the DEcompenSation of Cirrhosis with non-selective beta-blockers

• Cooperative, multicenter, placebo-controlled, randomized clinical trial

• Population studied: compensated cirrhotics with HVPG ≥ 10 mmHg (CSPH), without varices requiring treatment or previous decompensation (n=210)

Acute HVPG response to iv Propranolol*:

acute responders Propranolol non-responders Carvedilol vs placebo vs placebo

* 0.15 mg/Kg IV; Acute Responders: HVPG ≥ 10% of baseline

Villanueva C…Bosch J. Lancet (in press) ¿?=)

8

New!

Propranolol/Carvedilol (according to HVPG response) prevents decompensation of cirrhosis: The PREDESCI Study

First clinical decompensation

New!

Villanueva C…Bosch J. Lancet (in press)

Ascites: HR 0.44 (0.20-0.97), p=0.037

Primary and Secondary Prophylaxis of Variceal Bleeding Current Guidelines

Prevention of First Variceal Bleeding

NSBB (non-selective beta-blocker)*

or

Endoscopic band ligation (EBL)

Prevention of Recurrent Variceal Bleeding

NSBB (non-selective beta-blocker)

plus

Endoscopic band ligation (EBL)

De Franchis et al Baveno VI, J Hepatol 2015

*Only recommended therapy for high-risk small varices

“There is no indication for treating patients without high-risk varices”

% Rebleeding

0

20

40

60

80

100

HVPG decrease > 20% from

baseline

HVPG decrease to < 12 mmHg

~0%

46-65%

7-13%

No change in HVPG

Modified from Bosch and García-Pagán, Lancet 2003 361:952

Efficacy of propranolol mainly depends on the decrease in portal pressure (HVPG)

months

967248240

Cu

mu

lative

pro

ba

bility

1,0

,8

,6

,4

,2

0,0

p=0.013

Responders

non-responders

months

967248240

cu

mu

lative

su

rviv

al

1,0

,8

,6

,4

,2

0,0

p=0.003

Responders

non-responders

Variceal bleeding Survival

months

967248240

cu

mu

lative

pro

ba

biliy

1,0

,8

,6

,4

,2

0,0

p=0.025

Responders

non-responders

Ascites

Abraldes….Bosch, Hepatology 2003

HVPG responders to NSBB for preventing variceal rebleeding also have a lower incidence of ascites

and improved survival

Endoscopic band ligation of esophageal varices

Banded varix

- Repeat sessions of band ligation (mean 3 sessions, every 2-4 weeks), 4-8 bands per session

- Relatively safe severe complications: 8% bleeding, 20% ulcers, 20% dysphagia (transient)

- Variceal recurrence: 60-90% at 1-year Need of long-term endoscopic surveillance

Beta-blockers are of key importance in the prevention of variceal rebleeding: Individual patient meta‐analysis

Survival analysis (Kaplan Meyer)

Albillos et al, Hepatology 2017

Take home message: Don’t use EVL alone in Child B and C patients!

New!

Statins: protecting the heart only?

p=0.583

Rebleeding

Simvastatin associated with NSBB + EVL improves prognosis after variceal bleeding (BLEPS Study)

Outcomes

Simvastatin

Placebo

Addition of simvastatin to standard therapy did not improve there was a marked survival benefit with simvastatin

p=0.030

Survival*

Simvastatin

Placebo

HR: 0.387 (0.152 to 0.986)

* less deaths due to bleeding and infections

Abraldes, Villanueva,… Bosch. Gastroenterology 2016

Simvastatin prevents ACLF in cirrhotic rats (Tripathi et al, Gastroenterology 2018) Simvastatin prtects the liver from LPS induced injury (LaMura et al, Hepatology 2013) Simvastation protects the cirrhotic liver from acute bleeding (Meireles et al, Shock 2016)

New!

Statins are associated with a decreased risk of decompensation and death in compensated HCV

cirrhosis*

P value: <0.001

P value: <0.001

No. at risk

User 685 386 154 48 13

Nonuser 2062 924 333 92 22

No. at risk

User 685 399 165 53 17

Nonuser 2062 991 370 107 27

Statin user

Statin user Non-user

Non-user

Decompensation Death

HR 0.55 (0.39, 0.77) Ascites: HR 0.59; p=0.02

Bleeding: HR 0.39; p=0.01 HR 0.56 (0.46 - 0.69)

Mohanty, et al. Gastroenterology 2016 *Propensity score matched study

New!

Failure of combination pharmacological/endoscopic

therapy

EBL continue drug for 5 days

Rescue TIPS, discontinue drug

Endoscopy within 12 hours esophageal variceal bleeding

Modified from García-Tsao G, Bosch J. N Eng J Med 2010

S–B tube/ Esophageal Stent if necessary

Safe vasoactive drug (terlipressin, SMT, octreotide, 2–5 days), Antibiotic prophylaxis (5–7 days)

Cautious volume resuscitation (aim at Hb 7-8 g/dl) +

Current management of acute variceal bleeding

* Child-Pugh C (<14 points) and Child-Pugh B + active bleeding

High risk* early TIPS (<72h)

Low risk continue on treatment New!

Transjugular Intrahepatic Portosystemic Shunt

A non-surgical calibrated shunt, very effective in decreasing portal pressure. Lower morbidity and mortality than surgical shunts, feasible in poor

surgical candidates (Child–Pugh C)

TIPS (Transjugular Intrahepatic Portal-systemic Shunt)

TIPS as rescue therapy in acute variceal hemorrhage

n % of total

Child C Control of bleeding

Mortality (4-6

weeks) McCormick-1994 20 8% 60% 100% 60%

Jalan-1995 19 16% 84% 100% 42%

Sanyal-1996 30 23% 73% 100% 40%

Bañares-1998 57 16% 41% 95% 28%

Azoulay-2001 58 16% “majority” 90% 29%

“Treatment was a success but the patient died”

Success rate vs Funeral rate

Early TIPS improved outcomes in high-risk patients with variceal hemorrhage

• 63 patients, Child–Pugh C (score < 14), or Child–Pugh B with active bleeding

• TIPS within 72 hours of admission vs. current recommended therapy (vasoactive drugs plus endoscopic band ligation plus antibiotics)

García-Pagán … Bosch. N Engl J Med. 2010;362:2370–9.

Primary End-Point: failure to control bleeding or rebleeding

100

80

60

40

20

0

p < 0.001 (log-rank)

Early PTFE-TIPS

Standard*

86%

60%

96%

67%

6 weeks 1 year

18 12 6 0 24

TIPS is first-choice therapy in patients Child–Pugh C (<14 points)

Pre-emptive TIPS patients had better survival than those on standard therapy

*7 patients had TIPS placed as rescue therapy, 4 (57%) died

Time to death (months)

Findings confirmed by

two larger observational

studies

Time (months)

García-Pagán … Bosch. N Engl J Med. 2010;362:2370–9.

• Coated self-expanding metal stents

• Special introducer allowing placement without endoscopic control

• Safely removed endoscopically

• Higly effective in uncontrolled series

Expandable esophageal stents a new and safer way of tamponade

Zehetner J, et al. Surg Endosc. 2008;22:2149–52. Hubmann R, et al. Endoscopy. 2006;38:896–901.

Results of a RCT confirm that ES is better and safer than

tamponade

Escorsell et al, Hepatology 2016

New! Balloon Tamponade Vs Esophageal Stent in Refractory Variceal Bleeding

Probability of being free of failure*

0

0.2

0.4

0.6

0.8

Balloon Tamponade 13%

Esophageal Stent 46%

0 1 3 4 5 15 2 Days from inclusion

p=0.04

* Absence of death, of bleeding and of severe complications at day 15

Escorsell… Bosch, Hepatology 2016

Efficacy and safety of macitentan in patients with portopulmonary hypertension: the randomized, placebo controlled PORTICO trial

Michael Krowka1, Emmanuelle Cottreel2, Pascal de Groote3, Marius M Hoeper4, Nick Kim5, Nicolas Martin2, Laurent Savale6, Oliver Sitbon6, Jaume Bosch7

1Mayo Clinic, Rochester, MN, USA. 2Actelion Pharmaceuticals Ltd, Allschwil, Switzerland. 3Service de cardiologie, Institut Coeur-Poumon, CHRU Lille – INSERM U1167, Institut Pasteur de Lille, F - 59000 Lille, France. 4Hannover Medical School and German Center of Lung Research, Hannover, Germany. 5University of California San Diego, La Jolla, CA, USA. 6 APHP, Hôpital Bicêtre, INSERM UMR_S999, Université Paris-Sud, Université Paris-Saclay, Le Kremlin-Bicêtre, France. 7 Hospital Clinic-IDIBAPs and CIBERehd, University of Barcelona, Spain, and Swiss Liver, Inselspital, Berne University, Switzerland

Study sponsor: Actelion Pharmaceuticals Ltd.

New! AASLD 2018

The PORTICO study

Randomized, double-blind, placebo-controlled, prospective, multicenter study to assess the efficacy and safety of macitentan in patients with PoPH (NCT02382016)

Week 12 Week 24 Randomization

Screening 28 days

Double-blind period Open-label period

Macitentan 10 mg

Safety follow-up 30 days

Macitentan 10 mg

Placebo

Primary endpoint – PVR at Week 12 expressed as ratio of baseline

35% reduction in PVR for macitentan vs. placebo

Model-adjusted* ratio of geometric means (95% CI) for macitentan:placebo 0.65 (0.59, 0.72); p<0.0001

Horizontal lines in figure are geometric mean PVR ratio Week 12:baseline for each treatment; imputed values shown in red. *ANCOVA model on loge-transformed ratio of baseline PVR with terms for treatment, region and background PAH therapy and with loge-transformed PVR at baseline as a covariate.

Acknowledgements

www.swissliver.ch

Collaborations Spanish Cooperative group Baveno Cooperation Dr Garcia-Tsao, Yale University Dr Juan Abraldes, Edmonton Dr A.Luca & G.Pietrosi, ISMETT

G Stirnimann A De Gottardi A Berzigotti J-F Dufour S Casu N Semmo