How do you distinguish compensated from decompensated cirrhosis?

Compensated cirrhosis presents without ascites, hepatic encephalopathy, variceal hemorrhage, or jaundice, typically corresponding to Child-Pugh A or MELD <12. Decompensation is defined by the first occurrence of any of these four events. Five-year survival drops from over 85% in compensated cirrhosis to roughly 14-35% after decompensation, making this boundary decisive for both drug labeling and trial endpoint selection.

Why does CSPH use HVPG ≥10 mmHg as the diagnostic threshold?

Above HVPG 10 mmHg, the risk of varices, ascites, and hepatic encephalopathy rises sharply, which is why this threshold defines clinically significant portal hypertension. Baveno VII (2022) added non-invasive criteria: liver stiffness ≥25 kPa rules CSPH in (PPV >90%), while LSM ≤15 kPa with platelets ≥150 G/L rules it out, sparing many patients from invasive HVPG measurement.

What changed between MELD-Na and MELD 3.0?

MELD 3.0 (OPTN Board action in June 2022; allocation implementation on July 13, 2023) added a female sex coefficient and serum albumin term, lowered the creatinine ceiling to 3 mg/dL, and introduced bilirubin-sodium and creatinine-albumin interaction terms. The c-statistic improved to 0.869 versus 0.862 for MELD-Na, reclassifying 8.8% of waitlist deaths into higher urgency tiers and narrowing the female mortality disparity.

Why is Resmetirom not approved for F4 cirrhotic patients?

Rezdiffra labeling restricts use to noncirrhotic MASH with F2-F3 fibrosis, with avoidance in decompensated cirrhosis listed as a Limitation of Use. The MAESTRO-NASH Phase 3 (NEJM 2024) enrolled F1B/F2/F3 only, so no F4 efficacy or safety data exist. F4 efficacy in well-compensated NASH cirrhosis is being tested in the ongoing MAESTRO-NASH-OUTCOMES trial (NCT05500222, n=700, active, not recruiting).

How do MASH trials measure decompensation prevention?

MAESTRO-NASH-OUTCOMES uses a composite primary endpoint: liver-related and cardiovascular death, liver transplant, hepatic decompensation events (ascites, encephalopathy, variceal bleeding), and confirmed MELD increase from <12 to ≥15. ESSENCE Part 2 (week 240) tracks progression to cirrhosis, liver failure, and HCC. F2-F3 trials use histologic fibrosis improvement; F4 trials measure these clinical events.

Compensated vs Decompensated Cirrhosis: MASH Endpoint Guide

Q: Why does CSPH use HVPG ≥10 mmHg as the diagnostic threshold?

Above HVPG 10 mmHg, the risk of varices, ascites, and hepatic encephalopathy rises sharply, which is why this threshold defines clinically significant portal hypertension. Baveno VII (2022) added non-invasive criteria: liver stiffness ≥25 kPa rules CSPH in (PPV >90%), while LSM ≤15 kPa with platelets ≥150 G/L rules it out, sparing many patients from invasive HVPG measurement.

The compensated/decompensated boundary is the central strategic axis in MASH drug development

Patient selection, primary endpoints, and label language for every MASH Phase 3 program orbit a single biological boundary: the transition from compensated to decompensated cirrhosis. Resmetirom's F2-F3 label, semaglutide/Wegovy receiving FDA accelerated approval in August 2025 for noncirrhotic MASH with F2-F3 fibrosis based on ESSENCE Part 1 week 72 histology, and Resmetirom's separate long-term outcomes trial (MAESTRO-NASH-OUTCOMES) measuring clinical event suppression in F4 all derive from this boundary. Understanding it is non-negotiable for endpoint design.

This guide aligns three quantitative tools — HVPG, MELD 3.0, and Child-Pugh — with the latest Baveno VII (2022) and AASLD Practice Guidance (2024) frameworks, then maps the logic onto current Phase 3 endpoint selection. For complementary context, see our coverage of the MASLD/MAFLD nomenclature update, PRO-C3 and ECM turnover biomarkers, the ELF score, and the comprehensive MASLD/MASH biomarker review.

1. Why decompensation is the kingpin of MASH Phase 3 strategy

The natural history of cirrhosis follows a sharply non-linear trajectory. Annual mortality in compensated cirrhosis sits at 1-3%, but the first decompensation event (ascites, hepatic encephalopathy, variceal hemorrhage, or jaundice) drives annual mortality to 20-57%. Five-year survival collapses from above 85% in compensated patients to roughly 14-35% after decompensation³.

This non-linearity bifurcates MASH drug strategy.

F2-F3 (compensated, including cACLD): histologic fibrosis improvement (≥1 stage) and MASH resolution serve as the registrational endpoints. Resmetirom (MAESTRO-NASH¹) and semaglutide (ESSENCE²) follow this template.
F4 (well-compensated cirrhosis): long-term clinical event suppression — decompensation prevention, MELD progression, and transplant-free survival — replaces histology. MAESTRO-NASH-OUTCOMES (NCT05500222) is the prototype.

F4 trials therefore measure clinically meaningful disease progression, not fibrosis regression. This is why "F4 readiness" has become the central differentiation axis for the next wave of MASH compounds. See the MASH combination therapy pipeline, liver antifibrotic landscape 2026, MASH landscape 2025, and fibrosis market size 2026 for sponsor-level positioning.

For researchers tracking fibrosis & inflammation R&D

FDA approval alerts, trial readouts, preclinical model selection, and assay optimization — curated signal for bench-to-pipeline readers. 2 emails/month max.

2. Clinical definition: compensated vs decompensated

The four decompensating events

Event	Pathophysiology	Bedside assessment
Ascites	Portal hypertension + hypoalbuminemia	SAAG ≥1.1 g/dL, diagnostic paracentesis
Hepatic encephalopathy (HE)	Ammonia and neurotoxin accumulation	West Haven grading, CHE/OHE distinction
Variceal hemorrhage	Esophageal/gastric variceal rupture	Endoscopic F1-F3 screening
Jaundice	Persistent bilirubin elevation	Total bilirubin ≥3 mg/dL (sustained)

The first incident of any of these reclassifies a patient as decompensated, typically pushing Child-Pugh into B/C and MELD above 15.

The cACLD framework

AASLD 2024 Practice Guidance³ formalized compensated advanced chronic liver disease (cACLD) — patients near or at cirrhosis identified by liver stiffness measurement (LSM) and platelet count, without mandatory biopsy. The guidance positions CSPH as the dominant predictor of decompensation risk, expanding the F4 population from biopsy-confirmed METAVIR/Ishak stages to non-invasively estimated cohorts. Staging fundamentals are covered in our METAVIR vs Ishak comparison, and upstream biology in the TGF-β/Smad pathway explainer.

3. HVPG and CSPH: the Baveno VII update

HVPG threshold hierarchy

The hepatic venous pressure gradient (HVPG) is the standard surrogate for portal pressure, with a well-established threshold ladder⁴.

HVPG	Clinical state
<5 mmHg	Normal
5-9 mmHg	Subclinical portal hypertension
≥10 mmHg	CSPH (clinically significant portal hypertension)
≥12 mmHg	Variceal bleeding risk
≥16 mmHg	Increased mortality
≥20 mmHg	Difficult hemostasis during acute bleeding

CSPH at ≥10 mmHg is the inflection point at which varices, ascites, and encephalopathy risk rises steeply.

Baveno VII non-invasive CSPH criteria

Because HVPG measurement is invasive and operator-dependent, Baveno VII⁴ introduced non-invasive surrogates:

CSPH rule out: LSM ≤15 kPa AND platelets ≥150 × 10⁹/L
CSPH rule in: LSM ≥25 kPa (PPV >90% in viral/alcohol-related cACLD and non-obese NASH/MASH; PPV drops to ~63% in obese MASLD/MASH, so use ANTICIPATE±NASH ≥75% as an adjunct)
Grey zone: LSM 15-25 kPa or platelets <150 → spleen stiffness or the ANTICIPATE±NASH model for further refinement

These criteria are now standard stratification variables for MASH Phase 3 enrollment. AASLD 2024³ endorses early carvedilol — the preferred non-selective beta-blocker (NSBB) — once CSPH is confirmed, framing primary decompensation prevention as a clinical management goal.

4. MELD 3.0 vs Child-Pugh: which to use when

Child-Pugh classification (1973 origin)

The Child-Pugh score combines ascites, encephalopathy, bilirubin, albumin, and PT/INR into A/B/C tiers. It remains the standard for surgical risk assessment and MASH trial inclusion criteria because it is simple, outpatient-friendly, and embedded in decades of historical data.

Child-Pugh	Score	1-year survival
A	5-6	~95%
B	7-9	~80%
C	10-15	~45%

MAESTRO-NASH-OUTCOMES restricts "well-compensated cirrhosis" to Child-Pugh A.

MELD 3.0 (OPTN Board action June 2022; allocation implementation July 13, 2023)

MELD 3.0⁵ extends MELD-Na by adding a female sex coefficient (+1.33) and a serum albumin term, lowering the creatinine ceiling to 3 mg/dL, and introducing bilirubin-sodium and creatinine-albumin interaction terms. The c-statistic improved to 0.869 (vs 0.862 for MELD-Na). Implementation reclassified 8.8% of waitlist deaths — predominantly women — into higher urgency strata, narrowing the female mortality gap.

Trial design: when to choose which

Application	Preferred tool	Rationale
Enrollment criteria (cACLD/F4)	Child-Pugh A	Simple, historically grounded
Severity stratification	MELD 3.0	Continuous variable, best prognostic discrimination
Endpoint event (worsening)	MELD <12 → ≥15 transition	Aligns with transplant listing threshold
Allocation	MELD 3.0	OPTN allocation standard since July 13, 2023

The "MELD <12 → ≥15" component of the MAESTRO-NASH-OUTCOMES composite endpoint matters because MELD 15 is the historical transplant listing threshold, making it a clinically meaningful surrogate for hepatic functional deterioration.

5. MAESTRO-NASH-OUTCOMES and ESSENCE: data on either side of the boundary

MAESTRO-NASH (F1B-F3, the registrational data)

The MAESTRO-NASH Phase 3¹ randomized 966 F1B/F2/F3 patients. The 100 mg arm achieved 29.9% MASH resolution and 25.9% one-stage fibrosis improvement (versus placebo at 9.7% and 14.2%), supporting accelerated FDA approval in March 2024. F4 is excluded from the label: Rezdiffra is restricted to "noncirrhotic MASH with moderate-to-advanced liver fibrosis (consistent with stages F2 to F3)" with avoidance in decompensated cirrhosis as a Limitation of Use⁶. See the Resmetirom approval analysis and the differentiated mechanism profiles of efruxifermin (FGF21) and pegozafermin (FGF21) for competitive context.

MAESTRO-NASH-OUTCOMES (closing the F4 evidence gap)

NCT05500222 enrolls 700 patients with well-compensated NASH cirrhosis, randomized 3:1 to resmetirom 80 mg vs placebo for approximately three years. The composite primary endpoint:

Liver-related and cardiovascular mortality
Liver transplant
Hepatic decompensation events (ascites, hepatic encephalopathy, variceal bleeding)
Confirmed MELD increase from <12 to ≥15

Unlike F2-F3 trials, the trial measures clinically meaningful event suppression rather than histologic regression. As of May 2026, ClinicalTrials.gov lists the trial as ACTIVE_NOT_RECRUITING (n=700 estimated, ~3 years), and interim data have not been disclosed.

ESSENCE (F2-F3, Part 1 interim 2025)

NCT04822181 randomized 1197 F2/F3 MASH patients 2:1 to subcutaneous semaglutide 2.4 mg weekly vs placebo for 240 weeks². The week 72 interim analysis (first 800 patients, NEJM 2025) reported:

MASH resolution without fibrosis worsening: 62.9% vs 34.3%
Fibrosis improvement without MASH worsening: 36.8% vs 22.4%
Combined endpoint: 32.7% vs 16.1%
Mean weight change: -10.5% vs -2.0%

ESSENCE Part 2 (week 240) will report long-term progression to cirrhosis, liver failure, and HCC. F4 is excluded. On August 15, 2025, the FDA granted Wegovy accelerated approval for noncirrhotic MASH with F2-F3 fibrosis (NDA 215256/s-024) based on this week 72 interim histology⁸, with Part 2 confirming clinical outcomes. Metabolic-pathway approaches in F4 will require dedicated Phase 3 designs.

Implications for F4 trial design

F4 outcomes trials are large (700-1500 patients), long (3-5 years of follow-up), and 2-3× more expensive than F2-F3 histologic trials. The reward — F4 labeling, premium pricing, and broad community uptake — is large enough that follower sponsors (Akero/efruxifermin, 89bio/pegozafermin, and others) are accelerating dedicated F4 Phase 3 programs.

References

1. Harrison SA, et al. A Phase 3, Randomized, Controlled Trial of Resmetirom in NASH with Liver Fibrosis (MAESTRO-NASH). N Engl J Med. 2024;390(6):497-509. PubMed / NCT03900429

2. Sanyal AJ, et al. Phase 3 Trial of Semaglutide in Metabolic Dysfunction-Associated Steatohepatitis (ESSENCE Part 1). N Engl J Med. 2025;392(21):2089-2099. PubMed / NCT04822181

3. Kaplan DE, Ripoll C, et al. AASLD Practice Guidance on risk stratification and management of portal hypertension and varices in cirrhosis. Hepatology. 2024;79(5):1180-1211. PubMed

4. de Franchis R, Bosch J, Garcia-Tsao G, et al. Baveno VII - Renewing consensus in portal hypertension. J Hepatol. 2022;76(4):959-974. PubMed

5. Kim WR, Mannalithara A, Heimbach JK, Kamath PS, et al. MELD 3.0: The Model for End-Stage Liver Disease Updated for the Modern Era. Gastroenterology. 2021;161(6):1887-1895.e4. PubMed

6. FDA. Rezdiffra (resmetirom) Prescribing Information. 2024 (current label revised 2025). FDA Label 2025

7. Madrigal Pharmaceuticals. MAESTRO-NASH-OUTCOMES Trial Initiation Press Release. 2022. NCT05500222

8. FDA. FDA Approves Treatment for Serious Liver Disease Known as MASH (Wegovy/semaglutide accelerated approval). 2025-08-15. FDA Press / Wegovy Label NDA 215256/s-024