Abstract
Background:
Recognition and management of metabolic dysfunction–associated steatohepatitis (MASH) are hindered by variability in diagnostic practices and limited real-world data on disease progression. This study assessed the use of non-invasive testing (NIT) in MASH identification and estimated incidence rates and time-to-disease progression.
Methods:
Adults in the TARGET-NASH cohort (enrolled 2016–2023) were included. MASH was confirmed via medical records, NITs, pathology reports, and imaging. Patients with MASH and significant/advanced fibrosis/cirrhosis (F2/F3/F4) were included. Outcomes included progression to cirrhosis, liver-related complications, and all-cause mortality.
Results:
Among 1964 patients (median age 59, median BMI 34), 79% had data to calculate Fibrosis-4 score (FIB-4), 36% had FibroScan, and 29% had biopsy. Among the 776 F2–F3 patients, 17% progressed to cirrhosis over a median follow-up (mFU) time of 64.6 months. In F2–F3 patients, one unit increase in FIB-4 was associated with 14% increased hazard of cirrhosis; similarly, incidence rates for cirrhosis in patients with liver stiffness measurements (LSMs) 10–15 kPa were 60.8/1000PY (per 1000 person-years; 3.5× higher than those with LSM <10 kPa). Among the 794 patients with compensated cirrhosis (cF4), 46% experienced progression events, 13% died, and 1.3% had a liver transplant (mFU=66.0 months). Among the 394 patients with decompensated cirrhosis (dF4), 29.4% died, and 9.6% had a liver transplant (mFU=62.0 months). Higher FIB-4 was associated with increased risk of decompensation and mortality in cF4 patients. Similarly, patients with LSM ≥15 kPa had a greater incidence rate of mortality (9.1/1000PY vs. 4.3/1000PY and 3.8/1000PY for LSM 10–15 kPa and <10 kPa, respectively).
Conclusions:
The burden of MASH underscores the urgency of standardizing NIT use and diagnosis. Expanding FibroScan use, automating FIB-4 calculation, and increasing awareness of NITs can improve early detection and risk stratification, thereby improving patient outcomes.
Keywords: Fibrosis-4 score, fibrosis, MASLD, non-invasive tests, TARGET-NASH
INTRODUCTION
Metabolic dysfunction–associated steatotic liver disease (MASLD), previously known as nonalcoholic fatty liver disease (NAFLD), 1 is defined as steatosis in the presence of one or more cardiometabolic risk factor(s) and the absence of harmful alcohol intake. The term MASLD encompasses those with metabolic dysfunction–associated steatohepatitis (MASH) and varying degrees of fibrosis, including patients with cirrhosis.
Current American Association for the Study of Liver Diseases (AASLD) and European Association for the Study of the Liver (EASL) guidelines recognize the importance of non-invasive testing for screening and risk stratification of MASH, but there is limited evidence on how these tests are used in the real-world setting. Retrospective studies in the United States are rare and have limitations; for example, a recent analysis of 18,396 adults diagnosed with MASH (2016–2023) was conducted using the MarketScan Commercial & Medicare Databases, showing high use of routine laboratory tests: more than 70% had common lab tests [alanine aminotransferase (ALT), aspartate aminotransferase (AST), complete blood count (CBC), glycated hemoglobin (HbA1c), lipids], and over 75% had lab data sufficient to calculate Fibrosis-4 score (FIB-4) about 2 years before diagnosis (via ICD-codes). Vibration-controlled transient elastography (VCTE) was less frequent and used in about 10% of the cases. However, the results may not be generalizable to US populations covered by other forms of insurance (eg, Medicaid) or the uninsured. 2
Conversely, previous studies that focused on disease progression in MASH have been based on clinical trials such as the STELLAR study which found that 13% of patients with F3 fibrosis progressed to cirrhosis within 4 years.3–5 Outside the clinical trial setting, past studies were small and not representative of the general population nor reflecting routine clinical practice: a meta-analysis of 559 subjects based in the United States, Sweden, and Denmark found that 25%–30% of patients with MASH and F3 fibrosis progressed to cirrhosis after 4 years and 50% after 6 years. 6 Such studies may overestimate disease progression for several reasons: they were often based on cohorts where follow-up liver biopsy was arranged due to clinical concerns, and were frequently based in tertiary centers. Studies examining the speed of progression from cirrhosis to decompensated cirrhosis demonstrated a range of estimates of 19%, 29.4% 7 and 31.0% 8 over 24 months, 53 months, and 8 years, respectively.
Thus, there is limited real-world evidence (RWE) 9 on disease progression in patients with MASH. The longitudinal observational TARGET-NASH study (NCT02815891) 10 is a high-quality real-world data source that was used to estimate disease progression rates across different MASH subpopulations (including CSPH) and identify factors that predicted progression. This study offers valuable insights into the natural history of MASH. By estimating disease progression to cirrhosis, decompensation, and mortality in TARGET-NASH, we expected to enhance disease understanding beyond clinical trial evidence and existing literature, ultimately informing clinical guidelines and future clinical trial planning.
METHODS
Study population
This analysis used data from an observational longitudinal cohort (TARGET-NASH) to define the progression of MASH in a real-world setting. TARGET-NASH (NCT02815891) was approved by the central [Copernicus Group Institutional Review Board (IRB)], local, and institutional review boards. The definitions of MASH, with and without cirrhosis, were developed by the TARGET-NASH steering committee as previously published11,12 and recently updated according to the new MASLD terminology 13 (Supplemental Table S1, http://links.lww.com/HC9/C353). At the date of enrollment into TARGET-NASH, participants signed IRB-approved consent forms providing access to their medical records for at least 3 years before the enrollment date and were followed prospectively for up to 5 years. Data from their electronic health records (EHRs), including patient narratives, laboratory results, pathology reports, and imaging data, were extracted and uploaded into a secure database.
This analysis focused on a subset of participants aged ≥18 years, who met criteria for MASH with significant fibrosis, advanced fibrosis, or cirrhosis (F2–3–4). These patients were further stratified into 6 different subgroups: MASH F2–3, MASH cirrhosis (F4), MASH with compensated cirrhosis (cF4), MASH with cF4 with or without clinically significant portal hypertension (CSPH), and MASH with decompensated cirrhosis (dF4). The baseline time point for each patient (index date) was defined as the earliest date of their classification into MASH. The follow-up duration was defined as the period beginning the day after the index date and continuing until the date of discontinuation from TARGET-NASH or the end of the study period, whichever occurred earlier. The study period was from August 1, 2016, to October 20, 2023. Specifically, patients in the MASH F2–F3 group were required to have either a liver biopsy with evidence of fibrosis stage 2 or 3 or a FibroScan with a liver stiffness value between 7 and 15.9 kPa.11–13 Among patients with MASH cF4, the presence of CSPH was defined as the presence of any of the following: (1) FibroScan stiffness >25 kPa 14 ; (2) FibroScan stiffness >20 kPa and platelets <150,000/μL; (3) evidence of splenomegaly, collaterals, or varices on CT/MRI or in patient’s medical history or adverse events; (4) gastric or esophageal varices on endoscopy. All patients with compensated cirrhosis who had evidence of CSPH on or before the index date were included in the compensated cirrhosis with CSPH cohort.
Risk factors and outcomes
Among possible risk factors associated with disease progression we included age, sex, race, ethnicity, site region, FIB-4, NAFLD fibrosis score (NFS), Aspartate Aminotransferase-to-Platelet Ratio Index (APRI), ALT, AST, platelets count, presence of obesity (yes/no), presence of type 2 diabetes mellitus (T2DM; yes/no), HbA1c, treatment for T2DM (yes/no), blood pressure, lipids, and other routinely collected biomarkers (eg, Estimated Glomerular Filtration Rate, eGFR). Secondary analysis was performed for the subgroup of patients with data available for FibroScan and biopsy. Calculation of FIB-4, APRI, or other indices was based on lab results obtained within a window of up to 12 weeks.
For patients with MASH F2–3 fibrosis, outcomes captured were progression to cirrhosis, development of major adverse cardiovascular events, and all-cause mortality. For patients with MASH compensated cirrhosis (F4), additional outcomes included progression to liver decompensation events and hepatocellular carcinoma (HCC), development of major adverse cardiovascular events, and all-cause mortality. Incidence proportions for all outcomes were calculated as the number of new patients experiencing the outcome during follow-up divided by the number of patients in each cohort at index date; incidence rates per 1000 person-years were calculated as number of patients experiencing the outcome divided by total person-time in years multiplied by 1000, with person-time calculated as time from index date to the earliest time the outcome is observed (for patients experiencing the outcome) or time from index date to the earliest of the date of discontinuation from TARGET-NASH or data cut date (for patients not experiencing the outcome). Model for End-stage Liver Disease (MELD) score change to >15 was evaluated only for patients with MASH cirrhosis at index and a MELD value <12 before index. Any new decompensation event was evaluated only for patients with compensated MASH cirrhosis.
Statistical analysis
Univariable and multivariable Fine–Gray subdistribution hazard models were used to evaluate the association between risk factors and time-to-event outcomes (ie, all-cause mortality, liver decompensation, cirrhosis) by estimating hazard ratios and 95% confidence intervals across patient cohorts (MASH F2/F3, compensated cirrhosis, compensated cirrhosis with CSPH, and decompensated cirrhosis). Death and discontinuation from the study were treated as competing risks. As this study reflected routine clinical practice, we calculated the missing values for each of the risk factors (eg, the biomarkers). Univariable models were performed for all risk factors; if a risk factor was statistically significantly associated with the outcome and had at least 60% non-missing values, the multivariable model was also performed, including the risk factor adjusted for age, sex, race, ethnicity, and site region.
RESULTS
After applying the inclusion/exclusion criteria to the 5745 adult participants enrolled in TARGET-NASH, 1964 adult patients with MASH F2–3–4 were eligible for this analysis (Figure 1), with a median follow-up (mFU) time of 64.9 months. At index date, 776 patients were classified as F2–F3 MASH, 794 as MASH cF4 (514 had CSPH, of which 187 had varices), and 394 as MASH dF4.
FIGURE 1.
Consort diagram. Abbreviations: CSPH, clinically significant portal hypertension; MASH, metabolic dysfunction–associated steatohepatitis; MASLD, metabolic dysfunction–associated steatotic liver disease.
Patients' characteristics
The median BMI was 34.0, with 94.1% classified as living with overweight or obesity, 58.8% had T2DM, and 50.5% had never smoked. Individuals with cirrhosis were older, more likely to be male, non-Hispanic White, users of Medicare, and had metabolic comorbidities (Table 1). Medications of interest at index date are provided in Supplemental Table S2, http://links.lww.com/HC9/C353.
TABLE 1.
Demographic and clinical patient-level characteristics by cohort at index date for adults >18 years enrolled in TARGET-NASH
| Patient characteristics | Patients with non-cirrhotic MASH [F2/F3] n=776) | All patients with cirrhosis (n=1188) | Patients with compensated cirrhosis (n=794) | Patients with compensated cirrhosis without CSPH (n=280) | Patients with compensated cirrhosis with CSPH (n=514) | Patients with decompensated cirrhosis (n=394) |
|---|---|---|---|---|---|---|
| Socio-demographic characteristics | ||||||
| Age at index date (years) | ||||||
| Median (n) | 55.0 (776) | 61.0 (1188) | 61.0 (794) | 62.0 (280) | 61.0 (394) | 61.0 (394) |
| Mean (SD) | 53.5 (13.3) | 59.9 (10.2) | 59.7 (10.6) | 60.8 (10.0) | 59.1 (10.8) | 60.3 (9.5) |
| Q1–Q3 (IQR) | 45.0–63.0 (18.0) | 54.0–67.0 (13.0) | 54.0–67.0 (13.0) | 55.0–67.0 (12.0) | 53.0–67.0 (14.0) | 55.0–67.0 (12.0) |
| Sex, n (%) | ||||||
| Female | 486 (62.6) | 675 (56.8) | 456 (57.4) | 181 (64.6) | 275 (53.5) | 219 (55.6) |
| Male | 290 (37.4) | 513 (43.2) | 338 (42.6) | 99 (35.4) | 239 (46.5) | 175 (44.4) |
| Race and ethnicity, n (%) | ||||||
| Asian | 43 (5.5) | 36 (3.0) | 34 (4.3) | 15 (5.4) | 19 (3.7) | 2 (0.5) |
| Hispanic/Latino | 84 (10.8) | 120 (10.1) | 76 (9.6) | 28 (10.0) | 48 (9.3) | 44 (11.2) |
| Non-Hispanic Black | 65 (8.4) | 27 (2.3) | 20 (2.5) | 10 (3.6) | 10 (1.9) | 7 (1.8) |
| Non-Hispanic White | 556 (71.6) | 960 (80.8) | 633 (79.7) | 219 (78.2) | 414 (80.5) | 327 (83.0) |
| Other/not reported | 28 (3.6) | 45 (3.8) | 31 (3.9) | 8 (2.9) | 23 (4.5) | 14 (3.6) |
| BMI (kg/m2) at index date | ||||||
| Median (n) | 34.0 (722) | 34.0 (1054) | 34.2 (695) | 33.3 (253) | 34.6 (442) | 33.8 (359) |
| Mean (SD) | 34.9 (7.3) | 35.1 (8.2) | 35.3 (8.2) | 34.3 (7.7) | 35.8 (8.4) | 34.8 (8.2) |
| Q1–Q3 (IQR) | 29.4–39.2 (9.8) | 29.4–39.7 (10.3) | 29.5-39.9 (10.3) | 28.9–39.3 (10.4) | 30.0 – 40.2 (10.2) | 29.0–39.1 (10.1) |
| BMI (kg/m2) at index date, categories, n (%) | ||||||
| <18.5 (underweight) | 2 (0.3) | 6 (0.6) | 3 (0.4) | 2 (0.8) | 1 (0.2) | 3 (0.8) |
| 18.5 to <25 (healthy weight) | 32 (4.4) | 64 (6.1) | 38 (5.5) | 15 (5.9) | 23 (5.2) | 26 (7.2) |
| 25 to <30 (Overweight) | 170 (23.5) | 224 (21.3) | 149 (21.4) | 61 (24.1) | 88 (19.9) | 75 (20.9) |
| 30 to <35 (obese class I) | 200 (27.7) | 292 (27.7) | 190 (27.3) | 71 (28.1) | 119 (26.9) | 102 (28.4) |
| 35 to <40 (obese class II) | 158 (21.9) | 219 (20.8) | 145 (20.9) | 48 (19.0) | 97 (21.9) | 74 (20.6) |
| >40 (obese class III) | 160 (22.2) | 249 (23.6) | 170 (24.5) | 56 (22.1) | 114 (25.8) | 79 (22.0) |
| Tobacco usage at enrollment, n (%) | ||||||
| Current | 61 (7.9) | 95 (8.0) | 64 (8.1) | 22 (7.9) | 42 (8.2) | 31 (7.9) |
| Former | 189 (24.4) | 368 (31.0) | 241 (30.4) | 91 (32.5) | 150 (29.2) | 127 (32.2) |
| Never | 421 (54.3) | 570 (48.0) | 379 (47.7) | 131 (46.8) | 248 (48.2) | 191 (48.5) |
| Unknown/not reported | 105 (13.5) | 155 (13.0) | 110 (13.9) | 36 (12.9) | 74 (14.4) | 45 (11.4) |
| AUDIT score at enrollment | ||||||
| Median (n) | 1.0 (715) | 0.0 (1107) | 0.0 (737) | 0.0 (257) | 0.0 (480) | 0.0 (370) |
| Mean (SD) | 1.4 (2.5) | 0.6 (1.7) | 0.7 (1.8) | 0.9 (2.3) | 0.6 (1.5) | 0.4 (1.5) |
| Q1–Q3 (IQR) | 0.0–2.0 (2.0) | 0.0–1.0 (1.0) | 0.0–1.0 (1.0) | 0.0–1.0 (1.0) | 0.0–1.0 (1.0) | 0.0–0.0 (0.0) |
| Alcohol usage, n (%) a | ||||||
| Current | 175 (22.6) | 118 (9.9) | 88 (11.1) | 36 (12.9) | 52 (10.1) | 30 (7.6) |
| Former | 26 (3.4) | 57 (4.8) | 31 (3.9) | 11 (3.9) | 20 (3.9) | 26 (6.6) |
| Never | 105 (13.5) | 214 (18.0) | 128 (16.1) | 46 (16.4) | 82 (16.0) | 86 (21.8) |
| Unknown/not reported | 470 (60.6) | 799 (67.3) | 547 (68.9) | 187 (66.8) | 360 (70.0) | 252 (64.0) |
| Use frequency per interval, n (%) | ||||||
| n | 201 | 175 | 119 | 47 | 72 | 56 |
| <1/day | 128 (63.7) | 94 (53.7) | 67 (56.3) | 26 (55.3) | 41 (56.9) | 27 (48.2) |
| 1–2/day | 25 (12.4) | 10 (5.7) | 9 (7.6) | 4 (8.5) | 5 (6.9) | 1 (1.8) |
| >2/day | 11 (5.5) | 14 (8.0) | 9 (7.6) | 4 (8.5) | 5 (6.9) | 5 (8.9) |
| Binge drinking | 0 (0.0) | 6 (3.4) | 1 (0.8) | 1 (2.1) | 0 (0.0) | 5 (8.9) |
| Unknown | 37 (18.4) | 51 (29.1) | 33 (27.7) | 12 (25.5) | 21 (29.2) | 18 (32.1) |
| Follow-up time (months) | ||||||
| Median (n) | 64.6 (776) | 65.0 (1188) | 66.0 (794) | 64.6 (280) | 66.8 (514) | 62.0 (394) |
| Mean (SD) | 61.9 (23.3) | 62.9 (25.2) | 63.8 (24.6) | 62.7 (23.3) | 64.4 (25.4) | 61.0 (26.3) |
| Q1–Q3 (IQR) | 44.6–78.6 (34.0) | 44.8–81.1 (36.2) | 46.2–82.4 (36.1) | 47.9–79.5 (31.6) | 44.9– 82.8 (37.9) | 42.8–80.0 (37.2) |
| Comorbidities present at the index date | ||||||
| Type 2 diabetes mellitus (T2DM), n (%) | 424 (54.6) | 731 (61.5) | 493 (62.1) | 173 (61.8) | 320 (62.3) | 238 (60.4) |
| Hyperlipidemia, n (%) | 490 (63.1) | 591 (49.7) | 422 (53.1) | 167 (59.6) | 255 (49.6) | 169 (42.9) |
| Hypertension, n (%) | 505 (65.1) | 867 (73.0) | 567 (71.4) | 203 (72.5) | 341 (66.3) | 300 (76.1) |
| Obesity, n (%) b | 518 (71.7) | 760 (72.1) | 505 (72.7) | 175 (69.2) | 330 (74.7) | 255 (71.0) |
| History of cardiovascular disease or event, n (%) | 137 (17.7) | 267 (22.5) | 165 (20.8) | 58 (20.7) | 107 (20.8) | 102 (25.9) |
| Patients with FibroScan availability, n (%) | Non-cirrhotic MASH (n=466) | With cirrhosis (n=239) | Compensated cirrhosis (n=212) | Compensated cirrhosis without CSPH (n=96) | Compensated cirrhosis with CSPH (n=116) | Decompensated cirrhosis (n=27) |
| LSM<10 | 297 (63.7%) | 37 (15.5%) | 31 (14.6%) | 8 (8.3%) | 23 (19.8%) | 6 (22.2%) |
| 10≤SM<15 | 153 (32.8%) | 42 (17.6%) | 38 (17.9%) | 19 (19.8%) | 19 (16.4%) | 4 (14.8%) |
| LSM≥15 | 16 (3.4%) | 160 (66.9%) | 143 (67.5%) | 69 (71.9%) | 74 (63.8%) | 17 (63.0%) |
| Patients with biopsy availability, n (%) | Non-cirrhotic MASH (n=256) | With cirrhosis (n=315) | Compensated cirrhosis (n=250) | Compensated cirrhosis without CSPH (n=145) | Compensated cirrhosis with CSPH (n=105) | Decompensated cirrhosis (n=65) |
| F0 c | 15 (5.9%) | 5 (1.6%) | 4 (1.6%) | 2 (1.4%) | 2 (1.9%) | 1 (1.5%) |
| F1 c | 45 (17.6%) | 19 (6.0%) | 15 (6.0%) | 5 (3.4%) | 10 (9.5%) | 4 (6.2%) |
| F2 | 104 (40.6%) | 17 (5.4%) | 14 (5.6%) | 6 (4.1%) | 8 (7.6%) | 3 (4.6%) |
| F3 | 90 (35.2%) | 65 (20.6%) | 56 (22.4%) | 25 (17.2%) | 31 (29.5%) | 9 (13.8%) |
| F4 | 2 (0.8%) | 209 (66.3%) | 161 (64.4%) | 107 (73.8%) | 54 (51.4%) | 48 (73.8%) |
Alcohol use is based on the closest assessment to the index, within +/= 6 months. If not available within 6 months of the index, then the most recent assessment at or before the index is used (if available).
Obesity is defined as a BMI >30. For all comorbidities involving obesity, the denominator used for percentage calculations is the number of patients with a non-missing BMI.
According to the Target definition and study design, patients with biopsy F0–1 were re-categorized in the MASH F2–F3 group due to FibroScan with a liver stiffness value between 7 and 15.9 kPa.
Abbreviations: AUDIT, Alcohol Use Disorders Identification Test; BMI, body mass index; CSPH, clinically significant portal hypertension; IQR, interquartile range; LSM, liver stiffness measurement; MASH, metabolic dysfunction–associated steatohepatitis; SD, standard deviation; T2DM, type 2 diabetes mellitus.
Frequency of use for non-invasive tests and biopsy
At baseline, 1554 patients (79%) had data on both AST, ALT, and platelet counts to calculate FIB-4. The FIB-4 score was calculated by the Target Study team a posteriori and for the scope of this study using the following formula: [Age (years) × AST (U/L)]/[[Platelet Count (109/L)] × [√ALT (U/L)]]. In addition, 705 (36%) had FibroScan measurements available, and 571 (29%) had a biopsy. Overall descriptive statistics on FibroScan and biopsy data availability are reported in Table 1, and for other non-invasive tests are reported in Supplemental Table S3, http://links.lww.com/HC9/C353.
Disease progression to health outcomes
Disease progression to health outcomes is summarized in Supplemental Table S4, http://links.lww.com/HC9/C353. Among patients with MASH F2/F3, 17.4% (n=135) progressed to cirrhosis, and 1.7% (n=13) died over an mFU period of 64.6 months [interquartile range (IQR) 44.6–78.6 months] (Figures 2, 3). Of the 135 patients with F2–F3 who progressed to cirrhosis, the median time to progression was 16.0 (IQR 7.5–25.0) months.
FIGURE 2.
(A) Cumulative development of cirrhosis for the F2/F3 group. (B) Progression from MASH (F2/F3) to cirrhosis. (C) Cumulative incidence of progression from MASH F2/F3 to all-cause mortality. Abbreviations: AUDIT, Alcohol Use Disorders Identification Test; CI, confidence interval; FIB-4, Fibrosis-4 score; HR, hazard ratio; MASH, metabolic dysfunction–associated steatohepatitis; NFS, NAFLD fibrosis score.
FIGURE 3.
(A) Cumulative all-cause mortality. (B) Cumulative development of decompensation. (C) Cumulative development of HCC. (D) Cumulative Incidence for MACE+. Abbreviations: CI, confidence interval; CSPH, clinically significant portal hypertension; HCC, hepatocellular carcinoma; HR, hazard ratio; MACE, major adverse cardiac events; MASH, metabolic dysfunction–associated steatohepatitis; NASH, nonalcoholic steatohepatitis.
For cF4 patients, 13.1% (n=104) died, while 29.7% (n=236) progressed to any new decompensation event over an mFU period of 66.0 months (IQR 46.2–82.4 months). Of the 104 cF4 patients who progressed to all-cause mortality, the median time to progression was 43.6 (IQR 26.8–58.6) months, while of the 236 patients who progressed to decompensation, the median time to progression was 15.6 (IQR 8.7–27.9) months. In all, 35% (n=180) of cF4 patients with CSPH progressed to a new decompensation event, with 13.6% (n=70) dying over an mFU period of 66.8 months (IQR 44.9–82.8 months). Of the cF4 patients without CSPH, 20.0% (n=56) progressed to a new decompensation event, and 12.1% (n=34) died over an mFU period of 64.6 months (IQR 47.9–79.5 months). Of the dF4 patients, 29.4% (n=116) died over an mFU period of 62.0 months (IQR 42.8–0.0 months). The incidence rate for death was 24.6, 25.4, and 58.0 per 1000 person-years for those with cF4, cF4 + CSPH, and dF4, respectively. The incidence rate for the development of HCC was 4.1, 4.4, and 9.2 per 1000 person-years for those with cF4, cF4 + CSPH, and dF4, respectively.
Cardiometabolic criteria used in the new definition of MASLD
Multivariable modeling showed that cardiometabolic criteria embedded in the new MASLD terminology were associated with disease progression (Supplemental Tables S5–S11, http://links.lww.com/HC9/C353): patients with MASH F2–3 and elevated HbA1c levels (>5.7% vs. ≤5.7) or on treatment for T2DM (vs. not) were more likely to progress to cirrhosis or a composite outcome of cirrhosis or all-cause mortality (Supplemental Tables S5, S6, http://links.lww.com/HC9/C353). Similar findings were observed in patients with cirrhosis: those with elevated HbA1c levels or on treatment for T2DM were at a greater risk for progression to all-cause mortality (Supplemental Table S7, http://links.lww.com/HC9/C353). Overall, among patients with compensated cirrhosis, lower lipid levels and lipid-lowering treatment use were associated with lower risk of progression to decompensation (Supplemental Tables S9, S10, http://links.lww.com/HC9/C353).
Risk factors of disease progression in MASH F2–3
Multivariable modeling among patients with MASH F2–3 fibrosis adjusted for age, sex, race, and site region, demonstrated that patients with diabetes, a higher FIB-4, a higher APRI, and an NFS >1.455 had a greater likelihood of progression to cirrhosis, while patients who never smoked, and higher platelet count had a lower likelihood of progression to MASH with cirrhosis (Figure 2B).
Multivariable models were not conducted for all-cause mortality among the MASH F2/3 group due to the small number of deaths. In univariable models, older age, history of osteoarthritis, history of cardiovascular disease (CVD), elevated serum alkaline phosphatase (ALP, IU/L), elevated serum gamma-glutamyl transferase (GGT, U/L), higher high-density lipoprotein (HDL, mg/dL), and history of beta blocker use were associated with a significantly higher likelihood of all-cause mortality. Variables associated with a decreased likelihood of all-cause mortality included: higher ALT (HR 0.98, 95% CI 0.96–1.00), albumin (HR 0.43, 95% CI 0.21–0.87), and eGFR (HR 0.95, 95% CI 0.93–0.97).
Risk factors of disease progression in MASH cF4 and dF4
Univariable analysis for patients with compensated cirrhosis showed that a higher ALP, higher total bilirubin, higher FIB-4, NFS category >0.676, higher APRI, and a history of opioid use were associated with an increased likelihood of progression to a decompensation event. A history of hyperlipidemia, history of sleep apnea, history of statin use, and higher ALT, hemoglobin, platelets, and albumin were all associated with a decreased likelihood of progression to a decompensation event. Multivariable modeling indicated that a higher ALP, total bilirubin, FIB-4, NFS >0.676, APRI, and history of opioid use were associated with a greater likelihood of progression to decompensation among patients with compensated cirrhosis, while a history of hyperlipidemia, history of sleep apnea, higher ALT, hemoglobin, platelets, albumin, and history of statin use were significantly associated with a lower likelihood of progression to decompensation (Figure 4).
FIGURE 4.
Progression to decompensation among patients with compensated cirrhosis. Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; APRI, Aspartate Aminotransferase-to-Platelet Ratio Index; AUDIT, Alcohol Use Disorders Identification Test; CI, confidence interval; FIB-4, Fibrosis-4 score; HR, hazard ratio; NFS, NAFLD fibrosis score.
Among patients with cirrhosis, Medicaid insurance (vs. private), a history of CVD, a higher ALP, total bilirubin, hemoglobin A1c, FIB-4, NFS category >0.676, and history of opioid use were significantly associated with increased risk of mortality, while a history of hyperlipidemia, higher ALT, hemoglobin, platelets, albumin, eGFR, and history of statin use were associated with a lower likelihood of mortality (results not shown). Among patients with compensated cirrhosis, a higher BMI, ALP, total bilirubin, FIB-4, APRI score, history of opioid use, and an NFS category were significantly associated with a greater likelihood of mortality, while history of hyperlipidemia and higher ALT, hemoglobin, platelets, and albumin were significantly associated with a lower risk of mortality.
In patients with compensated cirrhosis and CSPH, a higher ALP, total bilirubin, FIB-4, APRI, NFS >0.676, history of opioid use, and history of corticosteroid use were associated with a greater likelihood of progression to decompensation, while higher hemoglobin, platelets, and albumin were associated with a lower likelihood of decompensation (Supplemental Figure S1, http://links.lww.com/HC9/C353). A higher ALP, total bilirubin, FIB-4, APRI, and NFS category >0.676 were significantly associated with an increased likelihood of progressing to all-cause mortality, while a history of hyperlipidemia, a higher ALT, hemoglobin, platelets, and albumin were associated with a decreased likelihood of mortality (Supplemental Figure S2, http://links.lww.com/HC9/C353). Among patients with decompensated cirrhosis, Medicaid insurance (vs. private), and higher ALP were associated with an increased likelihood of progression to all-cause mortality, while higher albumin was associated with a lower likelihood of mortality (Figure 5).
FIGURE 5.
Progression to all-cause mortality among patients with decompensated cirrhosis. Abbreviations: ALP, alkaline phosphatase; CI, confidence interval; HR, hazard ratio.
Among patients with compensated cirrhosis with or without CSPH (Figure 4 and Supplemental Figure S1, http://links.lww.com/HC9/C353), a higher Alcohol Use Disorders Identification Test (AUDIT) score and current (vs. former) drinking appeared to be associated with a lower risk of decompensation. We believe this unexpected finding may be a product of reverse causality, whereby patients with more severe disease were more likely to have discontinued alcohol consumption and would therefore have a lower AUDIT score, falling in the “former” drinker category at enrollment. AUDIT score and the alcohol consumption information represent a snapshot at the time of enrollment rather than total alcohol consumption, which may have contributed to the disease. Moreover, the low “n” for “former” drinkers and missing data (Table 1) may skew the risk analysis. Future analyses using alcohol biomarkers measured at multiple time points may be better able to investigate this.
Subgroup analysis using FibroScan and biopsy data
A secondary analysis estimating incidence rates of progression to health outcomes was performed for the subset of patients with data available on FibroScan (N=705) and biopsy (N=571). Among those classified as F2–3 who had biopsy data, 43.8% were F2. Patients with a FibroScan score of >15 kPa versus lower scores (Supplemental Table S12, http://links.lww.com/HC9/C353) and biopsy stage F4 versus F2–3 (Supplemental Table S13, http://links.lww.com/HC9/C353) did report higher incidence rates of disease progression for most outcomes [eg, all-cause death, liver transplant, MELD score change to >15, cirrhosis, HCC, non-HCC cancers (non-skin cancers), major adverse cardiac events (MACE+)]. Cumulative incidence for these progression events is shown in Supplemental Figures S3–S5, http://links.lww.com/HC9/C353.
DISCUSSION
This study provides important RWE on rates of disease progression in patients with MASH, measuring progression to health outcomes typically used in the clinical trial setting. In a population of 1964 adult patients with MASH with significant fibrosis, advanced fibrosis, or cirrhosis (F2–F3–F4), nearly all patients were categorized as overweight or obese (94%), consistent with the new MASLD nomenclature.
The frequency of use of blood tests, and specifically those useful to calculate FIB-4 (AST, ALT, platelets) for MASH recognition, is similar to the frequency observed in other studies (eg, Gbadamosi et al) 2 ; this highlights opportunities for its automated calculation in EHRs for mortality prognosis. In our study, the use of FibroScan in about 1 out of 3 cases of MASH identification appeared to be higher than in Gbadamosi et al; this may be due to access, availability and the underlying differences in the data sources; one possible explanation is that some healthcare centers, especially academic liver clinics that comprise the majority in TARGET-NASH, have ready access to FibroScan devices and could have incorporated their use more routinely, while others (community clinics, primary care settings) may have limited or no access. Overall, studies conducted in specialized hepatology units or regions where FibroScan is standard have higher reported use. 15 Expanding the use of FibroScan, or similar non-invasive modalities to assess fibrosis, in MASH is important because it provides a safe, rapid, and highly accurate method to detect liver fibrosis, and it is also a reliable method to predict disease progression, without relying on invasive liver biopsies.
The main findings showed that in the MASH F2–F3 cohort, 17% progressed to cirrhosis in a median of 16 months. Among patients with MASH cF4, 46% experienced a composite of any progression event. In the MASH dF4 cohort, a more rapid disease progression to mortality was observed compared with the cF4 and F2–3 cohorts. Overall, patients with MASH who were older, with diabetes, smokers, with a higher FIB-4, a higher APRI, and an NFS >1.455 had a greater likelihood of progression to cirrhosis, mortality, or transplantation. Opioid use was associated with adverse outcomes, as previously reported, 16 while statin use may be protective despite its poor utilization in this population. 17 Disease regression was not assessed in this study due to the lack of standardized or available pharmacotherapy use and the limited number of patients in the study with multiple FibroScan measurements or biopsy over time (to accurately assess regression). Future studies without these limitations may be better able to assess disease regression.
Comparison with previous studies is complicated, given differences in study populations, design, and choice of outcomes. Historically, there has been limited data on the progression between MASH disease stages due to limited data sources following patients with MASH longitudinally for the length of time it often takes patients to progress in their disease state. A study in France between 2009 and 2015, which included 186,956 patients with NAFLD (old nomenclature), showed that over 24 months of the study, 19% of patients with compensated cirrhosis progressed to clinical events, including decompensation, HCC, or death. 18 Only 2 studies (focused on patients with compensated cirrhosis) provided estimates of rapid progression to decompensated cirrhosis with estimates of 29.4% 7 and 31.0%. 8 In this study, patients progressed from F2–F3 to cirrhosis in a median time of 16 months. The median time for progression to decompensation was 15.6 months, while the median time to all-cause mortality for patients with decompensation was 41.2 months. A study by Sanyal et al 19 found that 19% patients in a clinical trial with compensated cirrhosis progressed to a clinical event within the 30.9-month follow-up period; while a study by Boursier et al 18 found 19% of patients with compensated cirrhosis progressed to a clinical event (decompensated cirrhosis, HCC, or death) over a 24-month period.
Among patients with MASH F2–F3, a higher FIB-4 score, higher APRI, NFS >1.544, diabetes, and lower platelet count were important risk factors for disease progression to cirrhosis. 20 These trends align with the current understanding of MASH disease progression and its association with health outcomes: higher FIB-4 is indicative of worsening of fibrosis and progression to cirrhosis and mortality,21,22 while low platelet count levels may signal reduced liver activity due to disease progression. 23 Patients with CVD were also found to be at an increased risk of mortality. These findings are similar to those from a large retrospective study where patients were followed for an average of 12.6 years.
In patients with MASH cF4, a higher ALP, total bilirubin, FIB-4, NFS >0.676, and APRI were identified as risk factors for progression to decompensation. Conversely, hyperlipidemia, sleep apnea, higher ALT, hemoglobin, platelets, albumin, and statin use were significantly associated with a lower likelihood of progression to decompensation. Previous studies have found that the most significant predictors of decompensation are liver function, portal hypertension, and inflammation/fibrosis indicators. 24
Strengths and limitations
Our findings enhance the current knowledge on disease progression due to a longer follow-up time in comparison to previous studies, which were severely limited by sample size25,26 or short follow-up. 27 A strength of this study is that TARGET-NASH patients were identified in a real-world setting and diagnosed using non-invasive methods, biopsy, or both. This reflects how disease progression is assessed in routine clinical practice. In contrast, clinical trials or observational biopsy-based studies carry the risk of selection bias indication and availability of biopsy, which can overrepresent patients with comorbidity. 28 Our findings can thus be more broadly translated to other real-world populations in the United States. In addition, the TARGET-NASH dataset comprises an ethnically diverse makeup of patients within the United States (Non-Hispanic White = 77%, Hispanic/Latino = 10%, Non-Hispanic Black 5%). Finally, although this analysis used the initial TARGET-NASH disease severity definitions, with the additional criteria requiring F2–F3 MASH patients to have a FibroScan or biopsy, an overwhelming majority of these patients met the new MASLD criteria 13 (see Supplemental Tables S5–S10, http://links.lww.com/HC9/C353).
Some limitations associated with data collection methods should be acknowledged. The cumulative incidence mortality curves were similar between the MASH cF4 with CSPH group compared with the cF4 without CSPH group, and we could not exclude possible misclassification bias. For example, if the disease state were defined by liver stiffness measurements, high variance with inter-patient and intra-patient variability in measurements could occur. Furthermore, the 25 kPa liver stiffness threshold is known to have lower specificity for CSPH in patients with obesity. 29 This may have increased the likelihood of misclassification bias, because the most obvious sign, varices, is not yet present, or because early varices can be missed on endoscopy (or patients may not have had recent endoscopies).
Also, patients enrolled within TARGET-NASH consented to providing retrospective data dating from 3 years before enrollment. To maximize the amount of time during which to observe disease progression, the index date for this study was designated as the earliest date on which evidence of the disease was observed, as opposed to the date of enrollment. This provided up to an additional 3 years of data by which to potentially observe/detect disease progression. While this provides a greater period of time for observation, this practice may introduce immortal time bias and/or selection bias in the evaluation of exposure-outcome associations.
The timing of collecting the AUDIT score (at the time of enrollment, rather than at index) may have impacted the importance of this variable and thus was not included in the multivariable analyses. Another limitation associated with the data was that the cause of death was missing for about 50% of the 233 patients who died. Moreover, death data can have inaccuracies in real-world datasets. Hence, analyses were based on all-cause mortality and not separated into liver-related versus unrelated deaths.
TARGET-NASH includes a larger proportion of patients recruited from academic medical centers compared with community clinics or primary care settings. Due to referral bias, this also enriches the dataset with patients with more advanced diseases. While our study was not appropriately powered to evaluate the differences in outcomes between patients in academic medical centers compared with community clinics for each disease stage, future research on this topic may yield informative outcomes regarding disease management. Real-world data inherently offers an unfiltered view of ongoing treatment for patients. As there are no pre-determined points of data collection, and data from routine care is captured via the EHR, there is the potential for disease misclassification and limitations for generalizability. 30 In TARGET-NASH, misclassification is mitigated by medical record reviews conducted by expert adjudication as needed. There is a possibility of healthy entrant bias in TARGET-NASH, as with any observational study cohort.
In conclusion, the RWE generated by this study provides an assessment of both characteristics and the clinical course of patients with MASH and advanced fibrosis. This evidence can provide additional rationale and justification in clinical trial design, including enrollment criteria, trial duration, and the choice of endpoints. It also serves to inform health care providers and patients who are typically unaware of how MASH is interconnected with obesity, diabetes, and other metabolic diseases, or which MASH subpopulations may be more at risk of experiencing health outcomes. Lastly, this study highlights the prognostic utility of accessible biomarkers, including FIB-4, FibroScan, APRI, NFS, and platelets, in predicting health outcomes typically used in clinical trials. The burden of MASH underscores the urgency of standardizing non-invasive testing and diagnosis, particularly in primary care. Expanding FibroScan use and automating FIB-4 calculation could improve early detection and risk stratification, given the routine availability of relevant blood tests. Addressing deficits in access, awareness, and education on non-invasive tests is essential to enhance recognition and outcomes for MASH patients. FibroScan findings are especially relevant, as the FDA recently accepted a proposal for Vibration-Controlled Transient Elastography as a reasonably likely surrogate endpoint for clinical trials in adults with non-cirrhotic MASH with moderate-to-advanced liver fibrosis for all-cause mortality or liver-related events. 31
Supplementary Material
Acknowledgments
FUNDING INFORMATION
TARGET-NASH is supported in part by funding from Boehringer Ingelheim. TARGET-NASH is a study sponsored by Target RWE, headquartered in Durham, North Carolina.
CONFLICTS OF INTEREST
Heather L. Morris, Derek Gazis, Andrea R. Mospan, and Michael W. Fried are Target RWE employees. Philip N. Newsome—Grant/research support from Novo Nordisk; Speakers bureau—Eli Lilly; Consulting fees from AstraZeneca, Boehringer Ingelheim, GSK, Sagimet, Madrigal, MSD, Novo Nordisk, Shionogi, Zydus Therapeutics, and 89bio.
Footnotes
Abbreviations: AASLD, American Association for the Study of Liver Diseases; ALP, alkaline phosphatase; ALT, alanine aminotransferase; APRI, Aspartate Aminotransferase-to-Platelet Ratio Index; AST, aspartate aminotransferase; AUDIT, Alcohol Use Disorders Identification Test; CBC, complete blood count; CSPH, clinically significant portal hypertension; CVD, cardiovascular disease; EASL, European Association for the Study of the Liver; eGFR, estimated glomerular filtration rate; EHR, electronic health record; FIB-4, Fibrosis-4 score; HbA1c, glycated hemoglobin; HCC, hepatocellular carcinoma; LSM, liver stiffness measurement; MACE, major adverse cardiac events; MASH, metabolic dysfunction-associated steatohepatitis; MASLD, metabolic dysfunction-associated steatotic liver disease; MELD, Model for End-stage Liver Disease; mFU, median follow-up; NAFLD, nonalcoholic fatty liver disease; NFS, NAFLD fibrosis score; NIT, non-invasive test; RWE, real-world evidence; SLD, steatotic liver disease; T2DM, type 2 diabetes mellitus.
Philip N. Newsome and Claudio Sartini are joint first authors with equal contributions.
The authors meet the criteria for authorship as recommended by the International Committee of Medical Journal Editors. The authors did not receive any payment related to the development of this study. Boehringer Ingelheim was given the opportunity to review the manuscript for medical and scientific accuracy as well as intellectual property considerations. The study was supported and funded by Boehringer Ingelheim
Supplemental Digital Content is available for this article. Direct URL citations are provided in the HTML and PDF versions of this article on the journal’s website, www.hepcommjournal.com.
Contributor Information
Philip N. Newsome, Email: philip.newsome@kcl.ac.uk.
Claudio Sartini, Email: claudio.sartini@boehringer-ingelheim.com.
Heather L. Morris, Email: hlmorris27@gmail.com.
Derek Gazis, Email: dgazis@targetrwe.com.
Andrea R. Mospan, Email: andreamospan@gmail.com.
A. Sidney Barritt, IV, Email: barritt@med.unc.edu.
Michael W. Fried, Email: mfried@targetrwe.com.
Salvador Augustin, Email: salvador.augustin@boehringer-ingelheim.com.
Mandy Fraessdorf, Email: mandy.fraessdorf@boehringer-ingelheim.com.
Bryan Rudolph, Email: bryan.rudolph@boehringer-ingelheim.com.
Rohit Loomba, Email: roloomba@health.ucsd.edu.
Arun Sanyal, Email: arun.sanyal@vcuhealth.org.
REFERENCES
- 1.Rinella ME, Lazarus JV, Ratziu V, Francque SM, Sanyal AJ, Kanwal F, et al. A multisociety Delphi consensus statement on new fatty liver disease nomenclature. Hepatology. 2023;78:1966–1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gbadamosi SO, Evans KA, Brady BL, Hoovler A. Noninvasive tests and diagnostic pathways to MASH diagnosis in the United States: A retrospective observational study. J Med Econ. 2025;28:314–322. [DOI] [PubMed] [Google Scholar]
- 3.Wong VWS, Wong GLH, Choi PCL, Chan AWH, Li MKP, Chan HY, et al. Disease progression of non-alcoholic fatty liver disease: A prospective study with paired liver biopsies at 3 years. Gut. 2010;59:969–974. [DOI] [PubMed] [Google Scholar]
- 4.Harrison SA, Wong VWS, Okanoue T, Bzowej N, Vuppalanchi R, Younes Z, et al. Selonsertib for patients with bridging fibrosis or compensated cirrhosis due to NASH: Results from randomized phase III STELLAR trials. J Hepatol. 2020;73:26–39. [DOI] [PubMed] [Google Scholar]
- 5.Adams LA, Sanderson S, Lindor KD, Angulo P. The histological course of nonalcoholic fatty liver disease: A longitudinal study of 103 patients with sequential liver biopsies. J Hepatol. 2005;42:132–138. [DOI] [PubMed] [Google Scholar]
- 6.Musso G, Gambino R, Cassader M, Pagano G. Meta-analysis: Natural history of non-alcoholic fatty liver disease (NAFLD) and diagnostic accuracy of non-invasive tests for liver disease severity. Ann Med. 2011;43:617–649. [DOI] [PubMed] [Google Scholar]
- 7.Hernández-Gea V, Aracil C, Colomo A, Garupera I, Poca M, Torras X, et al. Development of ascites in compensated cirrhosis with severe portal hypertension treated with beta-blockers. Am J Gastroenterol. 2012;107:418–427. [DOI] [PubMed] [Google Scholar]
- 8.Fleming KM, Aithal GP, Card TR, West J. The rate of decompensation and clinical progression of disease in people with cirrhosis: A cohort study. Aliment Pharmacol Ther. 2010;32:1343–1350. [DOI] [PubMed] [Google Scholar]
- 9.FDA . Real-world evidence. US Food and Drug Administration. Accessed September 22, 2025. https://www.fda.gov/science-research/science-and-research-special-topics/real-world-evidence
- 10.Loomba R, Adams LA. The 20% rule of NASH progression: The natural history of advanced fibrosis and cirrhosis caused by NASH. Hepatology. 2019;70:1885–1888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Barritt AS, Gitlin N, Klein S, Lok AS, Loomba R, Malahias L, et al. Design and rationale for a real-world observational cohort of patients with nonalcoholic fatty liver disease: The TARGET-NASH study. Contemp Clin Trials. 2017;61:33–38. [DOI] [PubMed] [Google Scholar]
- 12.Kim HP, Idowu MO, Mospan AR, Allmon AG, Roden M, Newsome P, et al. Liver biopsy in the real world-reporting, expert concordance and correlation with a pragmatic clinical diagnosis. Aliment Pharmacol Ther. 2021;54(11–12):1472–1480. [DOI] [PubMed] [Google Scholar]
- 13.Barritt AS, Yu F, Mospan AR, Newsome PN, Roden M, Morris HL, et al. High concordance between nonalcoholic fatty liver disease and metabolic dysfunction–associated steatotic liver disease in the target-nash real-world cohort. Am J Gastroenterol. 2024;119:1624–1627. [DOI] [PubMed] [Google Scholar]
- 14.de Franchis R, Bosch J, Garcia-Tsao G, Reiberger T, Ripoll C, Baveno VIIF. Baveno VII—Renewing consensus in portal hypertension. J Hepatol. 2022;76:959–974. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.de Jong VD, Alings M, Bruha R, Cortez-Pinto H, Dedoussis GV, Doukas M, et al. Global research initiative for patient screening on MASH (GRIPonMASH) protocol: Rationale and design of a prospective multicentre study. BMJ Open. 2025;15:e092731. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Moon AM, Watkins SE, Lok AS, Firpi-Morell RJ, Trinh HN, Kupec JT, et al. Opioid use is more common in nonalcoholic fatty liver disease patients with cirrhosis, higher BMI, and psychiatric disease. Dig Dis. 2021;39:247–257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Thomson MJ, Serper M, Khungar V, Weiss LM, Trinh H, Firpi-Morell R, et al. Prevalence and factors associated with statin use among patients with nonalcoholic fatty liver disease in the TARGET-NASH study. Clin Gastroenterol Hepatol. 2022;20:458–460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Boursier J, Shreay S, Fabron C, Torreton E, Fraysse J. Hospitalization costs and risk of mortality in adults with nonalcoholic steatohepatitis: Analysis of a French national hospital database. EClinicalMedicine. 2020;25:100445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sanyal AJ, Harrison SA, Ratziu V, Abdelmalek MF, Diehl AM, Caldwell S, et al. The natural history of advanced fibrosis due to nonalcoholic steatohepatitis: Data from the simtuzumab trials. Hepatology. 2019;70:1913–1927. [DOI] [PubMed] [Google Scholar]
- 20.Huang DQ, Noureddin N, Ajmera V, Amangurbanova M, Bettencourt R, Truong E, et al. Type 2 diabetes, hepatic decompensation, and hepatocellular carcinoma in patients with non-alcoholic fatty liver disease: An individual participant-level data meta-analysis. Lancet Gastroenterol Hepatol. 2023;8:829–836. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mózes FE, Lee JA, Vali Y, Selvaraj EA, Jayaswal ANA, Boursier J, et al. Diagnostic accuracy of non-invasive tests to screen for at-risk MASH—An individual participant data meta-analysis. Liver Int. 2024;44:1872–1885. [DOI] [PubMed] [Google Scholar]
- 22.Kalas MA, Chavez L, Leon M, Taweesedt PT, Surani S. Abnormal liver enzymes: A review for clinicians. World J Hepatol. 2021;13:1688–1698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lim HI, Cuker A. Thrombocytopenia and liver disease: Pathophysiology and periprocedural management. Hematology Am Soc Hematol Educ Program. 2022;2022:296–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.D’Amico G, Perricone G. Prediction of decompensation in patients with compensated cirrhosis: Does etiology matter? Curr Hepatol Rep. 2019;18:144–156. [Google Scholar]
- 25.McPherson S, Hardy T, Henderson E, Burt AD, Day CP, Anstee QM. Evidence of NAFLD progression from steatosis to fibrosing-steatohepatitis using paired biopsies: Implications for prognosis and clinical management. J Hepatol. 2015;62:1148–1155. [DOI] [PubMed] [Google Scholar]
- 26.Noureddin N, Huang DQ, Bettencourt R, Siddiqi H, Majzoub AM, Nayfeh T, et al. Natural history of clinical outcomes and hepatic decompensation in metabolic dysfunction–associated steatotic liver disease. Aliment Pharmacol Ther. 2024;59:1521–1526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Pais R, Charlotte F, Fedchuk L, Bedossa P, Lebray P, Poynard T, et al. A systematic review of follow-up biopsies reveals disease progression in patients with non-alcoholic fatty liver. J Hepatol. 2013;59:550–556. [DOI] [PubMed] [Google Scholar]
- 28.Younossi ZM, Mangla KK, Berentzen TL, Grau K, Kjær MS, Ladelund S, et al. Liver histology is associated with long-term clinical outcomes in patients with metabolic dysfunction-associated steatohepatitis. Hepatol Commun. 2024;8:e0423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Bañares J, Pons M, Reiberger T, Mandorfer M, Jachs M, Semmler G, et al. Refining the Baveno VII criteria for clinically significant portal hypertension: An individual patient data meta-analysis. J Hepatol. 2026;84:543–553. [DOI] [PubMed] [Google Scholar]
- 30.Loomba R, Wong R, Fraysse J, Shreay S, Li S, Harrison S, et al. Nonalcoholic fatty liver disease progression rates to cirrhosis and progression of cirrhosis to decompensation and mortality: A real world analysis of Medicare data. Aliment Pharmacol Ther. 2020;51:1149–1159. [DOI] [PubMed] [Google Scholar]
- 31.FDA . FDA accepts proposal for reasonably likely surrogate endpoint for “MASH” all-cause mortality or liver-related events. US Food and Drug Administration; 2025. Assessed July 27, 2025, https://www.fda.gov/drugs/drug-safety-and-availability/fda-accepts-proposal-reasonably-likely-surrogate-endpoint-mash-all-cause-mortality-or-liver-related [Google Scholar]