69 - Patients With Non-Alcoholic Fatty Liver Disease Demonstrate Heterogenous Risk Factors but Similar Rate of Progression to Cirrhosis Across the Age Spectrum

Wednesday, October 26, 2022

9:30 AM – 9:45 AM ET

Location: Hall C1

Presenting Author(s)

Matthew J. Miller, MD

University of Michigan
Livonia, MI

PDF of Slides View Slides

Matthew J. Miller, MD¹, Emily Harding-Theobald, MD², Zhe Zhao, MS², Jacob V. DiBattista, MD², Daniel A. Burkholder, MD², Anna S. Lok, MD², Vincent L. Chen, MD²
¹University of Michigan, Livonia, MI; ²University of Michigan, Ann Arbor, MI

Introduction: Progression of non-alcoholic fatty liver disease (NAFLD) to cirrhosis occurs at varying rates. Reliably detecting advanced fibrosis in NAFLD patients across age groups and accurately predicting those at greatest risk of cirrhosis remains challenging. We aimed to characterize heterogeneity in metabolic and genetic risk factors for prevalent and incident cirrhosis among patients with NAFLD.

Methods: This was a single center analysis of patients with NAFLD seen at Michigan Medicine between 2010-2021. NAFLD was defined by hepatic steatosis on imaging, biopsy, or transient elastography in the absence of other chronic liver disease; the earliest date of hepatic steatosis was defined as the index date. Cirrhosis was determined by validated ICD-9/10 codes. Incident cirrhosis refers to new diagnosis > 1 year from the index date. In the subset of 4359 patients with genetic data, we evaluated frequency of known NAFLD risk alleles. We generated Fine-Gray competing risk models to compare risk of incident cirrhosis between ages.

Results: Of 31505 patients with NAFLD, there were 8252 age 18-< 40, 15035 age 40-< 60, and 8218 age ≥ 60 years at time of steatosis identification, of whom 804 had prevalent cirrhosis and 388 developed incident cirrhosis during 128090 person-years follow-up. Younger patients had a lower prevalence of diabetes but a higher prevalence of obesity class 3. They also had a higher frequency of the NAFLD-promoting PNPLA3-rs738409-G allele (GG genotype 11.8%, 9.6%, and 8.8% in age ranges 18-< 40, 40-< 60, and ≥ 60 years; p = 0.016). Frequency of NAFLD risk alleles in GCKR, HSD17B13, MBOAT7, and TM6SF2 was not different across age groups. APRI, FIB4, and NAFLD fibrosis score (NFS) demonstrated similar performance for identifying prevalent cirrhosis in patients aged 40-< 60 years, but NFS performed more poorly in the 18-< 40 and ≥ 60 years group (AUC 0.691 and 0.744, respectively, p < 0.05 compared to FIB4 and APRI; Fig 1). There was no significant difference in the 10-year risk of incident cirrhosis between age groups overall (p = 0.107) or among several pre-specified subgroups.

Discussion: In a large NAFLD cohort, a larger portion of younger patients were identified as having the high risk PNPLA3-rs738409-G allele. NFS was less accurate in identifying prevalent cirrhosis in those < 40 or ≥ 60 years old. Progression to cirrhosis was similar in patients < 40 years old compared to older patients, suggesting NAFLD in the young should not be considered more benign than in older patients.

Figure: Figure 1. Accuracy of Fibrosis-4, AST-to-platelet ratio index, and non-alcoholic fatty liver disease fibrosis score in predicting prevalent cirrhosis, stratified by age group. Point of maximal Youden J statistic is marked on each ROC curve. Optimal cutpoint (threshold) for each test is listed with area under the curve (AUC) in sub-table of each age group. APRI, AST-to-platelet ratio index. FIB4, Fibrosis-4. NFS, non-alcoholic fatty liver disease fibrosis score.

Predictor	HR (95% Cl)	P value
Age
18-< 40 years	(referent)
40-< 60 years	0.91 (0.66-1.24)	0.546
≥ 60 years	0.97 (0.66-1.44)	0.895
Diabetes	2.01 (1.51-2.67)	< 0.001
Hypertension	1.49 (1.09-2.02)	0.011
Hyperlipidemia	1.25 (0.91-1.70)	0.163
Body mass index
Normal	(referent)
Overweight	1.67 (0.84-3.33)	0.145
Class 1 obesity	1.88 (0.96-3.69)	0.067
Class 2 obesity	1.63 (0.81-3.29)	0.171
Class 3 obesity	2.80 (1.42-5.54)	0.003
Alanine aminotransferase
< ULN	(referent)
1-< 2x ULN	1.63 (0.99-2.66)	0.053
2-< 5x ULN	3.61 (2.24-5.82)	< 0.001
≥ 5x ULN	5.50 (3.16-9.58)	< 0.001

Table: Multivariable model for incident cirrhosis. Race and sex are not shown but were included. Hazard ratios (HR) and 95% confidence interval (CI) are reported. Normal body mass index (BMI) was defined as BMI <23 in Asians and BMI <25 in non-Asian patients. Upper limit of normal (ULN) for alanine aminotransferase was defined as 19 U/L for women or 30 U/L for men.

Disclosures:

Matthew Miller indicated no relevant financial relationships.

Emily Harding-Theobald indicated no relevant financial relationships.

Zhe Zhao indicated no relevant financial relationships.

Jacob DiBattista indicated no relevant financial relationships.

Daniel Burkholder indicated no relevant financial relationships.

Anna Lok: Novo Nordisk – DSMB chair.

Vincent Chen indicated no relevant financial relationships.

Matthew J. Miller, MD¹, Emily Harding-Theobald, MD², Zhe Zhao, MS², Jacob V. DiBattista, MD², Daniel A. Burkholder, MD², Anna S. Lok, MD², Vincent L. Chen, MD², 69, Patients With Non-Alcoholic Fatty Liver Disease Demonstrate Heterogenous Risk Factors but Similar Rate of Progression to Cirrhosis Across the Age Spectrum, ACG 2022 Annual Scientific Meeting Abstracts. Charlotte, NC: American College of Gastroenterology.