INTRODUCTION While higher body weight is closely associated with histologic severity in metabolic-associated steatotic liver disease (MASLD), lean MASLD (ie, MASLD in normal weight individuals) is a relatively distinct phenotype that has received considerable attention.1,2 However, there are knowledge gaps in our understanding of the interplay among weight categories, PNPLA3 genotype, and severity of liver histology. While there have been studies done previously to evaluate the interaction between body mass index (BMI) and MASLD severity, these have been somewhat limited by smaller study size and lack of liver histology.3,4 In this study, we examined clinical, PNPLA3 genotypic, and histological characteristics of MASLD across several BMI categories in 3386 adult participants of the NASH Clinic Research Network. METHODS We collected data for adult patients enrolled in various prospective studies between October 2004 and May 2024 at 9 participating NASH Clinic Research Network centers. Participants provided written informed consent. All patients had a liver biopsy centrally reviewed by the NASH Clinic Research Network Pathology Committee. Weight and height were measured at the screening visit and were the average of 2 measurements. We divided patients into the following BMI categories: normal weight (BMI <25 kg/m2), overweight (BMI 25–29.9 kg/m2), class 1 obesity (BMI 30–34.9 kg/m2) and ≥class 2 obesity (BMI ≥35 kg/m2). These weight categories were adjusted for Asian participants according to the WHO criteria, defined as normal weight BMI <23 kg/m2, overweight BMI 23–27.4 kg/m2, obese BMI 27.5–32.4 kg/m2, and severe obesity BMI ≥32.5 kg/m2. Selected characteristics of interest, including age, type 2 diabetes mellitus, metabolic syndrome, and PNPLA3 genotype, were collected. Histological phenotypes of interest included steatohepatitis, NAFLD Activity Score, advanced fibrosis (≥F3), and cirrhosis (F4). We compared the above clinical variables and histologic phenotypes between all 4 BMI categories. In addition, we made comparisons between the normal weight (BMI <25 kg/m2) and overweight (BMI 25–29.9 kg/m2) groups as well as between the nonobese (BMI <30 kg/m2) and obese (BMI ≥30 kg/m2) groups. p values for comparisons across the 4 categories of BMI were determined from ordinal logistic regression. p values for binary comparisons were determined from 2-sample t tests (Wilcoxon rank sum for variables not normally distributed) for continuous measures and χ2 tests for categorical measures. Statistical significance was defined as a 2-tailed p value of <0.05. Statistical analyses were completed with SAS and Stata. RESULTS Among 3386 adult participants included in this study, 107 (3.2%) adults had normal weight (mean BMI 23.2 kg/m2), 744 (22%) were overweight (mean BMI 27.7 kg/m2), 1128 (33.3%) with class 1 obesity (mean BMI 32.3 kg/m2), and 1407 (41.5%) with class≥2 obesity (mean BMI 40.5 kg/m2). The baseline characteristics of participants are outlined in Supplemental Table S1, https://links.lww.com/HC9/C348. The mean age of all participants was 50.4±12.2 years, and 62% were female. Across all groups, participants were 80% White, 3% Black, 7% Asian, and 13% of participants were of Hispanic ethnicity. As expected, there were significant differences among the prevalence of various metabolic disorders, such as hypertension, type 2 diabetes, and dyslipidemia across the weight groups (Supplemental Table S1, https://links.lww.com/HC9/C348). The prevalence of PNPLA3 G-allele was 74% in normal weight, 75% in overweight, 70% in class I obesity, and 63% in class≥2 obesity categories. The prevalence of liver histological outcomes across 4 weight groups is shown in Table 1. In the comparison between normal and overweight categories, the prevalence of definite steatohepatitis (50% vs. 50%), NAFLD Activity Score (4.1 vs. 4.1), advanced fibrosis (26% vs. 26%), or cirrhosis (8% vs. 8%). Interestingly, the prevalence of these histological phenotypes was progressively higher in class 1 and class 2 obesity categories, suggesting that BMI ~30 kg/m2 is a transition point for histological severity in this large cohort. TABLE 1 - Liver histology and PNPLA3 genotypes across the BMI categories BMI category (kg/m2)a Participant characteristics <25 25–29.9 30–34.9 ≥35 Total p N 107 744 1128 1407 3386 Biopsy length (cm), mean (SD) 21.7 (10.7) 20.7 (9.8) 20.9 (10.0) 20.8 (9.9) 20.9 (10.0) 0.87 Liver histology Fibrosis <0.001 0—none 33 (31) 227 (31) 258 (23) 272 (19) 790 (23) 1—a1, 1b, 1c 34 (32) 194 (26) 317 (28) 342 (24) 887 (26) 2—zone 3 and periportal 13 (12) 141 (19) 211 (19) 266 (19) 631 (19) 3—bridging 17 (16) 121 (16) 224 (20) 343 (24) 705 (21) 4—cirrhosis 8 (8) 60 (8) 112 (10) 178 (13) 358 (11) Steatohepatitis diagnosisb <0.001 NAFLD (not NASH) 33 (31) 221 (30) 245 (22) 269 (19) 768 (23) Suspicious/borderline/indeterminate 20 (19) 154 (21) 219 (19) 253 (18) 646 (19) Definite 54 (50) 369 (50) 663 (59) 884 (63) 1970 (58) NAFLD activity score (NAS)b, mean (SD) 4.1 (1.6) 4.1 (1.7) 4.4 (1.6) 4.4 (1.6) 4.3 (1.6) <0.001 PNPLA3 rs738409 <0.001 CC 20 (26) 140 (25) 250 (30) 378 (37) 788 (32) CG 38 (49) 264 (47) 362 (44) 453 (44) 1117 (45) GG 19 (25) 159 (28) 213 (26) 204 (20) 595 (24) Note: p values determined from ordinal logistic regression.aIf Asian, BMI<23=normal weight, 23–27.4=overweight, 27.5–32.4=obese, ≥32.5=severe obesity.bNASH CRN histology nomenclature has not been changed to the new SLD-MASLD nomenclature.Abbreviation: BMI, body mass index. When normal weight MASLD with (n=25) and without advanced fibrosis (n=82) were compared, individuals with normal weight with advanced fibrosis were older (58.5 vs. 49.8 y, p=0.005) with higher rates of type 2 diabetes (48% vs. 19%, p=0.003), hypertension (48% vs. 26%, p=0.04), and metabolic syndrome (46% vs. 19%, p=0.009). Notably, patients with normal weight with advanced fibrosis had higher rates of homozygosity for PNPLA3 G-allele when compared to those without advanced fibrosis (44% vs. 17%, p=0.02). However, when CG and GG alleles combined, the rate of G-allele carriage in the advanced fibrosis group was 78% versus 72% in the early fibrosis group (p=0.65). All-cause mortality and liver-related outcomes among 4 BMI categories are shown in Supplemental Table S2, https://links.lww.com/HC9/C348. Three individuals in the normal weight category died (0.53 per 100 person-years), but none had liver-related death. Only 2 individuals in the normal weight category had a hepatic decompensation event (0.36 per 100 person-years). While higher BMI categories were associated with higher rates of overall death and an increase in MELD to ≥15, they were not associated with liver-related deaths. Interestingly, there were fewer variceal bleeding events in higher BMI categories (Supplemental Table S2, https://links.lww.com/HC9/C348). DISCUSSION In this multicenter, biopsy-proven MASLD cohort, we found that histologic severity was similar between normal and overweight MASLD groups. In contrast, histologic severity increased once BMI reached ≥30 kg/m2. This inflection point is consistent with the well-documented role that obesity plays in MASLD progression, but challenges the idea that individuals in the overweight category are in an intermediate risk state when compared to lean and obese groups. Our results reflect those of prior studies demonstrating that lean MASLD and more broadly nonobese MASLD are not benign phenotypes, as evidenced by the prevalence rates of advanced fibrosis in these categories.5–7 In our study, 1 in 4 adults with normal weight with MASLD had advanced fibrosis, and 1 in 12 adults with normal weight with MASLD had cirrhosis. Along with metabolic risk factors, there are genetic risk factors to consider. In this cohort, there was enrichment of the PNPLA3 G-allele in patients with nonobese MASLD compared to obese patients with BMI≥30 kg/m2. The association between the PNPLA3 rs738409 (p.I148M) missense variant and the increased risk of hepatic steatosis, advanced fibrosis, and cirrhosis has been well described in the literature.8–10 Our study reflects this, as the prevalence of the PNPLA3 G-allele was also higher in patients with nonobese MASLD with advanced fibrosis compared to those without advanced fibrosis, which underscores the importance of genetic risk factors in the progression of disease independent of normal BMI. One important limitation of this analysis, inherent to all histology-based studies, is the selection bias of individuals who are referred to tertiary care centers, undergo liver biopsies, and participate in clinical studies. Another limitation pertains to the liver-related events outlined in Supplemental Table S2, https://links.lww.com/HC9/C348. There were a small number of liver-related events, and there was no adjustment for confounders, so these data need to be interpreted with caution. Moreover, we found that the rates of advanced fibrosis and cirrhosis are similar in patients with normal weight and overweight MASLD, and their histological severity worsens after BMI reaches 30 kg/m2. In those with normal weight, there are several metabolic and genetic factors that drive the progression of fibrosis irrespective of weight. These findings suggest that lean and overweight MASLD represent a shared phenotype and challenge the clinical distinction of “lean MASLD” as a unique entity.
Miller et al. (Thu,) studied this question.