Prevalence of Fibrosis in Hepatic Explants and Biopsies from Individuals with Urea Cycle Disorders

Abstract

Background:

Various forms of liver disease have been increasingly reported in individuals with urea cycle disorders (UCDs). In this study, we performed the first systematic and standardized histopathological assessment of the prevalence of fibrosis and steatosis in a large sample of hepatic explants and biopsies from individuals with UCDs at two liver transplantation centers.

Methods:

Sixty-seven hepatic tissue samples from 66 individuals with UCDs were staged by two pathologists for hepatic fibrosis and steatosis using standard scoring systems at two large liver transplantation centers in the United States. Histopathological findings were correlated with clinical parameters, including UCD type, laboratory parameters, and imaging findings.

Results:

Overall, 23% (n=15) of individuals demonstrated clinically significant hepatic fibrosis (≥ F2 according to Metavir staging). Of these, 12 were diagnosed with argininosuccinate lyase deficiency (ASLD) leading to an 80% prevalence of clinically significant fibrosis in this disorder in this cohort. Eighteen percent of the patients (n=12) had microvesicular and/or macrovesicular hepatic steatosis. No clinical parameters including routine laboratory testing or imaging were significantly associated with clinically significant hepatic fibrosis in individuals with ASLD.

Conclusion:

In this large study of hepatic histopathology in UCDs, our findings demonstrate a high prevalence of clinically significant hepatic fibrosis in ASLD. These findings emphasize the importance of monitoring for liver disease and promoting liver health in UCDs and may have implications for long-term monitoring for individuals with ASLD.

1. Introduction

Urea cycle disorders (UCDs) are inborn errors of hepatic metabolism that impact at least 1 in 35,000 individuals¹ and lead to reduced hepatic urea production and an associated risk for life-threatening elevations in ammonia. This group of disorders includes genetic deficiencies of one of the six enzymes required for urea synthesis or one of several transporters that shuttle urea cycle intermediates. Improvements in the treatment of hyperammonemia have led to improved survival and better recognition of long-term complications such as liver disease that had previously remained under-recognized.^2–4 Liver disease in UCDs can manifest in many ways including acute liver failure,^5,6 neonatal cholestasis,^7,8 hepatomegaly,^9,10 elevated hepatic transaminases,^3,9–11 hepatic fibrosis and cirrhosis leading to chronic liver failure,^10,12–14 and even hepatic tumors, such as hepatic adenoma and hepatocellular carcinoma.^9,15–19

Evaluation of liver explants from case reports or small cohorts of individuals with UCDs has identified many histopathological features including hepatic fibrosis and cirrhosis,^5,14,20–22 cholestasis,^17,21–23 hepatocyte enlargement,^20,21,23 nuclear glycogen and cytoplasmic hepatic glycogen accumulation,^{10,20,21,23,24} macrovesicular or microvesicular steatosis,^{10,14,20,21,23,24} increased or abnormal smooth surface endoplasmic reticulum,^12,23 and mitochondrial abnormalities.^20,23,24 However, results from such studies are likely to have a publication bias wherein abnormal findings are reported more frequently. To date, no large-scale, systematic studies of hepatic histopathology in individuals with UCDs have been performed. Thus, the overall prevalence of hepatic pathology in individuals with UCDs remains unknown.

Here, we performed a standardized histopathological assessment of the prevalence of fibrosis, steatosis, and excess glycogen accumulation in a large sample of hepatic explants and biopsies from children and adults with UCDs at two large pediatric liver transplantation centers in the United States. In addition, we correlated histopathological findings with clinical parameters, including diagnostic laboratory results and markers of disease severity.

Materials and Methods

Human Subjects Considerations.

This study was conducted by the National Institutes of Health (NIH) Rare Diseases Clinical Research Network (RDCRN) Urea Cycle Disorders Consortium (UCDC). The study protocol was approved by the NIH, Institutional Review Board (IRB) of record for the UCDC at Children’s National Medical Center, and IRBs of participating sites, Baylor College of Medicine (BCM) and Stanford University School of Medicine. Samples collected prior to 9/1/2021 were analyzed retrospectively with a waiver of consent at BCM and Stanford. Informed consent was obtained for all samples collected prospectively after 9/1/2021 at BCM. Stanford only participated in the retrospective portion of the study.

Study Procedures.

The inclusion criteria for this study were 1) diagnosis of a primary UCD based on enzyme activity, DNA testing, or metabolite analysis according to previously published criteria²⁵ and 2) history of liver transplantation or liver biopsy. The following primary UCDs were included in this study: arginase deficiency (ARG1D), argininosuccinate lyase deficiency or argininosuccinic aciduria (ASLD), argininosuccinate synthetase deficiency or citrullinemia (ASS1D), ornithine transcarbamylase deficiency (OTCD), carbamoyl phosphate synthetase 1 deficiency (CPS1D), N-acetylglutamate synthase deficiency (NAGSD), and citrullinemia type 2 (CITR). The diagnosis of a UCD may have occurred through newborn screening, family history, or symptomatic presentation. Thus, both early onset and late onset forms of disease were included in this cohort. The exclusion criteria were 1) unavailability of histopathology reports from the liver tissue sample or unavailability of liver tissue or slides from the biopsy or explant (retrospective arm) or anticipated inability to obtain pathology reports, liver tissue, tissue blocks, or pathology slides after liver biopsy or transplantation (prospective arm), and 2) a known history of a secondary cause of liver disease, such as chronic viral hepatitis, autoimmune liver disease, short gut, small bowel syndrome, alcohol-related liver disease, or total parenteral nutrition (TPN)-related cholestatic disease.

Data Collection and Storage.

All laboratory, pathology, and imaging data were collected as part of routine clinical care or as part of other Urea Cycle Disorders Consortium studies. The data were stored in a secure REDCap^® database managed by the Data Management and Coordinating Center (DMCC) of the NIH RDCRN^26,27. For each participant enrolled in the study, a chart review was performed to collect medical history including age, type of UCD, peak ammonia level, number of hyperammonemic episodes, dietary prescription for protein intake (includes both natural protein and essential amino acid supplements), and medications prescribed at the time of tissue collection. Participants were categorized as taking a nitrogen-scavenging agent if the medication list at the time of tissue collection included sodium benzoate, sodium phenylbutyrate, or glycerol phenylbutyrate. Dietary prescription for protein intake is reported as percent of the dietary reference intake for age²⁸. For children (2–20 years of age), BMI category was determined using the BMI %tile for age according to the definitions from the U.S. Centers for Disease Control and Prevention (CDC, www.CDC.gov) with <5^th %ile indicating underweight, 5^th %tile to less than 85^th %tile indicating normal weight, 85^th %tile to less than 95^th %tile indicating overweight, and 95^th%tile or greater indicating obesity. Similarly, the adult definitions of overweight and obesity from the CDC (www.CDC.gov) were used for individuals 20 years and older with a BMI less than 18.5 kg/m² indicating underweight, between 18.5 kg/m² and 24.9 kg/m² indicating normal weight, between 25 kg/m² and 29.9 kg/m² indicating overweight, and 30 kg/m² or above indicating obesity.

Laboratory parameters for routine markers of liver disease (total bilirubin, direct bilirubin, aspartate aminotransferase or AST, alanine aminotransferase or ALT, gamma-glutamyl transferase or GGT, platelet count, and international normalized ratio or INR) that were available in medical records and that had been collected prior to or at the time of liver tissue collection were recorded. Likewise, reports from imaging studies (liver ultrasound, CT scan, and/or MRI) that were available in the medical records and that had been performed prior to tissue collection were recorded. The AST-to-platelet (APRI) ratio was calculated with the following standard equation: APRI = (AST in units per liter/AST upper limit of normal in units per liter)/platelet count (10⁹ per liter).²⁹ For the APRI calculation, 40 units per liter was used as the upper limit of normal for AST.^30,31 For the purposes of this study, a participant was considered to be on “high dose” arginine if the arginine dose was > 250 mg/kg (if body weight was < 20 kg) or > 8.8 g/m² (if body weight was > 20 kg).³² All data points were not available for all individuals.

Imaging Review.

Regardless of the imaging modality (ultrasound, MRI, CT), the following findings listed on clinical imaging reports were considered important liver-related abnormalities: hepatomegaly, coarseness, increased echogenicity, hypertrophy of the caudate lobe, nodularity, splenomegaly, collateral vessels, varices, splenorenal shunt, reversal of portal blood flow, or recanalized umbilical vein.

Pathology Review.

Hepatic tissues or slides were available for all individuals in this study. If more than one sample was available from a single individual, all available samples were included in this study. Hepatic sections stained with hematoxylin and eosin (H&E), Masson’s trichrome, and periodic acid-Schiff (PAS) with and without diastase, if available, were reviewed by one of the two hepatopathologists. The samples were scored for fibrosis, steatosis, and the presence of tumors. In addition, a qualitative assessment of the presence or absence of increased glycogen was performed for each sample. Scoring for fibrosis was performed using the Metavir (F0–4) scoring system.³³ Scoring for steatosis was performed using a standard scoring system (S0–3) for steatotic liver disease.³⁴ Fibrosis and steatosis scoring systems used in this study are presented in Supplementary Tables 1 and 2.

Statistical Analysis.

Demographics and clinical characteristics were summarized for the overall sample and the UCD subtypes. Descriptive statistics were reported as median and range for numerical variables and frequency and proportion for categorical variables. Wilcoxon rank sum exact test for numerical covariates and Fisher’s exact test for categorical covariates were used to compare the group differences between clinically significant and non-clinically significant fibrosis in the ASLD group. A two-sided test with a significance level of 0.05 was carried out for all data analyses. The data were analyzed using SAS for Windows version 9.4 (SAS Institute Inc., Cary, NC, USA). Unless otherwise specified, when analyzing the individual with two available samples, the most recent sample was used. Percentages may not add up to 100% because of rounding.

Results

Study Population.

Tables 1 and 2 summarize the characteristics of the study population. A total of 61 hepatic explants and 6 hepatic biopsy samples were included in the analysis. Five of the six biopsies were performed for diagnostic purposes, and one was performed for follow-up of prior imaging findings. These samples were collected between 1996 and 2023 from 66 individuals with UCDs ranging in age from 6 days to 41.6 years. A total of 44 samples (from 43 individuals) were reviewed at BCM whereas 23 samples (from 23 individuals) were reviewed at Stanford. OTCD (males) was the most common diagnosis in our cohort (n=20, 30%) followed by ASLD (n=15, 23%). None of the individuals in this cohort had CITR or NAGSD. Of the individuals with available data for hyperammonemic events (n=59), all but one had a history of one or more episode(s) of hyperammonemia. Of the 58 individuals with available concomitant medications, 84% (n=49) were taking at least one nitrogen-scavenging agent at the time of tissue collection. Of the nine individuals who were not taking nitrogen-scavenging agents, two had OTCD, and the biopsy for one of these individuals was performed prior to the UCD diagnosis. Six individuals who were not taking a nitrogen-scavenging agent had ASLD or ASS1D managed with arginine, and one had ARG1D. At least a subset of laboratory values were available for 55 participants.

Table 1.

Characteristics of participants from whom explants/biopsies were examined

	Total	OTCD		CPS1D	ASS1D	ASLD	ARG1D
		Male	Female
Number of samples	67	211	7	11	8	15	5
Number of explants	61	18	6	11	7	15	4
Number of biopsies	6	3	1	0	1	0	1
Age range in years	0.02 – 41.6	0.02 – 4.5	1.6 – 19.3	0.3 – 41.6	0.4 – 21.7	0.7 – 38.8	1.2 – 16.2
Sex (Male/Female)	35 / 31	20 / 0	0 / 7	4 / 7	1 / 7	6 / 9	4 / 1
# overweight or obese / total number of individuals ≥ 2 years with data available 2	15 / 22 (68%)	0 / 2 (0%)	3 / 4 (75%)	2 / 3 (66%)	2 / 3 (66%)	6 / 6 (100%)	2 / 4 (50%)

¹One biopsy sample and one explant sample came from the same male with OTCD.

²BMI was available for 22 of the 29 individuals who were 2 years of age and older.

Table 2.

Hyperammonemia and use of nitrogen-scavenging agents in participants from whom biopsies/explants were examined

	Total	OTCD		CPS1D	ASS1D	ASLD	ARG1D
		Male	Female
Number (%) ≥1 HAE (n=59 with available data)	58 (98%)	19 (100%)	7 (100%)	10 (100%)	7 (100%)	11 (92%)	4 (100%)
Number (%) taking nitrogen-scavenging agents (n=58 with available data)	49 (84%)	17 (94%)	6 (86%)	9 (100%)	6 (86%)	8 (62%)	3 (75%)
% DRI1 for protein intake (n=45 with available data)	21 – 179%	55 – 133%	21 – 158%	63 – 113%	66 – 142%	47 – 179%	53 – 153%

¹Dietary reference intake for protein based on age²⁸

Hepatic Fibrosis.

Samples from 26 participants (39%) showed evidence of hepatic fibrosis (Table 3). F1 fibrosis was observed in seven participants (11%; n=1 CPS1D, n=2 OTCD females, n=4 OTCD males). F2 fibrosis was observed in six participants (9%, n=5 ASLD, n=1 OTCD male). F3 fibrosis was observed in eight participants (12%, n=1 ARG1D, n= 6 ASLD, n=1 OTCD male). F4 or cirrhosis was observed in one participant with ASLD. Other forms of hepatic fibrosis were observed in four samples (6%; n=1 perisinusoidal fibrosis in ARG1D, 1 pericentral venular fibrosis in ASLD, 1 pericentral venular fibrosis in ASS1D, and n=1 sinusoidal fibrosis in OTCD male). Fibrosis ≥ F2 is typically considered to be clinically significant. Using this definition, 23% (n=15) of the individuals in this cohort had clinically significant fibrosis. Interestingly, 80% (n=12, age range: <1 year to 38 years) of all participants with ASLD (n=15) in this cohort had clinically significant fibrosis compared to 20% (n=1) of the participants with ARG1D (age: 7.4 years), and 10% (n=2) of the males (<1 years of age) with OTCD. None of the participants with ASS1D or CPS1D and no females with OTCD had clinically significant fibrosis.

Table 3.

Prevalence of hepatic fibrosis and steatosis in the cohort

	Total (n=66)	OTCD			CPS1D (n=11)	ASS1D (n=8)	ASLD (n=15)	ARG1D (n=5)
		Male (n=20)		Female (n=7)
Hepatic Fibrosis
F0	40 (61%)	13 (65%)	5 (71%)		10 (91%)	7 (88%)	2 (13%)	3 (60%)
F1	7 (11%)	4 (20%)	2 (29%)		1 (9%)	0 (0%)	0 (0%)	0 (0%)
F2	6 (9%)	1 (5%)	0 (0%)		0 (0%)	0 (0%)	5 (33%)	0 (0%)
F3	8 (12%)	1 (5%)	0 (0%)		0 0%)	0 (0%)	6 (40%)	1 (20%)
F4	1 (2%)	0 (0%)	0 (0%)		0 (0%)	0 (0%)	1 (7%)	0 (0%)
F Other	4 (6%)	1 (5%)	0 (0%)		0 (0%)	1 (13%)	1 (7%)	1 (20%)
Hepatic Steatosis
S0 (<5%)	54 (82%)	18 (90%)	5 (71%)		10 (9%)	8 (100%)	11 (73%)	2 (40%)
S1 (5–33%)	8 (12%)	1 (5%)	1 (14%)		1 (9%)	0 (0%)	3 (20%)	2 (40%)
S2 (33–66%)	2 (3%)	1 (5%)	0 (0%)		0 (0%)	0 (0%)	0 (0%)	1 (20%)
S3 (>66%)	1 (2%)	0 (0%)	0 (0%)		0 (0%)	0 (0%)	1 (7%)	0 (0%)
S Other	1 (2%)	0 (5%)	1 (1%)		0 (0%)	0 (0%)	0 (0%)	0 (0%)
Tumor	0 (0%)	0 (0%)	0 (0%)		0 (0%)	0 (0%)	0 (0%)	0 (0%)
% Increased Hepatic Glycogen	65 (99%)	19 (95%)	7 (100%)		11 (100%)	8 (100%)	15 (100%)	5 (100%)

¹Percentages may not add up to 100% due to rounding.

²For this table, total number of individuals was used. For the individual with two samples, the sample that was most recently collected was used.

The 12 participants with ASLD and clinically significant hepatic fibrosis ranged in age from <1 year of age to 38 years, with 9 of these 12 being less than 6 years of age. Liver tissue samples from these individuals with ASLD were all hepatic explants collected between 1996–2022, and only a single tissue sample was available for each of these individuals. Representative pathology from several samples with clinically significant fibrosis are shown in Figure 1. Of these 12 participants, information about prior hyperammonemia was available for nine, and all nine had one or more episode(s) of hyperammonemia. However, at least one hyperammonemia episode was also reported in two of the three individuals with ASLD and no clinically significant hepatic fibrosis. Likewise, concomitant medications at the time of transplantation or biopsy were available for 10 of these 12 participants, and of these 10 participants, seven were taking at least one nitrogen scavenging agent. Lastly, arginine dosing was available for nine of these 12 participants, and only two were taking “high dose” arginine at the time of transplantation. Molecular variant information was only available for a subset of these individuals (Supplementary Table 3).

Figure 1. Hepatic Fibrosis in Individuals with ASL Deficiency.

A. H&E stain of hepatic explant tissue from a 10-month-old with ASL deficiency with F2 fibrosis (4x magnification). B. H&E stain of hepatic explant tissue from a 5-year-old with F2 fibrosis (4x magnification). C. H&E stain of hepatic explant tissue from a 15-month-old with ASL deficiency and F4 fibrosis (cirrhosis, 4x magnification). D. Trichrome stain of hepatic explant tissue from a 38-year-old with ASL deficiency (4x magnification) showing F3 fibrosis. E. Trichrome stain of hepatic explant tissue from a 2-year-old with ASLD showing F3 fibrosis (4x magnification). F. Trichrome stain of hepatic explant tissue from a 4-year-old with ASL showing F2 fibrosis (4x magnification).

Results from a hepatic ultrasound performed between < 1 month and up to 8 months prior to tissue collection were available for 11 participants with ≥ F2 fibrosis and any UCD diagnosis, and the reports were abnormal in 9 participants. The most common abnormalities were hepatomegaly (n=7) and increased echogenicity (n=4). For one individual with ASLD and F3 fibrosis on tissue examination, the ultrasound was interpreted as “probable hepatic cirrhosis.” However, hepatomegaly, abnormal echogenicity, and/or appearance of hepatic steatosis were also reported in individuals (n=12) with various UCDs but without clinically significant fibrosis by tissue examination. Notably, there were no specific reports of coarseness, nodularity, reversal of portal flow, collateral vessels or splenomegaly.

None of the participants with ≥ F2 fibrosis had an abdominal MRI report available for review. Only two participants (both with F3 fibrosis) had abdominal CT imaging reports available from studies performed prior to tissue collection. The CT scan for one of these individuals (ARG1D) reported only a liver size at the upper limits of normal (2 months prior to tissue collection). The CT scan of the other participant (ASLD) revealed an enlarged nodular liver. However, this scan was performed nearly 15 months prior to tissue collection.

No statistically significant associations were found between the categories of hepatic fibrosis in individuals with ASLD (i.e., clinically significant fibrosis vs. non-clinically significant fibrosis) and the following covariates: age, AST, ALT, GGT, INR, APRI, platelet count, arginine dose (mg/kg), history of hyperammonemia, abnormal imaging, nitrogen scavenging agent, and study site (Table 4; Supplementary Table 4). Alpha fetoprotein (AFP) was available in six individuals with ASLD and F2 or greater fibrosis and was either within normal limits or just above the normal range.

Table 4.

Covariates in individuals with ASLD with and without clinically significant fibrosis

Characteristics (total number with available data)	ASLD with F0, F1 or other Fibrosis	ASLD with F2, F3, F4 Fibrosis
Number of transplants/biopsies (n=15)	3 / 0	11 / 1
Previous hyperammonemia event (%, n=12)	2 (67%)	9 (100%)
Age at tissue collection (years, median, range)	3.7 (0.9 – 5.6)	3.4 (0.7 – 38.8)
AST (U/L, median with range, n=11)	91 (34 – 148)	80 (36 – 378)
ALT (U/L, median with range, n=11)	161 (15 – 306)	84 (46 – 797)
GGT (U/L, median with range, n=10)	19.5 (16 – 23)	36 (17 – 72)
INR (median with range, n=11)	1.2 (1.1 – 1.3)	1.1 (1 – 1.4)
Platelets (median with range, n=12)	399.5 (306.0 – 493.0)	409.5 (173.0 – 674.0)
AST/platelet ratio (median with range, n=11)	0.6 (0.3 – 0.8)	0.6 (0.3 – 2.3)
Abnormal hepatic ultrasound (n=11)	1 (50%)	8 (89%)
Number taking nitrogen scavenging agent (%, n=13)	1 (33%)	7 (70%)
%DRI for protein intake 1	Not available	47 – 179%

¹Dietary reference intake for protein based on age. Of the 3 individuals with F0, F1, or other fibrosis, % DRI for protein was only available for one individual and was 137%.

Hepatic Steatosis.

Samples from 12 participants (18%) showed some form of hepatic microvesicular or macrovesicular steatosis (Table 3). S1 to S3 steatosis was reported in both distal (n=4 with ASLD; n=3 with ARG1D) and proximal disorders (n=1 adult with CPS1D, 2 males with OTCD, and 1 female with OTCD). All four samples with steatosis from participants with ASLD also demonstrated F2 or F3 fibrosis, whereas two of the three samples with steatosis from individuals with ARG1D demonstrated fibrosis (F3 or perisinusoidal fibrosis). Of the 11 samples with S1 to S3 steatosis, only three demonstrated moderate or severe steatosis. Individuals with moderate or severe steatosis included two male infants (one with OTCD and one with ASLD) and one teenager with ARG1D who had a body mass index in the normal range for age.

Other Findings.

No tumors were identified in any of the 67 samples. Qualitatively, increased hepatic glycogen was noted in all but one sample (Supplementary Figure 1). The only sample that did not show increased hepatic glycogen was from a neonate with OTCD. In addition, of the individuals who had abdominal ultrasound findings available prior to transplantation or biopsy, 3 transplanted infants had a thrombus in the left portal vein on the ultrasound report prior to transplantation.

Discussion

With the implementation of universal newborn screening programs, the availability of hemodialysis or renal replacement therapy and intravenous nitrogen-scavenging medications for the treatment of hyperammonemia, advancements in dietary therapies and oral nitrogen-scavenging medications for the prevention of hyperammonemia, and the increased awareness of UCDs, the overall survival of individuals with UCDs has improved. Thus, one focus of individuals and families with UCDs, as well as their health care providers, is to optimize outcomes by preventing morbidity due to long-term complications. Liver disease is one complication that can have a direct impact on the care of individuals with UCDs. Whereas acute liver failure in OTCD can be a life-threatening condition,^5,6 chronic liver disease and hepatic fibrosis reported in UCDs can decrease hepatic reserve and have a negative impact on an already compromised urea cycle.^2,11,21 Furthermore, the efficacy of the most commonly used nitrogen-scavenging medications (e.g., sodium phenylbutyrate and glycerol phenylbutyrate) is dependent on effective beta-oxidation and conjugation in the liver. Thus, understanding the prevalence of liver fibrosis in UCDs is an important clinical question as it can influence the monitoring of individuals with UCDs and may support referral to a hepatologist. However, estimating the prevalence of fibrosis is difficult because of the invasive nature of liver biopsies.

By leveraging a cohort of samples from two large academic centers with expertise in treating individuals with UCDs including liver transplantation, we conducted this large histopathological study of liver explants and biopsies obtained during routine clinical care from individuals with UCDs. Our study demonstrated a very high prevalence (80%) of clinically significant fibrosis in individuals with ASLD. Whereas hepatic fibrosis and cirrhosis are known complications of ASLD,^{4,13,21,22,35,36} this is the first estimate of their prevalence among individuals with ASLD who have undergone liver biopsy or liver transplantation. While it is likely that the prevalence of fibrosis might increase with age, we only had hepatic tissue at one time point for all individuals. Thus, the rate of progression of this phenotype in ASLD could not be ascertained. However, interestingly, the youngest individual with hepatic fibrosis was eight months of age, alluding to the fact that hepatic fibrosis can develop early during the disease course.

The cause of the high prevalence of hepatic fibrosis in individuals with ASLD relative to other UCDs is unclear. One possible explanation is the accumulation of argininosuccinic acid (ASA) or its downstream metabolites in the liver because ASLD is the only UCD in which this metabolite accumulates. The hypothesis that ASA might contribute to the pathogenesis of liver disease is supported by a small randomized crossover clinical study evaluating the effects of a “high-dose” arginine vs. “low-dose” arginine plus sodium phenylbutyrate on hepatic functions in ASLD.³² In that study, ASA levels showed positive correlation with plasma levels of alanine and aspartate aminotransferases.³² In the current study, although we only identified two participants with ASLD who were on a “high dose” of arginine at time of transplantation, the dose of arginine at time of transplantation may not reflect long-term dosing. Thus, we were unable to draw conclusions regarding the role of arginine dosing and/or ASA levels and fibrosis or the impact of any other therapies on this finding. Alternatively, the role of ASL in nitric oxide production³⁷ or alterations in glutathione metabolism, which have recently been reported in ASLD, are other potential contributing factors to this phenotype.³⁸ Further studies are necessary to investigate the underlying mechanisms for these findings.

Likewise, the temporal progression of hepatic fibrosis in this population is unclear. Clinically significant hepatic fibrosis was detected in an infant less than 1 year of age and in an adult in this study. However, hepatic biopsies are not typically a component of routine clinical care for individuals with UCDs³⁹, and thus, the only liver sample available for most individuals in this study was their explant or single liver biopsy sample. Thus, future work should focus on exploring the utility of noninvasive imaging and biomarkers for liver disease in this population.

Although the cause of the increased prevalence of hepatic fibrosis and cirrhosis in ASLD is currently unclear, this high prevalence of hepatic fibrosis and cirrhosis has implications for the clinical care of individuals with ASLD. Despite the high prevalence of fibrosis, routine clinical laboratory tests and imaging were not predictive of the degree of liver disease. For instance, three of the individuals with ASLD and F3 or F4 fibrosis had only mild elevations (e.g., less than twice the upper limit of normal) in AST (36–65 U/L) and ALT (46–63 U/L). Likewise, although all six individuals with ASLD and F3 or F4 fibrosis with available ultrasound had abnormal hepatic imaging, hepatomegaly was the only abnormality noted in two of these individuals. Thus, until better noninvasive tools are available to diagnose (or predict) clinically significant fibrosis or cirrhosis, individuals with ASLD should be counseled regarding this potential risk and referral to hepatology for monitoring should be considered. Moreover, these individuals and their providers should be aware of early signs of advanced liver disease such as organomegaly, declining platelet count, persistent hepatitis, or easy bruising. Individuals with this diagnosis should be encouraged to be vaccinated against hepatotropic viruses and counseled to avoid medications associated with hepatotoxicity. Likewise, other risks such as alcohol consumption should be avoided and a healthy diet, albeit low in protein as prescribed by the primary metabolic physician, and healthy body weight should be encouraged given that these are independent risk factors for liver disease. In addition, liver function should be monitored, and our findings support the guidelines³⁹ that recommend hepatic ultrasonography for monitoring the liver in UCDs, especially as reports of hepatocellular carcinoma and hepatic adenoma continue to emerge in this group of disorders. Finally, the extent and severity of liver disease in ASLD may have implications for liver transplantation decision-making in this disorder.

We also observed clinically significant hepatic fibrosis in one individual with ARG1D and two males with OTCD less than 1 year of age. Hepatic fibrosis and cirrhosis have been reported previously in arginase deficiency^17,40,41 and may reflect the accumulation of arginine and its metabolites in the liver. Likewise, hepatic fibrosis and cirrhosis have been reported previously in citrin deficiency^18,42,43 with rare cases of this finding in ASS1D,¹² CPS1D,^10,44 and OTCD.^10,21

The findings of our study should be interpreted within the context of the strengths and limitations of the dataset. The large sample size for an orphan disease, systematic scoring of liver biopsy samples using standard methodologies by pathologists, and collection of clinical covariate data according to a manual of operations attest to the quality of our data. However, this study has some limitations. First, data regarding the reasons for transplantation or biopsy were not easily retrieved from the medical records of all participants with ASLD and F2 or greater fibrosis. However, for the five individuals for whom the reason for transplant was retrievable, liver disease was the reason for transplant for only one. Other reasons for transplantation included frequent hyperammonemia and quality of life/desire for diet liberalization. The lack of prior biopsies in all but one case suggests that the extent of liver fibrosis was likely unknown in most individuals. Moreover, for participants with available data, there was no evidence of thrombocytopenia or splenomegaly, factors that might suggest advanced liver disease. Thus, advanced liver disease was not the likely reason for transplantation in most individuals. However, the degree of fibrosis was suspected in at least two participants prior to biopsy. One adult participant with ASLD had a CT scan that showed an enlarged and nodular liver and FibroTest^™ that correctly predicted F3 fibrosis, and the reason for transplantation in this individual was suspected liver disease. Likewise, an ultrasound from another individual with ASLD suggested probable cirrhosis (tissue sample showed F3 fibrosis). Second, for some of the samples, particularly older ones, all clinical data were not available for analysis. Third, the study population may have referral bias as it might represent a more severely affected subset of individuals with ASLD evaluated at these tertiary transplant centers. However, although 91% of the individuals with ASLD had at least one hyperammonemia event prior to transplantation, the ammonia level ranged from 70 to > 1000 μmol/L suggesting a wide range in the severity of urea cycle dysfunction in our cohort. Moreover, the highest ammonia levels recorded for three individuals with ASLD and F2 or greater fibrosis were in the range of 70–102 μMol/L providing further data supporting a wide range of disease severity for ASLD. In addition, metabolic-associated steatotic liver disease (MASLD, formerly known as non-alcoholic fatty liver disease) was not an exclusion criterion for this study. Therefore, some histopathologic findings may be secondary to this entity especially given the high prevalence of overweight and obesity in this cohort. However, there is no reason to suspect that this diagnosis, if present in our population, would be more common in ASLD than in other UCDs, and thus, it is unlikely to explain the high prevalence of hepatic fibrosis in ASLD in this study especially since the prevalence of overweight and/or obesity was high across the entire cohort. Lastly, we had small sample sizes for some types of UCDs due to the rarity of the disorders and referral practices regarding transplantation at these two institutions. Similarly, the analysis of covariates in ASLD should be considered exploratory because the analysis was likely underpowered due to the relatively small sample size of the ASLD group (N=15) and the large proportion of individuals with clinically significant fibrosis in that group.

In summary, our study demonstrated a high prevalence of clinically significant hepatic fibrosis in individuals with ASLD who have undergone liver transplantation or liver biopsy. Although further studies of the underlying mechanism and risk factors for hepatic fibrosis in ASLD are necessary, counseling regarding the risk of liver disease and strategies for promoting liver health should be an important component of care for individuals with this disorder. Moreover, referral to hepatology should be considered for monitoring of the extent and severity of clinically significant liver disease. Finally, given the number of emerging genetic therapies for UCDs that are specifically targeting the liver, an understanding of this baseline risk for hepatic fibrosis in UCDs will be important for interpreting hepatic pathology after exposure to these new therapies.

Supplementary Material

Supplementary Figure. Supplementary Figure 1. Hepatic glycogen accumulation in hepatic explant tissue from an individual with ASL Deficiency.

A. PAS stain without diastase of a liver explant section from a 38-year-old female with ASLD (same individual as Figure 1D). B. PAS stain with diastase of a section of a liver explant from the same individual shows absence of the PAS stain after treatment with diastase.

Abbreviations:

AFP

alpha fetoprotein

ALT

alanine aminotransferase

APRI

AST-to-platelet ratio

ARG1D

arginase deficiency

ASA

argininosuccinic acid

ASLD

argininosuccinate lyase deficiency or argininosuccinic aciduria

ASS1D

argininosuccinate synthetase deficiency or citrullinemia

AST

aspartate aminotransferase

BCM

Baylor College of Medicine

CPS1D

carbamoyl phosphate synthetase 1 deficiency

DMCC

Data Management and Coordinating Center

GGT

gamma-glutamyl transferase

H&E

Hematoxylin and Eosin

IRB

Institutional Review Board

MASLD

metabolic associated steatotic liver disease

NAGSD

N-acetylglutamate synthase deficiency

NIH

National Institutes of Health

OTCD

ornithine transcarbamylase deficiency

PAS

Periodic acid–Schiff

RDCRN

Rare Diseases Clinical Research Network

UCD

urea cycle disorders

UCDC

Urea Cycle Disorders Consortium