Are asymptomatic carriers of OTC deficiency always asymptomatic? A multicentric retrospective study of risk using the UCDC longitudinal study database

Abstract

Background

Ornithine transcarbamylase deficiency (OTCD) due to an X‐linked OTC mutation, is responsible for moderate to severe hyperammonemia (HA) with substantial morbidity and mortality. About 80% of females with OTCD remain apparently “asymptomatic” with limited studies of their clinical characteristics and long‐term health vulnerabilities. Multimodal neuroimaging studies and executive function testing have shown that asymptomatic females exhibit limitations when stressed to perform at higher cognitive load and had reduced activation of the prefrontal cortex. This retrospective study aims to improve understanding of factors that might predict development of defined complications and serious illness in apparent asymptomatic females. A proband and her daughter are presented to highlight the utility of multimodal neuroimaging studies and to underscore that asymptomatic females with OTCD are not always asymptomatic.

Methods

We review data from 302 heterozygote females with OTCD enrolled in the Urea Cycle Disorders Consortium (UCDC) longitudinal natural history database. We apply multiple neuroimaging modalities in the workup of a proband and her daughter.

Results

Among the females in the database, 143 were noted as symptomatic at baseline (Sym). We focused on females who were asymptomatic (Asx, n = 111) and those who were asymptomatic initially upon enrollment in study but who later became symptomatic sometime during follow‐up (Asx/Sym, n = 22). The majority of Asx (86%) and Asx/Sym (75%) subjects did not restrict protein at baseline, and ~38% of Asx and 33% of Asx/Sym subjects suffered from mild to severe neuropsychiatric conditions such as mood disorder and sleep problems. The risk of mild to severe HA sometime later in life for the Asx and Asx/Sym subjects as a combined group was ~4% (5/133), with ammonia ranging from 77 to 470 μM and at least half (2/4) of subjects requiring hospital admission and nitrogen scavenger therapy. For this combined group, the median age of first HA crisis was 50 years, whereas the median age of first symptom which included neuropsychiatric and/or behavioral symptoms was 17 years. The multimodal neuroimaging studies in female heterozygotes with OTCD also underscore that asymptomatic female heterozygotes with OTCD (e.g., proband) are not always asymptomatic.

Conclusions

Analysis of Asx and Asx/Sym females with OTCD in this study suggests that future evidence‐based management guidelines and/or a clinical risk score calculator for this cohort could be useful management tools to reduce morbidity and improve long‐term quality of life.

Median age of first HA for the combined groups of females with OTCD (asymptomatic and asymptomatic at baseline but became symptomatic later during follow‐up) was 50 years, whereas median age of first symptom which included neuropsychiatric and/or behavioral symptoms was 17 years. Multimodal neuroimaging studies underscore that asymptomatic female heterozygotes with OTCD are not always asymptomatic.

1. INTRODUCTION

Ornithine transcarbamylase deficiency (OTCD) is the most prevalent urea cycle disorder (UCD), affecting as many as 1 in 14,000 to 1 in 70,000 live births annually, and is responsible for moderate to severe hyperammonemia (HA) resulting in substantial morbidity and mortality (Brusilow & Maestri, 1996; Dionisi‐Vici et al., 2002; Keskinen et al., 2008; Summar et al., 2013). Pathogenic variants in the X‐linked OTC gene cause OTCD, but an estimated 10%–25% of affected individuals with biochemically and/or enzymatically confirmed OTCD have no detectable pathogenic OTC variants (Caldovic et al., 2015; Ruegger et al., 2014; Shchelochkov et al., 2009). While approximately 80% of heterozygous females remain apparently “asymptomatic” (Batshaw et al., 1986), about 20% are thought to manifest symptoms, the risk of which is related to the time of onset, type of OTC variant, and severity of OTCD in their offspring (McCullough et al., 2000). The variability of the nature and onset of symptoms makes risk factor evaluation in heterozygous females especially challenging due, in part, to the unpredictability of skewed X‐inactivation (Lichter‐Konecki et al., 1993; Seker Yilmaz et al., 2022; Yorifuji et al., 1998) as well as the imperfect predictability of severity based on enzyme activity in liver biopsies (Musalkova et al., 2018). Genotype–phenotype correlations with the functional impact of 1570 individual amino acid substitutions in OTC using a yeast‐based functional complementation assay for human OTCD have been evaluated (Caldovic et al., 2015; Lo et al., 2023). Furthermore, outside of moderate to severe hyperammonemic events, there is significant association (p = 0.029) between median ammonia levels ≥50 micromole/liter (μM) and decrease in performance on Beery visual motor integration (Lichter‐Konecki et al., 2023). Some heterozygous females exhibit severe HA as neonates or infants, while others manifest behavioral, hepatic, or neurologic symptoms including HA later in life or well into adulthood (Chongsrisawat et al., 2018; Gropman, Prust, et al., 2013). Compared to unaffected controls, asymptomatic female carriers have difficulty with fine motor tasks, cognitive executive function, and cognitive flexibility (Sprouse et al., 2014). Thus, the prevalence of heterozygous females with symptoms may be higher than previously thought and early identification of those at high risk for future OTC‐related illness is crucial for prevention of complications.

Previous studies have shown that “asymptomatic” female heterozygotes with OTCD exhibit deficits in cognitive skills, fine motor dexterity, and mathematical reasoning when compared to healthy controls (Gyato et al., 2004), thereby suggesting that certain areas of brain are quite sensitive to mild elevations in ammonia or glutamine even in the absence of severe HA crisis. Others observed significant differences in measures of executive function (e.g., Comprehensive Trail‐Making Test and Stroop) and motor ability (Purdue Assembly) between symptomatic participants, asymptomatic carriers, and controls in a total cohort of 81 (Sprouse et al., 2014). It has been previously estimated that 5%–20% of asymptomatic heterozygote females with OTCD ultimately develop symptoms (Brassier et al., 2015). Study of this group has shown that they manifest a similar albeit milder picture on magnetic resonance spectroscopy (MRS), diffusion tensor imaging (DTI), and neurocognitive testing, especially when stressed to perform at higher cognitive load (Sprouse et al., 2014).

Such neuroimaging and neurocognitive findings raise the question as to whether female heterozygotes with OTCD are truly asymptomatic at such low frequency, and/or whether conditions that increase stress or cognitive loading could lead to metabolic decompensation and/or neurocognitive symptoms. Cognitive loading refers to the amount of mental effort or resources required to complete a task or process information. It can be affected by various factors, such as the complexity of the task, and the context in which the task is performed to determine the individual's cognitive abilities. High cognitive loading can lead to mental fatigue, decreased performance, and errors. In the context of inborn errors of metabolism, cognitive loading has been studied to understand the impact of these disorders on cognitive function. For example, individuals with phenylketonuria, a disorder that affects the body's ability to metabolize phenylalanine, may experience increased cognitive loading during tasks that require sustained attention or working memory (Christ et al., 2010).

In this study, we focused on reportedly asymptomatic female heterozygotes with OTCD and female heterozygotes with OTCD who were initially asymptomatic upon enrollment in the Urea Cycle Disorders Consortium (UCDC) longitudinal study but who became symptomatic sometime during follow‐up, with inclusion of mothers of affected overtly symptomatic offspring (neonate or child) and females symptomatic or asymptomatic with or without a detectable pathogenic variant in OTC. The goals of this retrospective study were to (1) improve understanding of individual factors that may predict the risk of serious illness, (2) evaluate the risk for an asymptomatic patient to develop HA, and/or (3) define more subtle behavioral and cognitive symptoms in apparently asymptomatic heterozygote females with OTCD, for future more comprehensive studies. This effort could help establish a basis for the creation of a risk score calculator to predict the risk of serious illness and/or the development of evidence‐based management guidelines for female heterozygotes with OTCD. In addition, we present the case of a proband and her daughter to highlight the utility of multimodal neuroimaging studies and executive function testing in female heterozygotes with OTCD as outcome measures and for a better understanding of the brain pathophysiology in such patients, respectively, and to underscore that asymptomatic female heterozygotes with OTCD are not always asymptomatic.

2. MATERIALS AND METHODS

2.1. Data collection

This was a retrospective study using the UCDC longitudinal natural history database. For this study, the team adopted a phased approach with two consecutive phases. Phase 1 involved describing the symptomatology (along with incidence, prevalence, timing, and severity), exploring severity groupings based on hospitalization records, and evaluating univariate association of symptomatology with baseline characteristics. Phase 2 involved incorporating time‐varying neuropsychiatric diagnostics collected during follow‐up to assess symptomatology and severity. Data from the longitudinal study were retrospective and used the most up‐to‐date information at study initiation, with last database query being August 31, 2019. The first participant in the longitudinal study was consented on February 26, 2006, and had the baseline visit on March 13, 2006. Subjects included in the study were limited to females with OTCD. Furthermore, the longitudinal study does not collect control data on unaffected individuals or siblings/relatives who do not have biochemical or molecular confirmed UCD such as OTCD.

2.2. Definition of ‘symptomatic’

In the UCDC longitudinal study manual of operations (version June 2018), a “symptomatic” patient is defined as a patient diagnosed with a UCD and meeting the following criteria:

1. Clinical symptoms associated with ammonia of >100 micromole/liter (μM) and/or

2. Liver transplantation due to UCD and/or

3. Meeting at least two of the eight clinical criteria below:

   • Recurrent vomiting (more frequent than once a month);

   • Protein intolerance (becoming physically ill with high protein intake on multiple occasions leading to a self‐imposed low protein diet);

   • Episodic lethargy;

   • Developmental and/or intellectual disabilities requiring special education or care (i.e. Individualized Education Plans and 504 Plans, which are plans of education for individuals with disability);

   • Abnormal neurological examination (hypotonia and/or spasticity and or hyperreflexia and/or clonus)

   • Brain edema (evidence on MRI or CT scan);

   • Chronic migraine headaches (more than once a month);

   • Psychosis (episodic).

In addition, if a participant became symptomatic sometime while enrolled in the study, they then were considered symptomatic from that specific date onward. In addition, asymptomatic females without a known pathogenic OTC variant but who were diagnosed through biochemical studies and/or pedigree analysis were also included in this study.

2.3. Behavioral symptoms and other symptomatology or queries from the longitudinal study

The checklist of behavioral symptoms within the UCDC longitudinal study database included sleeping problems, mood disorder, distractability, food refusal, hyperactivity, impulsivity, short attention, temper tantrums, aggressiveness, self‐injury, lying, pica, self‐stimulation, and stealing. Other symptomatology or queries of database included ammonia (including date), protein restriction at baseline (yes/no), nitrogen scavenger therapy (yes/no), hospitalization (yes/no including date), and liver transplantation (yes/no including date).

2.4. Statistical analysis

Subject characteristics were described by three symptomatology groups, determined based on timing of symptoms: (1) Those who were symptomatic at baseline/study entry (Sym); (2) those who did not have symptoms at baseline but developed symptoms during study follow‐up (Asx/Sym); and (3) those who were asymptomatic at baseline and at last follow‐up as well (Asx). For each of these groups, we described baseline characteristics, disposition, and hospitalization records. The description included counts and frequencies for categorical variables, and the four‐number summary for continuous variables (mean, standard deviation, median, and interquartile range). To compare each characteristic between symptomatology groups, we reported standardized mean differences (SMD) and p‐values (from t‐tests for continuous variables and from chi‐square tests or exact tests for categorical variables). SMD is the difference in mean outcome between groups divided by standard deviation of outcome among participants. SMD is a measure of effect of size and standardizes test results for comparison purposes. An SMD ≥0.8 in absolute value or a p‐value of <0.05 indicated large or evidence for statistically significant differences, respectively. We reported incidence and prevalence of symptoms, as well as incidence of first hospitalization for HA or symptomatology over the lifetime of patients using Kaplan–Meier curves. We used Jupyter Notebook for the statistical analysis and data processing, and Python as the programming language.

2.5. Neuroimaging modalities applied on case report proband and her daughter

Application of functional near‐infrared spectroscopy (fNIRS), fMRI, MRI and DTI, and ¹H MRS modalities on proband and her daughter were as described before (Gropman et al., 2008, 2010; Gropman, Prust, et al., 2013; Gropman, Shattuck, et al., 2013; Khaksari et al., 2022; Sprouse et al., 2014). An unaffected age‐matched normal female was used as control for the proband. Furthermore, a Boolean search of NCBI PubMed for UCDC publications related to neuroimaging in heterozygote females with OTCD was performed to put in context our results pertaining to neurocognitive symptoms.

3. CASE PRESENTATION

3.1. Carrier female (proband) with confusion and protein aversion following birth of daughter

Proband was well, seldom ill, and consumed a normal diet up until age 35 years. She was athletic and of normal stature and weight. She completed college and a master's degree without any difficulty. During her pregnancy, she initially consumed a normal diet, but later in her pregnancy developed hyperemesis gravidarum and protein aversion. After delivery, she was confused and violent and said to have postpartum psychosis. While the violent behavior continued, she received a psychiatric diagnosis until her daughter at age 4 years, was admitted to a children's hospital in hyperammonemic coma presumably triggered by a viral illness. Her daughter had been a picky eater, preferring carbohydrates and salads. Proband's daughter was diagnosed with OTCD (OTC c.1033T>C p.Tyr34His), where this tyrosine to histidine substitution results in 3.4% wildtype activity based on a yeast functional complementation assay (Lo et al., 2023). Proband's daughter was started on L‐citrulline and nitrogen scavenger therapy. Proband was tested and found to have baseline high glutamine at 1100 μM [Reference range (RR): 205–756 μM], low citrulline at 8 μM (RR: 12–55 μM), and ammonia at 55 μM (RR: 11–52 μM), but was not treated. Proband continued her self‐imposed protein restriction. The proband was subsequently determined to be carrier of the same OTC variant as her daughter. Both proband and her daughter participated in the multimodal neuroimaging protocol and fNIRS testing (Figure 1) at Georgetown University and Children's National Hospital, respectively, as described before (Gropman et al., 2008, 2010; Gropman, Prust, et al., 2013; Gropman, Shattuck, et al., 2013; Khaksari et al., 2022; Sprouse et al., 2014). Both individuals were found to have elevated glutamine, low myoinositol, and low choline on ¹H MRS (Figure 1e). After the imaging study, proband also started L‐citrulline and nitrogen scavenger, with subsequent reporting of “no more brain fog”. Extended family history was notable for proband's mother developing erratic behavior as she aged and a sister with “bipolar” disorder, with neither one ever being tested for OTCD or other inborn errors of metabolism. Proband's mother was diagnosed as having dementia, despite completely normal periods of cognitive function in between her episodes of erratic behaviors.

FIGURE 1.

Neuroimaging phenotype in proband and other female carriers with OTCD. (a) fNIRS: fNIRS after a Stroop task in age‐matched normal female controls (1) versus heterozygote females with OTCD (2) from Figure 6 of Anderson et al. (2020). (b) Structural MRI: Nonspecific white matter intensities (white arrows) found in both Asx and Sym female heterozygotes with OTCD on FLAIR imaging. This imaging modality does not offer specificity for this disorder; (c) fMRI: Normal female control showing frontal activation (green, bottom images) on a Stroop task, which is left side dominant, while the proband heterozygote with OTCD (red, top images) showing bilateral activation as well as recruitment of other regions; (d) DTI: Diffusion tensor imaging; and (e) ¹H MRS: Age‐matched normal control (green line), Asx female (blue line), and Sym female (red line). Choline (chol; blue arrow) levels are reduced in both the Sym and Asx female heterozygotes with OTCD, reflecting loss of membrane integrity. Interestingly, myoinositol (mI; yellow arrow) shows normal high levels in a normal control and low level in a Sym female heterozygote with OTCD. Asx female heterozygote with OTCD showing intermediate, but lower mI levels than normal control female (see Table 6). Asx, asymptomatic; Sym, symptomatic.

4. RESULTS

4.1. Subject categories, characteristics, and risks for HA crises

Data from this longitudinal multicentric study using the UCDC database demonstrates that the number of OTC heterozygote females who remained classified as asymptomatic (Asx; n = 111) and those who were initially asymptomatic but later became symptomatic (Asx/Sym; n = 22) taken together are comparable to the numbers of females who were symptomatic at baseline (n = 143) (Table 1). Table S1 details the age of diagnosis for the Asx, Asx/Sym, and Sym groups. The mean and median ages of the Sym group were much lower than those of the Asx and Asx/Sym groups (Table S1). Our data also shows that 14% (15/110) of heterozygotes in the Asx category self‐restricted protein and 25% (5/20) in the Asx/Sym category were on protein restriction (Table 2). The majority of Asx (86%, 95/110) and Asx/Sym (75%, 15/20) subjects did not restrict protein at baseline, but about 15% (20/130) of subjects from the combined Asx and Asx/Sym categories did (Table 2). The majority of Sym (94%, 132/140) subjects restricted protein intake (Table 2). Among subjects who are unknown whether Sym or Asx/Sym (Unk; Table 1), 94% (17/18) restricted protein but we do not have enough data to know if they could be better re‐classified as Sym or Asx/Sym (Table 2). Of the 136 Asx and Asx/Sym subjects with ammonia assessments, 5 developed HA (Table 3), with one patient from each category developing significant HA ranging 77–470 μM and requiring admission and treatment with nitrogen scavenger medication (Table 4). Kaplan Meier analysis with right censored data showed that the median age of first HA event among the Asx and Asx/Sym categories together was ~50 years (Figure 2a), whereas the median age of first symptom of same groups together was ~17 years (Figure 2b).

TABLE 1.

Subject categorization and abbreviation.

Group	Abbreviation	n	%
Asymptomatic at any time	Asx	111	37
Symptomatic at baseline	Sym	143	47
Asymptomatic at baseline but symptomatic later	Asx/Sym	22	7
Unknown whether Sym or Asx/Sym	Unk	18	6
Missing information	Miss	8	3
	Total	302	100

TABLE 2.

Protein restriction at baseline among groups. ^a

Group	Yes (%)	No (%)	Not known (%)	All (%)
Asx	15 (14)	95 (86)	1 (1)	111 (100)
Sym	132 (92)	8 (6)	3 (2)	143 (100)
Asx/Sym	5 (23)	15 (68)	2 (9)	22 (100)
Unk	17 (94)	1 (6)	0 (0)	18 (100)
Miss	0 (0)	3 (38)	5 (62)	8 (100)
Total	169 (56)	122 (40)	11 (4)	302 (100)

Abbreviations: Miss, missing information; Unk, unknown whether Sym or Asx/Sym; and for other abbreviations, see Table 1.

^a Chi‐square analysis; p < 0.0001 (1.81 × 10⁻⁵⁴).

TABLE 3.

Ammonia (μM; NH₄) among groups. ^a

Group	Mean	Min	Max	STD	n
Asx		470	470	NA	1
Sym	328	80	1280	189	119
Asx/Sym	164	77	220	66	4
Unk b	310	140	900	231	12

Abbreviations: STD, standard deviation; Unk, unknown whether Sym or Asx/Sym; and for other abbreviations see Table 1.

^a Subjects in group “Miss” (Table 1) did not have any recorded ammonia and had no hospitalizations. Among all hospitalized subjects, 2 Sym subjects did not have max ammonia recorded in the database. Total n = 136 with documented hyperammonemia, although symptom status of group Unk was not entered in database at time of data entry; those with hyperammonemia included Asx 1 of 111 initial subjects and Asx/Sym 4 of 22 initial subjects, from among total Asx + Asx/Sym = 133—see Table 1 of initial grouping; p = 0.19 by Kruskal–Wallis, cannot reject null.

^b This grouping is a consequence of incomplete or inadequate entry of symptom status in the UCDC database.

TABLE 4.

Nitrogen scavenger therapy among groups during hospitalization.

Group	Yes	No	Not known	All
Asx	1	0	0	1
Sym	78	40	25	143
Asx/Sym	1	2	19	22
Unk	7	4	7	18
Total	87	46	161	294

Abbreviations: Unk, unknown whether Sym or Asx/Sym; and for other abbreviations see Table 1.

FIGURE 2.

Kaplan Meier curves with right censored data of (a) age at first hyperammonemia and (b) age at first symptom, of the asymptomatic at any time (Asx) and asymptomatic at baseline who became symptomatic later while in the study (Asx/Sym) cohorts together.

4.2. Neurocognitive symptoms in cohorts

From among the 77% (17/22) of Asx/Sym subjects who had a behavioral symptom reported in the database survey, close to half (41%–53%) of females with average age of 46 (n = 9) and 43 (n = 7) years reported sleep problems and mood disorder, respectively (Table 5). From the Asx category (n = 37, with neurobehavioral data in the longitudinal study), the most commonly reported neurobehavioral symptoms were sleep disorder (41%), easy distractibility (41%), mood problems (35%), and short attention (32%) followed by food refusal (22%), impulsivity (22%), hyperactivity (16%), with a few exhibiting temper tantrums, self‐injurious and aggressive attributes (Table 5).

TABLE 5.

Behavioral symptoms checked in Asx and Asx/Sym groups.

Specific behavior symptom	Sleeping problems	Mood disorder	Food refusal	Distractibility	Short attention	Hyper‐activity	Temper tantrums	Aggressiveness	Impulsivity
Asx/Sym subjects who checked a behavioral symptom box (n = 17) a
Average visit age in years (min, max)	46 (1, 72)	43 (14, 69)	9 (3, 24)	29 (6, 60)	29 (9, 60)	20 (1, 50)	7 (1, 17)	9 (1, 17)	26 (6, 60)
n‐value	9	7	5	5	5	4	3	3	3
Behavior Sym (%)	53	41	29	29	29	24	18	18	18

Specific behavior symptom	Sleeping problems	Distractability	Mood disorder	Short attention	Food refusal	Impulsivity	Hyper‐activity	Temper tantrums	Lying	Self injury	Aggressiveness
Asx subjects who checked a behavioral symptom box (n = 37) b
Average visit age in years (min, max)	38 (9, 62)	33 (2, 70)	41 (10, 71)	32 (2, 59)	30 (2, 54)	37 (2, 62)	32 (2, 47)	35 (3, 47)	20 (3, 41)	24 (2, 45)	39 (39, 39)
n‐value	15	15	13	12	8	8	6	6	3	2	1
Behavior Sym (%)	41	41	35	32	22	22	16	16	8	5	3

^a Representing 77% (17/22) of total Asx/Sym group (see Table 1). Columns with missing data imply no responses provided by specific group.

^b Representing 33% (37/111) of total Asx group (see Table 1). Abbreviations: See Table 1.

4.3. Neuroimage findings in proband and other heterozygote females with OTCD

fMRI of normal female control showed frontal activation (Figure 1c; green, bottom images) on a Stroop task, which is left side dominant, while proband heterozygote with OTCD (Figure 1c; red, top images) showed bilateral activation as well as recruitment of other regions, not typically involved in neural activation after a Stroop task (Sprouse et al., 2014). The Stroop task is a measure that captures the ability to overcome automatic tendencies to respond in accordance with current goals and/or slowing due to parallel processing of multiple stimulus dimensions (Sprouse et al., 2014). DTI was collected, with discrete tracks integrated and voxels with significantly lower fractional anisotropy are shown in the proband heterozygote with OTCD and compared with a normal control (p < 0.001) overlaid on a standardized brain structural MR image (Figure 1d). Furthermore, the lower panel shows corticospinal tracts of normal caliber as this track is not affected in proband (Figure 1d). Subjects undergoing ¹H MRS (Figure 1e) are compared; age‐matched normal control undergoing ¹H MRS (green line), Asx female (not proband; blue line), and Sym female (red line). Of note, glutamine (gln; red and green arrows) levels are higher as expected in an affected Sym female heterozygote with OTCD but normal in Asx subjects and normal controls. Choline (chol; blue arrow) levels are reduced in both the Sym and Asx female heterozygotes with OTCD, reflecting loss of membrane integrity (Figure 1e). Interestingly, myoinositol (mI; yellow arrow) shows normal high levels in normal control and low level in a Sym female heterozygote with OTCD (Figure 1e). Asx female heterozygote with OTCD showed intermediate, but lower mI levels than normal control female (Figure 1e) (Table 6).

TABLE 6.

¹H MRS biomarkers for females with OTCD.

Group	Neuroimaging biomarkers
	Choline (Chol)	Myoinositol (mI)	Glutamine (Gln)
Asx	↓	N or ↓	N or ↑
Asx/Sym	↓	↓	↑
Sym	↓	↓↓	↑↑
Normal control	N	N	N

Note: Choline is a marker of membrane integrity. Diminished myoinositol in frontal and posterior white matter is associated with decreased non‐verbal intelligence.

Abbreviations: N, normal; for others abbreviations see Table 1.

5. DISCUSSION

OTCD is responsible for moderate to severe HA resulting in substantial morbidity and mortality (Brusilow & Maestri, 1996). The phenotypic spectrum in female heterozygotes varies from severe hyperammonemic encephalopathy in infancy to hepatic dysfunction or subtle neuropsychiatric symptoms in adulthood (Chongsrisawat et al., 2018; Gropman, Prust, et al., 2013). There is lack of understanding of the risk among the mildly symptomatic or “asymptomatic” females of developing more significant complications including HA later in life and in adulthood. Risk‐reduction management approaches in OTCD are well known (Lichter‐Konecki et al., 1993; Yorifuji et al., 1998), and using the n of 1 trial methodology, the efficacy of l‐arginine use for the long‐term management of mildly symptomatic middle‐aged adult female with OTCD reported (Hackett et al., 2008). In addition to acute decompensation with severe HA, female heterozygotes with OTCD are also at risk for several chronic and insidious neurobehavioral co‐morbidities. Identifying risk factors that predict development of complications in heterozygous females with OTCD would be highly valuable for long‐term clinical management of this disorder.

5.1. HA crises in females with OTCD

In this study, a symptomatic patient was defined as someone who developed clinical symptoms consistent with significant HA required a liver transplant, or had 2 of the gastrointestinal or neurobehavioral symptoms as noted above (see Methods). This emphasizes the need to identify biomarkers and risk stratification strategies for more predictable and systematic medical management of OTC heterozygotes. Since it is unknown when an Asx subject might 1 day become Asx/Sym, calculating the risk of mild to severe HA (77–470 μM) sometime in life for the Asx and Asx/Sym categories taken together (n = 133) seemed more relevant. For these two categories combined, the HA risk was at least ~4% (5/133), with about half (2/4) of subjects requiring hospital admission and nitrogen scavenger therapy (Tables 3 and 4). This HA risk estimation may be an underestimate for it does not include the Asx/Sym (n = 19) and Unk (n = 7) groups, where nitrogen scavenger therapy usage during any hospitalization is not clear (Table 4). Furthermore, the median age of first HA in the combined Asx and Asx/Sym categories was ~50 years whereas the median age of first symptom in this combined group was ~17 years (Figure 2). For most older children and adults, managing physicians generally make note of ammonia >70 μM. Age at recruitment and whether or not ammonia had always been checked with other previous possible symptomatic episode(s), could skew age of first ammonia documented in the UCDC database. However, analysis of this large cohort of females with OTCD with long‐term data suggests that evidence‐based guidelines are needed to reduce the mortality and morbidity in this subgroup.

5.2. Neurocognitive findings in females with OTCD

After multivariate logistic regression analyses adjusting for age, sex, and educational status, Geda et al. (Geda et al., 2008) found when testing about 2000 elder individuals (ages 70–89 years) for 12 neuropsychiatric domains including sleep problems, apathy, delusion, depression, anxiety, euphoria, agitation, eating/appetite changes, hallucination, disinhibition, irritability, and aberrant motor behavior, that nonpsychotic symptoms affected approximately 50% of subjects with mild cognitive impairment (MCI) and 25% of subjects with normal cognition, with psychotic symptoms being rare (Geda et al., 2008). Furthermore, among 1747 participating women ages 20 to 30 years from the US National Health and Nutrition Examination Survey (2009–2016), about 20% and 9% reported trouble sleeping and symptoms of depression, respectively (Barsha & Hossain, 2020). Interestingly, sleep problems, mood disorder, distractibility, and short attention span were found in >25% of subjects from either the Asx or Asx/Sym groups (Table 5), with approximately 40%–50% for some behavioral symptoms in those groups similar to elder subjects with MCI (Geda et al., 2008). Furthermore, sleep problems and mood disorders among Asx or Asx/Sym groups ranged from 35% to 53% of cases (Table 5), more than what was found among females in the general population presumably unaffected with OTCD by Barsha and Hossain (Barsha & Hossain, 2020). This could suggest that many heterozygote females with OTCD within the Asx and Asx/Sym groups could have unrecognized MCI at a much earlier age but this should be more thoroughly investigated.

There is no single MRI modality that adequately describes the degree of neurological damage in OTCD. UCDC studies have employed a multimodal brain imaging approach in studying the changes in anatomy, connectivity, and biochemistry to create a comprehensive framework for understanding the brain and its injury pattern in OTCD. These studies have included gender and age‐matched controls without OTCD, as well as symptomatic and asymptomatic female carriers. Multimodal neuroimaging in OTCD has shown value in identifying brain biomarkers of disease in female heterozygotes that correlated with measures of cognitive dysfunction in the domains of executive dysfunction. Neuroimaging studies have shown that the white matter tractography in frontal lobe is impacted by severe or mild HA, thereby providing a hypothesis for executive dysfunction in OTCD with partial enzyme activity (Sen et al., 2021, 2022). Studies using DTI have shown that individuals with OTCD have reduced white matter integrity in specific regions of the brain, which may be related to the cognitive and behavioral problems observed in these individuals (Gropman et al., 2010). Specifically, these studies enrolled controls as well as both symptomatic and asymptomatic females. Likewise, ¹H MRS studies comparing brain metabolites in asymptomatic and symptomatic sisters revealed decreased myoinositol in the posterior parietal white matter and frontal white matter in both subjects when compared to healthy control (Gropman et al., 2008). ¹H MRS identified low choline and low myoinositol in several “asymptomatic carriers” (Figure 1e and other data not shown) as well as elevated glutamine. Choline and myoinositol are two of the three biomarkers identified in symptomatic individuals with OTCD, with the third being elevated glutamine (even at baseline) (Figure 1e). Therefore, these biomarkers are useful in the clinical setting to screen at‐risk heterozygote carriers (Table 6). Diminished myoinositol was associated with decrease in nonverbal intelligence and constitutes a useful biomarker with which to discriminate females with a partial deficiency (Gropman et al., 2008). Functional MRI (fMRI) has also been used to investigate the effects of OTCD on brain function, specifically executive cognition. Studies have shown that individuals with OTCD have altered patterns of brain activity during tasks that require cognitive control and attention and involve the prefrontal cortex (PFC) (Gropman, Shattuck, et al., 2013). Because of the difficulties of performing fMRI given the technical concerns of movement, and the confinement in the scanner which can degrade the images, comparison with fNIRS, an optical, non‐invasive neuroimaging technique was used to measure changes in the concentration of oxygenated and deoxygenated hemoglobin in the brain. fNIRS has several advantages over other neuroimaging techniques like fMRI, including its portability, relatively low cost, and ability to measure brain activity in real time (Wagner et al., 2021). Using fNIRS, we measured the activation in PFC of the participants while performing the Stroop task. We have also previously investigated the difference in behavioral measures as well as brain activation in left and right PFC in symptomatic and asymptomatic female patients with OTCD and age‐matched controls. Results revealed a distinction in left PFC activation between controls and patients with OTCD including both symptomatic and asymptomatic female carriers who performed similarly but with “asymptomatic carriers” having more activation than symptomatic but less than controls. Control subjects showed significantly higher task‐related activation increase compared to OTCD patients overall. Subjects with OTCD also exhibited bilateral increase in PFC activation. The findings suggest alterations in neurocognitive function of PFC in OTCD that are also observed in carrier females, regardless of clinical manifestations (Anderson et al., 2020), confirming the fMRI outcomes observed in OTCD patients. A significant change in brain activation was also seen in Blood Oxygenation Level Dependent signal in the PFC of OTCD patients using fMRI compared to healthy controls (Gropman, Shattuck, et al., 2013). In summary, research using conventional and advanced neuroimaging has been critical to understanding the neurological consequences of OTCD and identifying potential treatment strategies.

5.3. Limitations

This study is limited by the incompleteness of the UCDC longitudinal natural history database and timely updates to specific subject data by various participating institutions, and by the choice of and missed queries of the database. Individuals were from the various UCDC institutions and normal range of ammonia may be different at different institutions. Symptomatology data on 9% (26/302) of subjects was either missing or unknown (Table 1). Protein restriction information at baseline also was either missing or unknown in 9% (26/302) of subjects (Table 2). Eight of 302 subjects (3%; representing the Miss group in Table 1) did not have any recorded ammonia and had no hospitalizations (Table 3). Some Sym subjects had missing ammonia information. Information about whether nitrogen scavenger therapy was used during any hospitalization was unknown for ~17% (25/143) of Sym and 86% (19/22) Asx/Sym subjects (Table 4). However, among subjects with this information known, 34% (40/118) of Sym and 67% (2/3) of Asx/Sym subjects were documented to have not received nitrogen scavenger therapy during a documented hospital admission (Table 4). Thus, the risk of HA among Asx/Sym (and by extension likely also the Asx subjects) may be an underestimate in this study.

5.4. Future directions

This effort sets the stage for the future development of a clinical risk score calculator in OTCD. Such calculators have been developed before to assess the likelihood of a specific condition, as well as to target molecular testing and/or medical intervention for several genetic disorders, including the R4S scoring system to gauge risk following newborn blood spot screening (Rinaldo, 2007), the Ghent nosology for Marfan syndrome (Loeys et al., 2010), and the clinical scoring system for selection of patients for PTEN testing for Cowden syndrome and Bannayan‐Riley‐Ruvalcaba syndrome (Tan et al., 2011). Our long‐term goal is to use our improved understanding of factors associated with risk of serious illness in symptomatic and asymptomatic female heterozygotes with OTCD, including mothers of affected offspring with and without a detectable pathogenic OTC variant, to develop a clinical scoring system utilizing data from survey questionnaire(s), executive function testing, and biochemical and neuroimaging data to improve prognosis and long‐term management of such individuals.

6. CONCLUSIONS

This study was conducted with goal of comprehensive review and analysis of available data in the UCDC longitudinal natural history study, the largest curated study of patients with OTCD, to evaluate specific candidate risk factors for their utility in predicting well‐defined complications in heterozygous females with OTCD with and without known pathogenic variants. Our findings demonstrate that the risk of moderate to severe HA crises sometime later in life for Asx and Asx/Sym categories taken together (n = 133) is at least 4% (5/133). Median age of first HA crisis was 50 years of age whereas the median age of first symptom which included neuropsychiatric and/or behavioral symptoms was 17 years of age. The majority of Asx (86%) and Asx/Sym (75%) patients did not restrict protein at baseline, and ~38% of Asx and 33% of Asx/Sym patients suffered from mild to severe psychiatric symptoms such as mood disorder, easy distractibility, short attention, and sleep problems. Findings of this retrospective study will lay the foundation for a prospective study with the goal of developing a risk calculator and enabling improved guidelines for management in female heterozygotes with OTCD. This is important since current treatment guidelines do not address carrier females, nor use neurological outcomes or neuromonitoring to inform decision‐making. By identifying the specific brain regions and neural pathways affected in females with OTCD, quantitative neuroimaging modalities, and executive function tests may provide risk coefficient(s) in a risk calculator and insights into potential targets for intervention, such as drugs that can reduce ammonia levels and/or therapies that could promote brain repair and recovery.

AUTHOR CONTRIBUTIONS

Rima Izem, Laura L. Konczal, Robert J. McCarter, and Jirair K. Bedoyan were involved in the study design. Rima Izem, Yuelin Long, Jiji Jiang, Robert J. McCarter, and Jirair K. Bedoyan were involved in data extraction and analysis. Rima Izem and Yuelin Long were involved in the statistical analysis of data. Kuntal Sen, Laura L. Konczal, Robert J. McCarter, Andrea L. Gropman, and Jirair K. Bedoyan were involved in manuscript preparation and critical data review. Jirair K. Bedoyan revised and approved the final version as submitted.

FUNDING INFORMATION

The Urea Cycle Disorders Consortium (UCDC; U54HD061221) is a part of the National Institutes of Health (NIH) Rare Disease Clinical Research Network (RDCRN), supported through collaboration between the Office of Rare Diseases Research (ORDR), the National Center for Advancing Translational Science (NCATS), the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), and the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). The UCDC is also supported by the O'Malley Foundation and the Kettering Fund.

CONFLICT OF INTEREST STATEMENT

The authors report no conflict of interest.

ETHICS STATEMENT

The research was conducted according to the ESC Clinical Practice Guidelines on ethics and the Helsinki Declaration and approved by the NIH as part of the RDCRN U54HD061221 UCDC natural history longitudinal study and by the IRB of each participating institution within the Urea Cycle Disorders Consortium (UCDC). Presented cases (proband and daughter) were consented and assented, respectively, as part of their enrollment into the Children's National Medical Center, Washington, DC, IRB‐approved study (IRB Protocol ID: 4837).

Supporting information

Table S1..

Sen, K. , Izem, R. , Long, Y. , Jiang, J. , Konczal, L. L. , McCarter, R. J. ; Gropman, A. L. , & Bedoyan, J. K. (2024). Are asymptomatic carriers of OTC deficiency always asymptomatic? A multicentric retrospective study of risk using the UCDC longitudinal study database. Molecular Genetics & Genomic Medicine, 12, e2443. 10.1002/mgg3.2443

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.