Received: 10 March 2020 Revised: 10 April 2020 Accepted: 27 April 2020 DOI: 10.1002/jimd.12247 ORIGINAL ARTICLE AADC deficiency from infancy to adulthood: Symptoms and developmental outcome in an international cohort of 63 patients Toni S. Pearson1 | Laura Gilbert1 | Thomas Opladen2 | Angeles Garcia-Cazorla3 | Mario Mastrangelo4 | Vincenzo Leuzzi4 | Stacy K. H. Tay5 | Jolanta Sykut-Cegielska6 | Roser Pons7 | Saadet Mercimek-Andrews8 | Mitsuhiro Kato9 | Thomas Lücke10 | Mari Oppebøen11 | Manju A. Kurian12 | Dora Steel12 | Filippo Manti4 | Kathleen D. Meeks1 | Kathrin Jeltsch2 | Lisa Flint13 1 Department of Neurology, Washington University School of Medicine, St. Louis, Missouri 2 Division of Child Neurology & Metabolic Medicine, University Children's Hospital, Heidelberg, Germany 3 Inborn Errors of Metabolism Unit, Institut de Recerca Sant Joan de Déu and CIBERER-ISCIII, Barcelona, Spain 4 Unit of Child Neurology and Psychiatry, Department of Human Neuroscience, Sapienza University of Rome, Rome, Italy 5 KTP-National University Children's Medical Institute, National University Health System, Singapore, Singapore 6 Department of Inborn Errors of Metabolism and Pediatrics, Institute of Mother and Child, Warsaw, Poland 7 First Department of Pediatrics, Aghia Sofia Hospital, University of Athens, Athens, Greece 8 Division of Clinical and Metabolic Genetics, Department of Pediatrics, University of Toronto, The Hospital for Sick Children, Toronto, Ontario, Canada 9 Department of Pediatrics, Showa University School of Medicine, Tokyo, Japan 10 University Children's Hospital, St. Josef-Hospital, Ruhr-University Bochum, Bochum, Germany 11 Division of Child Neurology, Oslo University Hospital, Oslo, Norway 12 Developmental Neurosciences, UCL Great Ormond Street-Institute of Child Health and Department of Neurology, Great Ormond Street Hospital, London, UK 13 AADC Research Trust, Caterham, UK Correspondence Toni S. Pearson, Department of Abstract Neurology, Campus Box 8111, 660 South Aromatic L-amino acid decarboxylase deficiency (AADCD) is a rare, autosomal Euclid Avenue, St Louis, MO 63110. recessive neurodevelopmental disorder characterized by impaired synthesis of Email: tpearson@wustl.edu dopamine, noradrenaline, adrenaline and serotonin, leading to a complex syn- Communicating Editor: Nenad Blau drome of motor, behavioral, and autonomic symptoms. This retrospective study assessed the symptoms and developmental outcome of a large interna- Funding information Fondo Europeo de desarrollo regional; tional cohort of patients with AADCD via physician and/or caregiver responses Instituto de Salud Carlos III; Pediatric to a detailed, standardized questionnaire. Sixty-three patients (60% female; ages Neurotransmitter Disease Association and 6 months-36 years, median 7 years; 58 living) from 23 individual countries par- NIH/NINDS ticipated. Common symptoms at onset (median age 3 months, range This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2020 The Authors. Journal of Inherited Metabolic Disease published by John Wiley & Sons Ltd on behalf of SSIEM J Inherit Metab Dis. 2020;43:1121–1130. wileyonlinelibrary.com/journal/jimd 1121 1122 PEARSON ET AL. 0-12 months) were hypotonia, developmental delay, and/or oculogyric crises. Oculogyric crises were present in 97% of patients aged 2 to 12 years, occurred in the majority of patients in all age groups, and tended to be most severe dur- ing early childhood. Prominent non-motor symptoms were sleep disturbance, irritable mood, and feeding difficulties. The majority of subjects (70%) had pro- found motor impairment characterized by absent head control and minimal voluntary movement, while 17% had mild motor impairment and were able to walk independently. Dopamine agonists were the medications most likely to produce some symptomatic benefit, but were associated with dose-limiting side effects (dyskinesia, insomnia, irritability, vomiting) that led to discontinuation 25% of the time. The age distribution of our cohort (70% of subjects under age 13 years) and the observation of a greater proportion of patients with a more severe disease phenotype in the younger compared to the older patients, both suggest a significant mortality risk during childhood for patients with severe disease. KEYWORDS dystonia-parkinsonism, gene therapy, natural history, neurotransmitter disorders, rare diseases 1 | INTRODUCTION two different stereotactic neurosurgical approaches to deliver gene vector either to the putamen5,10,11 or the Aromatic L-amino acid decarboxylase deficiency midbrain (ClinicalTrials.gov Identifier: NCT02852213). (AADCD) is a rare, autosomal recessive, neurometabolic In the context of new methods for early diagnosis disorder characterized by impaired synthesis of the cate- already implemented in some newborn screening pro- cholamines (dopamine, epinephrine, and norepineph- grams12 and these prospective novel therapeutic oppor- rine) and serotonin. AADCD typically presents in infancy tunities, an improved knowledge of the natural history with hypotonia, oculogyric crises, and developmental of the disease, modified by available medical treat- delay. Mood and sleep disturbance, autonomic dysfunc- ments, is critical. tion, and intellectual disability are additional disease fea- The objective of this study is to describe the initial tures. Since the initial description of the condition1 over symptoms, symptom course, treatment response, and 135 patients with this disorder have been described in the developmental outcomes in a large international cohort medical literature.2 of patients with AADCD. The majority of published case reports and case series focus on presenting clinical, biochemical and genetic disease features. Unanswered questions remain 2 | METHODS about symptom evolution, the full spectrum of develop- mental outcomes, and mortality. In Taiwan, where We created a written questionnaire to collect data about there is a higher disease prevalence due to the presence disease onset, symptom course, developmental outcome, of a founder mutation in the population,3 patients have and mortality (see Supporting Information). Participants an almost uniformly severe disease course with pro- were recruited via two sources: (1) The International found motor impairment4 and high risk of death in Working Group of Neurotransmitter-Related Disorders early childhood.5 Internationally, the majority of (iNTD) patient registry, which includes collaborating phy- reported patients have severe symptoms and motor dis- sicians from 32 centers in North America, Europe, and ability, but patients with mild disease and more favor- Asia13; and (2) The AADC Research Trust, a parent-run able response to medical treatment have also been foundation based in the United Kingdom. For participants described.6-8 recruited via the iNTD, the questionnaire was included as Most patients with AADCD experience limited ben- a module of the working group's de-identified online reg- efit from currently available medical therapies.9 Gene istry and was completed by the physician and/or parent/ therapy for this disorder is under investigation using caregiver. For participants recruited via the AADC Trust, PEARSON ET AL. 1123 an online English-language version of the questionnaire 3 | RESULTS was created using REDCap and was completed by a par- ent/caregiver. Families received an email invitation from 3.1 | Subject characteristics the AADC Trust with a link to participate in the REDCap survey. Translations of the questionnaire were available Completed questionnaires were received on behalf of in multiple languages to facilitate participation by non- 64 individuals with AADCD, comprised of 24 responses English speaking families. Participation was voluntary, through the iNTD registry, and 48 through the AADC and responses were anonymous. Limited demographic Research Trust (response rate 66% (48/73). Data was com- information (subject month and year of birth; country of bined for eight participants for whom we received responses residence) was collected to permit identification of poten- through both recruitment channels. One subject was tial duplicate responses. Questionnaires were completed excluded due to prior participation in a gene therapy clini- by physicians and caregivers between December 2015 and cal trial. The final number of participants included for anal- October 2018. ysis was 63 (38 female, 60%). Fifty-eight (58) of the If a respondent did not answer any questions for a 63 subjects were living at the time of questionnaire comple- given section of the questionnaire, they were excluded tion, and 5 were deceased. from that section's analysis. The total number of respon- Subjects' ages at time of questionnaire completion dents for each section is indicated in the Results (or death, when applicable) ranged from 6 months to section below. 36.8 years (median 7 years) (Figure 1A). Participants origi- Permission to use items from Your Child's Usual Way nated from 23 different countries in Europe (n = 27, 43%), of Moving Around to assess functional mobility was North America (n = 16, 25%), Asia (n = 12, 19%), South granted by Robert Palisano, CanChild Centre for Child- America (n = 4, 6.5%) and the Middle East (n = 4, 6.5%). hood Disability Research, McMaster University, Hamilton, Ontario, Canada. This study was approved by the Human Research 3.2 | Initial symptoms Protection Office of Washington University School of Medicine (protocol 201801179) and the local ethics com- The median age of onset of initial symptoms was mittee of the University Hospital Heidelberg (iNTD coor- 3 months (range: 0-12 months), and was ≤6 months in dinating center, application S-471/2014). all but three subjects. Latency from symptom onset to F I G U R E 1 Subject age distribution and latency to diagnosis. A, Subject age ranged from 6 months to 36 years. There is a skew toward younger age (48% of subjects younger than 6 years, 70% younger than 13 years). B, Latency from symptom onset to diagnosis has decreased in the past decade; most subjects over age 10 years experienced a delay of years between symptom onset and diagnosis 1124 PEARSON ET AL. diagnosis varied considerably, and tended to be longer in Non-motor symptoms were also prominent, including subjects older than age 10 years (8 months to 31 years sleepiness (50%), irritability (44%), excessive sweating n = 17) compared to those age ≤ 10 years (mean (44%), and nasal congestion (42%). (n = 52) (Figure 2A). ± SD = 0.9 ± 0.7 years) (Figure 1B). Initial misdiagnosis with epilepsy and/or a history treat- The most commonly reported initial symptoms were ment with anti-epileptic medication prior to confirmation hypotonia (75%), oculogyric crises (62%), developmental of the diagnosis of AADCD was reported for 14/52 sub- delay (62%), and feeding problems (54%) (Figure 2A). jects (27%), which is likely an underestimate. F I G U R E 2 Motor and non- motor symptoms. A, Reported prevalence of initial symptoms (n = 52). B-F, Reported prevalence of current non-motor symptoms across age groups (n = 51). Major (dark gray): frequent and/or severe, with significant impact on comfort or function; Minor (light gray): minor and/or infrequent PEARSON ET AL. 1125 3.3 | Non-motor symptoms for 6 to 12 year-olds (15/16), and 50% of respondents for those over 12 years old (8/16) cited ages ranging Respondents were asked to classify each of a standard list of from infancy to age 5 years. symptoms as either “major,” (frequent and/or severe) In a more detailed cross-sectional analysis (n = 42) “minor,” (infrequent and/or mild) or “absent” across differ- of the presently experienced characteristics of oculogyric ent ages. A cross-sectional analysis of symptoms for each crises in younger subjects (age < 6 years) vs older sub- subjects' current age was performed (n = 51, Figure 2B-F). jects (age ≥ 6 years), younger patients were more likely Insomnia was present in 86% of subjects aged 2 to 12 years to experience OGC episodes over 4 hours in duration (24/28), characterized as a major problem in 50% of chil- (84% vs 43%), and to have episodes at least three times dren in this age range (Figure 2B), while excessive sleepi- per week (55% vs 33%) (Figure 3B,C). A subset of ness was more prominent in subjects under age 2 years patients (n = 38) provided further details about episode (Figure 2C). Irritability was reported in 85% of children severity. Across all ages, the majority (29/38, 76%) aged 6-12 years, and over 40% of subjects in all other age experienced episodes characterized by dystonia of the groups (Figure 2D). Excessive sweating was reported in all trunk and/or limbs in addition to ocular deviation, children under age 2 years, and classified as a major symp- while 9/28 (24%) experienced symptoms limited to the tom in approximately 40% of subjects from infancy through head and neck region only (eye deviation with or with- age 17 years (Figure 2E). out involuntary movements of the lower face and head turning). A Mann-Whitney test indicated that the group with whole-body dystonic episodes was signifi- 3.4 | Oculogyric crises cantly younger (median age 5.2, range 1.3-19.7 years) than the group with symptoms limited to the head and Oculogyric crises (OGC) are episodes characterized by neck (median age 10.4, range 4.3-26.1 years) involuntary eye deviation that may be accompanied by (Z = 2.232, P = .026). Taken all together, these features involuntary movements of the face, neck, trunk, or suggest an association between more frequent, pro- limbs. Severe OGCs may cause whole body dystonia. longed and severe OGCs and younger age in this The occurrence of OGCs in the present, past, or both, cohort. was reported by 98% of subjects (56/57). OGCs were a Five (5) of 17 subjects aged ≥13 years do not presently present symptom in the majority of subjects across all experience OGCs. Four (4) of these 5 subjects have a mild age groups. Episodes were particularly prevalent in phenotype characterized by the ability to walk indepen- those age 2 to 12 years (97%, 30/31), and were less dently. Therefore, the reduced prevalence of OGCs in prevalent in patients over 18 years (60%, 6/10) adolescents and adults may be attributable either to older (Figure 3A). When asked to retrospectively state the age, milder disease phenotype (see “Developmental and age at which OGCs had been the most problematic (fre- Functional Outcome” below), or a combination of these quent, severe, and/or prolonged), 94% of respondents factors. F I G U R E 3 Oculogyric Crises (OGC). A, Reported current prevalence of OGC's across age groups (n = 57): 91% of subjects under 18, 60% of subjects over 18. B and C, OGC duration and frequency, respectively, in younger (age < 6 years, dark gray) vs older (age ≥ 6 years, light gray) subjects. Over 80% of younger subjects experienced prolonged episodes (> 4 hours duration), and the majority experienced episodes at least three times per week 1126 PEARSON ET AL. 3.5 | Developmental and functional 3.6 | Molecular genetic results and outcome genotype-phenotype correlation 3.5.1 | Motor development Molecular genetic data was known for 43 subjects, 11 of whom carried 14 variants that were not previously Forty-nine (49) respondents for subjects ≥12 months reported (Table 1). Overall, 71% of variants were classi- in age (range 1-36 years) reported information about fied as missense, 19% as splice site, 8% as frameshift, and attainment of motor developmental milestones. Head 1% as nonsense. The most frequent variant was the control was reported in 33% (16/49), and independent known common splice site variant, 714+4A>T (15% of sitting and walking was reported in 22% (11/49). All alleles in this cohort), which was detected in either 11 subjects who could sit independently could also homozygous (n = 3) or compound heterozygous (n = 7) walk; there were none who reported the ability to sit form in 10 patients of southern Chinese ethnicity. only. Motor function differed notably across age Each subject's motor phenotypic severity was broadly groups in this cohort: younger subjects had more classified based on responses to questions about mobility severe motor impairment than older subjects. None of and development (mild: able to walk independently with- the 19 subjects under 6 years of age could sit or walk out an assistive device; severe: minimal or no attainment independently. In contrast, 4/14 subjects (29%) aged of developmental milestones; moderate: intermediate; 6-12 years, and 7/16 subjects (44%) over age 12 years Table 1). Nine (9) of the 43 subjects with known genotype could sit and walk. This does not reflect simply a had a mild motor phenotype. These nine individuals all delay in the attainment of walking, since all but one carried compound heterozygous variants that included at of the 11 subjects who could walk had gained the least one missense variant. Specific variants that were ability before age 6 years (range 22 months-7 years). detected only in patients with mild or moderate pheno- Rather, the data suggest that a greater proportion of typic severity were: c.367G>A (p. G123R), c.446G>C the older patients in this cohort had a milder disease (p. S149T), c.734 C>T (p. T245I), c.853C>T (p. R285W), phenotype. c.876G>A (p. E292E), and c.1337T>C (p. L446P). An early regression in motor or developmental skills was reported in 15/63 patients (24%), typically occurring during infancy at the onset of oculogyric crises and other 3.7 | Treatment disease symptoms. The most commonly described lost skill was the ability to feed by mouth (n = 8). Others Fifty-nine (59) respondents answered questions about were loss of head control or other early motor skills such observed benefits and side effects of medications. Pyri- as rolling (n = 6), and speech (n = 2). doxine was the most commonly tried medication (78%, Feeding difficulties were common, reported in 75% of 46/59), but only 3/46 respondents (7%) reported notice- respondents (40/53) as either a past or current problem. able improvement (“increased energy” in all cases). No Swallowing difficulties and vomiting were the most fre- significant side effects were reported. quent problems reported (43% and 26%, respectively). In At least one dopamine agonist (DA) was tried by this cohort, 45% of subjects (24/53) had a gastrostomy 83% of subjects (49/59), including bromocriptine (46%, tube placed for feeding. 27/59), pramipexole (41%, 24/59), rotigotine (12/59, 20%), and ropinirole (14%, 8/59). The DA most com- monly reported to be beneficial was the rotigotine 3.5.2 | Activities of daily living and patch, judged to be beneficial in 82% of those who had adaptive behavior tried it (10/12). Reported rates of benefit among the other DA's were 29% for pramipexole (7/24), 26% for Functional independence for routine activities of daily bromocriptine (7/27), and 13% for ropinirole (1/8). The living including mobility, feeding, bathing, and dressing most commonly described benefits (by caregivers who was analyzed for 38 subjects age 5 years and older. volunteered additional details) were improvements in Respondents classified 11% (4/38) as completely indepen- tone or spontaneous movements, improved alertness, dent, 18% (7/38) as partially independent, and 71% and decreased oculogyric crises. However, caregivers (27/38) as completely dependent. Of 29 subjects age 5 to reported adverse effects in association with all DA's: 18 years, 62% (18/29) attended school. Of 8 young adults 30% with bromocriptine, 38% with pramipexole, and over age 18 years, 2 participated in work outside 50% with rotigotine and ropinirole. Dyskinesia and the home. insomnia were the two most commonly reported side PEARSON ET AL. 1127 TABLE 1 Patient demographics and genotype Geographic Motor phenotypic ID Age (y) Sex region Allele 1 genotype Allele 2 genotype severityc 1 8 F North America c.179T>C (p.V60A) 714+4A>T Moderate 2 26.1 M Europe c.476C>T (p. A159V) ? Moderate 3 6.6 F North America Severe 4 10.9 M Europe c.73G>A (p.E25K) c.1073G>A (p. R358H) Severe 5 15 M North America Moderate 6 4 M North America Severe a 7 18.5 F North America c.140C>A (p. P47H) Homozygous Severe 8 8.8 M North America Mild 9 7.3 M South America c.1040G>A (p. R347Q) Homozygous Severe 10 3.7 M Middle East Severe 11 6.7 M Europe Mild 12 2.1 M North America c.19C>T (p. R7*) c.214C>T (p. H72Y) Severe b 13 21 F North America Severe 14 1.6 F South America c.330_334dupCGATC (p.Q112fs*13) Homozygous Severe 15 4.3 M Europe c.231C>A (p. F77L) Homozygous Severe 16 3.9 F North America c.286G>A (p. G96R) c.665T>C (p. L222P) Severe b 17 2.5 F Asia Severe 18 4.1 F Europe c.323G>A (p. S108N) c.1041+1G>C Severe 19 5.2 F North America 714+4A>T Homozygous Severe b 20 2.4 F Europe Severe 21 5.3 F Europe c.73G>A (p. E25K) c.315G>C (p. W105C) Severe 22 12.5 M Europe c.782G>T (p. C261F) c.1060G>A (p. G354S) Severe 23 1.4 F Middle East Severe 24 16.7 F Middle East c.242C>T (p. P81L) Homozygous Severe 25 10.7 F North America Severe 26 8.5 M South America c.568_569insCGAT (p. Q190Pfs) c.1040G>A (p. R347Q) Severe 27 0.8 M South America Severe 28 11.9 F Middle East c.1040G>A (p. R347Q) Homozygous Severe 29 7.9 F Europe c.139C>G (p. P47A) ? Severe 30 24.2 F Europe c.367G>A (p. G123R) c.876G>A (p. E292E) Mild 31 4 M North America Severe b 32 7 M North America Severe 33 5 M Europe Moderate 34 1.3 M Europe c.73G>A (p. E25K) c.624delC (p. I209Sfs*26) Severe 35 13.1 M Europe c.367G>A (p. G123R) c.876G>A (p. E292E) Mild 36 9 M Europe c.1A>G (p. M1V) c.181G>A (p. E61K) Severe b 37 13 F Europe Severe 38 14.8 M Europe c.1073G>A (p. R358H) Homozygous Severe 39 3.4 M Asia 714+4A>T Homozygous Severe 40 3.8 F Asia 714+4A>T c.1234C>T (p. R412W) Severe 41 1.4 F Asia c.179T>C (p. V60A) c.1234C>T (p. R412W) Severe (Continues) 1128 PEARSON ET AL. TABLE 1 (Continued) Geographic Motor phenotypic ID Age (y) Sex region Allele 1 genotype Allele 2 genotype severityc 42 3.3 F Asia 714+4A>T c.1297dupA (p. I433Nfs*60) Severe 43 3.2 F Asia c.106G>A (p. G36R) 714+4A>T Severe 44 4.6 F Asia c.170A>G (p. I57T) c.1234C>T (p. R412W) Severe 45 2 M Asia c.106G>A (p. G36R) 714+4A>T Severe 46 1 F Asia 714+4A>T Homozygous Severe 47 10.1 F Europe c.206C>T (p. T69M) 1337T>C (p. L446P) Mild 48 25.4 F Europe c.206C>T (p. T69M) c.439A>C (p. S147R) Severe 49 5.7 F Europe c.260C>T (p. P87L) c.799T>C (p. W267R) Severe a 50 19.7 F Europe c.214C>T (p. H72Y) Homozygous Severe 51 1.5 F Europe c.201+5G>C Homozygous Moderate 52 8.8 F Europe c.367G>A (p. G123R) c.734C>T (p. T245I) Mild 53 0.5 F Europe Severe a 54 36.8 F Europe c.105delC (p. Y37T fs*5) c.710T>C (p. F237S) Mild 55 15 M Europe c.843C>G (p. C281W) c.1085T>C (p. M362T) Mild 56 0.5 F Europe c.322A>C (p. S108R) c.812A>T (p. D271V) Severe 57 0.8 M Europe Moderate a 58 18.2 M Asia Moderate 59a 25.4 F Asia 714+4A>T c.853C>T (p. R285W) Mild a 60 22.7 F Asia 714+4A>T c.853C>T (p. R285W) Mild 61 14.2 F North America c.260C>T (p. P87L) c.446G>C (p. S149T) Mild 62 4.3 F North America c.260C>T (p. P87L) c.446G>C (p. S149T) Moderate 63 10.4 M North America Severe Note: Novel variants are italicized. a Previously published cases: 716; 5017; 546; 5818; 59 to 60.8 b Deceased. c Phenotypic severity: mild = able to walk independently; severe = minimal or no attainment of developmental milestones; moderate = intermediate. *Denoting a nonsense variant in the gene. effects (27% and 14%, respectively), followed by 3.8 | Mortality vomiting and irritability. These adverse effects were reported at similar frequencies across all the DA's, and The parents of five individuals who had died responded prompted discontinuation of the medication in 25% of to the survey. Ages of death were 2 years (n = 2), 7 years, subjects. 13 years, and 21 years, respectively. Causes of death A trial of therapy with at least one monoamine oxidase reported by parents were pneumonia (n = 2), acute com- inhibitor (MAO-I) was reported in 62% of subjects (37/59): plications during an OGC (n = 2), and “myocardial either selegiline (n = 20), tranylcypromine (n = 11), or both infarction” (n = 1). (n = 6). Beneficial effects were reported in 35% for selegiline and 18% for tranylcypromine (better alertness, improved tone or spontaneous movements). The reported frequency 4 | DISCUSSION of side effects was lower for both medications than was reported for any DA (27% for selegiline, 6% for This retrospective analysis of disease-related symptoms tranylcypromine), and the most commonly described side and developmental outcome in an international cohort of effect was insomnia. Side effects led to discontinuation of patients, aged 6 months to 36 years, revealed that the medication in a small proportion of subjects (12% for AADCD is associated with a variety of complex motor selegiline, 6% for tranylcypromine). and non-motor symptoms throughout the lifespan, PEARSON ET AL. 1129 including oculogyric crises, sleep disorder, mood distur- The age distribution of subjects in this study raises ques- bance, gastrointestinal symptoms, and autonomic dys- tions about survival. Younger subjects outnumbered older function. It also confirmed that, while the majority of subjects: nearly half were younger than 6 years, and 71% currently diagnosed patients have profound motor devel- were younger than 13 years. Furthermore, a significant pro- opmental impairment, some (20% of patients over age portion of the adult patients in our cohort exhibited a 12 months in this cohort) have milder motor impairment milder disease phenotype than the younger patients, char- and the ability to walk independently. acterized by milder motor impairment and the absence of Symptom onset occurred in early infancy, almost oculogyric crises. These observations suggest a significant always within the first 6 months. The most commonly childhood mortality risk for patients with severe disease, as reported initial symptoms were hypotonia, oculogyric cri- has been previously described for patients in Taiwan.5 Five ses, delayed motor development, and feeding difficulties, families shared information about the end of their child's consistent with previous descriptions in the literature.2 life, providing some insight into potential causes of death. The most prominent non-motor symptoms were sleep Two of the subjects died of complications of pneumonia, disturbance and irritability, both of which typically while the other three suffered an acute life-terminating emerge during infancy, but are experienced by patients of event that occurred in the context of an OGC in two cases. all ages. The sleep disturbance seems to evolve with age: The retrospective nature of this study prohibited precise infants have excessive sleepiness, while many children analysis of the childhood mortality risk associated with and adolescents exhibit prominent insomnia. AADCD. An additional limitation of the study design was Our findings confirm that OGCs are a near-universal the possibility of under-reporting some disease features due disease feature, and that they occur in the majority of to partially incomplete responses for some subjects. patients of all ages. In our cohort, they were most preva- Our findings illustrate that AADCD is associated with lent (97%) in children between ages 2 and 12 years. Epi- severe functional impairment in the majority of patients, sodes were reported to peak in duration, frequency and and that prominent non-motor and autonomic symptoms severity before age 6 years in many patients. Indeed, epi- occur throughout the lifespan. Disease features are sodes lasting longer than 4 hours were typical for 80% of refractory to existing medical therapies, particularly in subjects under age 6 years. Of note, for 2 of the deceased severe cases, which motivates the search for novel effec- subjects, parents reported an OGC as the apparent proxi- tive therapeutic strategies. mate cause of death, highlighting that OGCs represent not only a weekly burden of hours of symptom management, ACKNOWLEDGMENT but may also have the potential to be life-threatening. This study was supported by the Pediatric Neurotransmitter Our results provided additional insight into patients' Disease Association and NIH/NINDS (5 R01-NS094292). experiences with medications, previously described to be AGC is funded by FIS: PI18/0111 (Instituto de Salud Carlos of limited benefit for most patients.2,9 Pyridoxine was III: ISCIII and Fondo Europeo de desarrollo regional, well-tolerated, but not clearly observed to be beneficial. FEDER). Dopamine agonists, particularly rotigotine, were associ- ated with observed benefits in tone, movement, alertness, CONFLICT OF INTEREST and OGCs, but side effects including dyskinesia, insom- Dr. Pearson reports that she is an investigator on a NIH- nia, irritability, and vomiting were dose-limiting, leading funded gene therapy clinical trial for AADC deficiency. to discontinuation of the medication about 25% of the Dr. Opladen reports grants from Dietmar Hopp Founda- time. MAO-inhibitors were associated with a lower rate tion, during the conduct of the study. Dr. Garcia-Cazorla of side effects than DAs, but were also reported to be ben- reports personal fees from PTC Therapeutics, outside the eficial in only a minority of subjects. submitted work. Dr. Kato reports grants from Among the genotypes of our subjects were 14 novel Nobelpharma Co., Ltd., outside the submitted work. variants (base upon the recent review by.14 Those novel Dr. Gilbert, Dr. Mastrangelo, Dr. Leuzzi, Dr. Tay, variants were associated with a severe phenotype in all Dr. Sykut-Cegielska, Dr. Pons, Dr. Mercimek-Andrews, but one subject with a mild phenotype who had the het- Dr. Lücke, Dr. Oppebøen, Dr. Kurian, Dr. Steel, erozygous novel c.734C>T (p. T245I) variant and a known Dr. Manti, Dr. Jeltsch, Ms. Black, and Ms. Flint declare heterozygous c.367G>A (p. G123R) variant. In subjects of that they have no conflict of interest. southern Chinese ethnicity, the known common c. 714 +4A>T variant was the most frequently detected. A sec- AUTHOR CONTRIBUTIONS ond recurrent variant (c.1234C>T; p.R412W), recently Toni S. Pearson: Study design, data analysis, manuscript reported in multiple patients of Chinese ethnicity15 was initial draft, manuscript revision. Thomas Opladen, Lisa detected in three subjects in our cohort. Flint: Study design, data interpretation, manuscript 1130 PEARSON ET AL. revision. Laura Gilbert, Kathleen Meeks: Data analysis and 8. Tay SK, Poh KS, Hyland K, et al. Unusually mild phenotype of interpretation, manuscript revision. Angeles Garcia- AADC deficiency in 2 siblings. Mol Genet Metab. 2007;91:374- Cazorla, Mario Mastrangelo, Vincenzo Leuzzi, Stacy K. H. 378. https://doi.org/10.1016/j.ymgme.2007.04.006. 9. Brun L, Ngu LH, Keng WT, et al. Clinical and biochemical features Tay, Jolanta Sykut-Cegielska, Roser Pons, Saadet of aromatic L-amino acid decarboxylase deficiency. Neurology. Mercimek-Andrews, Mitsuhiro Kato, Thomas Lücke, Mari 2010;75:64-71. https://doi.org/10.1212/WNL.0b013e3181e620ae. Oppebøen, Manju Kurian, Dora Steel, Filippo Manti, 10. Chien Y-H, Lee N-C, Tseng S-H, et al. Efficacy and safety of Kathrin Jeltsch: Data collection (completed questionnaire) AAV2 gene therapy in children with aromatic L-amino acid for at least subject, data interpretation, manuscript revision. decarboxylase deficiency: an open-label, phase 1/2 trial. Lancet All authors approved the final version of the manuscript. Child Adolesc Health. 2017;1:265-273. https://doi.org/10.1016/ S2352-4642(17)30125-6. 11. Kojima K, Nakajima T, Taga N, et al. Gene therapy improves E TH IC S ST A T EME N T S motor and mental function of aromatic l-amino acid decarbox- This study was approved by the Human Research Protec- ylase deficiency. Brain. 2019;142:322-333. https://doi.org/10. tion Office of Washington University School of Medicine 1093/brain/awy331. (Protocol 201 801 179, approved 26 March 2018), and by 12. Brennenstuhl H, Kohlmüller D, Gramer G, et al. High through- the local ethics committees of the University Hospital Hei- put newborn screening for aromatic L-amino-acid decarboxyl- delberg (iNTD coordinating center, application S-471/2014, ase deficiency by analysis of concentrations of 3-O-methyldopa approved December 22, 2014), and Great Ormond Street from dried blood spots. J Inherit Metab Dis.2020;43(3):602–610. Hospital (protocol 19NM23). SM-A received an institu- https://doi.org/10.1002/jimd.12208. 13. Opladen T, Cortès-Saladelafont E, Mastrangelo M, et al. The tional case report consent form to share her patient's infor- international working group on neurotransmitter related disor- mation. This consent is available upon request. All ders (iNTD): a worldwide research project focused on primary procedures followed were in accordance with the ethical and secondary neurotransmitter disorders. Mol Genet Metab standards of the responsible committee on human experi- Rep. 2016;9:61-66. https://doi.org/10.1016/j.ymgmr.2016.09.006. mentation (institutional and national) and with the Hel- 14. Himmelreich N, Montioli R, Bertoldi M, et al. Aromatic amino sinki Declaration of 1975, as revised in 2000 (5). Informed acid decarboxylase deficiency: molecular and metabolic basis consent was obtained from all patients for being included and therapeutic outlook. Mol Genet Metab. 2019;127:12-22. in the study. This consent is available upon request. https://doi.org/10.1016/j.ymgme.2019.03.009. 15. Dai W, Lu D, Gu X, et al. Aromatic L-amino acid decarboxylase deficiency in 17 mainland China patients: clinical phenotype, ORCID molecular spectrum, and therapy overview. Mol Genet Genomic Toni S. Pearson https://orcid.org/0000-0002-3679-7707 Med. 2020;8(3):e1143. https://doi.org/10.1002/mgg3.1143. 16. Pons R, Ford B, Chiriboga CA, et al. Aromatic L-amino acid R EF E RE N C E S decarboxylase deficiency: clinical features, treatment, and prog- 1. Hyland K, Clayton PT. Aromatic amino acid decarboxylase nosis. Neurology. 2004;62:1058-1065. deficiency in twins. J Inherit Metab Dis. 1990;13:301-304. 17. Manegold C, Hoffmann GF, Degen I, et al. Aromatic L-amino 2. Wassenberg T, Molero-Luis M, Jeltsch K, et al. Consensus acid decarboxylase deficiency: clinical features, drug therapy guideline for the diagnosis and treatment of aromatic l-amino and follow-up. J Inherit Metab Dis. 2009;32:371-380. https://doi. acid decarboxylase (AADC) deficiency. Orphanet J Rare Dis. org/10.1007/s10545-009-1076-1. 2017;12:12. https://doi.org/10.1186/s13023-016-0522-z. 18. Ide S, Sasaki M, Kato M, et al. Abnormal glucose metabolism in 3. Lee HF, Tsai CR, Chi CS, et al. Aromatic L-amino acid decar- aromatic L-amino acid decarboxylase deficiency. Brain Dev. boxylase deficiency in Taiwan. Eur J Paediatr Neurol. 2009;13: 2010;32:506-510. https://doi.org/10.1016/j.braindev.2009.05.004. 135-140. https://doi.org/10.1016/j.ejpn.2008.03.008. 4. Hwu WL, Chien YH, Lee NC, Li MH. Natural history of aro- matic L-amino acid decarboxylase deficiency in Taiwan. JIMD SU PP O R TI N G I N F O RMA TI O N Rep. 2018;40:1–6. https://doi.org/10.1007/8904_2017_54. Additional supporting information may be found online 5. Hwu WL, Muramatsu S, Tseng SH, et al. Gene therapy for aro- in the Supporting Information section at the end of this matic L-amino acid decarboxylase deficiency. Sci Transl Med. article. 2012;4:134ra61. https://doi.org/10.1126/scitranslmed.3003640. 6. Leuzzi V, Mastrangelo M, Polizzi A, et al. Report of two never treated adult sisters with aromatic L-amino acid decarboxylase How to cite this article: Pearson TS, Gilbert L, deficiency: a portrait of the natural history of the disease or an Opladen T, et al. AADC deficiency from infancy to expanding phenotype? JIMD Rep. 2015;15:39-45. https://doi. org/10.1007/8904_2014_295. adulthood: Symptoms and developmental outcome 7. Mastrangelo M, Manti F, Patanè L, et al. Successful pregnancy in an international cohort of 63 patients. J Inherit in a patient with L-amino acid decarboxylase deficiency: thera- Metab Dis. 2020;43:1121–1130. https://doi.org/10. peutic management and clinical outcome. Mov Disord Clin 1002/jimd.12247 Pract. 2018;5:446-447. https://doi.org/10.1002/mdc3.12622.