Variation of the clinical spectrum and genotype-phenotype associations in Coenzyme Q10 deficiency associated glomerulopathy

Primary Coenzyme Q10 deficiency is a rare mitochondriopathy with a wide spectrum of organ involvement, including steroid-resistant nephrotic syndrome mainly associated with disease-causing variants in the genes COQ2, COQ6 or COQ8B. We performed a systematic literature review, PodoNet, mitoNET, and CCGKDD registries queries and an online survey, collecting comprehensive clinical and genetic data of 251 patients spanning 173 published (47 updated) and 78 new cases. Kidney disease was first diagnosed at median age 1.0, 1.2 and 9.8 years in individuals with disease-causing variants in COQ2, COQ6 and COQ8B, respectively. Isolated kidney involvement at diagnosis occurred in 34% of COQ2, 10.8% of COQ6 and 70.7% of COQ8B variant individuals. Classic infantile multiorgan involvement comprised 22% of the COQ2 variant cohort while 47% of them developed neurological symptoms at median age 2.7 years. The association of steroid-resistant nephrotic syndrome and sensorineural hearing loss was confirmed as the distinctive phenotype of COQ6 variants, with hearing impairment manifesting at average age three years. None of the patients with COQ8B variants, but 50% of patients with COQ2 and COQ6 variants progressed to kidney failure by age five. At adult age, kidney survival was equally poor (20-25%) across all disorders. A number of sequence variants, including putative local founder mutations, had divergent clinical presentations, in terms of onset age, kidney and non-kidney manifestations and kidney survival. Milder kidney phenotype was present in those with biallelic truncating variants within the COQ8B variant cohort. Thus, significant intra- and inter-familial phenotype variability was observed, suggesting both genetic and non-genetic modifiers of disease severity. Primary Coenzyme Q10 deficiency is a rare mitochondriopathy with a wide spectrum of organ involvement, including steroid-resistant nephrotic syndrome mainly associated with disease-causing variants in the genes COQ2, COQ6 or COQ8B. We performed a systematic literature review, PodoNet, mitoNET, and CCGKDD registries queries and an online survey, collecting comprehensive clinical and genetic data of 251 patients spanning 173 published (47 updated) and 78 new cases. Kidney disease was first diagnosed at median age 1.0, 1.2 and 9.8 years in individuals with disease-causing variants in COQ2, COQ6 and COQ8B, respectively. Isolated kidney involvement at diagnosis occurred in 34% of COQ2, 10.8% of COQ6 and 70.7% of COQ8B variant individuals. Classic infantile multiorgan involvement comprised 22% of the COQ2 variant cohort while 47% of them developed neurological symptoms at median age 2.7 years. The association of steroid-resistant nephrotic syndrome and sensorineural hearing loss was confirmed as the distinctive phenotype of COQ6 variants, with hearing impairment manifesting at average age three years. None of the patients with COQ8B variants, but 50% of patients with COQ2 and COQ6 variants progressed to kidney failure by age five. At adult age, kidney survival was equally poor (20-25%) across all disorders. A number of sequence variants, including putative local founder mutations, had divergent clinical presentations, in terms of onset age, kidney and non-kidney manifestations and kidney survival. Milder kidney phenotype was present in those with biallelic truncating variants within the COQ8B variant cohort. Thus, significant intra- and inter-familial phenotype variability was observed, suggesting both genetic and non-genetic modifiers of disease severity. Steroid-resistant nephrotic syndrome (SRNS) is one of the main causes of kidney failure in the first 2 decades of life.1Trautmann A. Schnaidt S. Lipska-Zietkiewicz B.S. et al.Long-term outcome of steroid-resistant nephrotic syndrome in children.J Am Soc Nephrol. 2017; 28: 3055-3065Crossref PubMed Scopus (137) Google Scholar,2Johansen K.L. Chertow G.M. Foley R.N. et al.US Renal Data System 2020 annual data report: epidemiology of kidney disease in the United States.Am J Kidney Dis. 2021; 77: A7-A8Abstract Full Text Full Text PDF PubMed Scopus (267) Google Scholar More than 60 causative genes, encoding podocyte-associated proteins, have been identified, explaining the disease etiology in up to 30% of pediatric SRNS cases. Primary coenzyme Q10 (CoQ10, ubiquinone) deficiency is the underlying disease cause in 1%–2.7% of SRNS cases, and in up to 10% of those in whom a genetic cause is identified.3Trautmann A. Bodria M. Ozaltin F. et al.Spectrum of steroid-resistant and congenital nephrotic syndrome in children: the PodoNet registry cohort.Clin J Am Soc Nephrol. 2015; 10: 592-600Crossref PubMed Scopus (217) Google Scholar,4Sadowski C.E. Lovric S. Ashraf S. et al.A single-gene cause in 29.5% of cases of steroid-resistant nephrotic syndrome.J Am Soc Nephrol. 2015; 26: 1279-1289Crossref PubMed Scopus (456) Google Scholar The term primary CoQ10 deficiency defines a group of rare mitochondrial disorders caused by recessive disease-causing variants in genes encoding proteins of the CoQ10 biosynthesis pathway. CoQ10 is a lipid component of the respiratory chain. Its deficiency leads to bioenergetic defects, H2S depletion, and oxidative stress in multiple cell types with a wide and variable spectrum of organ involvement and disease severity, ranging from single-organ disease to complex syndromic phenotypes. Defects of the podocyte lead to proteinuria and progressive loss of glomerular function. Among the 10 genes encoding proteins involved in CoQ10 biosynthesis, a kidney phenotype has mainly been associated with biallelic disease-causing variants in COQ2, COQ6, and COQ8B (previously referred to as ADCK4).5Schijvens A.M. van de Kar N.C. Bootsma-Robroeks C.M. et al.Mitochondrial disease and the kidney with a special focus on CoQ10 deficiency.Kidney Int Rep. 2020; 5: 2146-2159Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar, 6Tan W. Airik R. Primary coenzyme Q10 nephropathy, a potentially treatable form of steroid-resistant nephrotic syndrome.Pediatr Nephrol. 2021; 36: 3515-3527Crossref PubMed Scopus (7) Google Scholar, 7Alcazar-Fabra M. Trevisson E. Brea-Calvo G. Clinical syndromes associated with coenzyme Q10 deficiency.Essays Biochem. 2018; 62: 377-398Crossref PubMed Scopus (78) Google Scholar The CoQ10 deficiency–associated glomerulopathies are of particular interest as they are potentially amenable to therapeutic intervention with oral CoQ10 supplementation, through targeting of the underlying molecular defect. Up to now, the rarity of the primary CoQ10 deficiency disorders has limited the assessment of their spectrum of clinical presentation, their genotype–phenotype correlation, and their natural history. To advance the state of knowledge, we systematically reviewed 251 (173 published and 78 previously unreported) cases of CoQ10 deficiency–associated glomerulopathy. A comprehensive search strategy was applied to compile the study cohort from different sources (for details, see Figure 1). A systematic search of the PubMed database yielded 44 published articles with 174 patients. Three patient registries were queried: the international PodoNet registry8International PodoNet registrywww.podonet.orgDate accessed: March 31, 2021Google Scholar for children with primary steroid-resistant nephrotic syndrome,3Trautmann A. Bodria M. Ozaltin F. et al.Spectrum of steroid-resistant and congenital nephrotic syndrome in children: the PodoNet registry cohort.Clin J Am Soc Nephrol. 2015; 10: 592-600Crossref PubMed Scopus (217) Google Scholar the Chinese Children Genetic Kidney Disease Database9Chinese Children Genetic Kidney Disease Database.www.ccgkdd.com.cnDate accessed: March 31, 2021Google Scholar (CCGKDD); and the German Network for Mitochondrial Disorders (mitoNET10mitoNETGerman Network for Mitochondrial Disorders.www.mitonet.orgDate accessed: March 31, 2021Google Scholar).11Rao J. Liu X. Mao J. et al.Genetic spectrum of renal disease for 1001 Chinese children based on a multicenter registration system.Clin Genet. 2019; 96: 402-410Crossref PubMed Scopus (47) Google Scholar A total of 20, 17, and 7 new cases, respectively, were identified in these registries, and updated information was provided on 20,13, and 2 previously published cases. Finally, invitations to an online survey were sent to all members of the European Rare Kidney Disease Network (ERKNet), the PodoNet Consortium, the ESCAPE Clinical Research Network, and the European and Asian Societies for Pediatric Nephrology (ESPN and AsPN). Through this effort, 34 novel and 32 previously published patients, in 24 countries, were identified and documented. All patient-related data were collected in a completely deidentified manner. Names, initials, birth dates, and hospital-specific patient identifiers were not retrieved. Center identifiers were deleted from the database after checking for duplicate entries was completed, and only the country or residence was retained for demographic studies. Also, times and ages (e.g., age at diagnosis, time since diagnosis) were reported instead of calendar dates. As a result of these measures, all analyses were performed on completely anonymized datasets. All molecular diagnoses were re-evaluated by an expert geneticist, following the best practice recommendations of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) for the interpretation of sequence variants.12Richards S. Aziz N. Bale S. et al.Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.Genet Med. 2015; 17: 405-424Abstract Full Text Full Text PDF PubMed Scopus (17807) Google Scholar Variants classified as (likely) pathogenetic, per ACMG criteria, were considered causative and were included in the study. Two COQ2 variants with conflicting interpretations of pathogenicity according to the ClinVar database—namely NM_015697.8:c.288dupC [rs759310292] and c.683A>G [rs121918232]—were considered disease-causing due to their increased prevalence among affected individuals, compared to that among the general population, their presence in trans with another (likely) pathogenic variant in patients with highly specific phenotype, and the absence of homozygous cases in control databases. A previously published case with a homozygous COQ6 NM_182476.3:c.41G>A [rs17094161] variant was excluded from the analysis, as the variant was reclassified as benign because it is relatively frequent, is also in a homozygous state, in general populations, has been reported as benign in the CliniVar database, and is located upstream of the standard start codon of the canonical isoform of the gene. Descriptive data are presented as median and interquartile range. Patient and kidney survival rates were calculated using Kaplan–Meier lifetable analysis, with log–rank testing for analysis of significant differences. Fisher's exact test was performed for genotype-based pairwise comparisons, with 2 × 2 and 2 × 3 contingency tables of dichotomized data. Intra- and interfamilial phenotype variability was assessed by the coefficient of variation (CV = SD/mean) of the age at kidney-disease onset. A value of P < 0.05 was considered statistically significant. Statistical analyses were performed using Prism, version 8, data analysis software system (GraphPad Software). Clinical data were available for 63, 48, and 140 patients with disease-causing variants in COQ2, COQ6, and COQ8B, respectively. A summary of the phenotype characteristics and clinical outcomes per genetic diagnosis is given in Table 1. Patient-level genotype and phenotype characteristics are provided in Supplementary Table S1.Table 1Patient characteristics and clinical outcomesPatient characteristicCOQ2COQ6COQ8BTotal number of patients (females)63 (30)48 (16)140 (65)First disease manifestationAge at first symptoms, yr1 (0.3–2)1.2 (0.6–3.4)9.8 (5–15)Kidney involvement93.6 (59/63)97.9 (47/48)100 (140/140)Kidney disease presentationAge at first kidney disease manifestation, yr1 (0.5–2)2 (0.9–4.5)9.9 (5.3–14.4)Nephrotic range proteinuria85.7 (36/42)86.1 (31/36)71.7 (86/120)Non–nephrotic range proteinuria14.3 (6/42)13.9 (5/36)28.3 (34/120)Asymptomatic proteinuria0 (0/63)18.7 (9/48)23.6 (33/140)Hypertension28.6 (12/42)21 (4/19)39 (32/82)Oedema86.6 (39/45)47.8 (11/23)40.2 (33/82)Microhematuria6.8 (4/59)8.5 (4/47)18 (13/72)CKD stage 171.8 (23/32)42.1 (8/19)34.3(35/102)CKD stage 2–418.7 (6/32)36.8 (7/19)32.3 (33/102)ESKD9.4 (3/32)21 (4/19)33.3 (34/102)Renal histopathologic findings FSGS69.4 (25/36)72.2 (26/36)77.1 (64/83)FSGS, not otherwise specified76 (19/25)88.4 (23/26)89 (57/64)FSGS, collapsing subtype24 (6/25)11.5 (3/26)9.4 (6/64)FSGS, tip-lesion0 (0/25)0 (0/26)1.5 (1/64) Global glomerulosclerosis11.1 (4/36)8.3 (3/36)15.6 (13/83) Mesangioproliferative glomerulonephritis11.1 (4/36)5.6 (2/36)7.2 (6/83) Minimal change disease5.6 (2/36)5.6 (2/36)0 (0/83) Dysmorphic mitochondria30.5 (11/36)25 (9/36)10.8 (9/83)Extrarenal features Any extrarenal involvement77.9 (46/59)89.1 (41/46)29.3 (41/140) Intrauterine abnormalities/preterm delivery13.6 (8/59)2.2 (1/46)0.7 (1/140) Infantile multisystemic disease/multi-organ failure22 (13/59)0 (0/46)0 (0/140) Neurologic findings47 (28/59)21.7 (10/46)12.1 (17/140)Encephalopathy/seizures42 (25/59)8.7 (4/46)7.1 (10/140)Developmental delay/cognitive impairment5 (3/59)13 (6/46)5 (7/140) Retinopathy/ocular abnormalities20.3 (12/59)17.4 (8/46)5 (7/140) Hearing abnormalities1.7 (1/59)73.9 (34/46)0 (0/140) Myopathy20.3 (12/59)8.7 (4/46)0 (0/140) Cardiovascular abnormalities15.2 (9/59)8.7 (4/46)7.1 (10/140) Liver dysfunction13.6 (8/59)2.2 (1/46)0 (0/140) Growth retardation11.8 (7/59)8.7 (4/46)3.6 (5/140) Facial/body dysmorphisms3.4 (2/59)4.3 (2/46)1.4 (2/140)Clinical outcome (status at last follow-up) Median follow-up time, yr1.5 (0.3–4.5)1.7 (0–5.4)3.9 (1.3–6.9) Deceased25.4 (16/63)10.4 (5/48)4.3 (6/140) Median age at death, yr0.5 (0.3–0.9)6.5 (5.7–12)13.9 (12.6–19.5) Survival rate at age 1 yr, %77.2100100 Survival rate at age 10 yr, %69.990.899.2 ESKD50.8 (32/63)56.2 (27/48)66.4 (93/140) Median age at ESKD, yr2.5 (0.7–5.8)3.4 (1.7–6.3)13 (10–16.7) Median time from first manifestation to ESKD, yr0.5 (0–1.6)1.0 (0.3–2.1)1.0 (0–4.2) Probability of ESKD at age 5 yr, %47.647.83.1 Probability of ESKD at age 18 yr, %80.872.374.2 Kidney transplantation26.9 (17/63)29.2 (14/48)26.4 (37/140) Median age at kidney transplantation, yr8.3 (5.2–17.2)6.4 (5.1–7.5)15.1 (11–16)CKD, chronic kidney disease; ESKD, end-stage kidney disease; FSGS, focal segmental glomerulosclerosis.Values are % (number of affected patients/informative number of patients) or median (interquartile range), as appropriate. Open table in a new tab CKD, chronic kidney disease; ESKD, end-stage kidney disease; FSGS, focal segmental glomerulosclerosis. Values are % (number of affected patients/informative number of patients) or median (interquartile range), as appropriate. Patient survival was markedly compromised with COQ2 deficiency (Table 1), with 77.3% (95% confidence interval [CI], 67.8%–84.2%) of patients being alive at 1 year of age, and 69.9% (95% CI, 55.1%–80.6%) being alive at 10 years of age. Ten-year survival from birth was 90.8% (95% CI, 67.7%–97.6%) for COQ6, and 99.2% (95% CI, 94.7%–99.8%) for COQ8B disease (P < 0.0001; Figure 2). The leading cause of death in COQ2 disease patients was severe multisystemic involvement leading to multiorgan failure (n = 13) or progressive neurologic deterioration (n = 2). In patients with COQ6 and COQ8B, disease sepsis was the most common cause of death. The age at disease onset differed markedly depending on the gene involved. Whereas 50% of the COQ2 and COQ6 cohort manifested first symptoms of kidney disease within the first 15 months of life, and 85% were symptomatic at 3 and 5 years of age, respectively, 50% of COQ8B disease patients remained asymptomatic until age 9 years (Figure 3a). A total of 14.3% of all patients presented kidney disease beyond 16 years of age. The oldest documented ages at kidney disease diagnosis were 11.6 years, 27 years, and 32.2 years in patients with COQ6-, COQ2-, and COQ8B-related disease, respectively. Isolated kidney involvement at diagnosis was found in 34% of the COQ2 cohort, 10.8% of the COQ6 cohort, and 70.7% of the COQ8B cohort. Kidney involvement was observed in 98% (246 of 251) of patients (Figure 3b). In 4 of 63 children with COQ2 disease, no kidney manifestations were reported; 3 were deceased within the first 6 months of life. Additionally, 1 of 48 with confirmed COQ6 disease, a subject diagnosed by family screening at 10 years of age, presented with hearing impairment but still had no renal symptoms at last observation at age 17 years. With the term end-stage kidney disease (ESKD), we identified patients with a glomerular filtration rate <15/ml per 1.73 m2 (chronic kidney disease [CKD] stage 5, as delineated by Kidney Disease: Improving Global Outcomes [KDIGO]). Nephrotic-range proteinuria was reported in 85.7% of the COQ2 disease patients, 86.1% of the COQ6 disease patients, and 71.7% of the COQ8B disease patients. Although the COQ2 group typically presented with edema and hypertension but normal kidney function, nearly 40% of children with COQ6 disease presented with mild-to-moderate CKD (CKD 2–4), and up to 20% presented with ESKD (CKD 5) at the time of diagnosis. Children with COQ8B disease presented with the most-advanced CKD, with one third already in ESKD at the time of diagnosis. Notably, asymptomatic proteinuria was present in 18.7% and 23.6% of children with COQ6 and COQ8B disease, respectively, including a few younger siblings of index cases (2 of 9 with COQ6 disease and 5 of 33 with COQ8B disease). Concomitant hematuria was reported at diagnosis in only 6.8% and 8.5%, respectively, of COQ2 and COQ6 disease patients, but in 18% of the COQ8B cohort. Kidney biopsy was performed in 155 patients (61.7%). Focal segmental glomerulosclerosis accounted for 74.2% of the histopathologic diagnoses. Among those patients with focal segmental glomerulosclerosis in whom a specific subtype was reported, 94% displayed collapsing variant and 6% displayed tip lesions. Dysmorphic mitochondria were observed in 80.5% of biopsies in which electron microscopy results were reported (18.7% of all biopsies), namely in 30% of the COQ2 cohort, 25% of the COQ6 cohort, and 10% of the COQ8B cohort patients. In 3 cases, mitochondrial abnormalities identified by electron microscopy of kidney tissue led to the diagnosis of a mitochondriopathy prior to genetic testing. Approximately 60% of individuals progressed to ESKD during follow-up (Table 1). Although the COQ2 and COQ6 cohorts tended to progress faster than the COQ8B cohort in the first decade of life, COQ8B disease patients were more likely to develop ESKD in the second decade (Figure 4). Interestingly, for all 3 groups, a significant probability of ESKD persists beyond 25 years of age. Overall, 70 surviving patients were on dialysis at last observation (7 in the COQ2 cohort, 8 in the COQ6 cohort, and 55 in the COQ8B cohort), and 68 had received a kidney transplant. Disease recurrence in the allograft was never observed. Extrarenal disease manifestations were most frequent and severe in the COQ2 cohorts and least common in COQ8B disease (Table 1; Figure 3). Neurologic manifestations occurred in 47% of the COQ2 cohort patients (60.8% of patients with extrarenal manifestations); these included encephalopathy, ataxia, seizures, nystagmus, and any degree of psychomotor delay or intellectual disability. Although neurologic symptoms manifested within the first 3 years of life in 87% of patients (Figure 3c), 2 subjects showed late-onset symptoms—that is, headache with phono- and photophobia at 17 years, and myoclonic epilepsy at 18 years. Neuroimaging revealed a wide spectrum of abnormalities, including cerebral or cerebellar cortical atrophy, stroke-like lesions, basal ganglia involvement, hemorrhage (1 of 42), and intracranial calcifications (1 of 42). Thirteen COQ2 disease cases (20.6%) were characterized by severe infantile multiorgan disease, encompassing some degree of encephalopathy, myopathy, cardiomyopathy, retinopathy, and/or metabolic disorders (hyperlactatemia and/or diabetes mellitus), which led to rapidly progressive clinical deterioration and death at a median age of 6 months. In contrast to these severe courses, 22 patients (34%) were reported as not showing any extrarenal symptoms at the time of diagnosis. However, extrarenal symptoms developed in 9 of 22 patients with initially isolated kidney disease—specifically, in 6 patients within 3 years, and in a further 3 patients after up to 20 years of follow-up. The association of SRNS and sensorineural hearing loss was confirmed as the distinct phenotype of COQ6 disease. Hearing impairment was present in 73.9% (34 of 48) of the COQ6 cohort, but in only one child with COQ2, and in none of those with COQ8B disease. Among COQ6 disease patients, 4 cases had congenital deafness and 19 (19.6%) showed later-onset sensorineural hearing loss, with a median age at diagnosis of 5 years (interquartile range: 2.9–6 years; Figure 3d). In 11 patients (24%), hearing loss preceded the onset of renal symptoms. Almost a quarter of the COQ6 cohort showed other neurologic abnormalities, namely seizures and/or cognitive impairment; severe encephalopathy was reported in one case. If present, neurologic symptoms usually occurred within the first 2 years of life (Figure 3c). Individuals with COQ8B disease showed a much milder extrarenal phenotype; the most common findings were neurologic abnormalities (12.1%), including mostly moderate neurocognitive impairment, developmental delay, and seizures. A total of 39 (21 novel, hitherto unreported as [likely] pathogenic) different variants in COQ2, 16 (5 novel) variants plus one deletion encompassing the entire gene sequence in COQ6, and 40 variants (8 novel) in COQ8B were identified (Figure 5a–c; Supplementary Table S2). For COQ2, the detected variants were distributed randomly throughout the coding sequence of the gene. By contrast, all COQ6 missense variants clustered to 2 gene regions coding fragments involved directly in the binding of oxidized flavin adenine dinucleotide (FAD) (residues 194–300 and 340–425; NP_872282.1; Figure 5b). For COQ8B, 3 hot-spot regions were recognized: residues 147–154 (loop between GQα1 and GQα2 helices), 174–184 (GQα3 helix), and 246–253 (GQα5 helix); NP_079152.3; Figure 5c (helices nomenclature according to Stefely et al.13Stefely J.A. Reidenbach A.G. Ulbrich A. et al.Mitochondrial ADCK3 employs an atypical protein kinase-like fold to enable coenzyme Q biosynthesis.Mol Cell. 2015; 57: 83-94Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar The 2 former are also frequently mutated in the twin protein COQ8A,14Traschutz A. Schirinzi T. Laugwitz L. et al.Clinico-genetic, ımaging and molecular delineation of COQ8A-ataxia: a multicenter study of 59 patients.Ann Neurol. 2020; 88: 251-263Crossref PubMed Scopus (45) Google Scholar whereas the variants located in the latter (GQα5) are specific for COQ8B. See Supplementary Material S1 and Supplementary Movie S1 for further COQ8B structural characteristics. The geographical clustering of the recurrent variants to certain regions suggests founder effects in several populations (Figure 5d). Intra- and interfamilial phenotype variability was globally assessed for variants identified in at least 2 homozygous individuals from at least 2 different families, which was the case in 60 patients from 16 COQ8B and 6 COQ6 families harboring 7 different variants. A synopsis of the clinical timelines by gene, variant, and family is provided in Supplementary Figure S1. Age at disease onset was least variable among members of families sharing the same causative variant (intra-family, intra-variant CV: 35%) and similar among family members across families harboring a different causative variant (intra-family, inter-variant CV: 40%). Disease onset was more variable when different families shared the same variant (intra-variant, inter-family CV: 64%) and most diverse between families with different causative variants (inter-variant, inter-family CV: 106%; Figure 6). Specific genotype–phenotype associations were explored for recurrent variants reported in at least 5 homozygous individuals or in at least 5 unrelated families (Supplementary Tables S3A–C). Patients who were homozygous for the most frequent COQ2 variant c.683A>G (p.Asn178Ser) had a lower risk of disease onset in the first 2 years of life (odds ratio [OR] 0.07; 95% CI 0.0–0.56, P = 0.0292). The homozygous variant c.437G>A (p.Ser96Asn), present in 6 non-related Turkish children, was associated with a severe disease course characterized by higher risk of disease onset in the first 6 months of life (P = 0.0006), higher risk of death (P < 0.0001), obligate neurologic symptoms, and multiorgan failure, which was much less commonly observed in the rest of the COQ2 cohort (P < 0.0001). Five of the 6 children received genetic testing due to a phenotype suggestive of mitochondriopathy, while SRNS was the leading complaint in the remaining COQ2 disease patients. The subgroup carrying the most frequent variant c.1058C>A (p.Ala353Asp), clustering in Kazak, Turkish, and Iranian populations, showed later disease onset, with a lower risk of symptoms within the first 15 months of life (P = 0.0036). Conversely, homozygosity for c.1078C>T (p.Arg360Trp), present in Central/Eastern Europe and China, associated with a higher risk of neurologic involvement, myopathy, cardiomyopathy, and growth retardation (P < 0.05). The c.763G>A (p.Gly255Arg) variant, predominant in the Middle East, presented with severe phenotype, early disease onset (P = 0.0207), and early-onset ESKD (within the age of 2 years, OR 13.6, 95% CI 1.4–173, P = 0.0297), as well as higher odds of mortality (P = 0.0104). Patients with the variant NM_024876.4:c.748G>C (p.Asp250His), observed in China, showed a higher risk of developing kidney disease within the first 5 years of life (OR 5.7; 95% CI 1.5–18.7, P = 0.0116). Patients homozygous for variant c.1339dupG, common in patients of Turkish and Kurd descent, demonstrated later disease onset (>10 years, P = 0.0044) and later ESKD (>12 years, P = 0.0055), with higher odds of microhematuria at diagnosis (P = 0.0497). The variant c.1199dup, also common in patients of Turkish ancestry, associated with slower disease progression (P = 0.0165) and ESKD attained at a median age of 16 years (interquartile range: 14.4–17.6 years). Individuals with c.293T>G (p.Leu98Arg), another variant prevalent in patients of Turkish descent, had a higher prevalence of extrarenal symptoms (OR 6.7; 95% CI 1.3–34.5, P = 0.0228). In individuals carrying c.532C>T (p.Arg178Trp), ocular anomalies were reported more commonly. Patients with COQ2 and COQ6 disease carried at least one missense variant, except for 4 individuals with biallelic truncating variants located close to 3'UTR that are known to be associated with some residual enzymatic activity.15Desbats M.A. Morbidoni V. Silic-Benussi M. et al.The COQ2 genotype predicts the severity of coenzyme Q10 deficiency.Hum Mol Genet. 2016; 25: 4256-4265Crossref PubMed Scopus (46) Google Scholar On the contrary, one third (47 of 140) of COQ8B disease patients had biallelic truncating variants (Supplementary Table S4). Patients without biallelic truncating alleles in COQ8B showed earlier disease onset (onset before age 10 years: OR = 2.2; 95% CI 1.0–4.8, P = 0.0384) and earlier kidney failure (ESKD before age 15 years: OR = 7.7; 95% CI 2.9–20.6, P < 0.0001). Patients carrying at least one allele with a disease-causing missense variant affecting residues in the GQα1/2 or GQα3 motif were compared with the rest of the COQ8B population, excluding patients with biallelic truncating variants. ESKD was reached in 83% of patients with a variant in the GQα1/2 or GQα3 motif during the period of observation, as compared to 61% in the remaining population (P = 0.0475). Similar analysis was performed for patients carrying a missense variant in the GQα5 motif on at least one allele. Patients with a missense variant affecting GQα5 reached ESKD at an earlier age; however, the observation was not confirmed in the restricted group of patients carrying biallelic missense variants affecting residues within the GQα5 motif. We present the largest international cohort collected thus far of primary CoQ10-deficiency patients carrying biallelic disease-causing variants in COQ2, COQ6, and COQ8B, the 3 causative genes associated with kidney involvement and SRNS. Although the estimation of the exact prevalence of CoQ10 deficiencies is beyond the scope of this study, published registry reports suggest that these disorders account for ∼0.5% of mitochondriopathies, ∼2% of childhood SRNS/persistent proteinuria, and up to 5% of molecularly confirmed genetic SRNS cases.4Sadowski C.E. Lovric S. Ashraf S. et al.A single-gene cause in 29.5% of cases of steroid-resistant nephrotic syndrome.J Am Soc Nephrol. 2015; 26: 1279-1289Crossref PubMed Scopus (456) Google Scholar,16Trautmann A. Lipska-Zietkiewicz B.S. Schaefer F. Exploring the clinical and genetic spectrum of steroid resistant nephrotic syndrome: the PodoNet Registry.Front Pediatr. 2018; 6: 200Crossref PubMed Scopus (74) Google Scholar We found significant geographical divergence, with a globally higher incidence of CoQ10-related pathogenic variants observed in Asian populations. Given that these ethnicities are currently underrepresented in mostly Caucasian-based registries, global prevalence figures might differ notably. Already, the recent multicenter Chinese study reported COQ8B