Genetic Screening of Patients With Inherited Fanconi Syndrome Authors: Yuta Inoki, Nana Sakakibara, Asahi Yamamoto, Yuka Kimura, Shuhei Aoyama, Atsushi Kondo, Hiroaki Hanafusa, Tomoko Horinouchi, Tomohiko Yamamura, China Nagano, Shingo Ishimori, Shinya Asato, Akihiro Fukuda, Shun Hashimoto, Tsubasa Murase, Koji Nagatani, Izaya Nakaya, Takayuki Okamoto, Shunsuke Takayanagi, Hiroshi Tanaka, Koji Tsugawa, Tatsuo Tsukamoto, Takeshi Yamada, Masaki Yamamoto, Kandai Nozu Published: 2026 ABSTRACT AbstractIntroductionThe clinical and genetic spectrum of inherited Fanconi renotubular syndrome (FRTS) remains incompletely characterized, and the role of recently recognized genes such as GATM has not been systematically assessed.MethodsThis retrospective study included 20 families with FRTS (14 pediatric and 6 adult index cases) who underwent targeted panel sequencing between 2016 and 2024. The variant spectrum and clinical characteristics were evaluated. Previously reported cases were reviewed to elucidate the phenotypic spectrum and renal prognosis in patients with GATM variants.ResultsPathogenic variants were identified in 13 of 20 index cases (65%; 4 pediatric, 9 adult cases). Disease-causing genes detected included GATM (n = 7), CLCN5 (n = 2), HNF4A (n = 1), BCS1L (n = 1), CTNS (n = 1), and EHHADH (n = 1). Notably, 4 of the 7 GATM variants were completely novel, and 2 others were first reported by our group. The review, incorporating our cases, revealed that patients with GATM variants frequently exhibited incomplete FRTS, expanding the known clinical spectrum. Kidney function exhibited progressive decline, with combined analyses of our cohort and published cases indicating a median kidney survival age of 51.9 years, establishing GATM-associated FRTS as progressive tubulopathy.ConclusionOver half of the patients with primary FRTS displayed detectable monogenic etiologies, with GATM variants being unexpectedly prevalent. Our findings provide new insights into the genetic and clinical spectrum of FRTS with GATM variants, highlighting its under-recognition, phenotypic variability, and progressive kidney dysfunction. GATM variants should be considered in patients with incomplete proximal tubular abnormalities or unexplained chronic kidney disease (CKD).Keywords: Fanconi syndrome, genetic kidney disease, GATM, targeted panel sequence FRTS is characterized by generalized dysfunction of the renal proximal tubules, resulting in excessive urinary excretion of amino acids, glucose, phosphate, uric acid, bicarbonate, low-molecular-weight proteins, and other solutes.1 The etiology of FRTS is categorized into acquired causes, exogenous substances, and hereditary forms. Hereditary FRTS comprises a heterogeneous group of disorders, including primary and secondary FRTS (e.g., Lowe syndrome, Dent disease, cystinosis, and mitochondrial disorders).1, 2, 3 Primary FRTS includes FRTS types 1to 5, identified over the past decade, and idiopathic FRTS with an unknown genetic mechanism.4, 5, 6, 7, 8 The responsible genes and their inheritance patterns are GATM (type 1, autosomal dominant; online Mendelian inheritance in man [OMIM]: 134600), SLC34A1 (type 2, autosomal recessive; OMIM: 613388), EHHADH (type 3, autosomal dominant; OMIM: 615605), HNF4A (type 4, autosomal dominant; OMIM: 616026), and NDUFAF6 (type 5, autosomal recessive; OMIM: 618913). With the widespread adoption of next-generation sequencing, comprehensive genetic testing has revealed that a considerable proportion of patients with unexplained CKD or tubular disorders harbor pathogenic variants in genes causing inherited kidney disease. Recent studies have reported that a genetic etiology can be identified in approximately 1 in 5 adults and 1 in 3 children with unexplained CKD, including FRTS.9, 10, 11, 12 Among primary FRTS entities, FRTS because of GATM variants has only been reported in a limited number of families worldwide, and available data indicate autosomal dominant inheritance with slowly progressive kidney dysfunction.3,8,13, 14, 15, 16, 17 Despite these advancements, the overall contribution of monogenic etiologies to clinically diagnosed FRTS remains unclear, and no study has systematically evaluated the diagnostic yield of genetic testing specifically in patients with inherited FRTS. Furthermore, the phenotypic spectrum, penetrance, and kidney prognosis associated with each causative gene—particularly recently recognized genes such as GATM, EHHADH, and HNF4A—have not been fully elucidated. For GATM specifically, available data emanate almost exclusively from single families or small case series, and no large cohort study has systematically evaluated its clinical and renal manifestations. Consequently, the true frequency, phenotypic variability, and long-term kidney outcomes of patients with GATM variants remain largely unknown, and early clinical manifestations may be subtle or incomplete, raising concern that such cases are potentially underdiagnosed. In this context, this study aimed to systematically investigate the genetic landscape of inherited FRTS in a well-defined Japanese cohort using targeted panel sequencing. Specifically, we sought to determine the detection rate and spectrum of variants in known causative genes and to characterize the associated clinical features. Additionally, by integrating newly identified patients with previously published cases, we focused on clarifying phenotypic diversity and kidney prognosis associated with FRTS because of GATM variants. Methods Study Design and Patient Population This retrospective observational study included patients diagnosed with FRTS who underwent genetic analysis at Kobe University Graduate School of Medicine from January 2016 to December 2024. FRTS was characterized by low-molecular-weight proteinuria in conjunction with 1 or more of the following symptoms: generalized aminoaciduria, glucosuria, hypophosphatemia, metabolic acidosis, or hyperuricosuria. Patients diagnosed with acquired FRTS or those with a history of exogenous substance use were excluded. Patients clinically diagnosed with Lowe syndrome, characterized by congenital cataracts, hypotonia, and severe developmental delay, were excluded, because at our institution these cases are separately enrolled in a Lowe syndrome cohort rather than being classified as idiopathic FRTS. Data Collection and Analysis The following clinical data were extracted from medical records: sex, age at symptoms onset, age at genetic testing, initial symptoms, height and body weight at the initial and last visits, presence of extrarenal symptoms, and family history. Laboratory data and pathological findings were also extracted from medical records. Creatinine based-estimated glomerular filtration rate (Cr-eGFR, ml/min per 1.73 m2) was calculated using an equation specific to Japanese patients aged 2 to 18 years18 and a different equation for those aged ≥ 19 years.19 The final follow-up was conducted on December 1, 2024. All patients were analyzed using a targeted panel sequence. However, some patients were initially analyzed with an older panel that did not include newly identified causative genes such as GATM and BCSL1. These patients were re-analyzed in 2024 using an updated targeted panel sequence that included all of these genes. For the patient with BCSL1 variants (Patient 9), the variants were not identified using the initial targeted panel sequence but were first detected through whole exome sequencing. These variants were subsequently confirmed using our updated targeted panel sequence during re-analysis. The identification rate of disease-causing pathogenic variants and the clinical course were assessed. For cases with GATM variants, clinical information, from previously reported cases published in PubMed, was collected to facilitate comparison with our cohort. Genetic Analysis Genomic DNA was extracted from the peripheral blood leukocytes of patients using the Quick Gene Mini 80 System or QuickGene-Auto 12S (Kurabo Industries Ltd., Tokyo, Japan), according to the manufacturer’s protocol. Targeted panel sequencing was conducted for genes associated with inherited kidney diseases including GATM, SLC34A1, EHHADH, HNF4A, NDUFAF6, SLC2A2, CLCN5, OCRL, CTNS, BCS1L, and EHD1 (Supplementary Table S1). Sequencing samples were prepared using the Haloplex or Sure Select target enrichment system kit (Agilent Technologies, Santa Clara, CA) according to the manufacturer’s instructions. All indexed DNA samples were amplified via polymerase chain reaction and sequenced using the MiSeq platform (Illumina, San Diego, CA). Sequence data were processed and analyzed using 2 pipelines based on the sequencing date. For data generated up to May 2024, reads were processed using SureCall software (Agilent Technologies, Santa Clara, CA), encompassing alignment through variant calling and categorization. For data generated from June 2024 onward, variant calling was performed according to the Genome Analysis Toolkit Best Practices workflows developed by the Broad Institute (Cambridge, MA),20 and variants were annotated using ANNOVAR.21 The pathogenicity of each variant was assessed according to the American College of Medical Genetics consensus guidelines.22 In silico analyses were conducted using the following prediction tools: Sorting Intolerant from Tolerant (http://sift.bii.a-star.edu.sg), PolyPhen2 (http://genetics.bwh.harvard.edu/pph2/), Mutation Taster (https://www.mutationtaster.org/), Combined Annotation Dependent Depletion (https://cadd.gs.washington.edu/snv), and Rare Exome Variant Ensemble Learner (https://sites.google.com/site/revelgenomics). Additionally, when a suspected splicing site variant was detected, a “minigene” in vitro splicing assay was conducted to assess its splicing pattern as previously reported.23 Statistical Analyses Data are expressed as medians with interquartile range for continuous variables or as percentages for categorical variables. The Kaplan–Meier method was employed to assess the occurrence of events (median age for developing kidney failure), expressed as the median survival time. Analyses were performed using JMP (version 14.0,SAS Institute Japan, Tokyo, Japan). Results During the study period, 20 families met the inclusion criteria and were enrolled. Disease-causing variants were identified in 13 of the 20 families (65.0%), 4 of which had been previously reported by our group.14,24, 25, 26 The disease-causing genes identified were GATM (n = 7), CLCN5 (n = 2), HNF4A (n = 1), BCS1L (n = 1), CTNS (n = 1), and EHHADH (n = 1). Intellectual disability was observed in 1 patient with a BCS1L variant (1/13) among those with detected variants, whereas it was identified in 4 of 7 patients without pathogenic variants. The detected pathogenic variants are summarized in Table 1.27, 28, 29, 30, 31, 32, 33 GATM was the gene most frequently found to harbor pathogenic variants. Among the detected GATM variants, 4 were completely novel. Two additional variants (c.802C>G and c.888T>A) were initially identified in this cohort and have been recently published as case reports by our group.14,26 The HNF4A variant (Patient 8), BCS1L variant (Patient 9), CTNS variant (Patient 10), CLCN5 variants (Patients 11 and 12), and the EHHADH variant (Patient 13) had all been previously reported. Table 1Genetic findingsFamilyGeneMOIExonNucleotide change (zygosity)Amino acid changeType of variantgnomAD allele frequencyMissense pathogenic predictorsClinVar Clinical SignificancePrevious reportACMG criteriaSIFTCADDMutation TasterPolyPhen2REVEL1GATM (NM_001482.3)ADExon 5c.802C>Gp.(Gln268Glu)Missense–Tolerated26.2Benignprobably damaging0.532NATakayanagi et al.26 (This case)Likely pathogenic (PS3, PM1, PM2, PP1, PP3)2GATM (NM_001482.3)ADExon 6c.958C>Ap.(Pro320Thr)Missense–Deleterious24.0Deleteriouspossibly damaging0.552NANot reportedLikely pathogenic (PM1, PM2, PM5, PP3, PP4)3GATM (NM_001482.3)ADExon 6c.888T>Ap.(Phe296Leu)Missense–Deleterious24.6Deleteriousprobably damaging0.271NAKudo et al.14(This case)Likely pathogenic (PM1, PM2, PP1, PP3, PP4)4GATM (NM_001482.3)ADExon 8c.1079T>Ap.(Met360Lys)Missense6.85 ∗ 10-7Tolerated27.3Deleteriousprobably damaging0.679NANot reportedLikely pathogenic (PM1, PM2, PP3, PP4)5GATM (NM_001482.3)ADExon 3c.484G>Ap.(Gly162Ser)Missense–Tolerated33.0Deleteriousprobably damaging0.732NANot reportedLikely pathogenic (PS2, PM1, PM2, PP4)6GATM (NM_001482.3)ADExon 6c.958C>Tp.(Pro320Ser)Missense–Deleterious24.0Deleteriouspossibly damaging0.356Pathogenic (917493)Reichold et al.8Likely pathogenic (PM1, PM2, PP3, PP5)7GATM (NM_001482.3)ADIntron 2c.288+8A>Tp.(Gly97_Arg98insValLys)Splicing––––––NANot reportedLikely pathogenic (PS3, PM2)8HNF4A (NM_175914.4)ADExon 2c.187C>Tp.(Arg63Trp)Missense–Deleterious29.2Deleteriousprobably damaging0.949Pathogenic (156152)Flanagan et al.27Pathogenic (PS2, PM1, PM2, PP3, PP5)9BCS1L (NM_004328.5)ARExon 3c.268C>Tp.(Arg90Cys)Missense8.05 ∗ 10-6Deleterious24.5Deleteriousbenign0.684Likely pathogenic (625207)Calvoet al.28Likely pathogenic (PM1, PM2, PM5, PP3, PP5, BP4)Exon 7c.821delp.(Pro274Argfs∗26)Nonsense1.86 ∗ 10-6–––––Pathogenic (555982)Shigematsu et al.29Pathogenic (PVS1, PM2, PP5)10CTNS (NM_004937.3)ARIntron 6c.329+2T>C–Splicing1.37 ∗ 10-6–––––Pathogenic (371171)Higashi et al.30Pathogenic (PVS1, PM2, PP5)Exon 11c.922G>Ap.(Gly308Arg)Missense2.97 ∗ 10-5Deleterious31.0Deleteriousprobably damaging0.991Pathogenic (276310)Shotelersuk et al.31Likely pathogenic (PM1, PM2, PM3, PP3, PP5)11CLCN5 (NM_000084.5)X-linkedExon 7c731C>Tp.(Ser244Leu)Missense9.15 ∗ 10-7Deleterious26.9DeleteriousNo result0.958Pathogenic (11802)Lloyd et al.32Likely pathogenic (PM1, PM2, PP3, PP5)12CLCN5 (NM_000084.5)X-linkedExon 2c.100C>Tp.(Arg34∗)Nonsense––––––Pathogenic (207994)Hoopes et al.33Pathogenic (PVS1, PM2, PP5)13EHHADH (NM_001966.3)ADExon 1c.7 G>Ap. (Glu3Lys)Missense–Tolerated22.9Deleteriousbenign0.669Pathogenic (96711)Klootwijket al.6Pathogenic (PS3, PM2, PP5, BP4)AD, autosomal dominant; AR, autosomal recessive; MOI, mode of inheritance; NA, not available. Genetic findings AD, autosomal dominant; AR, autosomal recessive; MOI, mode of inheritance; NA, not available. Figure 1 summarizes both the previously reported GATM variants and those identified in our study. In our cohort, among the families with GATM variants, 6 possessed missense variants, whereas 1 family carried an intronic variant. A minigene assay revealed that the intronic GATM variant (c.288+8A>T) caused an in-frame insertion of 6 nucleotides (p.[Gly97_Arg98insValLys]) (Supplementary Figure S1). To further support the causality of the identified GATM variants, we performed segregation analysis on available family members in 5 families (Patients 1–5) (Supplementary Figure S2). The pathogenic alleles cosegregated with the disease phenotype in Families 1 and 3, where affected relatives were confirmed to harbor the same variants. In Families 2 and 4, the variants were confirmed to be absent in the tested unaffected parents. Furthermore, trio analysis demonstrated that the variant in patient 5 occurred de novo. Detailed characteristics of the previously reported GATM variants are summarized in Supplementary Table S2. Localization of GATM variants detected in the analyzed cohort with Fanconi renotubular syndrome or chronic kidney disease. The GATM reference sequence used was NM_001482.3 (GRCh38 assembly), corresponding to the ENSEMBL transcript ENST00000396659.8 (ENSG00000171766.17). The genomic structure of the GATM gene is shown, with boxes representing exons (numbered 1 to 9). Novel GATM variants identified for the first time in this study are highlighted in red. Previously reported variants derived from the literature are shown in black. The origins and detailed characteristics of the previously reported variants are summarized in Supplementary Table S2. Clinical Features of Patients With GATM Variants Clinical characteristics of the 7 patients with GATM variants in this study, along with summary data from previously reported cases (n = 35), are presented in Table 2.34,35 Furthermore, the detailed clinical phenotypes of these previously reported patients are comprehensively summarized in Supplementary Table S3. Among our patients, the clinical details of 2 individuals (Patients 1 and 3) were previously described.14,26 Four of the 7 patients had a family history of FRTS. The detailed pedigrees and the results of this segregation analysis are shown in Supplementary Figure S2. Table 2Clinical characteristics of patients with GATM variantsPatient No1234567Previous reports3,8,13, 14, 15, 16, 17,44,45Age at onset (years)362131.70.91512Age at diagnosis (years)45225716.539-Primary complaint (initial symptoms)Kidney dysfunction (health checkup)RicketsProteinuria (health checkup)Glycosuria (health checkup)Short staturePoor weight gainGlycosuria (health checkup)Proteinuria (health checkup)Symptoms Rickets–+–+–––6/33 Nephrocalcinosis–––++–4/6 Intellectual disability–––––––0/35 Multiple fractures–––––––2/8 Short stature–+––+–+8/11 Poor weight gain–––++––4/8 Laboratory findings Proteinuria+++++++35/35 Nephrotic range proteinuria–––––––0/15 Hypercalciuria––––+––4/4 Hematuria–––––––0/5 AminoaciduriaNA+NANA+NANA33/33 Glycosuria+++++++35/35 HyperphosphaturiaNANANANANANANA35/35 Hypouricemia+––++––3/3 Hypophosphatemia+–++++–17/18 Hypokalemia++++++–7/11 Metabolic acidosis++–++––35/35Pathological findingsFocal global sclerosis, Interstitial FibrosisNot performedNot performedNot performedNot performedNot performedNot performedNA, data not available. Clinical characteristics of patients with GATM variants NA, data not available. Among 3 cases presenting with rickets and poor weight gain, FRTS was diagnosed before the age of 2 years. In contrast, cases without subjective symptoms during early childhood were identified later during school-age years or adulthood, typically through routine health checkups. Rickets was observed in 2 cases in this study and in 7 of 35 cases in previous reports. Short stature was observed in 3 cases in this study and in 9 out of 14 cases in previous reports. Intellectual disability was not seen in any cases in this study or in previously reported cases. Regarding laboratory findings, neither severe proteinuria nor hematuria was observed. All 7 patients exhibited low-molecular-weight proteinuria accompanied by glycosuria and, in most cases, aminoaciduria, consistent with the high frequencies reported in earlier case series (Table 2). Conversely, metabolic acidosis and hypophosphatemia, which had been documented in almost all previously reported patients, were present in only 4 of 7 and 5 of 7 patients, respectively, in our cohort. One patient with a GATM variant who had progressed to stage 4 CKD (Patient 1) exhibited focal global glomerulosclerosis on kidney biopsy. The trend of kidney function by age in patients and families with the pathogenic GATM variant is shown in Figure 2. Overall, kidney function tended to decline with age, although the rate and extent of this decline varied even among members of the same family, as exemplified by the family of Patient 1, in whom the degree of decline in kidney function varied among affected relatives. In the combined cohort, which included a total of 41 patients from this study and previously reported cases, the median kidney survival age was 51.9 years (95% confidence interval, 46–66 years) (Figure 3). Longitudinal changes in kidney function in patients with GATM variants. Creatinine-based estimated GFR (Cr-eGFR, mL/min/1.73 m2) is plotted against age (years) for each patient, with each colored line representing an individual patient. Patient 1-1 is the daughter of Patient 1, Patient 3-1 is the mother of Patient 3, and Patient 7-1 is the grandmother of Patient 7. Note that this figure includes only the probands and those affected family members for whom precise longitudinal data on renal function were available. Kidney survival in patients with pathogenic GATM variants in this study and in previously reported cases. Kaplan–Meier curve showing the cumulative probability of remaining free from end-stage kidney disease (ESKD) according to age. This analysis was performed on a total of 41 patients. The median kidney survival age was 51.9 years (95% confidence interval, 46–66 years). Clinical Features of Patients With Variants in Other Genes Patient 8, with an HNF4A variant, was diagnosed with FRTS after proteinuria was detected on urinalysis at 3 years of age. The pathogenic HNF4A variant (c.187C>T [p.Arg63Trp]) was identified when the patient was 14 years of age. At that time, he had developed short stature, nephrocalcinosis, and kidney dysfunction with a Cr-eGFR of approximately 60 ml/min per 1.73 m2. At the final observation at 18 years of age, no further deterioration in kidney function was observed. The clinical details of the patient with BCS1L variants (Patient 9), who had compound heterozygous variants c.268C>T (p.Arg90Cys) and c.821del (p.Pro274Argfs∗26), have been previously reported.25 Her twin brothers had also been diagnosed with FRTS in the same report.25 At 1.5 years of age, she exhibited motor developmental delay. Subsequent evaluation revealed FRTS and rickets. At the final observation at 8 years of age, she continued to exhibit growth retardation (height 97.1 cm, -5.5 SD; weight 14.1 kg, -5.0 SD), hearing loss, motor developmental delay, and severe intellectual disability, with only babbling speech. Kidney function was mildly impaired, with a Cr-eGFR of 83 ml/min per 1.73 m2. The patient with typical infantile nephropathic cystinosis (Patient 10) was identified with FRTS at 8 months of age during a comprehensive evaluation for poor weight gain. She was immediately treated with cysteamine following the genetic diagnosis at 1 year of age. At the last follow-up (2 years of age), there were no indications of extrarenal complications or kidney dysfunction, although FRTS persisted. Two male patients with Dent disease type 1 (Patients 11 and 12) harbored hemizygous pathogenic variants in CLCN5, c.731C>T (p.Ser244Leu) and c.100C>T (p.Arg34∗), respectively, and were diagnosed after proteinuria was detected during routine medical examinations at 3 years of age. One patient (Patient 11) already exhibited short stature (−2.3 SD), kidney dysfunction with a Cr-eGFR of 78 ml/min per 1.73 m2, and nephrocalcinosis at the time of genetic analysis (5 years of age). By the time of the final observation at 12 years of age, he still had short stature (−2.6 SD), and his Cr-eGFR had declined to 43 ml/min per 1.73 m2, indicating progressive kidney dysfunction. Both patients exhibited incomplete FRTS without severe metabolic acidosis. Patient 13, with an EHHADH variant, was referred at 4 years of age after mild proteinuria had been detected during a 3-year-old urinary screening. The clinical course and pathological analyses supporting the pathogenicity of this heterozygous EHHADH missense variant (c.7G>A, p.Glu3Lys) in Patient 13 have been previously reported.24 At presentation, urinary β2-microglobulin was markedly elevated (85520 μg/l) and hypercalciuria was present, whereas kidney function was normal and kidney biopsy findings were unremarkable. During follow-up, low-molecular-weight proteinuria and hypercalciuria persisted and glucosuria gradually appeared. At the last follow-up at 28 years of age, low-molecular-weight proteinuria and hypercalciuria persisted with mild kidney dysfunction (Cr-eGFR 76.3 ml/min per 1.73 m2). Genetic testing revealed a previously reported heterozygous EHHADH missense variant (c.7G > A, p.Glu3Lys). Her father, with similar low-molecular-weight proteinuria and mild kidney impairment, carried the same variant. Discussion In this study, monogenic disease-causing variants were identified in 65.0% (13/20) of families with FRTS. Pathogenic variants were detected in 6 genes as follows: GATM, CLCN5, HNF4A, BCS1L, CTNS, and EHHADH. Among these, GATM was the most frequently affected gene, with 4 completely novel heterozygous GATM variants identified, along with 2 variants recently reported by our group. Patients with GATM variants exhibited a progressive decline in kidney function over time, reaching kidney failure at a median age of approximately 52 years. Some patients with GATM variants were identified early because of symptomatic presentations such as rickets, whereas others were detected through routine urinalysis, indicating pronounced variability in the clinical severity of FRTS. There have been no reports that specifically address the identification rate of disease-causing variants in patients with FRTS. A previous study of patients with hypophosphatemic rickets demonstrated that whole exome sequencing detected pathogenic variants in all 29 patients with rickets and tubular dysfunction, all of whom exhibited both upper and lower limb deformities.17 Additionally, over the past few years, numerous studies have reported detection rates of monogenic disease-causing variants in patients with CKD using whole exome sequencing or targeted panel sequencing, and some patients were found to carry variants in CTNS, CLCN5, and HNF4A, which are known to be causative genes of FRTS.9,10,36 These findings indicate that FRTS-related genetic variants are present not only in symptomatic cases, such as those with rickets, but also in patients with idiopathic CKD. Among the genetic causes of FRTS, cystinosis is the most common globally.2 In contrast to this global trend, GATM variants were the most prevalent in our study, with only 1 case of cystinosis identified. Overall, our results indicate that many patients with idiopathic FRTS may actually have an underlying monogenic etiology, highlighting the importance of genetic testing in this population. Our study provides several new clinical and genetic insights into FRTS because of GATM variants. First, by integrating our cohort with all previously reported cases, we performed the first kidney survival analysis in a relatively large series of patients with FRTS because of GATM variants, demonstrating that this condition is a progressive tubulopathy leading to kidney failure, with a median kidney survival age of approximately 52 years, consistent with earlier observations.8 This quantitative evidence strengthens the concept that heterozygous GATM variants cause a chronic, age-dependent decline in kidney function rather than a static tubulopathy. Second, unlike earlier reports that largely described patients with complete biochemical features of FRTS, we observed that incomplete proximal tubular dysfunction was common, with metabolic acidosis and hypophosphatemia absent in a substantial subset of affected patients. In the initial report by Reichold et al.,8 patients with FRTS because of GATM variants were described as not exhibiting rickets, skeletal deformities, or extrarenal clinical features. In contrast, in our combined series of cases from this study and previously reported cohorts, although no intellectual disability or other consistent extrarenal manifestations were observed, rickets was present in approximately 20% of patients, including severe vitamin D–resistant forms,13 and some patients did not exhibit hypophosphatemia or hypokalemia. Collectively, these findings expand the known phenotypic spectrum and indicate that GATM variants should be considered even in the absence of classical biochemical features of Fanconi syndrome or when rickets is the primary manifestation. Further, the identification of multiple novel variants in a relatively small cohort suggests the possibility that FRTS because of GATM variants has been historically underdiagnosed, particularly in cases with subtle or incomplete biochemical abnormalities. GATM encodes the mitochondrial enzyme glycine amidinotransferase, and homozygous loss-of-function variants cause cerebral creatine deficiency syndrome characterized by severe neurological impairment but preserved kidney function.37,38 In contrast, heterozygous missense variants clustered in a specific region have been shown to cause childhood-onset FRTS through intramitochondrial aggregation and mitochondrial damage.8 In our cohort, GATM was the most frequently identified causative gene, with 4 of the 7 heterozygous variants being novel, including an intronic variant (c.288+8A>T) that resulted in an in-frame 6-base pairs insertion. Importantly, most of these identified variants are absent from large population databases such as gnomAD (Table 1), further supporting the notion that these are rare and highly likely to be disease-causing variants. These data suggest that not only missense but also in-frame splice-generated variants can be pathogenic and that heterozygous GATM variants—particularly missense or intronic changes—have been underrecognized in the context of genetic FRTS. Collectively, these findings refine the clinical spectrum of FRTS because of GATM variants and underscore the importance of genetic testing in patients with unexplained proximal tubular dysfunction or progressive CKD, even in the absence of extrarenal manifestations. In this study, only 1 patient was diagnosed with cystinosis. The prevalence of cystinosis is estimated to be 1 in 100,000 to 200,000 in the United States and Europe.39 However, it is much lower in Asia, including Japan, because the most common variant—a 57-kilobase deletion in the CTNS gene—is largely absent in Middle Eastern and East Asian populations.40 Our patient with cystinosis commenced cysteamine treatment before the onset of extrarenal symptoms and before reaching 1 year of age. Notably, several studies have demonstrated that earlier initiation of cysteamine therapy is associated with a better renal prognosis, highlighting the importance of prompt genetic diagnosis and early intervention in managing cystinosis.—41, 42, 43, 44 A recent study indicated that patients who initiated cysteamine treatment before the age of 2 years experienced a longer time to kidney failure.43 To date, our patient’s kidney function is well preserved despite the short observation period, which is consistent with previous reports. In this study, 2 patients with Dent disease displayed incomplete FRTS at diagnosis, and 1 patient (Patient 11) already presented with nephrocalcinosis and kidney dysfunction. A recent review article reported that incomplete FRTS is observed in approximately half of the cases, whereas complete FRTS is found in approximately 5% of cases.45 In Japan, patients with Dent disease often do not present with typical FRTS at the time of diagnosis.46 This tendency might be partly attributable to the early detection of the disease through school urinalysis. Consistent with the low prevalence of complete FRTS globally, all Dent disease patients in our present cohort exhibited only incomplete FRTS. The BCS1L variant identified in our cohort is associated with mitochondrial complex III deficiency, which is known to cause a wide range of systemic manifestations, including lactic acidosis, developmental delay, and hearing loss.47,48 FRTS has been described as part of this multisystem phenotype, and our patient similarly exhibited significant developmental delay in addition to FRTS, supporting the concept that proximal tubular cells are particularly susceptible to mitochondrial dysfunction. Similarly, EHHADH variants induce FRTS through mislocalization of a peroxisomal enzyme to mitochondria, leading to impaired mitochondrial function.6 In our cohort, the affected patient exhibited slowly progressive proximal tubular dysfunction, including persistent low-molecular-weight proteinuria and hypercalciuria. Notably, the patient’s father, who carried the same variant, also exhibited mild renal impairment, highlighting the broad phenotypic variability associated with EHHADH and suggesting that this gene should be considered in cases of slowly progressive or apparently isolated proximal tubular dysfunction. In contrast, HNF4A variants are believed to cause FRTS through a different mechanism, specifically via impaired transcriptional regulation of multiple proximal tubular transporters rather than primary defects in energy metabolism.4 The patient in our cohort had nephrocalcinosis and mild kidney dysfunction but did not display rapid disease progression. This observation suggests the possibility that FRTS because of HNF4A follows a more indolent course than mitochondrial-associated forms, although further studies are required to elucidate its natural history. Our study has some limitations. First, the cohort was small and confined to Japanese patients, which may limit the generalizability of our findings to other populations. Second, our approach was unable to detect relevant deep intronic variants, some copy number variants, large deletions, pathogenic variants in mitochondrial DNA, or variants in as-yet-unidentified disease-causing genes, because the targeted panel did not systematically cover these regions. Among the 7 cases in which genetic testing failed to establish a diagnosis, there were patients with intellectual disability and 1 patient in whom kidney biopsy revealed mitochondrial inclusions in proximal tubules, a finding suggestive of an underlying mitochondrial DNA variant. Advances in whole-genome sequencing are expected to enhance the detection of diagnostic variants in these regions and may also facilitate the discovery of novel causative genes associated with FRTS. In conclusion, our findings suggest that monogenic disease-causing variants account for a relatively high proportion of patients with inherited FRTS, with FRTS because of GATM variants emerging as a major subtype. By demonstrating its prevalence, wide phenotypic variability, and association with progressive renal decline, our study indicates that FRTS because of GATM variants has likely been underrecognized in routine clinical practice. These results support the inclusion of GATM in the genetic evaluation of patients with incomplete proximal tubular dysfunction or unexplained CKD and highlight the need for expanded application of genetic testing to improve early detection and guide management of this progressive tubulopathy. Disclosure KN received grants-in-aid for Scientific Research (KAKENHI, 23K07698), childhood-onset, rare and intractable kidney diseases in Japan, research on rare and intractable diseases, health, labor and welfare sciences research grants (23FC1047), the Japan Agency for Medical Research and Development (AMED) (Grant Numbers 24015773, 22810094, and 23ek0109617s1702). All the other authors declared no competing interests. Acknowledgments The authors thank all study participants and their families. Funding This study was supported by Grant-in-Aid for Scientific Research from KAKENHI 23K07698 (to KN). Data Availability Statement The data underlying this article will be shared by the corresponding author upon reasonable request. Footnotes Supplementary File (PDF) Supplementary Reference. Figure S1. Minigene splicing assay demonstrating aberrant splicing caused by the GATM c.288+8A>T variant in patient 7. Figure S2. Pedigrees of Families with GATM Variants. Table S1. List of kidney disease-related genes included in targeted sequencing analysis within a clinically approved gene panel test developed in our laboratory. Table S2. Summary of previously reported GATM variants. Table S3. Clinical characteristics of previously reported patients with GATM variants. STREGA Checklist. Supplementary Material Supplementary File (PDF) Supplementary Reference. Figure S1. Minigene splicing assay demonstrating aberrant splicing caused by the GATM c.288+8A>T variant in patient 7. Figure S2. Pedigrees of Families with GATM Variants. Table S1. List of kidney disease-related genes included in targeted sequencing analysis within a clinically approved gene panel test developed in our laboratory. Table S2. Summary of previously reported GATM variants. Table S3. Clinical characteristics of previously reported patients with GATM variants. STREGA Checklist. References Associated Data Supplementary Materials Supplementary File (PDF) Supplementary Reference. Figure S1. Minigene splicing assay demonstrating aberrant splicing caused by the GATM c.288+8A>T variant in patient 7. Figure S2. Pedigrees of Families with GATM Variants. Table S1. List of kidney disease-related genes included in targeted sequencing analysis within a clinically approved gene panel test developed in our laboratory. Table S2. Summary of previously reported GATM variants. Table S3. Clinical characteristics of previously reported patients with GATM variants. STREGA Checklist. Data Availability Statement The data underlying this article will be shared by the corresponding author upon reasonable request.