The exchange centers on the kind of questions that determine whether a study’s conclusions can be trusted: how variables were measured, how confounding factors were handled, and whether the analytical framework actually supports the causal interpretation the authors intended. In the competitive and high-stakes field of pediatric obesity research, where findings can shape clinical guidelines and public health policy for millions of children, these methodological disputes are far from academic quibbles. They are the mechanism by which the scientific community polices the reliability of the evidence base.

Wu and Luo open their response by thanking their critics for what they describe as a careful methodological review, a tone that reflects the collegial norms of scientific correspondence even when the substance of the disagreement is pointed. The authors state that they address each point concisely in their letter, indicating that the critique touched on multiple distinct aspects of the original analysis rather than a single flaw. Such multi-point critiques often arise when independent researchers attempt to reproduce or extend a published finding and encounter ambiguities in the analytic pipeline.

Among the works cited in the exchange is a recent study published in JAMA Network Open by Münte and colleagues examining the prevalence of extremely severe obesity and metabolic dysfunction among children and adolescents in the United States. That research, published in 2025, documented the scale of severe pediatric obesity and its metabolic consequences, underscoring why methodological rigor in this field matters so much. As rates of severe obesity climb in pediatric populations worldwide, the precision of prevalence estimates and the correct characterization of metabolic dysfunction become foundational to everything from screening recommendations to the design of intervention trials.

The correspondence also engages with the technical literature on causal mediation analysis, a statistical framework that allows researchers to decompose an observed relationship between an exposure and an outcome into direct and indirect pathways. The foundational work here traces back to a 2010 paper by Imai, Keele, and Yamamoto in Statistical Science, which established identification conditions, inference procedures, and sensitivity analysis methods for causal mediation effects. Mediation analysis has become a staple of observational epidemiology because it promises to reveal not just whether two variables are linked, but through what mechanism the link operates.

A more recent methodological contribution cited in the exchange comes from Derkach, Kantor, Sampson, and Pfeiffer, published in Statistics in Medicine in 2024, which tackles a persistent practical problem: conducting mediation analysis when the available information from publicly accessible data sources is incomplete. This is an everyday reality for research groups analyzing large national health surveys, where key mediators or confounders may be measured inconsistently, missing at random or not at all, or simply absent from the public-use files. The Derkach paper provides tools for drawing valid inferences under exactly these constraints, and its appearance in the debate suggests the critique and the response both grapple with how much confidence incomplete real-world data can support.

The framing of the exchange as a letter to the editor, published in the pediatrics category of the journal, reflects a long tradition in scientific publishing. Correspondence sections serve as a rapid, visible venue for post-publication peer review, allowing the broader community to watch methodological disputes unfold in real time. Unlike formal retraction or correction processes, which are reserved for demonstrable errors, letters and responses typically concern matters of interpretation, analytical choice, and the strength of the inferences that a dataset can bear. The published response from Wu and Luo indicates the authors stood by their analytical approach while engaging substantively with the specific objections raised.

Behind the technical details lies a substantive scientific question of genuine urgency. Childhood obesity, and especially its extreme forms, is associated with early-onset metabolic dysfunction including insulin resistance, dyslipidemia, and hypertension, conditions that were once seen almost exclusively in adults. Understanding the causal pathways that connect obesity to these outcomes, and the factors that mediate them, is central to designing interventions that work. If a mediating mechanism can be reliably identified, it becomes a potential target for treatment; if the mediation analysis is flawed, an apparent target may be an artifact of confounding or measurement error. This is precisely the terrain on which the current exchange takes place.

The work of Wu and Luo is supported by the National Key Research and Development Program of China, reflecting the substantial national investment that China has made in pediatric health research as it confronts rapidly rising rates of childhood obesity. Both authors are based at the Children’s Hospital of Fudan University, a National Children’s Medical Center, where Feihong Luo serves as the corresponding author. According to the contributor statement, Luo was responsible for conceptualization, manuscript review, and revision, while Wu drafted the response manuscript. The authors declare no competing interests.

For readers outside the field, an exchange like this one may appear esoteric, but it illustrates how scientific knowledge actually consolidates. A published finding is not the end of the process but the beginning of a public stress test in which independent researchers probe the assumptions, methods, and interpretations behind the result. When authors respond point by point, as Wu and Luo have done, the resulting correspondence becomes part of the permanent record, allowing future researchers to weigh both the finding and the objections to it. In a field where the evidence base directly informs how clinicians and policymakers confront one of the most pressing pediatric health challenges of the era, that transparency is not a formality. It is the difference between a conclusion that endures and one that quietly fades under scrutiny.

The publication timeline of the exchange offers its own insight into how journals manage post-publication dialogue. The response was received on 10 July 2026, revised just over two weeks later on 28 July, accepted on 19 August, and appeared as the version of record on 11 September. That rapid turnaround, roughly two months from submission to publication, is characteristic of correspondence sections, which are deliberately kept nimble so that debates over published work remain timely rather than becoming stale. The quick revision date also suggests the authors were prepared for the critique and able to address it without extensive new data collection, consistent with a defense of existing analytical choices rather than a reanalysis from scratch.

Readers unfamiliar with mediation analysis may benefit from understanding why it has attracted both enthusiasm and skepticism in epidemiology. The framework’s appeal is that it separates an exposure’s total effect into a portion that flows through an intermediate variable and a portion that operates by other routes. In the obesity context, an intermediate variable might be a metabolic marker, an inflammatory signal, or a behavioral factor, and identifying which pathway carries the effect can suggest where intervention is most likely to succeed. The caution is that these decompositions rest on strong assumptions, notably that no unmeasured confounding influences the mediator-outcome relationship, an assumption that can rarely be verified directly in observational survey data. Sensitivity analysis methods, such as those formalized in the 2010 Statistical Science paper cited in the exchange, exist precisely to quantify how violations of this assumption would distort the conclusions.

The challenge is compounded when analyses rely on publicly available national survey data, as the 2024 Statistics in Medicine contribution acknowledges. Public-use files are often stripped of direct identifiers and sometimes of sensitive variables, and complex survey designs introduce sampling weights and stratification that must be handled correctly for estimates and their variances to be meaningful. When a mediation analysis is layered on top of these constraints, the number of assumptions multiplies, and reasonable analysts can disagree about the appropriate handling of missing data, weighting, and model specification. That such disagreements surface in published correspondence, rather than remaining buried in peer review, is arguably a strength of the process.

The clinical stakes of these statistical debates should not be understated. Estimates of extremely severe obesity prevalence among children and adolescents inform resource planning for pediatric weight-management clinics, decisions about which patients warrant pharmacotherapy or metabolic surgery, and surveillance of early metabolic disease. If prevalence is overstated or understated, or if the metabolic burden is mischaracterized, the consequences ripple through guidelines and funding priorities. Correspondence exchanges like this one, though brief, contribute to the calibration of those estimates by forcing explicit articulation of the methods behind the numbers.

Finally, the institutional context is worth noting. The response originates from a National Children’s Medical Center in Shanghai, part of a clinical and research infrastructure that has expanded rapidly as China’s pediatric obesity rates have risen. Engagement by such centers with international methodological debate, in a journal published by Springer Nature, reflects the increasingly global character of the evidence base on childhood obesity, and the shared standards of causal inference and statistical transparency to which researchers across countries are now held.

Subject of Research: Methodological defense of a pediatric obesity and metabolic dysfunction study using causal mediation analysis

Article Title: Response to the comments by noor un nisa and colleagues

Article References: Wu, J., & Luo, F. (2026). Response to the comments by noor un nisa and colleagues. International Journal of Obesity. https://doi.org/10.1038/s41366-026-02205-0

Image Credits: AI Generated

DOI: 10.1038/s41366-026-02205-0

Keywords: childhood obesity, pediatric endocrinology, metabolic dysfunction, mediation analysis, causal inference, methodological critique, International Journal of Obesity, scientific correspondence, statistical methods, severe obesity, Fudan University, peer review

The research team, led by investigators at Soochow University’s Suzhou Medical College and the Children’s Hospital of Soochow University, designed the study around a cohort of children who had already undergone the most rigorous diagnostic sequencing available. Trio-based whole-exome sequencing, in which the affected child and both parents are sequenced to identify de novo and inherited variants, has become the gold standard for unraveling unexplained genetic conditions. Yet in many cases of short stature it returns no definitive answer. The researchers therefore turned their attention to the children left in this diagnostic gap, hypothesizing that their condition might arise from the collective effect of multiple rare variants distributed across growth-related genes rather than from a single damaging mutation in one gene. To test this idea, they compared the 112 undiagnosed patients against two control groups: 352 healthy adults of normal stature and 4,327 internal samples drawn from the Exome Aggregation Consortium database, one of the largest publicly available catalogs of human protein-coding variation.

The analytical engine of the study was an optimized version of TRAPD, a framework for testing rare variants using public data. Gene-based burden testing works by collapsing all of the rare, functionally significant variants within a single gene and asking whether cases carry a greater burden of such variants than controls would be expected to. This approach has substantially more statistical power than testing variants one at a time, because individual rare variants are too uncommon for any single one to reach significance in a modest cohort. The team applied the method under both dominant and recessive inheritance models, and the results were striking in their breadth. Under a dominant model, in which a single copy of a damaging variant is sufficient to influence the trait, 3,907 genes showed significant enrichment of rare variants at a threshold of P less than 0.05. Under a recessive model, which requires both copies of a gene to carry damaging variants, 85 genes were significantly enriched. Among the top ten most significantly associated genes identified through the primary modeling, four stood out with extraordinarily strong signals: FCGBP, FRAS1, MPDZ, and OBSCN, each reaching significance levels beyond P less than 1 × 10⁻⁹.

Each of the four implicated genes offers a distinct window into the biology of linear growth. OBSCN encodes obscurin, a gigantic cytoskeletal protein that scaffolds the sarcolemma and sarcoplasmic reticulum in muscle cells and helps organize muscle metabolism. Its connection to stature may run through the mechanical and metabolic dialogue between muscle and bone during development, a relationship increasingly recognized as a driver of skeletal growth. FRAS1 encodes a component of the extracellular matrix that is essential for the structural integrity of basement membranes, the thin sheets of specialized extracellular matrix that underlie epithelial tissues; mutations in FRAS1 are classically associated with Fraser syndrome, a congenital disorder that can include skeletal abnormalities. MPDZ is a scaffolding protein known to promote DLL4-induced Notch signaling during angiogenesis, the formation of new blood vessels, a process on which the vascular supply of the growth plate depends. FCGBP, or IgGFc-binding protein, is a large mucin-associated protein produced by goblet-like cells in the gut that participates in mucosal barrier function and wound healing, an intriguing connection given emerging evidence linking gut health and inflammation to growth outcomes in children.

To make sense of how these and the other enriched genes might converge on growth, the researchers performed functional enrichment analyses using Kyoto Encyclopedia of Genes and Genomes and Gene Ontology pathway frameworks. The results revealed significant enrichment in four metabolic and signaling pathways: steroid hormone biosynthesis, ascorbate and aldarate metabolism, pentose and glucuronate interconversions, and porphyrin metabolism. The steroid hormone biosynthesis signal is perhaps the most intuitive, given the central role of hormones from the hypothalamic-pituitary-gonadal and adrenal axes in regulating the growth plate. The ascorbate-related findings carry particular weight in light of recent research showing that vitamin C epigenetically controls osteogenesis and bone mineralization, meaning that the pathways governing ascorbate metabolism could directly influence how cartilage and bone mature. Pentose and glucuronate interconversions, meanwhile, connect to glycosaminoglycan synthesis and extracellular matrix remodeling, both critical for growth plate architecture.

The pathway picture also reinforces the idea that linear growth is not simply a matter of growth hormone and insulin-like growth factor acting on cartilage, but rather an integrated output of hormonal signaling, extracellular matrix integrity, and muscle-skeletal mechanical interaction. Obscurin, FRAS1, and MPDZ each appeared in multiple enriched pathways, suggesting that these genes occupy hub positions within a network whose perturbation impairs growth. This systems-level view aligns with a broader paradigm shift in growth biology, in which height is understood as a quantitative trait shaped by the coordinated activity of hundreds of genes across multiple tissue types. It also provides a mechanistic rationale for why sequencing a limited panel of canonical growth genes—SHOX, NPR2, NPPC, ACAN, IHH, FGFR3, and others—leaves so many children without a diagnosis. If susceptibility is distributed across the genome, then single-gene analysis will inevitably miss much of the picture.

The technical significance of the study lies in part in its adaptation of burden testing to a clinical cohort using public control data. Traditionally, association studies demand large, locally sequenced control cohorts, a resource that few clinical centers possess. The TRAPD framework circumvents this obstacle by allowing researchers to compare the variant burden observed in their patient exomes against allele frequency expectations derived from public repositories such as the Exome Aggregation Consortium. The optimized implementation used here accounted for gene-specific mutation rates and variant annotation, allowing a fair comparison across genes of very different sizes and constraint profiles. The approach had already proven useful for identifying rare-variant contributions to other quantitative traits, and this study demonstrates its utility in a pediatric endocrinology setting, where cohorts are inherently limited in size because short stature of unexplained origin, while common in clinics, is heterogeneous and difficult to recruit in the thousands.

Clinically, the findings carry several implications. First, they suggest that children with ISS may harbor a polygenic load of rare variants that current diagnostic frameworks do not capture, and that gene-based burden testing could become a complementary tool in the evaluation of such patients, layered on top of standard trio sequencing. Second, the identification of steroid hormone biosynthesis as an enriched pathway hints that some children may have subtle, subclinical perturbations of hormonal metabolism that standard endocrine workups do not detect, opening the possibility of more refined metabolic phenotyping. Third, the involvement of extracellular matrix and muscle-related genes raises the prospect of therapeutic strategies aimed not at the growth hormone axis itself but at the structural and mechanical environment of the growth plate. The authors are careful to frame these as susceptibility genes rather than deterministic causes, which is consistent with the variable expressivity expected of a quantitative trait.

The study also sits within a growing literature that connects growth to biology far beyond the skeleton. Recent work has documented gut microbiome dysbiosis with enriched pro-inflammatory species in children with idiopathic short stature, and the FCGBP finding provides a possible genetic thread connecting mucosal immune function to growth outcomes. Similarly, the observation that obscurin and its paralog OBSCN-family relatives maintain sarcolemmal integrity and muscle metabolism dovetails with evidence that skeletal muscle growth differences across individuals have a genetic basis that parallels bone growth. Height, in this emerging framework, is a trait of the whole organism: it depends on vasculature supplying the growth plate, on basement membranes structuring developing tissues, on muscle generating the mechanical signals that sculpt bone, and on hormones whose synthesis depends on metabolic pathways not traditionally considered in the endocrinology clinic.

The researchers acknowledge that their findings represent association rather than proof of causation, and that functional validation in model systems will be needed to confirm how rare variant burdens in OBSCN, FCGBP, FRAS1, and MPDZ translate into altered growth. Larger cohorts and replication in independent populations will also be essential to consolidate the signals. Nevertheless, the study represents a meaningful expansion of the known genetic landscape of isolated short stature, moving the field from a catalogue of single-gene defects toward an appreciation of the polygenic architecture that underlies one of the most common referrals in pediatric endocrinology. For the many families who leave genetics clinics without an answer, the message is that their child’s short stature may not be idiopathic at all—it may simply reflect a pattern of rare genetic variation that medicine is only now learning to read.

NR5A1, also known as steroidogenic factor-1 or SF-1, sits on chromosome 9q33 and encodes a 461-amino-acid nuclear receptor that functions as a transcription factor — a protein that binds specific DNA sequences and switches other genes on. The protein is built from several functionally distinct parts: a DNA-binding domain with two zinc-finger motifs, an auxiliary hinge region, a ligand-binding domain, and two activation-function domains that recruit the cellular machinery needed to drive transcription. During embryonic development, NR5A1 acts upstream of nearly every major event in male sex determination. It is expressed in both Leydig cells, which produce testosterone, and Sertoli cells, which nurture developing germ cells and secrete anti-Müllerian hormone, the signal that causes the Müllerian ducts — the precursors of the uterus and upper vagina — to regress in male embryos. Its expression even precedes that of SRY, the gene that initiates testis formation. Variants in NR5A1 are inherited in an autosomal dominant manner and account for roughly 10 to 15 percent of 46,XY disorders of sex development, a spectrum of conditions in which individuals with one X and one Y chromosome develop genitalia that do not align with typical male anatomy.

The research team, led by investigators at Shanghai Children’s Hospital affiliated with Shanghai Jiao Tong University, identified four naturally occurring NR5A1 variants in children evaluated between 2018 and 2023. Two of the variants — p.Cys65Ser and p.His310Arg — had never been reported before, while the other two, p.Cys30Ser and p.Gln329*, were previously known but had never been functionally characterized in depth. All four patients presented with severe proximal hypospadias, in which the urethral opening sits abnormally far down the penis, and a bifid scrotum, reflecting incomplete fusion of the genital folds. Three of the four also had undescended testes confined to the groin. Endocrine testing painted a consistent picture of primary gonadal insufficiency: follicle-stimulating hormone levels were elevated at baseline in all four children and climbed sharply after stimulation, while anti-Müllerian hormone — a marker of Sertoli cell mass — was measurable in every patient but varied widely, from 11.82 to 46.45 pmol/L. The child carrying the novel p.Cys65Ser variant had the lowest AMH and inhibin B values and the most severe external masculinization score, suggesting the greatest compromise of Sertoli cell function. Notably, ultrasound found no Müllerian structures in any patient, indicating that even the lowest AMH level had been sufficient during fetal life to trigger complete duct regression. Two children initially assigned female at birth were later reassigned and raised as boys after hormonal testing confirmed functional testicular tissue and an adequate androgen response to human chorionic gonadotropin stimulation.

Before moving into the laboratory, the team examined the evolutionary and structural context of each mutation. All four altered amino acids sit at positions that are highly conserved across species, a hallmark of functional importance. Computational modeling using SWISS-MODEL and visualization in PyMOL showed that the p.Gln329* nonsense variant — which introduces a premature stop codon — truncates the protein by 132 amino acids at its C-terminus, deleting two beta-sheets, six alpha-helices, and the entire activation-function-2 domain needed to recruit co-activator proteins. The two cysteine substitutions, p.Cys30Ser and p.Cys65Ser, strike the first and second zinc-finger regions of the DNA-binding domain respectively; while chemically subtle, each substitution introduces new polar interactions, including one with the zinc ion that stabilizes the DNA-binding fold, that may subtly distort the geometry required for high-affinity sequence recognition. The p.His310Arg substitution in the ligand-binding domain swaps a histidine for a bulkier arginine while preserving most local hydrogen bonds, but adds an unexpected new contact with a neighboring valine residue.

The functional consequences were unambiguous. When the researchers expressed wild-type or mutant NR5A1 in HEK293T cells — a line chosen in part because it does not produce NR5A1 endogenously, allowing clean measurement of transfected constructs — western blotting revealed that all four variants yielded substantially less protein than the wild-type receptor. Confocal immunofluorescence microscopy then showed that whereas the normal protein accumulates exclusively in the nucleus, where it must reside to bind DNA, all four mutant proteins showed markedly reduced nuclear signal and a corresponding shift into the cytoplasm. A transcription factor trapped outside the nucleus is, in effect, a transcription factor lost: the dual defect in protein abundance and localization provides a direct mechanistic account of why each variant behaves as a loss-of-function allele despite sometimes modest changes at the amino acid level.

To map the downstream wreckage, the team turned to the p.Gln329* truncating variant, which had shown the strongest effects in earlier assays and carries clear clinical relevance. RNA sequencing of cells expressing the mutant versus the wild-type protein identified 359 differentially expressed genes — 85 upregulated and, tellingly, 274 downregulated. Among the most strongly suppressed were canonical pillars of steroidogenesis: CYP11A1, which catalyzes the first committed step converting cholesterol into pregnenolone inside mitochondria; STAR, which shuttles cholesterol to that enzymatic machinery; and CYP17A1 and CYP21A2, which carry steroid synthesis through subsequent steps. Gene Ontology and KEGG pathway analyses confirmed that the downregulated genes clustered overwhelmingly in steroid biosynthesis, steroid and cholesterol metabolism, cortisol and aldosterone synthesis pathways, and cholesterol transport — exactly the molecular programs a developing testis depends on to masculinize the fetus.

The decisive technical advance came from combining the transcriptomic data with CUT&Tag — Cleavage Under Targets and Tagmentation — a technique that uses an antibody-guided transposase to map where a protein of interest contacts the genome. CUT&Tag on the wild-type receptor revealed binding peaks enriched in promoter regions and a recurring “CGCC” motif that likely represents part of the NR5A1 consensus binding element. When the same experiment was run with the p.Gln329* mutant, locus-specific enrichment collapsed at the promoters of two genes that had also emerged as strongly downregulated in the RNA-seq data: AMHR2, which encodes the receptor that mediates anti-Müllerian hormone signaling, and STARD8, a less-studied member of the steroidogenic acute regulatory-related lipid transfer domain family. Quantitative PCR validated the reduced messenger RNA levels of both genes, and the convergence of expression loss with loss of genomic binding made AMHR2 and STARD8 the study’s prime candidates for novel NR5A1 targets.

Dual-luciferase reporter assays then tested whether the variants actually impair the ability of NR5A1 to activate these promoters. Wild-type NR5A1 robustly increased the activity of the CYP11A1, AMHR2, and STARD8 promoter constructs compared with an empty vector. All four patient variants significantly diminished activation of the CYP11A1 and AMHR2 promoters, confirming these as targets vulnerable across the variant spectrum. STARD8 told a subtler story: the three missense variants left its promoter activity essentially untouched, while only the truncating p.Gln329* variant caused a marked decrease. This variant-specific pattern is intriguing because it suggests that STARD8 regulation is exquisitely sensitive to the integrity of the receptor’s C-terminal activation-function-2 region, and it may help explain why truncating variants can produce clinical phenotypes that missense variants in the same domain do not.

The clinical implications run in several directions. AMHR2 dysfunction is already known to cause persistent Müllerian duct syndrome, and the finding that NR5A1 variants suppress AMHR2 expression may clarify why a minority — roughly 17 percent — of NR5A1-related patients retain Müllerian structures despite measurable AMH: the hormone may be present, but its receptor may be transcriptionally starved. STARD8 has a more mysterious history; a previous case report described two sisters with 46,XY gonadal dysgenesis carrying variants in the gene, and work in fruit flies supports a conserved role for the protein’s family in testis development, but direct functional evidence in humans had been lacking. The new data position STARD8 as a context-dependent NR5A1 target that may participate in Leydig cell differentiation and lipid handling during steroidogenesis. The authors caution that their experiments were performed in a non-gonadal cell line, that CUT&Tag demonstrates physical association rather than definitive functional binding, and that the transcriptomic and CUT&Tag analyses covered only the truncating variant — validation in physiologically relevant gonadal cell models and larger patient cohorts will be needed. Even so, by connecting specific molecular lesions to reduced protein output, failed nuclear import, and the selective silencing of hormone-synthesis and sex-development genes, the study converts a list of genetic variants into a mechanistic narrative — and takes a substantial step toward predicting, from a DNA sequence alone, how deeply a given NR5A1 mutation will disturb the intricate choreography of human sex development.