Whole exome sequencing: clinical utility in pediatric immune disorders

It can change infection prophylaxis, clarify transplant decisions, identify relatives at risk, and redirect surveillance toward complications that may not yet be visible.

The diagnostic yield is substantial but not uniform. Across 29 studies involving 5,847 patients with suspected pediatric primary immunodeficiencies, pooled next-generation sequencing achieved a diagnostic yield of 42% (95% CI 0.29–0.54). In a family-history-positive subgroup, the reported yield was 58%. These are cohort-specific findings, not fixed performance limits for every child or a universal prediction for an individual test. Presentation, pedigree structure, age, phenotype quality, test design, and interpretation resources all influence the result.

That distinction matters. A family with affected siblings and a consistent phenotype creates a different pretest setting from an infant with a severe immune phenotype but no known affected relatives. WES is most useful when the sequencing result is interpreted as part of that clinical architecture rather than as an isolated laboratory event.

Diagnostic Yield and the Role of Family History in WES

The pooled 42% yield reflects the genetic structure of pediatric IEI. Many clinically severe childhood presentations are caused by single-gene disorders, and a large share of the relevant variants lie in coding regions that standard exome enrichment is designed to capture. But the pooled number combines different diseases, referral pathways, sequencing strategies, and definitions of a confirmed diagnosis. It should be read as an overview of performance across studies, not as a guaranteed probability for a child sitting in a clinic.

Family history is one of the variables that can shift that context. Consanguinity, affected siblings, recurrent severe disease across generations, or a recognizable pattern among relatives can raise suspicion for a monogenic disorder and provide additional evidence during variant interpretation. In the family-history-positive subgroup represented in the available analysis, the reported diagnostic yield was 58%. That result supports the value of pedigree information, but it does not establish 58% as an upper boundary for WES or as a target that every family with a positive history should expect.

Conversely, the absence of a family history does not make WES low value. Severe IEI may appear sporadically because of a de novo variant, an X-linked disorder in a child without previously recognized affected males, incomplete penetrance, a small family size, early death in an undiagnosed relative, or simple lack of diagnostic recognition. A negative pedigree is therefore not the same as evidence against genetic disease.

The practical interpretation is more measured than a single percentage:

• A positive family history can increase the prior probability of a monogenic diagnosis and help laboratories evaluate segregation.
• A negative family history does not support assigning a specific lower diagnostic yield to an individual child.
• The child’s phenotype remains central: age at onset, infection pattern, lymphocyte and immunoglobulin findings, vaccine responses, inflammatory features, autoimmunity, cytopenias, and organ involvement may be more informative than the pedigree alone.
• Trio sequencing, when available, can clarify whether a candidate variant is de novo, inherited, or absent from both parents. That information often matters more than simply widening the list of genes analyzed.

The contrast between family-history-positive and family-history-negative cohorts should therefore be used to frame expectations, not to calculate a fixed bedside estimate. If an infant has absent B cells, recurrent opportunistic infections, and a compelling immunologic phenotype, WES may be appropriate even when no relative has a similar diagnosis. The decision rests on the strength of the phenotype and the consequences of finding—or not finding—a molecular cause.

Family history changes the evidentiary landscape, but it does not turn a cohort yield into an individual prediction.

Pediatric cohorts often show stronger Mendelian architecture than adult immune deficiency cohorts because severe inherited disorders tend to present earlier. That makes pediatric data more relevant for children than adult yield estimates. Even within pediatrics, however, the spectrum ranges from recognizable syndromic disorders to incomplete, atypical, or age-dependent phenotypes. A result from one cohort should not be transferred mechanically to another.

The clinical question is not simply whether sequencing can identify a variant. It is whether the suspected diagnosis is sufficiently plausible, whether the laboratory can interpret the relevant variant classes, and whether a result would change care. Those questions should be discussed before testing rather than after an uncertain result arrives.

Comparative Efficacy: Targeted Gene Panels Versus Whole Exome Sequencing

Targeted gene panels and WES are not interchangeable versions of the same test. They make different trade-offs between scope, interpretation, turnaround, and the likelihood of encountering variants whose significance is unclear.

A longitudinal analysis of 878 patients with suspected primary immunodeficiency reported a 56% diagnostic yield for targeted panel sequencing. Reflex WES identified 18 additional diagnoses, bringing the cumulative yield in that cohort to 58%. The important point is not that WES universally adds only two percentage points. It is that, in this particular panel-first pathway, most diagnoses were already found within the targeted test, while WES provided additional diagnoses for a smaller group whose cases remained unresolved.

Panel testing can be efficient when the clinical phenotype is sharply defined. A child with a classic pattern of X-linked agammaglobulinemia, a strongly suggestive SCID phenotype, or chronic granulomatous disease supported by abnormal functional testing may fit naturally within a curated diagnostic pathway. The smaller search space can simplify annotation and reduce the number of unrelated findings returned to the clinical team.

WES is more attractive when the phenotype does not point cleanly to one pathway. It can examine genes outside a predefined panel, support discovery of unexpected diagnoses, and provide a broader starting point when the patient has overlapping immune, inflammatory, neurologic, gastrointestinal, or developmental features. That breadth can be valuable in children whose presentation does not respect the boundaries of a traditional immunology category.

The comparison also depends on how current the panel is. A narrow or outdated panel may perform differently from a comprehensive contemporary panel that includes newly established IEI genes and relevant copy-number analysis. Conversely, WES is not automatically broad in practice: capture quality, pipeline design, coverage of difficult exons, CNV calling, trio availability, and the laboratory’s gene-disease knowledge all determine what the test can actually deliver.

The reported panel-plus-reflex-WES figures illustrate a workflow, not a universal hierarchy. In one setting, a panel may be a sensible first test. In another, starting with WES may avoid delay and repeated testing, particularly when the phenotype is multisystem or the differential diagnosis is wide. Genetic sequencing versus targeted gene panels is therefore a clinical design question, not a contest with one winner.

A prospective single-center study of Czech children with undiagnosed rare diseases using singleton WES reported a 43% diagnostic yield and a 76% clinical utility rate. The difference between those figures is instructive. Clinical utility does not require every child to receive a definitive molecular label. A partially resolved finding, a narrowed differential diagnosis, or a result that changes surveillance can have practical value even when the sequence does not produce a fully confirmed diagnosis.

Translating Molecular Findings into Clinical Management Changes

A molecular diagnosis matters because it can alter what happens next. The value of WES is not exhausted when a report classifies a variant as pathogenic or likely pathogenic. The result must be connected to treatment, prevention, transplantation, surveillance, and family care.

Across the pediatric cohorts summarized in the literature, reported clinical management changes after a definitive molecular diagnosis range from 50% to 76%. A South African eight-year evaluation reported a management change in 67% of diagnosed patients, while the Czech prospective cohort reported 76% clinical utility. These rates describe the populations studied and the way each study defined a management change; they should not be treated as a guaranteed outcome for every positive result.

The changes can be immediate or cumulative:

• Treatment selection. A molecular diagnosis may support a pathway-specific therapy, including a small-molecule inhibitor for a signaling defect, a biologic directed at a relevant cytokine pathway, or a replacement strategy suited to the underlying disorder. The gene result does not replace functional assessment, but it can make a previously broad treatment discussion much more precise.
• Infection prevention. The diagnosis may lead to antimicrobial prophylaxis, a change in its intensity, or a decision to stop an intervention that is no longer justified. It can also clarify when live vaccines are contraindicated and when blood products require special handling, such as irradiation in selected T-cell disorders.
• Transplant planning. Genotype-confirmed disease can affect the urgency of hematopoietic stem cell transplantation, the assessment of expected disease course, and donor evaluation. Testing siblings may be particularly important when the inheritance pattern indicates that a relative could be affected or could carry the relevant variant.
• Family counseling. A result can establish or refine recurrence-risk counseling, guide parental testing, and inform reproductive options. It may also prompt evaluation of relatives whose symptoms were previously considered unrelated or too mild to suggest IEI.
• Long-term surveillance. Some molecular diagnoses change the monitoring plan for malignancy, autoimmunity, pulmonary disease, inflammatory complications, or organ-specific damage. The appropriate surveillance depends on the disorder and the child’s current phenotype; a gene name alone is not a substitute for individualized follow-up.

The Czech cohort reported an average of two management changes per diagnosed patient. That finding captures an important feature of pediatric immune disorders: one result may affect several parts of care at once. It may change prophylaxis, trigger family testing, and alter the timing of transplant assessment. In that sense, clinical utility is broader than the binary distinction between diagnosed and undiagnosed.

At the same time, not every molecular finding is actionable. A pathogenic variant in a gene unrelated to the child’s phenotype may be incidental. A variant in a gene with incomplete penetrance may require cautious counseling. A diagnosis can also explain the disease without immediately offering a targeted therapy. Clinical utility should therefore be assessed in relation to the child’s phenotype, available interventions, and the information needed by the family—not inferred from the technical sophistication of the assay.

Navigating Variants of Uncertain Significance in Pediatric Care

The central interpretive problem in WES is not finding variants. It is deciding which variants explain the child’s disease.

Coding single-nucleotide variants and small insertions or deletions in well-characterized IEI genes may be classified with greater confidence when population frequency, predicted effect, clinical phenotype, segregation, and functional data point in the same direction. Genes such as RAG1, RAG2, IL2RG, BTK, STAT1, STAT3, ATM, and NBN illustrate the range of disorders in which detailed genotype–phenotype knowledge can support interpretation. But even in familiar genes, the same variant can have different implications depending on inheritance, residual function, age, and clinical context.

Outside well-established gene–disease relationships, the number of variants of uncertain significance (VUS) increases. A VUS is not a concealed diagnosis and should not be used alone to justify a major treatment decision. It represents insufficient evidence for classification, not evidence that the variant is harmless or causative.

Pediatric interpretation has several particular complications:

• Trio information may be decisive. Parental sequencing can establish de novo status, reveal inheritance, or show that a candidate variant is present in an unaffected parent. None of these findings automatically settles pathogenicity, but each can materially change the evidence assessment.
• Age-dependent disease can obscure segregation. A parent who appears healthy may later develop manifestations of a disorder associated with incomplete penetrance or variable expressivity. This is relevant to genes such as STAT1, NFKB1, and CTLA4, where disease expression can vary across relatives and across time.
• Functional evidence may lag behind clinical decisions. Laboratory assays can be technically demanding and may not be available at the moment a treatment or transplant decision is being considered. A VUS should not be converted into a diagnosis merely because the clinical situation is urgent.
• Reanalysis is part of the test’s useful life. Gene–disease relationships evolve, databases are updated, and additional family or functional data may become available. A previously uncertain result can be reclassified, but reanalysis is not guaranteed to produce a diagnosis.

When a VUS occurs in a high-priority gene, the appropriate response is usually structured investigation rather than either dismissal or overinterpretation. Depending on the gene and phenotype, that may include segregation analysis, protein-expression studies, pathway assays, immune-cell functional testing, review of the patient’s longitudinal record, and periodic literature or laboratory reanalysis. For example, a STAT1 VUS may prompt a relevant phosphorylation or functional assessment, while a RAG1 or RAG2 finding may require a recombination-focused assay where clinically available.

The strongest interpretation combines several lines of evidence. A variant that fits the phenotype, follows the expected inheritance pattern, is absent or extremely rare in population databases, affects a critical region, and has supporting functional data is different from a computationally predicted change with no segregation or laboratory evidence. The report should make that distinction visible to the treating team and the family.

A VUS is a prompt for disciplined follow-up, not permission to treat uncertainty as a diagnosis.

Limitations of Exome-Focused Sequencing in Rare Immune Conditions

WES is broad across coding regions, but it is not genome sequencing. Standard exome approaches are optimized for protein-coding exons and nearby splice regions. They can miss disease-causing changes outside those regions or fail to characterize variants that require a different technical method.

Important limitations include:

• Deep intronic variants. Changes far from canonical splice sites can disrupt regulatory elements, create cryptic splice sites, or alter transcript processing. These variants are generally outside the reliable capture and interpretation range of standard WES.
• Promoter and enhancer variants. Regulatory elements controlling tissue-specific or developmentally timed gene expression may lie outside the exome. A child can therefore have a strong phenotype and a plausible candidate gene without a detectable coding variant.
• Structural and copy-number variants. Exome depth-of-coverage methods may identify some exon-level deletions or duplications, but sensitivity varies. Balanced translocations, inversions, and many larger or more complex structural variants often require genome sequencing, chromosomal microarray, or another dedicated assay.
• Repeat expansions. Repeat expansion disorders are not reliably detected through routine exome analysis.
• Mitochondrial variants. Standard nuclear exome enrichment does not provide comprehensive assessment of the mitochondrial genome.
• Difficult genomic regions. Poorly captured exons, pseudogenes, repetitive sequences, and regions with inadequate read depth can create technical gaps even within coding genes.

These limitations matter most when the clinical suspicion remains high after a negative result. A negative WES report does not establish a non-genetic cause, and it does not necessarily exclude the gene that clinicians already suspect. The next step should be guided by the phenotype and by what the original assay could and could not assess.

For a child with a strong monogenic IEI pattern and non-diagnostic WES, reasonable follow-up may include reanalysis of existing data, trio testing if parental samples were not initially available, targeted testing for a suspected structural or intronic variant, RNA sequencing, or whole genome sequencing. Functional immunology remains important throughout. Flow cytometry, protein expression, lymphocyte proliferation, oxidative burst testing, cytokine assays, and other studies can refine the diagnosis even when sequencing has not yet produced a definitive answer.

The timing of that escalation depends on clinical urgency. A child with severe infections, profound lymphocyte abnormalities, or a phenotype compatible with a transplantable disorder cannot wait passively for every variant to be resolved. Management may need to proceed on the basis of immunologic and clinical evidence while genomic investigation continues.

Clinical Utility Assessment

The clinical utility of WES in pediatric immune disorders comes from the relationship between three questions: what the test can detect, how confidently the result can be interpreted, and whether the result changes care.

Reported diagnostic yields vary across pediatric IEI cohorts. The pooled estimate of 42% across 29 studies provides a broad reference point, while the 58% result in a family-history-positive subgroup and the 56% panel yield followed by 18 additional diagnoses with reflex WES describe specific cohort and workflow settings. None of these figures should be presented as a universal ceiling, floor, or individual prediction. The variation is itself clinically meaningful: patient selection and test design shape the result.

The choice between a panel-first strategy and WES-first strategy should follow the phenotype. A targeted panel may be efficient when the clinical picture maps closely to established genes and rapid interpretation is important. WES may be preferable when the phenotype is atypical, multisystem, unusually severe, or not well explained by the available panel. It is also a logical next step when a panel is negative but the immunologic evidence continues to support a genetic disorder.

The main cost of WES breadth is interpretive complexity. Trio sequencing, functional validation, expert review, and reanalysis are not decorative additions to the laboratory report. They are part of the infrastructure required to turn a list of sequence changes into a clinically defensible conclusion. Without that infrastructure, a broader assay can produce more uncertainty without delivering more care.

A negative exome should also be understood as a transition point, not a final verdict. When suspicion remains strong, the question becomes whether the unresolved mechanism is non-coding, structural, mitochondrial, repeat-mediated, technically missed, or not yet recognized as a disease-causing gene. Whole genome sequencing, RNA-based analysis, targeted deep-intronic testing, and continued functional immunology may each be more informative than simply repeating the same exome.

Whole exome sequencing for pediatric immune disorders is therefore most valuable as part of a diagnostic system. It can identify a molecular cause, reveal an unexpected pathway, refine risk for relatives, and change management across several domains. Its clinical utility depends on appropriate patient selection, a phenotype-led interpretation, honest handling of uncertainty, and a clear plan for what happens after both positive and negative results.