A genome-first study of more than 640,000 people suggests systemic autoinflammatory diseases are more common, and substantially less recognised, than previously thought.
Researchers have uncovered a significant burden of undiagnosed systemic autoinflammatory disease, with almost all individuals carrying high-confidence pathogenic variants lacking a formal diagnosis despite half showing evidence of inflammatory disease.
The findings, published in The Lancet Rheumatology, suggest monogenic autoinflammatory diseases may affect approximately one in 4000 people, challenging assumptions about their rarity and raising questions about current approaches to genetic testing.
“Historically, genetic testing for systemic autoinflammatory disorders has been limited to individuals with canonical symptoms,” the researchers wrote.
“Therefore, the current understanding of the clinical spectrum of disease in autoinflammatory disorders, similar to other genetic diseases, is limited to those with prototypic manifestations.
“The current Eurofever–PRINTO classification criteria for hereditary recurrent fevers rely on age of onset, ancestry, and presence of prototypic symptoms like fever and serositis.
“This phenotypic bias results in the forgoing of genetic testing in affected individuals with less common presentations.”
The retrospective cohort study analysed whole-exome sequencing and linked electronic health records from 640,092 participants across two large biobanks in the US and UK.
Researchers identified 159 individuals carrying high-confidence pathogenic or likely pathogenic variants in 17 genes associated with systemic autoinflammatory disease, representing an estimated prevalence of one in 4026 people (0.02%).
Of these, 81 participants (51%) had clinical evidence of inflammatory disease, yet just one had a documented diagnosis of a systemic autoinflammatory disorder.
The researchers said the findings highlighted a substantial diagnostic deficit and suggested clinicians may be overlooking genetically driven inflammatory diseases in patients who do not present with classical symptoms.
“Despite carrying disease-associated variants, almost no affected individuals had a documented systemic autoinflammatory disease diagnoses, pointing to a substantial gap in disease recognition and a missed opportunity to identify individuals with systemic inflammation who could benefit from genetic evaluation to guide therapy,” the researchers wrote.
The study, led by researchers from NYU School of Medicine, used a genome-first approach to investigate the prevalence, penetrance and clinical spectrum of systemic autoinflammatory diseases.
Rather than selecting patients based on characteristic symptoms such as recurrent fever, the investigators screened genomic data for pathogenic variants across 28 genes, including 27 associated with dominantly inherited conditions and UBA1, which is associated with the acquired autoinflammatory disorder VEXAS syndrome.
Participants were drawn from the Geisinger MyCode Community Health Initiative in Pennsylvania, which included 170,503 individuals, and the UK Biobank, which contributed 469,589 participants.
Pathogenic and likely pathogenic variants were classified according to American College of Medical Genetics and Genomics guidelines and underwent additional filtering to identify variants with the strongest evidence of disease causality.
The investigators initially identified 755 individuals with potentially disease-associated variants but restricted their primary analyses of clinical manifestations to 159 participants carrying the highest-confidence variants.
Disease penetrance, defined as the presence of at least one autoinflammatory disease-associated ICD-10 diagnostic code group, differed substantially between cohorts.
In the Geisinger cohort, 50 of 63 variant carriers (79%) had evidence of inflammatory disease, compared with 31 of 96 (32%) in the UK Biobank.
The researchers suggested this discrepancy could partly reflect differences in clinical data capture, particularly the absence of outpatient electronic health records in the UK Biobank, as well as healthy volunteer bias.
The clinical manifestations extended well beyond the recurrent fevers traditionally associated with hereditary autoinflammatory syndromes.
Variant carriers demonstrated inflammatory involvement across multiple organ systems, particularly the lungs and skin, alongside rheumatic autoimmune disorders, vasculitis, and arthritis.
In the UK Biobank, individuals carrying high-confidence variants had significantly increased odds of rheumatic autoimmune disorders compared with controls (OR 12.28; 95% CI 4.12–30.68).
They also had higher odds of arthritis (OR 3.70; 95% CI 1.70–7.99), lower airway inflammation (OR 9.76; 95% CI 2.98–25.84), fever (OR 7.51; 95% CI 2.25–19.87), and other pulmonary disorders (OR 5.29; 95% CI 1.93–12.67).
In the Geisinger cohort, variant carriers had increased odds of vasculitis (OR 5.06; 95% CI 1.01–15.54) and rheumatic autoimmune disorders (OR 2.29; 95% CI 0.88–5.04), although these associations were not statistically significant after correction for multiple comparisons.
The researchers also found evidence of increased treatment for inflammatory disease.
Related
Geisinger participants carrying pathogenic variants were nearly four times more likely to have received immunosuppressive medications than controls (OR 3.98; 95% CI 1.40–9.23).
However, the increased use was observed across the broader immunosuppressant category rather than being driven by any particular class of disease-modifying antirheumatic drug.
The findings suggested that some patients with genetically mediated autoinflammatory disease may already be receiving treatment for inflammatory manifestations without recognition of their underlying molecular diagnosis, the researchers said.
They also identified considerable variation in clinical presentations among individuals carrying variants in the same gene.
Notably, variants in NLRP3 and TNFRSF1A, which were associated with established periodic fever syndromes, were not consistently linked to fever-related diagnostic codes.
Instead, some carriers presented with pulmonary, cutaneous, or other inflammatory manifestations that would not necessarily prompt investigation for hereditary autoinflammatory disease.
The researchers found no consistent clinical phenotype that reliably distinguished one disease-associated gene from another across both cohorts.
They suggested this phenotypic overlap and variability could explain why conventional diagnostic approaches, which often depend on recognising characteristic symptom patterns, fail to identify many affected individuals.
They said the findings had potential implications for rheumatologists managing patients with unexplained systemic inflammation, atypical autoimmune presentations, or inflammatory disease involving multiple organ systems, and that genetic evaluation should not be restricted to patients presenting with classical features of hereditary recurrent fever syndromes.
They also highlighted the potential value of molecular diagnosis in guiding targeted treatment and identifying relatives who may carry disease-associated variants.
However, the researchers acknowledged several limitations, including reliance on diagnostic coding as a surrogate for clinical disease, incomplete medication records in the UK Biobank and differences in the availability of longitudinal clinical data.
The cohorts were also predominantly White, potentially limiting the generalisability of the findings to more diverse populations.
The study did not comprehensively assess recessively inherited autoinflammatory diseases, somatic variants outside UBA1 or polygenic disorders, suggesting the overall burden of autoinflammatory disease could be higher than estimated.
Importantly, the presence of a pathogenic variant alongside an inflammatory diagnostic code does not necessarily establish that the variant caused the recorded clinical condition.
Despite this, the researchers concluded that population-based genomic screening revealed a broader spectrum of autoinflammatory disease than is currently recognised in clinical practice.
“This study shows the need for a lower threshold to obtain genetic testing and increased provider awareness of these diseases and their heterogenous phenotypes,” they concluded.
“Identification of genetic disorders can impact disease morbidity by guiding targeted therapies and lead to better outcomes.”
