NHS genomics hubs lag on mandatory ClinVar submissions as rare disease families wait for answers
Three of England’s seven NHS Genomics Laboratory Hubs say they are not yet routinely submitting genetic variants to ClinVar, despite professional guidance that makes sharing a requirement. The gap is not an abstract governance issue. For families navigating rare disease, it can mean years spent with a suspicious result that never reaches the wider evidence base capable of changing its meaning.
Sophie Muir’s experience with her son Calvin shows how much can hinge on that missing link. Calvin’s early developmental delay, hypotonia, speech difficulty, later drooling, uncontrolled movements, and cardiac symptoms triggered years of testing for suspected syndromes including Angelman and fragile X, yet those results were negative. The eventual answer came only after research participation revealed a CACNA1C variant and the diagnosis was finally made at age 15, long after the variant had first been seen in the NHS laboratory as a variant of uncertain significance, or VUS.
A VUS is not a diagnosis. It is a statement of uncertainty, and in a system where most people carry millions of genomic variants, uncertainty is often the correct first reading. The problem emerges when the interpretation stays trapped inside a single laboratory instead of being shared into a public database where other cases can accumulate, compare, and eventually shift the clinical meaning.
Calvin Muir’s diagnosis arrived only after the broader evidence base caught up
Calvin’s case illustrates the biology and the bureaucracy at the same time. CACNA1C encodes part of a calcium channel, and pathogenic changes in this gene can produce a disorder spectrum that includes Timothy syndrome. When a variant sits in the uncertain category, it may not be disclosed through routine NHS reporting if there is no evidence strong enough to classify it as disease-causing. Yet once another patient in a different country is found with the same change and a different classification, the cumulative evidence can push the result over the threshold from uncertainty to likely pathogenicity.
That is exactly what happened here. Muir later found a patient in France with the same variant listed as likely pathogenic on ClinVar. Calvin’s own variant was eventually reclassified as likely pathogenic when he was 15, six years after Muir first learned of its existence and ten years after he entered the first research study. Reclassification did not merely tidy a file. It opened the door to a diagnosis, brought coherence to a decade of symptoms, and clarified the family’s understanding of what had been happening all along.
For rare disease families, that delay carries a clinical cost. The wait for reclassification can shape testing decisions for siblings, influence surveillance, and determine whether clinicians can connect apparently unrelated symptoms into a single inherited disorder. In Calvin’s case, the lost time stretched across childhood and adolescence, during the years when developmental support and medical planning often matter most.
FOI responses show uneven submission practices across England
Responses from the seven Genomics Laboratory Hubs, gathered through freedom of information requests at the start of the year, show that variant sharing is neither uniform nor complete. Three hubs — South East, South West, and North East and Yorkshire — said they were not routinely submitting variants to ClinVar, the public database named in the 2024 Association for Clinical Genomic Science guidance as the required destination for NHS submissions. One of those hubs said it had no staff member assigned responsibility for ClinVar submissions.
Even the four hubs that said they were making routine submissions — North Thames, East, North West, and Central and South — had backlogs of unsubmitted variants and gaps in operational documentation. Only those four reported internal policies, standard operating procedures, or governance documents tied to ClinVar submission. The picture that emerges is not one of a fully integrated national system but of variable local capacity, uneven workflow design, and a persistent mismatch between guidance and practice.
ClinVar is only one part of a wider data ecosystem. Some laboratories may also share data with gnomAD, Decipher, CanVIG, or CanVar-UK, and the updated guidance now names ClinVar, Decipher, and CanVar-UK as public databases to which variants should be shared as relevant. Even so, the existence of multiple repositories creates its own problem when they do not communicate cleanly with one another. Variant interpretation depends on comparison, and comparison weakens when the evidence base is fragmented.
Broken feedback loops slow variant reclassification and diagnosis
The value of submitting variants lies in the feedback loop. A single laboratory may see only one child or one adult with a given change. A shared database can show that the same variant has appeared elsewhere, under similar clinical circumstances or with a different classification. As the evidence accumulates, an uncertain result may move toward likely pathogenic or benign. For patients, that shift can change surveillance, family testing, reproductive counseling, and in some conditions access to treatment.
John Sayer, clinical professor of renal medicine at Newcastle University, argued that the system loses power when laboratories fail to contribute their data back into the shared pool. He described submission as a duty of care and said patients are missing out when variants are not entered into ClinVar. Jack Underwood, a forensic psychiatry registrar working with Muir on the Timothy Syndrome Alliance, described a broken loop in which variant entry can open treatment possibilities for different groups of patients.
The problem is also temporal. Sayer has found that upgrade or downgrade from VUS status typically takes two to four years, though he acknowledged that can still feel like an exceptionally long time to a family waiting for a result that may guide testing in other children or help access treatment. Two years is not a trivial delay when a child is still moving through diagnostic workup and medical surveillance.
Labs say staffing and workflow limits are slowing implementation
The Genomics Laboratory Hubs pointed to insufficient resource, staffing constraints, and the lack of semi-automation for exporting data to ClinVar as reasons for the delay. ClinVar itself was described by experts as clunky, unaccommodating, and inconsistent in data quality. Laboratories are also being asked to submit information to NHS England and the National Disease Registration Service, adding another layer to already crowded data workflows.
Those operational pressures do not erase the professional obligation to share, but they do explain why implementation has been uneven. The laboratories are dealing with a growing number of databases, different data standards, and the need to extract the right information without creating errors or losing clinically relevant detail. The result is a system in which some hubs are moving, some are backlogged, and some are still not routinely submitting at all.
Helen Firth, consultant in clinical genetics and genomic medicine at Cambridge University Hospitals NHS Foundation Trust, supported data sharing but drew a careful line around consent. She said there is a difference between encouraging patients to share data with an understanding of why it is useful and assuming all clinical data should be shared openly worldwide simply because a test was performed. Her point reaches beyond software and staff time. It asks what patients believe they are agreeing to when they enter a genomics pathway run with public funds.
Public investment is central to the dispute. Muir said it is not right for publicly funded genetic testing to produce clinically interpreted data that is then left unshared. Sayer made the same point from another angle: if the state pays for sequencing and interpretation, the value of that work is diminished when the resulting knowledge is not returned to the wider system.
For Calvin and families like his, the consequence is concrete. A VUS that never leaves the laboratory can remain invisible to the larger community that might reclassify it sooner. A diagnosis delayed by years is not just a matter of administrative friction; it is lost time in a disorder where every additional case can strengthen the evidence that turns uncertainty into clarity.