NHS inherited cancer register exposes gaps in genetic testing

Posted in partnership with Jeen Health

NHS England has launched a National Inherited Cancer Predisposition Register, a national record of people carrying gene variants that raise their cancer risk. It covers around 120 cancer susceptibility genes.

People on it are tracked, invited for screening and offered faster access to personalised treatment and trials, and those with BRCA variants are automatically called for breast screening.

It is a good piece of infrastructure. It is also an answer to the second question rather than the first.

A declaration of interest before going further: Jeen Health provides hereditary cancer genetic testing privately, and so sees people who sit just outside NHS criteria routinely. That vantage point shapes what follows.

The precedent for the register is strong.

It builds on the existing Lynch syndrome register, which between July 2023 and November 2025 saw 12,245 people receive additional bowel screening, and NHS England estimates that screening prevents between 40 and 60 bowel cancers for every 100 people with Lynch syndrome.

Applied across 120 genes, that is a serious prevention instrument.

But a register is a downstream tool. It can only hold the people the system has already identified, and identification is the part of this pathway that works least well.

There are only two realistic routes in

In practice, people reach a genetics service one of two ways. Either they are diagnosed with cancer and tested as part of their care, or they have a family history strong enough to prompt a referral.

The second route carries most of the preventive weight, because it is the one that can reach people before they are ill. It also rests almost entirely on family history.

The national framework for assessing and referring people with a family history of breast cancer is set out in NICE’s familial breast cancer guideline, and the pattern it describes is familiar: a GP asks who in the family had cancer, which side they were on, what type, and at what age.

Several close relatives on the same side with the same or related cancers constitutes a strong history. Without one, testing is usually judged unlikely to be informative, and the referral does not happen.

Worth noting that NICE has the assessment and genetic testing sections of that guideline under review. The criteria are not treated as settled even by the body that wrote them.

Family history is a proxy, and a leaky one

None of this is a criticism of the thresholds. They exist because specialist genetics capacity is finite and has to be rationed somehow, and family history is the cheapest available signal.

But a threshold designed to allocate a scarce resource is not the same thing as a threshold designed to find everyone at risk, and the difference matters.

Family history fails in predictable ways. Small families generate less signal, so a variant can segregate quietly through two or three people and never look like a pattern.

Paternal transmission is routinely underweighted, despite BRCA variants passing through fathers exactly as readily as mothers. Relatives who died young of something else never register as the cancer cases they might have become.

Adoption, estrangement, migration and fractured medical records all erase the data the assessment depends on.

The result is a cohort of people who carry a clinically actionable variant, meet no referral criterion, and have no reason to suspect anything. The system is not failing them through error. It simply cannot see them.

The register raises the stakes of being missed

Before the register, being unidentified meant not knowing your risk. That was bad enough.

Now it means something sharper. It means exclusion from a national surveillance system that actively contacts people, schedules their screening and routes them towards trials and targeted treatment.

The distance between identified and unidentified stops being a gap in knowledge and becomes a gap in care, with one group receiving structured lifelong follow-up and the other receiving nothing at all.

That is worth naming plainly, because it is the kind of inequity that builds quietly inside a well-designed programme. The better the register performs, the more consequential it becomes to be outside it.

Identification is a counselling problem before it is a testing problem

The sequencing is not the bottleneck. Sequencing is cheap, fast and getting cheaper.

The constraint is everything wrapped around it.

Someone has to take a proper family history, decide which test is appropriate, explain what a variant of uncertain significance means, interpret a negative result that does not actually rule much out, and support the conversations with relatives that follow a positive one.

That work is skilled, slow and delivered by a workforce the NHS does not have enough of. Any proposal to widen identification that does not also address genetic counselling capacity is not a proposal, it is a wish.

There are routes that ease this.

Mainstreaming, where testing happens within cancer services rather than exclusively through regional genetics clinics, has already shown it can widen access without collapsing quality.

Better digital tools for family history capture would help. So would a serious national conversation about where criteria-based testing should give way to broader testing for specific genes and populations.

What the private sector should not claim

It would be convenient for a private provider to present private testing as the answer here. It is not, and the sector should say so.

Private testing reaches people who know enough to go looking and can afford to act. Those are not the people the national pathway is missing most acutely.

A commercial market sitting alongside a rationed public one does not close an equity gap, and may widen it.

The defensible role for independent providers is narrower. Put counselling at the front rather than selling kits with a report attached.

Be honest about what a negative result does not exclude. Contribute outcome data to the bodies setting standards instead of holding it as commercial advantage.

And resist the direct-to-consumer model that hands people a risk figure with nobody qualified to explain it.

The register is the right answer. It is just an answer to the second question.

The first one, which is how the system finds people before cancer does, remains largely unsolved.