
The letter arrives around a birthday. It knows how old you are, and that's about all it knows. Not whether your mother or sister had breast cancer, not how dense your breast tissue is, not what your genes say. In England the first invitation comes between 50 and 53, then every three years; in the trial I'm about to describe, the comparison was a mammogram every year from 40. Either way, the invitation is set by the calendar.
For a decade, researchers have asked whether it could be set by risk instead: screen the women most likely to develop an aggressive cancer more often, and those at very low risk less often. In March this year, the largest trial to test that idea reported its results. They're more interesting than a simple yes or no.
What the trial found
WISDOM recruited 28,372 women aged 40 to 74 from all fifty US states between 2016 and 2023, entirely through an online platform (Esserman et al., JAMA, 2026). Half were assigned to the usual yearly mammogram. The other half had their risk worked out: a genetic test covering nine breast cancer genes, a "polygenic risk score" that adds up many small, common variants, and a standard calculator using age, family history and other factors.
Then came the tailored plan. The highest-risk group — a five-year risk of 6% or more, or a high-risk gene — alternated mammography and MRI every six months, with counselling. An elevated-risk group, the top 2.5% for their age, kept yearly mammography plus counselling on reducing their risk. Average-risk women went to every two years. And women in their forties whose five-year risk was below 1.3% were told: no screening yet; we'll revisit when your risk rises, or at 50.
The trial had two main questions. First, safety: would risk-based screening let more advanced cancers — stage IIB or worse — slip through? After a median of 5.1 years, no. Advanced cancers were found at 30.0 per 100,000 woman-years in the risk-based group versus 48.0 in the annual group: if anything fewer, and within the pre-set margin for "no worse".
Second, the promised benefit: fewer biopsies. Here the answer was also no. Despite roughly 3,800 fewer mammograms per 100,000 woman-years, the biopsy rate in the risk-based group was not lower; the central estimate actually pointed slightly the other way, though the uncertainty range included zero.
One more figure is telling. Women who declined to be randomised could still join an observational arm and pick their own approach. Of those 18,031 women, 89% chose risk-based screening.
What's still uncertain
Five years is a short time in breast cancer. The trial counted advanced-stage cancers as a stand-in for what we really care about — deaths — because deaths take ten or more years to count. That's a reasonable proxy. It isn't the thing itself.
Why didn't biopsies fall? The paper's abstract doesn't say. One plausible reason is that the highest-risk women received MRI, which tends to find more things worth a second look. Rates of cancer, biopsy, mammography and MRI all rose with risk category — which is what you'd expect, and partly the point.
Who was in it. Three-quarters of participants were non-Hispanic White. Polygenic scores were mostly developed in people of European ancestry and perform less well elsewhere, so the "low risk, skip screening for now" call may be less reliable for other women. And the funding source isn't stated in the abstract — worth checking before leaning on the result too hard.
It's a US trial. NHS breast screening invites women from 50 up to their 71st birthday, every three years, and women over 71 can ask to continue. Britain, in other words, already sits closer to WISDOM's "average risk" arm than to its "annual" arm. For a UK reader the lesson runs the other way: the trial supports screening higher-risk women more intensively, and shows it can be done safely.
What you can do
The inputs to a risk assessment are mostly things you already know. Write down who in your family has had breast or ovarian cancer and at roughly what age, and bring the list to your next appointment. Then ask your doctor whether that history — or a previous note about dense breast tissue — puts you in a group that should be assessed for earlier or more frequent screening.
The invitation knows your birthday. You know the rest. Between the two, you can have a much better conversation.
Curiosity first.
— Dr. Brugal
Related on Dr. Brugal
For more on making appointments count: The one question to ask your GP that changes the appointment and The AI reading your scans before the radiologist does. On the wider question of how trials get read and misread: Why scientists still can't agree on fasting — and what would settle it.
References
All studies retrieved via PubMed.
- Esserman LJ, Fiscalini AS, Naeim A, et al. Risk-based vs annual breast cancer screening: the WISDOM randomized clinical trial. JAMA. 2026;335(9):763-774. Pragmatic randomised trial in people: 28,372 women aged 40-74, median follow-up 5.1 years. doi:10.1001/jama.2025.24784 (PMID 41385349)
Awareness content — not medical advice. Always talk to your doctor or pharmacist before making changes.