Inherited Cancer Risk: Common Enough to Plan For, Mostly Untested

Most cancers are not inherited, but a meaningful minority are linked to a variant passed down in a family: in one large study of tumours from 33 cancer types, about 8% of people carried a harmful inherited variant^[1]. In a large US population sequencing study, most carriers had never been tested, and whether someone ever finds out still depends heavily on which cancer they have and where they live.

Inherited risk reaches beyond one person. A single test result can carry meaning for parents, siblings and children, and it often arrives in the middle of treatment, or years before any cancer at all. That is why the gap between who carries a high-risk variant and who has been offered a test matters so much to families and to the charities that support them. This piece sets out what the population-level evidence shows about BRCA1, BRCA2 and Lynch syndrome, how large the risks are and how uncertain those estimates can be, and how unevenly testing reaches people around the world. It does not interpret anyone's family history, and it is not a guide to whether any individual should be tested.

About one cancer in twelve was linked to a known inherited variant in a study of 10,389 adults

The clearest single estimate comes from the Cancer Genome Atlas, a large US-led research collection of tumour and matched normal samples. Sequencing the inherited (germline) DNA of 10,389 adults with 33 types of cancer, researchers found 853 harmful or likely harmful variants in about 8% of cases^[1]. That is a research cohort, not a random sample of everyone with cancer, and the share differs a great deal between cancer types, so it is best read as an order of magnitude: most cancers arise without a high-risk inherited variant, but a sizeable minority carry one.

For particular cancers the share is higher. The US Preventive Services Task Force (USPSTF) states that harmful BRCA1 and BRCA2 variants account for 5% to 10% of breast cancer cases and 15% of ovarian cancer cases^[4]. Lynch syndrome, caused by variants in the DNA mismatch repair genes MLH1, MSH2, MSH6 and PMS2, is linked mainly to bowel and womb (endometrial) cancers. For the wider picture on these cancers, see our pieces on breast cancer, bowel cancer and gynaecological cancers.

It is worth setting this alongside the other things known to shape cancer risk. Modelled estimates link a considerably larger share of cancers to factors societies can change, such as tobacco, alcohol and infections; we cover those in how much cancer is linked to changeable risk factors. Inherited variants are a different kind of risk: present from birth, not chosen, and shared within families. Neither kind is a matter of blame.

Carrying a high-risk variant is uncommon for any one person, but not rare across a population

Estimates of how many people carry these variants come from different methods and populations, so they are not directly comparable, but together they show the scale:

• BRCA1 and BRCA2: the USPSTF puts harmful variants at an estimated 1 in 300 to 500 women in the general population (0.2% to 0.3%), and 2.1% among Ashkenazi Jewish women^[4]. When a US health system in Pennsylvania sequenced 50,726 adult volunteers, 267 (0.5%) carried a BRCA1 or BRCA2 variant expected to be harmful^[2].
• Lynch syndrome: a modelling study, based on the families of 5,744 people with bowel cancer from population registries in the US, Canada and Australia, estimated that 1 in 279 people in the population carry a variant in a mismatch repair gene, with PMS2 (1 in 714) and MSH6 (1 in 758) more common than MLH1 (1 in 1,946) and MSH2 (1 in 2,841)^[5]. These are model-based estimates rather than a direct count, and other methods give different figures.

Purely as an illustration (our arithmetic, not a study figure), in a town of 100,000 people these estimates would imply about 360 people carrying a Lynch syndrome variant and, among the town's roughly 50,000 women alone, 100 to 170 carrying a harmful BRCA variant, most of them living without cancer. That framing matters for services: inherited risk is not a niche concern for a handful of families, but a population-wide pattern that health systems can plan for.

Two cautions apply. First, prevalence differs by ancestry, as the Ashkenazi Jewish figure shows, and most of the studies cited here drew on populations in North America, Europe and Australasia, so the figures may not transfer to other populations. Second, being a carrier is not the same as having or developing cancer. Many carriers never do, which is why the size of the risk, and the uncertainty around it, matters.

For BRCA carriers, estimated breast cancer risk by age 80 is around 70%, while ovarian risk differs sharply between the two genes

"Penetrance" is the share of carriers who go on to develop a cancer by a given age. Estimates depend on how carriers were found and followed. A widely cited estimate comes from following carriers forward in time (a prospective cohort): a study pooling three carrier consortia, including the International BRCA1/2 Carrier Cohort Study, followed 6,036 women carrying BRCA1 variants and 3,820 carrying BRCA2 variants, recruited in 1997-2011, mostly through national studies in the UK, the Netherlands and France, for a median of five years^[6].

In that cohort, the estimated cumulative risk of breast cancer by age 80 was 72% (95% confidence interval 65% to 79%) for BRCA1 carriers and 69% (61% to 77%) for BRCA2 carriers. For ovarian cancer it was 44% (36% to 53%) for BRCA1 and 17% (11% to 25%) for BRCA2. Among women who had already had breast cancer, the estimated risk of a cancer in the other breast within 20 years was 40% (35% to 45%) for BRCA1 and 26% (20% to 33%) for BRCA2^[6].

Three points keep these figures honest. They are averages across a cohort, not a forecast for any one woman. The same study found that risk was higher for women with two or more affected close relatives and varied with where in the gene a variant lay^[6]. And the cohort was drawn mainly from three western European countries, so the estimates may not describe carriers elsewhere, or carriers found through population sequencing: in the Pennsylvania study, 44 of 89 previously untested carriers with complete history data (49.4%) did not meet published guidelines for clinical testing^[2].

Lynch syndrome risk differs sharply by gene, from around half to around one in ten for bowel cancer

Lynch syndrome is not one risk but four, one for each mismatch repair gene. The Prospective Lynch Syndrome Database pooled 6,350 carriers followed by expert centres in 18 countries, mostly in Europe, the Americas and Australasia, over 51,646 years of follow-up, with 1,808 cancers observed^[7]. The table shows the observed cumulative incidence by age 75, with 95% confidence intervals.

Source: Dominguez-Valentin et al., Genetics in Medicine 2020, Table 1^[7]. Observed cumulative incidence among carriers followed at specialist centres, so it reflects the care those carriers received; not a lifetime risk or a prediction for any individual.

The pattern is striking when read beside the prevalence estimates above. The two genes thought to be most common in the population, PMS2 and MSH6, carry the lowest estimated bowel cancer risks, while MLH1 and MSH2, rarer in the population, carry the highest^[5][7]. The confidence intervals for PMS2 and MSH6 are also very wide, because fewer carriers and cancers were observed: a PMS2 bowel cancer interval running from 2.9% to 40.8% is a plain statement of how much is still unknown.

That uncertainty is itself an evidence gap. Families affected by the less studied genes deserve estimates as precise as those for the better studied ones, and that depends on long-term follow-up of carriers in more countries.

Most carriers have never been tested, and testing after a cancer diagnosis is the exception

The clearest evidence that most carriers are unaware comes from population sequencing. Among the 50,726 adult volunteers in the Pennsylvania health-system study, 219 of the 267 BRCA1/2 carriers (82.0%) had never had clinical genetic testing, and sequencing found about five times as many carriers as previous clinical care had^[2]. One in five carriers (56 of 267, 20.9%) already had a related cancer diagnosis.

Testing is also uncommon among people who already have cancer. Linking cancer registry records to laboratory results for 1,369,602 people diagnosed in California and Georgia in 2013-2019, researchers found that only 93,052 (6.8%) had germline genetic testing by March 2021^[3]. The share varied enormously by cancer type, from half of men with breast cancer to 0.3% of people with lung cancer.

The same study found racial and ethnic gaps among people with breast or ovarian cancer: 31% (95% CI 30%-31%) of non-Hispanic White patients were tested, compared with 25% (24%-25%) of Black, 23% (23%-23%) of Hispanic and 22% (21%-22%) of Asian patients^[3]. These figures describe two US states in one period, not other countries, but they show the shape of the problem: even where tests are available, who gets tested is uneven, and the gaps run along racial and ethnic lines.

Systematic testing can close the gap quickly, but access across the world is very uneven

England offers a worked example of what a deliberate system change can do. Testing tumours for mismatch repair deficiency is a way to pick out people with bowel or womb cancer who may have Lynch syndrome, without relying on family history. A national transformation project in England set up national and regional project teams in June 2021 and trained 276 "Lynch syndrome champions" across 248 bowel and womb cancer multidisciplinary teams, covering more than 95% of teams nationally^[8]. According to the published evaluation:

• tumour mismatch repair testing rose from 43% to 94% for bowel cancer and from 19% to 95% for womb cancer, comparing national registry data from 2020 with the project's own audits of 30 consecutive diagnoses per team in 2022-2023 (different data sources, so an indicative comparison);
• new Lynch syndrome diagnoses rose from 545 in 2020 (535 in the paper's results text) to 1,394 in 2024, about two and a half times the 2020 total (the paper calls this a 255% increase; strictly a 156% rise), against a target of a 50% increase;
• where cancer teams arranged germline testing themselves ("mainstreaming"), the time from tumour test to germline test was 21 days, against at least 180 days for people referred on to regional clinical genetics services, in 109 cases in one region (North Thames); mainstreaming had been set up in 46% of bowel and 41% of womb cancer teams^[8].

This is a single country's programme, funded by NHS England and evaluated largely through its own audit data, and it measures testing and diagnoses rather than cancer outcomes. But it shows that low testing rates are not fixed: they respond to organisation, training and clear responsibility.

Elsewhere, evidence on access is patchier. The MAGENTA survey, circulated through patient organisations in nine countries and funded by AstraZeneca and Merck Sharp & Dohme (which the authors say had no role in its design or analysis), drew 1,524 responses from people with breast cancer, 1,176 of them analysed. The share who had been tested ranged from 82.7% of Australian respondents and 79.0% in Latin America to 45.3% in Asia and 9.9% in Egypt^[9].

The most common reason respondents gave for not being tested was simply that no one had offered a test (71.6% of those untested); cost was named as a reason by 15.4% of those untested, and as a perceived barrier by 49.4% of those asked^[9]. The authors note that people who had been tested were more likely to respond, so the figures probably overstate testing, and population rates in several of these countries may well be lower. Because this one self-selected, industry-funded survey is the only multi-country evidence here, it shows the shape of the access gap rather than measuring it.

What the evidence asks of us

The figures point to one gap above all: between the number of people who carry a high-risk inherited variant and the number who have ever been offered the chance to find out. Closing it is a matter for services and policy, not for individuals to navigate alone.

• Make testing systematic where guidance already supports it. England's Lynch syndrome project shows that when tumour testing becomes routine and responsibility is clear, testing rates can rise from under half to around 94-95% within a few years^[8].
• Bring testing closer to cancer care. In one English region's evaluation, mainstreamed testing cut the time from tumour test to germline test from at least 180 days to 21^[8]. For people deciding on treatment, and for relatives who may be affected, months matter.
• Close gaps in who is offered a test. The lower testing rates for Black, Hispanic and Asian people with breast or ovarian cancer in California and Georgia^[3], and the finding that not being offered a test was the commonest reason for not having one across nine countries^[9], call for monitoring of testing by ethnicity, region and income, and for action where gaps appear.
• Recognise the limits of family-history criteria. The USPSTF recommends assessing women with a personal or family history of breast, ovarian, tubal or peritoneal cancer, or ancestry linked to BRCA variants, and recommends against routine testing for women without these^[4]. Yet population sequencing has found many carriers who would not have met such criteria^[2]. Guideline bodies and researchers need to keep weighing who is missed by current criteria.
• Fund research in more populations. Most of the risk estimates here come from cohorts in Europe, North America and Australasia. Charities and funders can press for carrier studies, penetrance estimates and access data in more countries, including countries where, as the Egyptian survey results suggest, testing remains scarce^[9].

Language: this piece says "people with cancer" and "people living with and after cancer", and describes the illness and its treatment plainly, without combat metaphors. Clinical terms such as "stage", "incidence" and "net survival" are used in the sense the cited source defines them.

Why we're publishing this

Charities working with families affected by inherited cancer risk often have to make the case for testing services with scattered figures. We have brought the key population-level evidence together in one place, with the uncertainty left in, so that it can be cited accurately. The case these figures make belongs to people living with and after cancer and their families: that inherited risk is common enough to plan for, and that whether someone finds out should not depend on their cancer type, ethnicity or country. More pieces in this series are collected in our cancer research library.

How to read this data

Prevalence figures (how many people carry a variant) and penetrance figures (how many carriers develop cancer by a given age) answer different questions and come from different study designs. Penetrance figures are cumulative risks to a stated age in a stated cohort, with 95% confidence intervals showing the range consistent with the data; wide intervals mean few carriers or cancers were observed. They describe groups, not individuals. Testing-uptake figures come from two US states, one national programme in England (evaluated by the programme itself) and a self-selected, industry-funded nine-country survey, so they are snapshots, not a global rate. Several sources were read as abstracts only; we have used only the figures stated there.

Use this data

Free to cite with attribution to Better Health For Everyone. Charities, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.

Important: This article summarises published research for general information. It is not medical advice and cannot be used to diagnose, treat or predict anything about an individual. If you have concerns about your own or someone else's health, please speak to a qualified health professional.

Sources

1. Huang KL, Mashl RJ, Wu Y, Ritter DI, et al. Pathogenic germline variants in 10,389 adult cancers. Cell, 2018. europepmc.org/article/MED/29625052
2. Manickam K, Buchanan AH, Schwartz MLB, et al. Exome sequencing-based screening for BRCA1/2 expected pathogenic variants among adult biobank participants. JAMA Network Open, 2018. europepmc.org/article/MED/30646163
3. Kurian AW, et al. Germline genetic testing after cancer diagnosis. JAMA, 2023. europepmc.org/article/MED/37276540
4. US Preventive Services Task Force. BRCA-related cancer: risk assessment, genetic counseling, and genetic testing. Final recommendation statement. USPSTF, 2019. uspreventiveservicestaskforce.org/uspstf/recommendation/brca-related-cancer-risk-assessment-genetic-counseling-and-genetic-testing
5. Win AK, Jenkins MA, Dowty JG, et al. Prevalence and penetrance of major genes and polygenes for colorectal cancer. Cancer Epidemiology, Biomarkers & Prevention, 2017. europepmc.org/article/MED/27799157
6. Kuchenbaecker KB, Hopper JL, Barnes DR, et al. Risks of breast, ovarian, and contralateral breast cancer for BRCA1 and BRCA2 mutation carriers. JAMA, 2017. europepmc.org/article/MED/28632866
7. Dominguez-Valentin M, et al. Cancer risks by gene, age, and gender in 6350 carriers of pathogenic mismatch repair variants: findings from the Prospective Lynch Syndrome Database. Genetics in Medicine, 2020. europepmc.org/article/PMC/PMC7371626
8. Monahan KJ, Fleming P, Ryan NAJ, Monje-Garcia L, Armstrong R, et al. Outcomes from the English National Lynch Syndrome transformation project. International Journal of Cancer, 2026. europepmc.org/article/MED/41521736
9. Powell S, Artigas M, Borovova I, et al. MAGENTA: a multinational patient survey assessing the awareness, perceptions and unmet needs in genetic testing and counselling among patients with breast cancer. Frontiers in Oncology, 2024. frontiersin.org/journals/oncology/articles/10.3389/fonc.2024.1380349/full