Medical disclaimer: This article is educational and is not medical advice. Cancer genetic testing decisions should be made with an oncologist or certified genetic counselor. Never make treatment or screening decisions based on consumer genetic testing alone.
"Genomic testing for cancer" sounds like one thing. It is two completely different things, run on two different genomes, answering two different questions, for two different purposes.
Confusing them is the single most common and most consequential mistake in this entire field. It leads people to believe they have been screened for hereditary cancer risk when they have not, and to believe a tumour test tells them what to tell their children when it does not.
Germline testing reads the DNA you were born with, in every cell of your body. It answers: am I at inherited risk of developing cancer, and is my family?
Somatic (tumour) testing reads the DNA inside a cancer that already exists. It answers: what is driving this specific tumour, and which drugs might work against it?
Germline testing: the genome you inherited
Every cell in your body carries the genome you were born with — half from each biological parent. That is your germline. It never changes, it is the same in your blood as in your skin as in your saliva, and it is what you pass to your children.
Some people inherit a variant in a gene whose job is to suppress tumours or repair DNA damage. BRCA1 and BRCA2 are the famous examples, but the list is much longer: clinically validated hereditary breast and ovarian cancer genes include ATM, BARD1, BRCA1, BRCA2, BRIP1, CDH1, CHEK2, PALB2, PTEN, RAD51C, RAD51D, STK11 and TP53. Lynch syndrome adds MLH1, MSH2, MSH6, PMS2 and EPCAM. Polyposis syndromes add APC, MUTYH, POLE, POLD1 and NTHL1.
Carrying such a variant does not mean you have cancer or will get it. It means your cells start life with one of their safety mechanisms already impaired, so fewer additional things need to go wrong before a cancer can develop.
What germline testing is for
- Risk assessment. Quantifying your lifetime risk of specific cancers.
- Screening decisions. Earlier, more frequent, or different-modality screening — MRI instead of mammography, colonoscopy starting decades earlier.
- Risk-reducing surgery. For some high-penetrance variants, prophylactic surgery is an evidence-based option.
- Cascade testing. This is the one people underrate. A germline finding is not just about you — siblings, children and parents each have a meaningful chance of carrying the same variant, and they can be tested for that single specific variant cheaply.
- Treatment selection. Increasingly, germline status affects treatment. PARP inhibitors, for example, are used in BRCA-associated cancers.
How common is hereditary cancer?
Approximately 5–10% of all cancers are associated with germline pathogenic variants. That is a minority — but in absolute terms it is an enormous number of people, and it is the subset where prevention actually has purchase.
The under-testing problem is severe. One study found that fewer than one in five women with a history of breast or ovarian cancer had undergone genetic testing, despite meeting NCCN criteria. In a community screening programme of 14,192 women, 23% met NCCN criteria, only about half of those opted for testing, and 7.6% of those tested carried a pathogenic or likely pathogenic variant across 18 different genes.
The barriers identified include provider awareness, insurance coverage, financial burden, racial bias, limited access to genetic counseling, and geography.
That 7.6% figure deserves a moment. Among women who met criteria and got tested, roughly one in thirteen carried a variant that changed their medical management. These are people whose screening and prevention plans changed as a direct result. And most of the eligible population never got there.
Somatic testing: the genome the cancer built
A tumour is a population of your own cells that has accumulated mutations. Those mutations are somatic — acquired during your lifetime, present only in the tumour, and not inheritable. You did not get them from your parents and you cannot pass them to your children.
Somatic testing sequences the tumour itself, typically from a biopsy sample or, increasingly, from circulating tumour DNA in a blood draw (a "liquid biopsy").
What somatic testing is for
- Targeted therapy selection. The central purpose. If a lung tumour carries an EGFR mutation, there are drugs specifically designed for that. If it does not, those drugs will not work.
- Immunotherapy eligibility. Mismatch repair deficiency and high tumour mutational burden predict response to immune checkpoint inhibitors.
- Prognosis. Some mutation profiles predict a more or less aggressive course.
- Clinical trial matching. Many trials enrol by molecular profile rather than by tumour site.
- Monitoring. Tracking tumour DNA over time can reveal recurrence or acquired resistance.
The key point: somatic testing is done after a cancer diagnosis. It is a treatment-planning tool, not a screening tool. It cannot tell you whether you will get cancer, because it requires a cancer to exist first.
Where the two intersect
They are not entirely separate, and the interaction is clinically important.
A tumour test sequences tumour DNA — but tumour cells still carry the germline genome underneath their acquired mutations. So a tumour test can incidentally reveal an inherited variant. This is common enough that reflex protocols exist.
Lynch syndrome is the canonical example, and the workflow is instructive:
NCCN recommends tumour screening for mismatch repair deficiency for all colorectal and endometrial cancers, regardless of age at diagnosis.
If MLH1/PMS2 protein expression is lost in the tumour, ASCO guidelines recommend first testing for a BRAF V600E variant or MLH1 promoter methylation to rule out a sporadic cause. If the tumour is mismatch-repair deficient and neither is found, germline testing is indicated. If any other protein (MSH2, MSH6, PMS2) is lost, germline testing should be carried out for the corresponding genes.
Read that carefully and you can see the logic. The tumour test flags something suspicious. A second tumour test distinguishes "this happened by chance in this tumour" from "this person may have inherited it." Only then does germline testing follow. It is an elegant funnel — and it is why a tumour report saying "MSH2 loss" is a reason to ask about germline testing, not a substitute for it.
The side-by-side
| Germline testing | Somatic / tumour testing | |
|---|---|---|
| What it reads | The DNA you were born with | The DNA inside an existing tumour |
| Sample | Blood or saliva | Tumour biopsy, or blood (liquid biopsy) |
| When | Any time — often before any cancer | Only after a cancer diagnosis |
| Main question | Am I at inherited risk? | Which treatment will work? |
| Heritable? | Yes — 50% chance per first-degree relative | No |
| Changes family's care? | Yes — cascade testing | No |
| Changes screening? | Yes | No |
| Changes treatment? | Sometimes (e.g. PARP inhibitors) | Yes — that is the point |
Who should have germline testing?
NCCN criteria are the standard, and they have been progressively widening. Broadly, testing is indicated for people with:
- A cancer diagnosis at an unusually young age. Criteria expanded to include people aged 45–49 at breast cancer diagnosis.
- Multiple primary cancers in one person
- Several relatives with related cancers
- Specific tumour types with high hereditary rates — ovarian, pancreatic, male breast cancer
- A known familial variant
- Certain ancestries with founder variants, notably Ashkenazi Jewish ancestry
- A personal history of 20 or more cumulative colorectal adenomas, or more than 10 adenomas, triggering polyposis panel testing
- A tumour showing mismatch repair deficiency without a sporadic explanation
An important caveat on criteria: research comparing guideline-based and alternative criteria has found comparable diagnostic yields between patients who met NCCN/ACMG criteria and those who did not, suggesting current criteria may miss carriers. The guidelines are a floor, not a ceiling. If your family history worries you and you do not strictly meet criteria, that is a conversation worth having with a genetic counselor rather than a closed door.
Where consumer genetic testing fits — and does not
This is the part that causes real harm. A consumer DNA test that reports "BRCA1/BRCA2" typically checks a small number of specific variants — commonly three founder variants most prevalent in Ashkenazi Jewish populations. There are thousands of known pathogenic BRCA variants. A "negative" result from such a test tells you almost nothing about whether you carry one of the others.
People have received a reassuring consumer result, skipped clinical testing, and later been diagnosed with a hereditary cancer caused by a variant the test never looked for. If you have a family history that concerns you, a consumer test is not a substitute for clinical germline testing. Full stop.
Whole genome sequencing occupies a genuinely better position here — it reads the entire gene rather than a handful of pre-selected spots, so it does not have the same blind spot by design. But it still is not a clinical hereditary cancer panel: it is not ordered by a physician, not accompanied by genetic counseling, and typically not validated to clinical standards for the specific reporting of pathogenic variants. We cover the technical reasons in detail in our panel vs. WGS comparison.
Understand your genome before you need to
30x whole genome sequencing reads every cancer-risk gene end to end — not just a handful of pre-selected variants. Full raw data included, so a genetic counselor can work with it directly.
Explore Dante Labs Whole Genome Sequencing → Use code GENOME for 10% off · Affiliate link · Not a substitute for clinical testingThe bottom line
If someone tells you they are getting "genomic testing for cancer," the useful next question is: germline or somatic?
If they have cancer and are choosing treatment, they probably mean somatic. If they are worried about family history and their own future risk, they mean germline. The two are not interchangeable, and treating one as if it were the other is how people end up with false reassurance about the thing that actually mattered.
Sources
- Carelon Medical Benefits Management. Appropriate Use Criteria: Hereditary Cancer Testing, 2026.
- UnitedHealthcare. Genetic Testing for Hereditary Cancer, 2026 — summarising NCCN and ESMO guidance.
- Identification of Patients at Elevated Cancer Risk through a Community-Based Genetic Testing Program. PMC.
- Multiple Primary Cancers as an Independent Criterion for Germline Testing. PMC.