What Does Circulating Tumor DNA Mean?

Everyone's blood holds cell-free DNA, or cfDNA. These are short broken pieces of DNA that float in plasma after cells die and spill their contents. In a person with cancer, some of those pieces came from tumor cells. That fraction is circulating tumor DNA, or ctDNA.

The FDA defines it as tumor-derived fragmented DNA shed into the bloodstream that is not attached to any cell. The same guidance is blunt about the catch. How much ctDNA is present varies by tumor type and location. It also varies by stage, by how much tumor there is, and by how the cancer is responding.

That variability is the whole story behind why these tests can be powerful and unreliable in the same week.

A ctDNA test is not one thing, but the FDA-approved lung cancer assay Guardant360 CDx shows the scale involved.

It needs at least 5 mL of whole blood. The blood goes into a special tube that keeps white blood cells from bursting and flooding the sample with normal DNA. The test then reads that DNA. It looks for single-letter changes in 55 genes, extra copies of 2 genes, and fusions in 4 genes.

Its published limit of detection is 0.2% to 1.5% mutant allele fraction. Which end applies depends on how much DNA the lab could recover. Mutant allele fraction, also called variant allele frequency, is the share of DNA pieces at that spot that carry the change. At 0.2%, the test is hunting two changed pieces among a thousand.

If your report lists a variant with a percentage next to it, that percentage is this number. A very low fraction is not automatically less serious, but it is closer to the edge of what the assay can see.

This is the single most important line in the approval document, and it is worth quoting: "A negative result from a plasma specimen does not assure that the patient's tumor is negative for genomic findings."

The FDA's instruction follows from that. Lung cancer patients who test negative on plasma for the listed biomarkers should be reflexed to tissue biopsy testing.

A tumor may put almost nothing into the blood. That happens when it is small, when it sheds slowly, or when it sits behind the blood-brain barrier. So a negative blood test means the test did not find it. It does not mean it is not there. A positive result is far more informative than a negative one.

Three different jobs get bundled under the same phrase, and they carry different levels of evidence.

Choosing a drug right now. This is the established use. Several plasma tests are FDA-approved companion diagnostics. That means the agency cleared them to pick patients for one named drug.

Finding molecular residual disease. MRD means ctDNA that is still there after surgery, radiation, or chemotherapy meant to clear the cancer. Finding it says disease persists, even when scans look clean.

Tracking response over time. Watching ctDNA rise or fall during treatment is under study as an early signal. The FDA says plainly that more data from randomized trials is needed first. Until then, a change in ctDNA is not an accepted trial endpoint.

The FDA maintains a public list of cleared and approved companion diagnostics. Three plasma-based tests appear on it.

Guardant360 CDx covers four EGFR changes: exon 19 deletions, exon 21 L858R, T790M, and exon 20 insertions. It also covers KRAS G12C, two ERBB2 changes, and ESR1 missense changes. Its paired drugs include osimertinib, amivantamab, sotorasib, fam-trastuzumab deruxtecan, and elacestrant.

FoundationOne Liquid CDx covers a wider set. It includes EGFR, BRAF V600E, MET exon 14 skipping, and ROS1, NTRK, and ALK gene changes. It also covers set PIK3CA changes and BRCA1 and BRCA2 changes. Its drugs include gefitinib, osimertinib, erlotinib, encorafenib with binimetinib, rucaparib, capmatinib, entrectinib, alectinib, and alpelisib.

cobas EGFR Mutation Test v2 covers EGFR exon 19 deletions, L858R, and T790M, paired with osimertinib and erlotinib.

Notice what this list is not. None of these are approved to tell you whether you have cancer. None are approved to decide on chemotherapy after surgery. That gap between what is approved and what gets ordered is worth raising.

The FDA guidance splits ctDNA assays into two designs, and the difference shows up in your results.

A tumor-informed test reads your actual tumor first. It picks a personal set of changes, then hunts only for those in blood. The payoff is precision, since anything found was already proven to be in your tumor. The weakness is that it cannot see changes that showed up later. That includes the ones that drive drug resistance.

A tumor-naive test uses a fixed gene panel and no prior look at your tumor. It is faster, since nothing has to be read first. But it may not include the gene that matters in your case.

If you are told the ctDNA test was negative, ask which type it was. The two fail in different ways. One may never have included the gene. The other may have missed a change present at a very low level.

For MRD use, the FDA sets two different bars. Cutting back treatment on a negative result demands high sensitivity and a high negative predictive value. Adding treatment on a positive result demands high specificity and a high positive predictive value.

That gap is the practical point. Using ctDNA to add treatment is not the same act as using it to hold treatment back. One test may be good enough for one and not the other.

The guidance also stresses steady timing and handling. When the blood was drawn and how it was processed both change the answer.

• Which test was run, by name, and was it tumor-informed or tumor-naive
• Is this test FDA-approved for this purpose, or is it being used off-label
• What was the mutant allele fraction for each variant reported
• If it was negative, is tissue testing the next step
• What decision changes based on this result, and what happens if it is wrong
• When would a repeat draw be meaningful

https://www.fda.gov/regulatory-information/search-fda-guidance-documents/use-circulating-tumor-deoxyribonucleic-acid-early-stage-solid-tumor-drug-development-guidance https://www.fda.gov/media/183874/download https://www.accessdata.fda.gov/cdrh_docs/pdf20/P200010B.pdf https://www.fda.gov/medical-devices/in-vitro-diagnostics/list-cleared-or-approved-companion-diagnostic-devices-in-vitro-and-imaging-tools https://www.cancer.gov/about-cancer/treatment/types/biomarker-testing-cancer-treatment