MRD stands for measurable residual disease (also called minimal residual disease). It refers to a very small number of cancer cells that remain in the body during or after treatment, too few to show up on standard tests like blood counts or imaging scans but detectable with specialized laboratory methods. These tests can find as few as one cancer cell among a million normal cells. MRD testing is used primarily in blood cancers like leukemia and multiple myeloma to gauge how well treatment is working and predict whether cancer is likely to return.
Why It Matters After Treatment
When someone with a blood cancer finishes a round of chemotherapy or other treatment, doctors typically check whether the cancer is in remission. Traditional methods look at blood samples or bone marrow under a microscope and can confirm that the cancer appears to be gone. But “appears to be gone” and “actually gone” are two different things. Tiny clusters of cancer cells can survive treatment and eventually grow back into a full relapse.
MRD testing fills that gap. By detecting residual cancer cells far below what a microscope can see, it gives doctors a much clearer picture of how deep the remission actually goes. A patient who tests MRD-negative (no detectable residual disease) generally has a significantly better outlook than someone who tests MRD-positive. In multiple myeloma, for example, patients who achieved MRD negativity had a median progression-free survival of 49.2 months compared to 18.4 months for those who remained MRD-positive. Their overall survival was substantially longer as well.
How the Tests Work
Two main technologies are used for MRD testing, each with different strengths.
Flow cytometry works by tagging cells with fluorescent markers that bind to specific proteins on their surface. Cancer cells carry a different combination of surface proteins than normal cells, so the test can sort through millions of cells and flag the abnormal ones. The most advanced version, called next-generation flow, uses an eight-color system and can detect one cancer cell in 100,000 normal cells (a sensitivity level of 10⁻⁵). It delivers results quickly, costs less, and doesn’t require a sample from the time of diagnosis for comparison.
Next-generation sequencing (NGS) takes a DNA-based approach. Cancer cells carry unique genetic rearrangements that act like a fingerprint. The test sequences millions of DNA fragments and searches for that fingerprint among normal cells. NGS can reach a sensitivity of one cancer cell in a million (10⁻⁶), making it 10 to 100 times more sensitive than standard flow cytometry. It also works on frozen or stored samples, not just fresh ones. The only FDA-cleared MRD test, the clonoSEQ assay by Adaptive Biotechnologies, uses this NGS approach and is approved for multiple myeloma and acute lymphoblastic leukemia.
Both methods are considered adequate at the standard benchmark of 10⁻⁵ sensitivity, though NGS has the advantage when deeper detection is needed.
What the Sensitivity Levels Mean
You’ll often see MRD sensitivity described with numbers like 10⁻⁴, 10⁻⁵, or 10⁻⁶. These represent how rare a cancer cell can be and still get picked up by the test:
- 10⁻⁴ (0.01%): one cancer cell per 10,000 normal cells. This is the level most standard flow cytometry assays reach.
- 10⁻⁵ (0.001%): one cancer cell per 100,000. This is the current international benchmark for what qualifies as a true MRD assessment.
- 10⁻⁶ (0.0001%): one cancer cell per million. The deepest level currently available through NGS testing.
Deeper sensitivity matters clinically. In one study of acute lymphoblastic leukemia, none of the patients who tested MRD-negative at the 10⁻⁶ level after their first round of chemotherapy went on to relapse. At the less sensitive 10⁻⁴ cutoff, 16% of patients who tested negative still relapsed within five years.
When MRD Testing Happens
MRD is not a one-time test. Doctors check it at several points during and after treatment, with each time point providing different information. The most common testing windows are during or immediately after the initial treatment phase (induction), before a stem cell transplant, and after transplant. Some protocols also check MRD during the consolidation phase, which is the treatment given after the first remission to deepen the response.
In childhood acute lymphoblastic leukemia, MRD results at two specific time points shape the entire treatment plan. The first check comes at the end of induction therapy (roughly four to six weeks into treatment). The second comes around week 10 to 12. These two results divide patients into three risk groups: those with low or undetectable MRD at both checks have the best prognosis, those who start with high MRD but clear it by the second check fall in the middle, and those still MRD-positive at the second time point face the highest risk of relapse. Children in that high-risk group are moved to more intensive treatment regimens.
How MRD Guides Treatment Decisions
MRD results increasingly influence what happens next in a patient’s care. The core idea is straightforward: patients whose cancer responds deeply enough may be able to avoid the side effects of more aggressive therapy, while those with persistent disease can be identified early and switched to stronger treatments before a full relapse occurs.
In acute myeloid leukemia, the European LeukemiaNet published updated guidelines in 2025 that now categorize MRD results into three response levels: optimal, warning, or high risk of treatment failure. These categories are tailored to specific genetic subtypes of the disease, recognizing that the same MRD result can mean different things depending on the biology of the cancer. For certain genetic subtypes, ultra-high-sensitivity NGS testing is now specifically recommended before transplant.
In multiple myeloma, MRD status is becoming a key factor in deciding whether to continue, intensify, or eventually stop maintenance therapy. Clinical trials are actively using MRD negativity as a primary endpoint, which may eventually allow MRD results to support faster drug approvals.
What the Test Involves
For most blood cancers, MRD testing requires a bone marrow sample, collected through a bone marrow aspirate. This is a procedure where a needle is inserted into the back of the hip bone to draw out a small amount of liquid marrow. It’s done under local anesthesia and takes about 15 to 30 minutes, though the area can be sore for a few days afterward.
Researchers have explored whether a simple blood draw could replace bone marrow sampling, at least in some cases. In multiple myeloma, a comparison of matched blood and bone marrow samples found that blood testing had perfect specificity: if the blood test detected residual disease, it was always confirmed in the marrow. However, blood testing missed a large proportion of cases that the bone marrow test caught. About a third of patients who were MRD-positive in their marrow tested negative in their blood. This means a positive blood result is reliable and could spare some patients a bone marrow procedure, but a negative blood result can’t rule out residual disease on its own. For now, bone marrow remains the standard sample type for MRD assessment.
Which Cancers Use MRD Testing
MRD testing is most established in blood cancers where the disease lives primarily in the bone marrow and blood, making it accessible to sampling. The cancers where MRD testing plays the largest role include acute lymphoblastic leukemia (particularly in children, where it has been used for treatment decisions for over a decade), acute myeloid leukemia, multiple myeloma, and chronic lymphocytic leukemia. Research into MRD-like approaches for solid tumors is growing, often using circulating tumor DNA detected in blood samples, though these applications are less standardized than in blood cancers.
The shift in terminology from “minimal” to “measurable” residual disease reflects an important point: the disease isn’t necessarily minimal in its threat. What matters is that it’s measurable, and that measurement gives patients and their doctors concrete information to act on.