What Does Minimal Residual Disease Mean?

A pathologist reads a bone marrow slide under a microscope. At best, they can spot about one cancer cell among 100 normal ones. The FDA states that limit plainly. Standard visual detection in blood cancers has a threshold of one tumor cell in 100 cells.

So "complete remission" has always meant something narrower than it sounds. It means nothing visible at 1 in 100.

MRD testing works far below that. You may also see the letters written out as measurable residual disease. It is the same test.

MRD results come as a fraction or as a negative power of ten. Translate them:

• 10⁻⁴ — 1 cancer cell in 10,000, the same as 0.01%
• 10⁻⁵ — 1 in 100,000
• 10⁻⁶ — 1 in 1,000,000

A result of "MRD negative at 10⁻⁵" does not mean zero cells. It means none were found at a sensitivity of one in 100,000. That sensitivity is part of the result. Always ask which level was used.

The guidance recognizes four platforms:

• Flow cytometry — tags cells with glowing markers, then sorts them by the proteins on their surface
• Next-generation sequencing — reads DNA or RNA to find the tumor's own unique sequence
• RT-qPCR — copies and counts one specific fusion gene signal
• ASO-PCR — copies and counts one exact known sequence

The FDA says it does not favor one platform. It does require that the choice be set out in advance. That matters for you too. Results from different platforms are not interchangeable. Ask whether follow-up testing will use the same method as your first test.

Here is a detail worth borrowing. The FDA says a test's detection floor should sit at least 10-fold below the level used for decisions. If a decision is made at 1 × 10⁻⁵, the test should reach 1 × 10⁻⁶ where possible.

The reason is simple. A test working right at its own limit gives shaky answers at the exact point you care about.

There is no single MRD cutoff. The FDA guidance sets them separately.

Acute lymphoblastic leukemia. The FDA has accepted an MRD level of 0.1% or more as a marker of high relapse risk. That applies to patients in a first or second complete remission. In relapsed or refractory ALL, levels under 0.01% have been accepted as supporting evidence that a drug works.

Acute promyelocytic leukemia. An MRD level below 0.01% is generally counted as negative here. That applies after first-line induction built on arsenic and tretinoin. Testing at the end of consolidation is preferred.

Chronic lymphocytic leukemia. The test must reach a detection floor of at least 10⁻⁴, which is 0.01%. Testing applies to patients in complete response.

Chronic myeloid leukemia. CML uses a different yardstick, the International Scale. Major molecular response means BCR-ABL below 0.1% on that scale. The FDA advises PCR tests sensitive to more than a 4.5-log reduction.

Acute myeloid leukemia. Testing is done at complete response, once blood counts have recovered. There is an extra step here. Drug sponsors must show the markers reflect real leukemia. They must rule out clonal hematopoiesis, a common change in aging blood cells that is not leukemia.

The FDA names bone marrow as the preferred substrate for ALL, AML, and APL. Peripheral blood can be used with appropriate justification.

In multiple myeloma the rule is firmer. The FDA's myeloma guidance states that a bone marrow aspirate is required to call MRD negative.

So an MRD check usually means another marrow procedure. If your MRD is being followed, ask how often marrow will be sampled.

In multiple myeloma the FDA laid out specifics:

• Sensitivity of at least 1 in 10⁵ tumor cells, so one cell in 100,000
• Flow cytometry or sequencing-based methods
• Bone marrow aspirate required
• The patient must first reach complete response or stringent complete response
• MRD checked at that point, or within a set window such as plus or minus 3 months

The FDA also records a vote. Its Oncologic Drugs Advisory Committee agreed, with no dissent, that MRD is acceptable as an endpoint. It can support accelerated approval of drugs for multiple myeloma. That is a real shift, and it explains why the term now comes up so often.

Sustained MRD negativity means staying negative over time. That is a different measure. The FDA notes limited data on it, and suggests it stay a secondary endpoint for now.

MRD is strongest in combination, not alone. NCI's clinician summary for childhood ALL shows the range. Gene findings combined with MRD can define groups with event-free survival above 95%. The same pairing defines other groups with event-free survival of 50% or lower.

Two cautions belong with any negative result. First, marrow sampling checks one site, and disease can be patchy. Second, the FDA is explicit on validation. The link between MRD and real benefit must be proven separately in each setting. That includes newly diagnosed after transplant, relapsed or refractory, and transplant-ineligible.

So a negative MRD is good news about one sample, at one sensitivity, on one date. It is not a discharge from follow-up.

• Which platform was used, and at what sensitivity was the result reported?
• Was this bone marrow or blood?
• What threshold is my team using to make decisions, and where does my result fall?
• Will the next test use the same assay, so the numbers can be compared?
• Does an MRD result here change my treatment, or is it being tracked for information?

Remission and Recurrence covers the wider vocabulary. What Does Circulating Tumor DNA Mean? covers the blood-based cousin used in solid tumors. Questions to Ask at End of Treatment helps with what comes next.

• U.S. Food and Drug Administration — Hematologic Malignancies: Regulatory Considerations for Use of Minimal Residual Disease in Development of Drug and Biological Products for Treatment
• U.S. Food and Drug Administration — Minimal Residual Disease and Complete Response in Multiple Myeloma: Use as Endpoints to Support Accelerated Approval
• NCI PDQ — Childhood Acute Lymphoblastic Leukemia Treatment (Health Professional Version)