Biomarkers in Acute Myeloid Leukemia

Print this article

A biomarker is a measurable feature of a cancer that tells doctors something useful about how it will behave or how to treat it. In acute myeloid leukemia, almost all biomarkers are genetic changes inside the leukemia cells, found by testing bone marrow or blood. They matter more in this disease than in most cancers, because the tumor cannot be measured on a scan. The genetic result is what defines the subtype, sets the risk group, and opens treatment options.

This article gives an overview of the biomarkers reported in acute myeloid leukemia in adults and what each one is used for. This site has several detailed articles on each, linked below. For an explanation of the pathology report itself, see Acute myeloid leukemia in adults. Biomarkers in children differ substantially and are covered in Acute myeloid leukemia in children.

How biomarker testing is done in acute myeloid leukemia

Biomarker testing in acute myeloid leukemia is performed on the same bone marrow or blood sample used to make the diagnosis. No separate procedure is needed. Several methods are used together, because no single test finds everything.

Chromosome analysis. A karyotype examines all chromosomes under a microscope and finds large changes, such as missing chromosomes or exchanged pieces.
FISH. Uses fluorescent probes to look for one specific change at a time. Faster than a karyotype and able to find changes a karyotype misses.
PCR. Copies and measures a specific piece of genetic material. It is very sensitive, which makes it the main method for tracking small amounts of leukemia after treatment.
Next-generation sequencing. Reads many genes at once and finds mutations that the other methods cannot. Most centers run a panel covering the genes listed in this article.
Flow cytometry. Measures proteins on the surface of individual cells rather than genes. It identifies CD33 and is one of the two ways residual disease is measured.

Results usually arrive in stages. Flow cytometry and FISH return within a day or two, while sequencing can take one to two weeks. Treatment often begins before all results are back, and the plan is adjusted as the remaining results arrive.

Biomarkers that define the subtype

Some biomarkers in acute myeloid leukemia define the leukemia. These findings name the subtype on the pathology report, and several place the leukemia directly into a favorable or adverse risk group.

PML::RARA. Defines acute promyelocytic leukemia. This result is treated as an emergency because the leukemia causes severe clotting and bleeding problems. Treatment with all-trans retinoic acid and arsenic trioxide is highly effective and differs from all other subtypes.
RUNX1::RUNX1T1 and CBFB::MYH11. Together, these make up core binding factor leukemia, reported as t(8;21) and inv(16) or t(16;16). Both are favorable and respond well to chemotherapy containing high-dose cytarabine.
KMT2A rearrangement. The KMT2A gene joins one of many partner genes. Most partners place the leukemia in the adverse group. This is also a drug target, described below.
Myelodysplasia-related mutations. Mutations in ASXL1, BCOR, EZH2, RUNX1, SF3B1, SRSF2, STAG2, U2AF1, or ZRSR2. Any one of these places the leukemia in the adverse group, whether or not there was a preceding blood disorder.
Chromosome changes. Loss of chromosome 7, loss of part of chromosome 5, a MECOM rearrangement, and a complex karyotype are all adverse findings.

FLT3

FLT3 is a gene that helps blood stem cells grow and survive. Two types of change occur in acute myeloid leukemia. An internal tandem duplication, written FLT3-ITD, inserts an extra copy of part of the gene, leaving the FLT3 protein permanently switched on. A point mutation in the tyrosine kinase domain, written FLT3-TKD, has a similar but weaker effect. FLT3-ITD is found in roughly a quarter of adults with acute myeloid leukemia.

Clinicians test FLT3 in everyone at diagnosis for two reasons. It affects the risk group, and it is one of the strongest drug targets in this disease. Midostaurin is added to standard induction chemotherapy for newly diagnosed FLT3-mutated leukemia and improves survival compared with chemotherapy alone. Quizartinib is also approved for newly diagnosed FLT3-ITD-positive leukemia alongside induction and consolidation, and as maintenance afterward. Gilteritinib is used for leukemia that has come back or has not responded.

One point is confusing. Older reports and older sources describe a FLT3 allelic ratio, meaning the proportion of leukemia cells carrying the change. That measure was removed from risk classification in 2022, so any FLT3-ITD without adverse-risk changes now sits in the intermediate group regardless of the amount present. Our detailed article is titled FLT3 mutations in acute myeloid leukemia.

NPM1

NPM1 is the most commonly mutated gene in adult acute myeloid leukemia, found in roughly 30 percent of people. The mutation causes the nucleophosmin protein to sit in the wrong part of the cell, which disrupts normal blood cell development. Because nearly all NPM1 mutations affect the same small region of the gene, they are unusually easy to measure precisely.

An NPM1 mutation does three things. It defines a subtype. It places the leukemia in the favorable risk group when no FLT3-ITD is present. And it provides the most sensitive residual disease test available in this disease, because PCR can measure the mutation at levels far below what any microscope can see. Rising NPM1 levels after treatment predict relapse weeks in advance.

Since 2025, an NPM1 mutation has also been a drug target, through the menin inhibitors described below. Our detailed article is NPM1 mutations in acute myeloid leukemia.

Menin inhibitors: NPM1 and KMT2A as drug targets

Menin inhibitors are the newest class of drugs in acute myeloid leukemia and are relevant to two of the biomarkers above. Menin is a protein that leukemia cells with an NPM1 mutation or a KMT2A rearrangement depend on to stay in an immature, dividing state. Blocking menin allows those cells to mature and die. Together,r these two groups account for roughly 40 percent of acute myeloid leukemia, so a large share of people carry a target for this class.

Two drugs are approved. Revumenib was approved in November 2024 for relapsed or refractory acute leukemia with a KMT2A translocation, in adults and in children aged 1 and older. In October 2025, it was also approved for relapsed or refractory NPM1-mutated leukemia. Ziftomenib was approved in November 2025 for adults with relapsed or refractory NPM1-mutated leukemia. Both are taken by mouth.

These approvals are for leukemia that has returned or has not responded to earlier treatment. Trials combining menin inhibitors with standard chemotherapy at diagnosis are underway, so this may change. The class has a characteristic side effect called differentiation syndrome, in which maturing leukemia cells cause fever, breathlessness, and fluid retention, and it needs prompt treatment.

IDH1 and IDH2

IDH1 and IDH2 are genes that normally make enzymes involved in how cells produce energy. When mutated in acute myeloid leukemia, these enzymes produce an abnormal substance that blocks blood cells from maturing, so blasts accumulate. IDH1 mutations occur in roughly 6 to 10 percent of adults with this disease, and IDH2 mutations in roughly 8 to 15 percent.

Both are tested at diagnosis because approved inhibitors exist for each. Ivosidenib targets IDH1 and is approved for newly diagnosed IDH1-mutated leukemia in people who cannot have intensive chemotherapy, often combined with azacitidine, and for disease that has returned. Enasidenib targets IDH2 in relapsed or refractory disease. Olutasidenib is a second IDH1 inhibitor approved after relapse. Differentiation syndrome can occur with these drugs as well. Our detailed article is IDH1 and IDH2 mutations in acute myeloid leukemia.

TP53

TP53 normally acts as a brake on uncontrolled cell growth, detecting damaged DNA and triggering the cell to die. When it is mutated in acute myeloid leukemia, which happens in roughly 5 to 10 percent of cases, that brake is lost. The leukemia cells then survive chemotherapy that would otherwise kill them.

A TP53 mutation places the leukemia in the adverse risk group and predicts lower remission rates and shorter survival with standard treatment. No approved drug targets mutated TP53. Drugs designed to restore the mutated protein’s function have been tested but have not shown the hoped-for benefit in trials. Clinical trial enrollment is often discussed early for this group. Our detailed article is TP53 mutations in blood cancers.

CD33 and BCL2

Two biomarkers in acute myeloid leukemia are proteins rather than genetic changes, and both are drug targets.

CD33 is a protein on the surface of leukemia blasts in most people with this disease, and flow cytometry measures it at diagnosis. Gemtuzumab ozogamicin is an antibody that finds CD33-positive cells and delivers a chemotherapy payload directly into them. Adding it to standard chemotherapy improves outcomes, and the benefit is clearest in core-binding factor leukemia.

BCL2 is a protein that keeps cells alive by blocking the signal that would normally kill them. Leukemia cells rely on it. Venetoclax blocks BCL2 and removes that protection. It is not used alone but is combined with azacitidine or decitabine. That combination is the standard first treatment for adults who cannot tolerate intensive chemotherapy, including most adults over 75. It is also used after relapse.

CEBPA and inherited predisposition

Some biomarkers in acute myeloid leukemia point beyond the leukemia to the rest of the family. CEBPA is the clearest example. An in-frame mutation in the bZIP region of CEBPA is a favorable prognostic finding, and a proportion of CEBPA mutations are inherited rather than acquired.

Other genes on a standard panel carry the same implication. DDX41 mutations are found in roughly 2 to 5 percent of adults with acute myeloid leukemia and are frequently inherited. They often cause disease later in life than most inherited cancer syndromes. RUNX1, GATA2, ANKRD26, and ETV6 behave similarly. Finding one of these on the leukemia panel does not by itself prove it is inherited, and a separate test on normal tissue is needed to tell the difference.

This matters for two reasons. Relatives may be offered testing and monitoring. If a stem cell transplant is planned and a brother or sister may be the donor, that sibling is tested first. A donor carrying the same inherited change cannot be used. You can read more in What is a hereditary cancer syndrome?

Biomarkers used to follow treatment.

Some biomarkers in acute myeloid leukemia are measured again after treatment starts, to detect very small amounts of remaining leukemia. This is called measurable residual disease (MRD), previously called minimal residual disease. Flow cytometry measures it in almost everyone, and PCR measures it more sensitively when the leukemia carries a suitable marker, such as an NPM1 mutation or a PML::RARA fusion.

MRD results are reported as negative or positive against a stated threshold, and they are used in three ways. Reaching MRD-negative status predicts a lower risk of relapse. A rising level after an initially clear result signals relapse before it becomes visible under the microscope. An MRD result can move the leukemia between risk groups, so a favorable-risk leukemia that stays MRD positive may be treated as intermediate risk. In acute promyelocytic leukemia, a positive PCR result after consolidation prompts treatment even with no symptoms.

Questions to ask your doctor

Which biomarker tests were performed on my bone marrow, and are any results still pending?
Which genetic changes were found, and which of them define my subtype?
Do I have a FLT3, NPM1, IDH1, IDH2, TP53, CEBPA, or KMT2A change?
Do any of my results open a targeted treatment option, and would I use it now or later?
Are my leukemia cells CD33 positive, and would gemtuzumab ozogamicin be added?
Which risk group do my biomarker results place me in?
Will my leukemia be followed with MRD testing, and by which method?
How often will MRD be measured, and what happens if it becomes positive?
Could any of the changes found be inherited, and should I have testing on normal tissue?
Should my relatives be tested, and would a sibling donor be tested before a transplant?
Are there open clinical trials that match my biomarker results?
Will testing be repeated if the leukemia comes back?