FLT3 mutated AML

Diagnosis, frontline therapy, MRD, and relapse

Ashwin Kishtagari, MBBS, MSCI

Division of Hematology/Oncology, Vanderbilt University Medical Center

2026-08-16

The target

FLT3 mutations in newly diagnosed AML

Roughly 30% of adults with newly diagnosed AML carry an activating FLT3 mutation; both subtypes let the receptor signal without ligand.

• ITD (internal tandem duplication, ~25% of AML). Juxtamembrane duplications of 3 to >100 amino acids; higher relapse risk, worse with a high allelic ratio.
• TKD (point mutations, mainly D835, ~7–8%). Prognostic effect uncertain, yet the subtype decides which drugs work.

Frequencies from the introductions of RATIFY and QuANTUM-First. ITD length and allelic-ratio prognosis from Stone et al. Both subtypes cause factor-independent growth in murine models and a fatal myeloproliferative neoplasm in mice.

Three inhibitors, two binding classes

Drug	Generation	Selectivity	ITD	TKD
Midostaurin	First	Multikinase (also KIT)	✓	✓
Quizartinib	Second	Selective, type II	✓	—
Gilteritinib	Second	Selective, type I (also AXL)	✓	✓

Type II agents bind the inactive kinase conformation, so TKD point mutations escape them. TKD clones emerging on quizartinib were an early described resistance route.

Single-agent first-generation inhibitors in relapsed disease gave transient blast reductions and rarely a CR (Stone, intro). Quizartinib’s KIT activity contributes to its myelosuppression (Perl, intro). Gilteritinib also inhibits AXL, implicated in FLT3-inhibitor resistance.

Frontline therapy, fit patients

Midostaurin with 7+3 (RATIFY)

Phase 3, 717 patients aged 18–59, ITD and TKD alike.

• Median OS 74.7 vs 25.6 mo; HR 0.78 (0.63–0.96), p = 0.009
• CR similar (58.9% vs 53.5%); the gain came after remission
• Benefit consistent across all FLT3 subtypes

Midostaurin 50 mg BID on days 8–21 of induction and consolidation, then 12 maintenance cycles. Trial therapy stopped at transplantation, so maintenance was effectively for non-transplanted patients. 57% underwent transplant at some point. Grade ≥3 anemia and rash were higher with midostaurin. The dramatic median OS gap partly reflects inflection points of the KM curves; the HR is the more reliable summary (per the authors).

Quizartinib with 7+3 (QuANTUM-First)

Phase 3, 539 patients aged 18–75, FLT3-ITD only (VAF ≥3%).

• HR 0.78 (0.62–0.98), p = 0.032
• CR duration 38.6 vs 12.4 mo
• Continuation up to 3 y, incl. post-HSCT
• Cardiac signal: QTcF >500 ms in 2%, two VF arrests

Discontinuation for adverse events 20% vs 9% (dose reductions for QT in 4%). Relapse by 24 months 31.2% vs 43.3%. Thirty-day mortality 6% vs 3%, driven by infections. Post-hoc age subgroups: HR 0.68 below 60, 0.91 at 60 or older. Quizartinib also tripled the depth of FLT3-ITD clearance at response (median VAF 0.01% vs 0.03%).

Midostaurin or quizartinib?

Never compared head to head, and the trials enrolled different populations, so cross-trial numbers deserve caution.

RATIFY

• Ages 18–59
• ITD and TKD
• Maintenance 12 cycles, stopped at transplant

QuANTUM-First

• Ages 18–75
• ITD only
• Continuation up to 3 years, transplant included

TKD disease points to midostaurin; age and the cardiac profile shape the ITD choice.

QuANTUM-First post-hoc age analysis: HR 0.68 below age 60, 0.91 at 60 or older; underpowered, hypothesis-generating. RATIFY enrolled to age 59, so older ITD patients have randomized evidence only for quizartinib.

Transplant in first remission

• In RATIFY, 28.1% vs 22.7% of patients were transplanted in CR1, and 57% at some point in the disease course
• Both trials censored survival at transplant as a sensitivity analysis, and the inhibitor’s benefit persisted (QuANTUM-First HR 0.75)
• Exposure to the inhibitor before transplant was brief in RATIFY, so the authors attribute much of the benefit to early reduction of disease burden

RATIFY censored analysis: 24.3% lower risk of death, 4-year OS 63.7% vs 55.7%, p = 0.08. These analyses argue the drug helps beyond simply delivering more patients to transplant. Post-transplant maintenance questions are taken up by dedicated trials (MORPHO, listed in the ELN-DAVID evidence table).

Unfit for intensive chemotherapy

Venetoclax regimens meet FLT3 resistance

Azacitidine plus venetoclax is the standard for patients unfit for intensive chemotherapy, and FLT3 mutations are a common route of resistance to it: FLT3 signaling drives BCL-xL and MCL-1, which bypass BCL-2 blockade.

With azacitidine plus venetoclax, median OS was 9.9 months in FLT3-ITD disease versus 14.7 months in wild type.

A frontline triplet for unfit patients

Aza + ven + gilteritinib 80 mg. Phase 1/2, newly diagnosed unfit, median age 71.

• No deaths in the first 60 days
• 18-mo OS 72%, RFS 71%
• 43% to HSCT in first remission

Thirty patients at one center, without randomization or approval; grade ≥3 infection 62%. Use within a trial when possible.

Cycle 1: azacitidine days 1–7, venetoclax 400 mg days 1–28, gilteritinib days 1–28, triazole prophylaxis. From cycle 2, azacitidine days 1–5 and venetoclax days 1–7 to limit myelosuppression; marrow held drugs if blasts <5% or aplastic. R/R cohort (n = 22): CR/CRi 27%, mCRc 73%.

Measurable residual disease

Residual FLT3-ITD predicts relapse

Persistent FLT3-ITD in morphologic remission after intensive therapy is associated with higher relapse and death across cohorts totaling thousands of patients (PreMEASURE, HOVON-SAKK, QuANTUM-First, MORPHO, and others).

• NGS-based FLT3-ITD MRD testing is now recommended by ELN and NCCN guidance, for both prognostic and predictive use
• In QuANTUM-First, patients in remission with ITD MRD below 10⁻⁴ had better survival regardless of arm (HR 0.57), and quizartinib tripled the depth of ITD clearance

What a reliable assay requires

• PCR-NGS on ideally 700 ng of DNA; diagnostic assays lack MRD sensitivity
• Limit of detection at least 10⁻⁴ (0.01%); relapse risk is elevated even for positivity below that threshold
• Detection across the full ITD spectrum, 3 to >200 bp; very long ITDs risk false negatives on short-read strategies
• ITDs called de novo, with every ITD and its allele fraction reported at each time point
• Results back within 3 weeks

Interpreting the result at the bedside

• Marrow carries a 5–10× higher ITD burden than blood; on the same assay, pretransplant MRD positivity was 46% in marrow vs 33% in blood
• Pre-transplant CR1 is the best-studied time point for acting on the result
• FLT3-ITD should not be the sole surveillance marker: with comutated NPM1, mutant NPM1 monitoring can guide, and serial flow may help during the highest-risk period
• Its role during less-intensive or maintenance therapy is still undefined

Relapsed or refractory disease

Gilteritinib at relapse (ADMIRAL)

Phase 3, 371 patients, gilteritinib 120 mg vs preselected salvage chemotherapy.

• Median OS 9.3 vs 5.6 mo; HR 0.64 (0.49–0.83), p < 0.001
• Benefit held censored at transplant (HR 0.58)
• Active in ITD, TKD, and after a prior FLT3 inhibitor

Median age 62 (range 19–85); 12% had received a prior FLT3 inhibitor (subgroup HR 0.70). With exposure adjustment, grade ≥3 and serious AEs were less frequent than with chemotherapy; most common grade ≥3 events were febrile neutropenia (45.9%), anemia (40.7%), thrombocytopenia (22.8%).

The three randomized trials at a glance

	RATIFY	QuANTUM-First	ADMIRAL
Drug	Midostaurin	Quizartinib	Gilteritinib
Setting	Newly diagnosed, + 7+3	Newly diagnosed ITD, + 7+3	R/R, monotherapy
Ages	18–59	18–75	≥18 (median 62)
FLT3 covered	ITD + TKD	ITD only	ITD + TKD
N	717	539	371
Median OS	74.7 vs 25.6 mo	31.9 vs 15.1 mo	9.3 vs 5.6 mo
HR for death	0.78 (0.63–0.96)	0.78 (0.62–0.98)	0.64 (0.49–0.83)
Post-HSCT drug	No (stopped at HSCT)	Yes (continuation ≤3 y)	Resumption 30–90 d post-HSCT

Populations, controls, and eras differ; the OS columns are within-trial comparisons, never across columns. RATIFY’s median OS gap is inflated by KM inflection points.

Takeaways

• Know both ITD and TKD status from diagnosis; the subtype picks the drug
• Fit: 7+3 plus a FLT3 inhibitor; quizartinib for ITD to age 75, midostaurin for TKD
• Unfit: aza + ven alone underperforms; the gilteritinib triplet is promising, still investigational
• Order NGS FLT3-ITD MRD (marrow, LOD ≥10⁻⁴), above all pre-transplant in CR1
• At relapse, gilteritinib beat salvage chemotherapy in ITD and TKD alike

References

Erba, Harry P., Pau Montesinos, Hee-Je Kim, et al. 2023. “Quizartinib Plus Chemotherapy in Newly Diagnosed Patients with FLT3-Internal-Tandem-Duplication-Positive Acute Myeloid Leukaemia (QuANTUM-First): A Randomised, Double-Blind, Placebo-Controlled, Phase 3 Trial.” The Lancet 401 (10388): 1571–83. https://doi.org/10.1016/S0140-6736(23)00464-6.

Hourigan, Christopher S., Lisanne Beugelink, Jad Othman, et al. 2026. “ELN-DAVID Recommendations for NGS-Based FLT3-ITD MRD Testing for Patients with Acute Myeloid Leukemia.” Blood 148 (7): 910–13. https://doi.org/10.1182/blood.2026033569.

Perl, Alexander E., Giovanni Martinelli, Jorge E. Cortes, et al. 2019. “Gilteritinib or Chemotherapy for Relapsed or Refractory FLT3-Mutated AML.” New England Journal of Medicine 381 (18): 1728–40. https://doi.org/10.1056/NEJMoa1902688.

Short, Nicholas J., Naval Daver, Courtney D. DiNardo, et al. 2024. “Azacitidine, Venetoclax, and Gilteritinib in Newly Diagnosed and Relapsed or Refractory FLT3-Mutated AML.” Journal of Clinical Oncology, ahead of print. https://doi.org/10.1200/JCO.23.01911.

Stone, Richard M., Sumithra J. Mandrekar, Ben L. Sanford, et al. 2017. “Midostaurin Plus Chemotherapy for Acute Myeloid Leukemia with a FLT3 Mutation.” New England Journal of Medicine 377 (5): 454–64. https://doi.org/10.1056/NEJMoa1614359.

Figures in this deck were redrawn from published summary statistics, never from patient-level data, and are for teaching. Verify against the primary sources before clinical use. This deck is not medical advice.