Skip to content
Bending the Curve

Blood and bone marrow

Acute myeloid leukaemia (including APL)

How AML treatment moved from cytarabine and anthracycline to risk-adapted, targeted and lower-intensity therapy, with APL as its own path.

5-year relative survival · United States, SEER 8 registries

5.45%→33.17%

1975 (observed value; the table column is headed 'Year') to 2018 (observed value; the table column is headed 'Year'). Read these with the caveats under “Then and now”.Source

24 milestones, by decade

  • The abstract of a 2008 review in Blood by Wang and Chen says that acute promyelocytic leukemia (APL) seems to be the most malignant form of acute leukemia, with a fatal course of only weeks, and summarises complete remission rates of 75% to 80% with chemotherapy in newly diagnosed patients and 90% to 95% with ATRA-based regimens, with 5-year disease-free survival of 74%. Population data on early death differ by source: a SEER-based study of 1,400 people diagnosed with APL in 1992 to 2007 reported an overall early death rate of 17.3% (Park 2011), whereas the HARMONY analysis of 1,438 people diagnosed in 1999 to 2022, pooled from two trials and four European registries, reported early death in 85 (5.9%) and 7-year overall survival of 91% with ATRA plus arsenic trioxide versus 81% with idarubicin-based chemotherapy (hazard ratio 2.14, 95% CI 1.51 to 3.05; the abstract does not state the direction of the ratio). The HARMONY comparison is a pooled cohort analysis, and its abstract does not describe randomisation to those regimens. The review's figures are its authors' summary of the literature; the two early-death figures come from different countries, periods and data sources, and the abstracts do not give the definition of early death, so whether the definitions match was not confirmed and the two rates are not compared.
  • On the SEER Cancer Stat Facts page for AML, observed five-year relative survival in the SEER 8 series is 5.45% for 1975, 17.54% for 1995, 37.29% for 2015 and 33.17% for 2018 (the page does not say whether the year is the year of diagnosis), while the US age-adjusted death rate is 2.54 per 100,000 in 1975, 2.13 in 1988, 2.87 in 2009 and 2.66 in 2024. The headline survival for SEER 21 (excluding IL) in 2016 to 2022 is 33.4%, a separate series. Survival after diagnosis and population death rates answer different questions and use different data; the page gives no reason why survival rose while the death rate stayed within 2.13 to 2.87, and none is offered here.
  • Gemtuzumab ozogamicin, an antibody-drug conjugate directed at CD33, was approved by the US FDA on 17 May 2000 under the accelerated approval regulations for CD33-positive AML in first relapse in people aged 60 years or older who were not considered candidates for other cytotoxic chemotherapy. A Federal Register notice records that on 21 May 2010 the FDA asked the sponsor to withdraw it voluntarily, after the required postapproval trial failed to verify clinical benefit and raised new safety concerns, and that approval was withdrawn effective 28 November 2011. The FDA then approved a new biologics licence application (BLA 761060) on 1 September 2017 for newly diagnosed CD33-positive AML in adults and, as a separate indication, relapsed or refractory CD33-positive AML in adults and in pediatric patients 2 years and older. The sources read give regulatory actions only; the trial results behind each step are in the milestone timeline.

Evidence current to October 2026 · Checked against primary sources to the extent possible

A guided tour of 12 milestones. You can pause or step through it yourself.

Filter milestonesshowing all 24
Treatment type
Setting

Chemotherapy

  1. 1957 · Origin · UnderstandingEarly reports of acute promyelocytic leukemia, 1957 and 1959Papers titled "Acute promyelocytic leukemia" appeared in Acta Medica Scandinavica in 1957 (Hillestad) and, in French, in the Schweizerische Medizinische Wochenschrift in 1959, where the title of Bernard and colleagues refers to a study of 20 cases. Only the titles are available, so what the papers reported, and who first described the disease, is not settled by these records.Hillestad, Acta Med Scand 1957, "Acute promyelocytic leukemia" (PubMed: no abstract; content not read). Bernard and colleagues, Schweiz Med Wochenschr 1959 (French; the English form of the title is a study made on 20 cases; no abstract; content not read). No clinical details are recorded. Wang and Chen (2008) describe APL as having a severe bleeding tendency and a fatal course of only weeks.

Molecular origins

  1. 1976 · Understanding · UnderstandingFAB proposals classify the acute leukaemias by morphologyIn 1976 seven haematologists from France, the United States and Britain examined about 200 cases of acute leukaemia under the microscope and proposed a common naming system. They divided acute leukaemia into lymphoblastic and myeloid types, with three and six subgroups, and described conditions that can be mistaken for acute myeloid leukaemia.Bennett 1976 (FAB co-operative group): seven haematologists reviewed peripheral blood and bone marrow films from some 200 acute leukaemia cases, independently and then together, using conventional morphological and cytochemical methods. They proposed lymphoblastic and myeloid groups, subdivided into three and six subgroups, and considered dysmyelopoietic syndromes that may be confused with AML. No outcome data are reported in the abstract.
  2. 1977 · Understanding · Understanding⚑ needs reviewt(15;17) and the PML-RARA fusion in acute promyelocytic leukemiaIn 1977 a short Lancet report titled "15/17 translocation, a consistent chromosomal change in acute promyelocytic leukaemia" reported the chromosome change. In 1990 and 1991 laboratories showed that the swap fuses the retinoic acid receptor alpha gene on chromosome 17 to a gene on chromosome 15, later named PML.Rowley, Golomb and Dougherty (Lancet 1977; indexed as a letter; no abstract available, title only): "15/17 translocation, a consistent chromosomal change" in APL. Borrow 1990: breakpoints in ten APL cases clustered in a 12 kb region within the first intron of RARA; the abstract states the translocation is present in almost 100 percent of cases. de Thé 1990: RARA and/or the chromosome 15 locus rearranged in 6 of 8 patients. Kakizuka and de Thé 1991 described the PML-RARA fusion.
  3. 1979 · Treatment · Cell therapy & transplant⚑ needs reviewAllogeneic transplant in first remission of AMLA 1979 paper reported marrow transplantation for acute non-lymphoblastic leukaemia in first remission. A 1995 European trial found the fewest relapses but the highest mortality after allogeneic transplant, with similar overall survival. A 2009 pooled analysis found a survival benefit of transplant for intermediate- and poor-risk disease, but not for good-risk disease.Zittoun 1995 (EORTC/GIMEMA; 623 in remission): allogeneic transplant assigned by sibling donor, others randomised to autologous transplant or intensive chemotherapy; projected 4-year DFS 55%, 48% and 30%, overall survival similar. Koreth 2009 meta-analysis (24 trials, 6007 patients), donor versus no donor: RFS HR 0.80 (95% CI 0.74-0.86), OS HR 0.90 (0.82-0.97); OS HR good-risk 1.07 (0.83-1.38), intermediate 0.83, poor 0.73. Thomas 1979: title only.
  4. 1981 · Treatment · Chemotherapy⚑ needs reviewCytarabine with an anthracycline: CALGB trials in AML, 1981-82In 1981 and 1982 the Cancer and Leukemia Group B reported randomised treatment of patients with acute myeloid leukaemia. The 1982 report gave cytarabine as a seven-day infusion with daunorubicin or doxorubicin on three days. Remission rates depended on the drug, its dose and age: about seven in ten for people under 60 on the best arm.Yates 1982 (CALGB): randomised daunorubicin 30 or 45 mg/m2 or adriamycin 30 mg/m2 on days 1-3 with cytarabine 100 mg/m2/day, 7-day infusion. Complete remission, age under 60: 72% (daunorubicin 45), 59% (daunorubicin 30), 58% (adriamycin 30). Age over 60: 47% (daunorubicin 30), 31% (daunorubicin 45), 35% (adriamycin 30). Induction mortality shifted correspondingly. Rai 1981 (CALGB, Blood) has no PubMed abstract.
  5. 1988 · Treatment · TargetedAll-trans retinoic acid induces complete remission in APLIn 1988 a paper in Blood described giving all-trans retinoic acid, a vitamin A derivative, to 24 people with acute promyelocytic leukemia. All 24 reached complete remission, and in laboratory cultures the leukemic cells of most patients tested matured into granulocytes. Eight later relapsed, after two to five months of remission. Two more series followed in 1990 and 1991.Uncontrolled series, abstracts only. Huang 1988: ATRA in 24 patients (16 untreated, 8 nonresponsive or resistant to prior chemotherapy), CR in all without marrow hypoplasia (one patient resistant to ATRA in vitro reached CR after low-dose cytarabine was added); 8 relapsed after 2 to 5 months of remission. Castaigne 1990: 22 patients, 14 CR, four transient responses, one failure, three early deaths. Warrell 1991: 11 patients given oral tretinoin, 9 CR, response linked to maturation of the leukemic clone.
  6. 1993 · Treatment · TargetedATRA in randomized trials for newly diagnosed APL, 1993-1997Randomized trials in 1993 and 1997 tested retinoic acid in newly diagnosed acute promyelocytic leukemia. In Europe, about 79 in 100 given retinoic acid with chemotherapy had no event after a year, against 50 in 100 on chemotherapy alone. In North America, 67 in 100 given retinoic acid first were alive at three years, against 50 in 100 given chemotherapy.European APL 91 (101 randomized when stopped early after the first interim analysis): 12-month EFS 79% vs 50% (P=.001); CR 91% vs 81%, difference not significant. North American Intergroup (346 randomized to induction): CR 72% vs 69% (P=0.56); ITT overall survival at 3 years 67% vs 50% (P=0.003). Long-term report (350 patients): 5-year DFS 69% vs 29%. Different eras, populations and endpoints.
  7. 1994 · Treatment · ChemotherapyHigh-dose cytarabine consolidation: CALGB randomised trial, 1994In a 1994 trial of 1,088 adults with AML, people in remission were randomly given one of three doses of cytarabine. Among those aged 60 or younger, about four in ten on the highest dose were still in remission after four years, against about one in four on the lowest dose. Over 60, 16% or less stayed disease-free.Mayer 1994 (CALGB): 1088 adults given daunorubicin and cytarabine; 596 of 693 in CR randomised to four courses of cytarabine 100 mg/m2, 400 mg/m2 or 3 g/m2 every 12 h. Median follow-up 52 months. DFS differed by dose (P=0.003); HR versus 100 mg/m2: 0.67 (95% CI 0.53-0.86) for 3 g, 0.75 (0.60-0.94) for 400 mg. Age 60 or younger: 4-year continuous CR 24%, 29%, 44% (P=0.002); older: 16% or less in each group.
  8. 1997 · Treatment · TargetedArsenic trioxide induces remission in relapsed APLIn 1997 a report in Blood described arsenic trioxide, given into a vein, in 15 people whose acute promyelocytic leukemia had returned after retinoic acid and chemotherapy. Nine of the ten given arsenic alone reached complete remission. The FDA approved the drug in 2000; a US multicenter study of 40 people, published in 2001, reported complete remission in 85%.Shen 1997: 15 relapsed patients, As2O3 10 mg/d IV; clinical CR in 9 of 10 given As2O3 alone (the abstract's wording on the five given combinations is ambiguous). Soignet 1998: 12 relapsed patients, 11 CR. Soignet 2001, US multicenter series: 40 patients, CR 85% (34 of 40; 70% by the stricter definition in the FDA label), 18-month OS 66% and RFS 56%, QT prolongation 63%. FDA approval 25 September 2000; EU authorisation 5 March 2002 under exceptional circumstances.

Targeted therapy

  1. 1998 · Understanding · UnderstandingDiagnostic cytogenetics defines AML prognostic groupsIn 1998 a British trial of 1,612 people with AML reported that chromosome changes found at diagnosis define three groups with different outlooks. A 2002 US study of 1,213 adults also defined three risk groups: five-year survival was about one in two, one in four and one in twenty across the three groups.Grimwade 1998 (MRC AML10, n=1612, up to age 55 including children): response, relapse risk and OS defined favourable, intermediate and adverse cytogenetic groups, which also determined outcome of autologous or allogeneic transplant in CR1. Byrd 2002 (CALGB 8461, 1213 adults, de novo AML): favourable, intermediate, adverse: CR 88%, 67%, 32%; 5-year CIR 54%, 67%, 92%; 5-year OS 55%, 24%, 5%.
  2. 2009 · Treatment · ChemotherapyAzacitidine and decitabine in older adults with AMLAzacitidine and decitabine were tested against conventional care in older adults with AML. A 2009 analysis in people with 20 to 30% blasts reported median survival of 24.5 months with azacitidine versus 16.0 months. A later trial with over 30% blasts found 10.4 versus 6.5 months, not statistically significant in its primary comparison.Three randomized comparisons (the first a subgroup of a trial designed in MDS). AZA-001 AML subgroup (n=113, 20-30% blasts): median OS 24.5 vs 16.0 months (HR 0.47, 95% CI 0.28-0.79; P=.005). DACO-016 (decitabine, n=485): median OS 7.7 vs 5.0 months (HR 0.85, 95% CI 0.69-1.04; P=.108). AZA-AML-001 (n=488, >30% blasts): median OS 10.4 vs 6.5 months (HR 0.85, 95% CI 0.69-1.03; stratified log-rank P=.1009).
  3. 2009 · Understanding · UnderstandingELN risk and WHO/ICC classification of AML (2009-2022)Since 2009 international expert panels have repeatedly updated how acute myeloid leukaemia is diagnosed and how its genetic risk is described. European LeukemiaNet recommendations appeared in 2009, 2016 and 2022. In 2022 two separate classifications of myeloid cancers were published, one from the World Health Organization and one from an international consensus group.Consensus documents, no trial data. ELN 2010 (online 2009): updated recommendations on diagnosis and management, with a standardised system for reporting cytogenetic and molecular data. ELN 2017: revised genetic categories, MRD-based response category, criteria for progressive disease. ELN 2022: revised genetic risk classification, response criteria and treatment recommendations. WHO 2016 revised the 2008 classification. In 2022 the ICC and the WHO 5th edition were published as separate classifications.
  4. 2010 · Reversal · Antibody–drug conjugateGemtuzumab ozogamicin: approved 2000, withdrawn 2010, back 2017Gemtuzumab ozogamicin joins an antibody to a cell-killing drug. It was approved in the United States in 2000 for older people with relapsed AML. In 2010, at the FDA's request, the sponsor moved to withdraw it after a required trial found no survival gain when added to chemotherapy. A new licence followed in 2017 for newly diagnosed and relapsed disease.FDA accelerated approval 17 May 2000 (first-relapse CD33-positive AML, age 60 or older; three phase 2 trials, 142 patients). SWOG S0106 (637 registered; daunorubicin-cytarabine with or without GO): CR 69% v 70% (P=.59); 5-year OS 46% v 50% (P=.85). FDA asked for withdrawal 21 May 2010. ALFA-0701 (n=280, age 50-70, fractionated 3 mg/m2): 2-year EFS 40.8% v 17.1% (HR 0.58, 95% CI 0.43-0.78). New BLA approved 1 Sep 2017.

Immunotherapy

  1. 2011 · Understanding · UnderstandingEarly death in APL: population data and pooled trial cohortsA 2011 analysis of US cancer-registry data found that about 17 in 100 people diagnosed with acute promyelocytic leukemia between 1992 and 2007 died early, a rate the authors said appeared higher than commonly reported in clinical trials. A 2025 analysis combining two trials and four European registries reported early death in about 6 in 100.Park 2011 (SEER, 1992-2007, 1400 patients): early death 17.3% overall and 24.2% at age 55 or older, with only a modest change over time; 3-year survival improved from 54.6% (1992-1995) to 70.1% (2002-2007), 46.4% at age 55 or older. Voso 2025 (HARMONY, trials plus registries, 1999-2022): early death 85 of 1438 (5.9%) at a median of 9 days; associated with age (OR 1.06, CI 1.04-1.08) and high Sanz risk score (OR 4.65, CI 2.55-8.51).
  2. 2013 · Treatment · TargetedAPL0406: ATRA plus arsenic trioxide versus ATRA plus chemotherapyA trial of 276 adults with newly diagnosed, lower-risk acute promyelocytic leukemia compared retinoic acid plus arsenic trioxide with retinoic acid plus chemotherapy. At about four years, about 97 in 100 on arsenic had no event such as relapse or death, against about 80 in 100 on chemotherapy. The EU approved the combination in 2016, the US in 2018.APL0406: phase 3, open-label noninferiority trial in adults aged 18 to 71 with low- or intermediate-risk APL (WBC ≤10 × 10^9 per liter); ATRA plus arsenic trioxide versus ATRA plus idarubicin-based chemotherapy. Primary report: 2-year EFS 97% vs 86% (P<0.001 noninferiority, P=0.02 superiority). Final analysis, 276 randomized, median follow-up 40.6 months: 50-month EFS 97.3% vs 80%, cumulative incidence of relapse 1.9% vs 13.9%, OS 99.2% vs 92.6%.
  3. 2016 · Understanding · UnderstandingMeasurable residual disease in AML: prognostic value and monitoringAfter treatment, sensitive tests can find leukaemia cells that a microscope misses; this is called measurable residual disease. Studies from 2016 linked it to a higher chance of relapse and shorter survival. A 2025 randomised trial of repeated testing during treatment found no overall survival gain, though one subgroup did better.Ivey 2016 (AML17, NPM1-mutated, n=346): persistent NPM1 transcripts after cycle 2 in 15%; 3-year relapse 82% v 30% (HR 4.80, 95% CI 2.95-7.80). Jongen-Lavrencic 2018 (n=482): non-DTA mutations in remission, 4-year relapse 55.4% v 31.9%. Short 2020 meta-analysis (81 publications, 11,151 patients): MRD negativity OS HR 0.36. Potter 2025 (randomised, n=637): monitoring v standard care, OS HR 1.11 (0.83-1.49); NPM1 with FLT3-ITD subgroup HR 0.53 (0.31-0.91).
  4. 2017 · Treatment · TargetedIDH1 and IDH2 inhibitors in AML: enasidenib, ivosidenib, olutasidenibFrom 2017 the US approved oral drugs aimed at IDH1 or IDH2 mutations in AML, beginning with enasidenib. In the AGILE trial, adults with an IDH1 mutation who could not have intensive chemotherapy had a median survival of about 24 months with ivosidenib plus azacitidine, against about 8 months with placebo plus azacitidine.Single-arm studies of enasidenib (IDH2; ORR 40.3%, median OS 9.3 months) and ivosidenib (IDH1; CR+CRh 30.4%, 95% CI 22.5-39.3, n=125) in relapsed or refractory AML preceded FDA approvals of each. AGILE: randomized phase 3, IDH1-mutated newly diagnosed AML ineligible for intensive chemotherapy (n=146): EFS HR 0.33 (95% CI 0.16-0.69; P=0.002); median OS 24.0 vs 7.9 months (HR 0.44, 95% CI 0.27-0.73; P=0.001).
  5. 2017 · Treatment · TargetedRATIFY: midostaurin added to chemotherapy in FLT3-mutated AMLIn 2017 the US approved midostaurin, a multitargeted kinase inhibitor, to add to standard chemotherapy for adults with newly diagnosed AML with a FLT3 mutation. In the RATIFY trial of people aged 18 to 59, adding it was associated with about a 22% lower risk of death than adding a placebo.RATIFY: randomized, placebo-controlled phase 3 in FLT3-mutated AML aged 18-59 (n=717; 360 midostaurin, 357 placebo); daunorubicin-cytarabine induction, high-dose cytarabine consolidation, then midostaurin or placebo maintenance; transplantation allowed. OS HR 0.78 (95% CI 0.63-0.96; one-sided P=0.009); EFS HR 0.78 (0.66-0.93; one-sided P=0.002). Censoring at transplantation: OS difference not significant (P=0.08).
  6. 2017 · Treatment · ChemotherapyCPX-351 versus 7+3 in older adults with high-risk or secondary AMLCPX-351 packs two chemotherapy drugs, daunorubicin and cytarabine, into a fat-based particle at a fixed ratio. In a trial of 309 people aged 60 to 75 with high-risk or secondary AML, median survival was about ten months with CPX-351 and about six months with standard chemotherapy. The US approved it in August 2017.Lancet JE 2018 (J Clin Oncol), open-label phase 3, 309 patients aged 60-75, newly diagnosed high-risk or secondary AML: median OS 9.56 v 5.95 months, CPX-351 v 7+3 (HR 0.69, 95% CI 0.52-0.90; one-sided P=.003); overall remission rate 47.7% v 33.3% (P=.016). Five-year follow-up (Lancet Haematol 2021): median OS 9.33 v 5.95 months (HR 0.70, 0.55-0.91); 5-year OS 18% v 8%. FDA approval 3 Aug 2017.
  7. 2018 · Treatment · TargetedADMIRAL: gilteritinib versus salvage chemotherapy, FLT3-mutated AMLIn 2018 the US approved gilteritinib, a FLT3 inhibitor taken by mouth, for AML that had come back or not responded to treatment and carried a FLT3 mutation. In the ADMIRAL trial, median survival was about 9 months with gilteritinib and about 6 months with salvage chemotherapy.ADMIRAL: randomized (2:1) phase 3 of gilteritinib 120 mg/day vs salvage chemotherapy in relapsed or refractory FLT3-mutated AML (371 eligible; 247 vs 124). Median OS 9.3 vs 5.6 months (HR 0.64, 95% CI 0.49-0.83; P<0.001); CR/CRh 34.0% vs 15.3%; CR 21.1% vs 10.5%; median EFS 2.8 vs 0.7 months (HR 0.79, 95% CI 0.58-1.09).
  8. 2018 · Treatment · TargetedVenetoclax with azacitidine for unfit AML (VIALE-A)In 2018 the US approved venetoclax, a tablet, together with azacitidine, decitabine or low-dose cytarabine for adults with newly diagnosed AML who could not have intensive chemotherapy. In the later VIALE-A trial, people given venetoclax with azacitidine lived a median of about 15 months, compared with about 10 months on azacitidine alone.VIALE-A: randomized, double-blind phase 3, AML ineligible for intensive induction (age 75 or older or comorbidities); ITT 431 (286 azacitidine-venetoclax, 145 azacitidine-placebo). Median OS 14.7 vs 9.6 months (HR 0.66, 95% CI 0.52-0.85; P<0.001), median follow-up 20.5 months; CR 36.7% vs 17.9%; CR+CRi 66.4% vs 28.3%. VIALE-C (venetoclax plus LDAC, n=211): planned OS analysis HR 0.75 (95% CI 0.52-1.07; P=.11).

Precision & engineered

  1. 2020 · Treatment · ChemotherapyQUAZAR: oral azacitidine maintenance after first remissionIn 2020 the US approved azacitidine tablets to continue treatment for adults with AML in a first remission after intensive chemotherapy who cannot complete intensive curative therapy. In the QUAZAR trial, median survival was about 25 months with the tablets and about 15 months with placebo.QUAZAR AML-001: randomized, double-blind, placebo-controlled phase 3 of oral azacitidine (CC-486 300 mg, 14 days per 28-day cycle) vs placebo in AML in first remission, aged 55 or older, not transplant candidates (n=472; 238 vs 234). Median OS from randomization 24.7 vs 14.8 months (P<0.001); median RFS 10.2 vs 4.8 months (P<0.001). CC-486 is not bioequivalent to injectable azacitidine.
  2. 2023 · Treatment · TargetedMenin inhibitors: revumenib and ziftomenib in KMT2A or NPM1 leukemiaMenin inhibitors are oral drugs for leukemias with KMT2A changes or NPM1 mutations. In single-arm studies of heavily pretreated people, about one in four reached a complete remission, sometimes with partial blood-count recovery. The US approved revumenib in 2024 and 2025 and ziftomenib in 2025 for relapsed or refractory disease.Single-arm studies, no comparator. Revumenib (AUGMENT-101 phase 2): KMT2A-rearranged acute leukemia, CR+CRh 22.8% (95% CI 12.7-35.8; n=57 evaluable) and NPM1-mutated AML, 23.4% (n=64), each tested against a 10% null (P=.0036; 1-sided P=.0014). Ziftomenib (KOMET-001 phase 2, NPM1-mutated, n=92): CR/CRh 22% (95% CI 14-32; P=.0058 against a 12% historical benchmark), median OS 6.6 months (95% CI 3.6-8.6).
  3. 2023 · Treatment · Targeted⚑ needs reviewQuANTUM-First: quizartinib with chemotherapy in FLT3-ITD AMLIn 2023 the US approved quizartinib, added to chemotherapy and then continued alone, for adults with newly diagnosed AML carrying a FLT3-ITD mutation. In the QuANTUM-First trial, median survival was about 32 months with quizartinib and about 15 months with placebo.QuANTUM-First: randomised, double-blind, placebo-controlled phase 3 in FLT3-ITD-positive newly diagnosed AML aged 18-75 (n=539; 268 quizartinib, 271 placebo), with 7 + 3 induction, consolidation and up to 3 years' continuation. Median OS 31.9 months (95% CI 21.0-not estimable) vs 15.1 months (13.2-26.2); HR 0.78 (95% CI 0.62-0.98; p=0.032); median follow-up 39.2 months.

Then and now

  • 5-year relative survival · United States, SEER 8 registries

    • 5.45%

      1975 (observed value; the table column is headed 'Year')

    People with acute myeloid leukemia, all races, both sexes (the chart note reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'); SEER 8 observed five-year relative survival by calendar year, on the SEER Cancer Stat Facts AML page. The page states no age range or subtype mix for this series, does not mention acute promyelocytic leukemia (APL) or give a separate APL series, and says there is no standard staging system for leukemia.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modeled trend value in the adjacent column (7.36% for 1975). The column is headed 'Year' and the chart note does not say that the year is the year of diagnosis. SEER 8 covers a limited set of registries and single-year values fluctuate (in this series 30.46% in 2014, 37.29% in 2015 and 31.53% in 2016). The page does not say how case definition, classification or the mix of subtypes and ages changed over 1975-2018, and it attributes the change in this series to no cause. Read beside the US age-adjusted death rate on the same page, which was 2.54 per 100,000 in 1975, 2.13 in 1988, 2.87 in 2009 and 2.66 in 2024: survival after diagnosis rose while the death rate ended close to where it began, the page does not explain the gap, and the two series use different data (SEER 8 registries versus US mortality data), so they are not subtracted. Compare only with other values from this same SEER 8 series, not with the SEER 21 value.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 12.84%

      1985 (observed value; the table column is headed 'Year')

    People with acute myeloid leukemia, all races, both sexes (the chart note reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'); SEER 8 observed five-year relative survival by calendar year, on the SEER Cancer Stat Facts AML page. The page states no age range or subtype mix for this series, does not mention acute promyelocytic leukemia (APL) or give a separate APL series, and says there is no standard staging system for leukemia.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modeled trend value in the adjacent column (11.90% for 1985). The column is headed 'Year' and the chart note does not say that the year is the year of diagnosis. SEER 8 covers a limited set of registries and single-year values fluctuate (in this series 30.46% in 2014, 37.29% in 2015 and 31.53% in 2016). The page does not say how case definition, classification or the mix of subtypes and ages changed over 1975-2018, and it attributes the change in this series to no cause. Read beside the US age-adjusted death rate on the same page, which was 2.54 per 100,000 in 1975, 2.13 in 1988, 2.87 in 2009 and 2.66 in 2024: survival after diagnosis rose while the death rate ended close to where it began, the page does not explain the gap, and the two series use different data (SEER 8 registries versus US mortality data), so they are not subtracted. Compare only with other values from this same SEER 8 series, not with the SEER 21 value.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 17.54%

      1995 (observed value; the table column is headed 'Year')

    People with acute myeloid leukemia, all races, both sexes (the chart note reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'); SEER 8 observed five-year relative survival by calendar year, on the SEER Cancer Stat Facts AML page. The page states no age range or subtype mix for this series, does not mention acute promyelocytic leukemia (APL) or give a separate APL series, and says there is no standard staging system for leukemia.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modeled trend value in the adjacent column (17.61% for 1995). The column is headed 'Year' and the chart note does not say that the year is the year of diagnosis. SEER 8 covers a limited set of registries and single-year values fluctuate (in this series 30.46% in 2014, 37.29% in 2015 and 31.53% in 2016). The page does not say how case definition, classification or the mix of subtypes and ages changed over 1975-2018, and it attributes the change in this series to no cause. Read beside the US age-adjusted death rate on the same page, which was 2.54 per 100,000 in 1975, 2.13 in 1988, 2.87 in 2009 and 2.66 in 2024: survival after diagnosis rose while the death rate ended close to where it began, the page does not explain the gap, and the two series use different data (SEER 8 registries versus US mortality data), so they are not subtracted. Compare only with other values from this same SEER 8 series, not with the SEER 21 value.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 26.3%

      2005 (observed value; the table column is headed 'Year')

    People with acute myeloid leukemia, all races, both sexes (the chart note reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'); SEER 8 observed five-year relative survival by calendar year, on the SEER Cancer Stat Facts AML page. The page states no age range or subtype mix for this series, does not mention acute promyelocytic leukemia (APL) or give a separate APL series, and says there is no standard staging system for leukemia.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modeled trend value in the adjacent column (24.24% for 2005). The column is headed 'Year' and the chart note does not say that the year is the year of diagnosis. SEER 8 covers a limited set of registries and single-year values fluctuate (in this series 30.46% in 2014, 37.29% in 2015 and 31.53% in 2016). The page does not say how case definition, classification or the mix of subtypes and ages changed over 1975-2018, and it attributes the change in this series to no cause. Read beside the US age-adjusted death rate on the same page, which was 2.54 per 100,000 in 1975, 2.13 in 1988, 2.87 in 2009 and 2.66 in 2024: survival after diagnosis rose while the death rate ended close to where it began, the page does not explain the gap, and the two series use different data (SEER 8 registries versus US mortality data), so they are not subtracted. Compare only with other values from this same SEER 8 series, not with the SEER 21 value.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 37.29%

      2015 (observed value; the table column is headed 'Year')

    People with acute myeloid leukemia, all races, both sexes (the chart note reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'); SEER 8 observed five-year relative survival by calendar year, on the SEER Cancer Stat Facts AML page. The page states no age range or subtype mix for this series, does not mention acute promyelocytic leukemia (APL) or give a separate APL series, and says there is no standard staging system for leukemia.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modeled trend value in the adjacent column (31.47% for 2015). The column is headed 'Year' and the chart note does not say that the year is the year of diagnosis. SEER 8 covers a limited set of registries and single-year values fluctuate (in this series 30.46% in 2014, 37.29% in 2015 and 31.53% in 2016). The page does not say how case definition, classification or the mix of subtypes and ages changed over 1975-2018, and it attributes the change in this series to no cause. Read beside the US age-adjusted death rate on the same page, which was 2.54 per 100,000 in 1975, 2.13 in 1988, 2.87 in 2009 and 2.66 in 2024: survival after diagnosis rose while the death rate ended close to where it began, the page does not explain the gap, and the two series use different data (SEER 8 registries versus US mortality data), so they are not subtracted. Compare only with other values from this same SEER 8 series, not with the SEER 21 value. The 2015 observed value is well above its neighbours and above the modeled trend (31.47%), so it should not be read as the level reached by 2015.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 33.17%

      2018 (observed value; the table column is headed 'Year')

    People with acute myeloid leukemia, all races, both sexes (the chart note reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'); SEER 8 observed five-year relative survival by calendar year, on the SEER Cancer Stat Facts AML page. The page states no age range or subtype mix for this series, does not mention acute promyelocytic leukemia (APL) or give a separate APL series, and says there is no standard staging system for leukemia.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modeled trend value in the adjacent column (33.70% for 2018). The column is headed 'Year' and the chart note does not say that the year is the year of diagnosis. SEER 8 covers a limited set of registries and single-year values fluctuate (in this series 30.46% in 2014, 37.29% in 2015 and 31.53% in 2016). The page does not say how case definition, classification or the mix of subtypes and ages changed over 1975-2018, and it attributes the change in this series to no cause. Read beside the US age-adjusted death rate on the same page, which was 2.54 per 100,000 in 1975, 2.13 in 1988, 2.87 in 2009 and 2.66 in 2024: survival after diagnosis rose while the death rate ended close to where it began, the page does not explain the gap, and the two series use different data (SEER 8 registries versus US mortality data), so they are not subtracted. Compare only with other values from this same SEER 8 series, not with the SEER 21 value. 2018 is the last year with an observed value in the table (2019 onward show a dash, with modeled trend values only). It is lower than the 2015 observed value, so the series is not monotone and no smooth line should be drawn from 1975 to 2018 on these points alone.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 33.4%

      2016-2022 (headline figure as printed; the page does not say whether these are diagnosis years)

    People with acute myeloid leukemia, the page's headline five-year relative survival figure ('Based on data from SEER 21 (Excluding IL) 2016-2022'). The page does not state sex, race or age range for this figure; its incidence and age tables use SEER 21 2019-2023, all races, both sexes.

    A different registry set and period from the SEER 8 series above, so the two are separate series: do not splice them or read 33.4% against the SEER 8 2018 value (33.17%) as a trend. Relative survival excludes deaths from other causes. The page gives no age-specific, subtype-specific (including APL) or stage-specific survival, and reports a median age at diagnosis of 70 (SEER 21, 2019-2023, with 26.6% of new cases at 65-74 years and 24.9% at 75-84 years), so the figure describes a population mostly of older adults and is not a figure for younger adults or children.

    Source

  • Age-standardised mortality · United States

    • 2.54per 100,000

      1975 (observed value)

    Deaths from acute myeloid leukemia in the US, all races, both sexes, age-adjusted rate per 100,000 (the page's 'Death Rate - U.S.' column; the page states that deaths come from U.S. Mortality, not from the SEER registries).

    Observed yearly value (modeled trend value for 1975: 2.56). First year in the table. The page does not say how AML deaths are coded (for example whether APL deaths are counted) or how coding changed over 1975-2024. The page's headline death rate is 2.7 per 100,000 for 2020-2024 deaths and it states that age-adjusted death rates have fallen on average 0.9% each year over 2015-2024; use the yearly values or the headline, not both. AML is not a screened cancer and the page shows no stage distribution, so there is no screening-related reading of these numbers. Shown beside survival on purpose: SEER 8 observed five-year relative survival rose from 5.45% (1975) to 33.17% (2018) while the observed death rate stayed between 2.13 and 2.87 per 100,000 (1975-2024); the page does not explain the gap, and the older age of people with AML (median age at death 74, US 2020-2024) is one thing the page reports but does not connect to it.

    Source

  • Age-standardised mortality · United States

    • 2.13per 100,000

      1988 (observed value)

    Deaths from acute myeloid leukemia in the US, all races, both sexes, age-adjusted rate per 100,000 (the page's 'Death Rate - U.S.' column; the page states that deaths come from U.S. Mortality, not from the SEER registries).

    Observed yearly value (modeled trend value for 1988: 2.15). Lowest observed value in the table. The page does not say how AML deaths are coded (for example whether APL deaths are counted) or how coding changed over 1975-2024. The page's headline death rate is 2.7 per 100,000 for 2020-2024 deaths and it states that age-adjusted death rates have fallen on average 0.9% each year over 2015-2024; use the yearly values or the headline, not both. AML is not a screened cancer and the page shows no stage distribution, so there is no screening-related reading of these numbers. Shown beside survival on purpose: SEER 8 observed five-year relative survival rose from 5.45% (1975) to 33.17% (2018) while the observed death rate stayed between 2.13 and 2.87 per 100,000 (1975-2024); the page does not explain the gap, and the older age of people with AML (median age at death 74, US 2020-2024) is one thing the page reports but does not connect to it.

    Source

  • Age-standardised mortality · United States

    • 2.87per 100,000

      2009 (observed value)

    Deaths from acute myeloid leukemia in the US, all races, both sexes, age-adjusted rate per 100,000 (the page's 'Death Rate - U.S.' column; the page states that deaths come from U.S. Mortality, not from the SEER registries).

    Observed yearly value (modeled trend value for 2009: 2.80). Highest observed value in the table. The page does not say how AML deaths are coded (for example whether APL deaths are counted) or how coding changed over 1975-2024. The page's headline death rate is 2.7 per 100,000 for 2020-2024 deaths and it states that age-adjusted death rates have fallen on average 0.9% each year over 2015-2024; use the yearly values or the headline, not both. AML is not a screened cancer and the page shows no stage distribution, so there is no screening-related reading of these numbers. Shown beside survival on purpose: SEER 8 observed five-year relative survival rose from 5.45% (1975) to 33.17% (2018) while the observed death rate stayed between 2.13 and 2.87 per 100,000 (1975-2024); the page does not explain the gap, and the older age of people with AML (median age at death 74, US 2020-2024) is one thing the page reports but does not connect to it.

    Source

  • Age-standardised mortality · United States

    • 2.66per 100,000

      2024 (observed value)

    Deaths from acute myeloid leukemia in the US, all races, both sexes, age-adjusted rate per 100,000 (the page's 'Death Rate - U.S.' column; the page states that deaths come from U.S. Mortality, not from the SEER registries).

    Observed yearly value (modeled trend value for 2024: 2.61). Last year in the table. The page does not say how AML deaths are coded (for example whether APL deaths are counted) or how coding changed over 1975-2024. The page's headline death rate is 2.7 per 100,000 for 2020-2024 deaths and it states that age-adjusted death rates have fallen on average 0.9% each year over 2015-2024; use the yearly values or the headline, not both. AML is not a screened cancer and the page shows no stage distribution, so there is no screening-related reading of these numbers. Shown beside survival on purpose: SEER 8 observed five-year relative survival rose from 5.45% (1975) to 33.17% (2018) while the observed death rate stayed between 2.13 and 2.87 per 100,000 (1975-2024); the page does not explain the gap, and the older age of people with AML (median age at death 74, US 2020-2024) is one thing the page reports but does not connect to it.

    Source

Survival figures are affected by earlier detection (lead time) and overdiagnosis, and by changes in who is diagnosed. For screened cancers, read them alongside mortality.

How outcomes moved, trial by trial

Every randomised comparison on this page in one chart: survival in the control arm beside survival in the experimental arm, by year, with one curve per subtype or setting. Switch endpoint, hide subtypes, and open any trial.

Each trial is compared only with its own control arm.

Medians and rates are as published. The curves join different trials in different eras, populations and follow-up, so a line rising over time is a visual guide to the sequence of trials, not a pooled estimate or proof that one treatment beat another.

020406080OS, median, months2010201520202025AZA-AML-001 +3.9DACO-016 +2.7AGILE +16.1RATIFY +49.1ADMIRAL +3.7VIALE-A +5.1VIALE-C +3.1QUAZAR AML-001 +9.9QuANTUM-First +16.8

Hover or focus a pair of bars to see the trial (names show on hover when there are many). Hatched bar: control arm. Solid bar and marker: experimental arm. The number after the name is the difference in months.

  • control arm
  • experimental arm
  • curve through experimental values, per subtype
  • Grey: trials outside the curated evidence chains (no curve)
  • Bars start at zero
Overall survival, median, months, 12 trial comparisons from 2009 to 2023. Subtypes: FLT3-mutated AML, newly diagnosed: added to intensive chemotherapy (2); Newly diagnosed AML in older adults or those unsuited to intensive chemotherapy (5); FLT3-mutated AML, relapsed or refractory (1); Maintenance after first remission (1); Any setting: other randomised trials (3).
Why these trials are not directly comparable
  • Newly diagnosed APL: induction. The two event-free survival values are not on the same footing: the European APL91 estimate is an interim 12-month figure from 101 randomized patients, with chemotherapy alone as control, whereas APL0406 reports a 24-month figure in lower-risk disease (white-cell count 10 x 10^9 per litre or less), with ATRA plus chemotherapy as control. The abstracts do not define an event or give the analysis population for these figures, so the entries sit at different landmarks and are not drawn on one scale. The North American Intergroup trial reported overall and disease-free survival, not event-free survival, so it is listed in the lane without an entry. Compare only each trial's own arms.
  • Gemtuzumab ozogamicin added to intensive induction. ALFA-0701 and SWOG S0106 differ in age range, gemtuzumab ozogamicin dose and schedule (3 mg/m2 on days 1, 4 and 7 versus 6 mg/m2 once on day 4) and, in S0106, in the daunorubicin dose (45 mg/m2 with gemtuzumab ozogamicin against 60 mg/m2 in the comparison arm). ALFA-0701 reports overall survival at 2 years and S0106 at 5 years, so the values sit at different landmarks and are not drawn on one scale. Compare only each trial's own two arms.
  • FLT3-mutated AML, newly diagnosed: added to intensive chemotherapy. RATIFY enrolled adults aged 18 to 59 with any FLT3 mutation (ITD or TKD), whereas QuANTUM-First enrolled FLT3-ITD-positive disease only, aged 18 to 75. The drugs, chemotherapy details, rates of allogeneic transplant and follow-up differ. The midostaurin median has a confidence interval that reaches not reached and the quizartinib median an upper limit that is not estimable, so both medians are imprecise. P values in RATIFY are one-sided as the design specified. Compare only each trial's own two arms.
  • Newly diagnosed AML in older adults or those unsuited to intensive chemotherapy. The control arms differ: conventional care regimens preselected per patient (supportive care, low-dose cytarabine or intensive chemotherapy) in AZA-AML-001, treatment choice (supportive care or low-dose cytarabine) in DACO-016, azacitidine plus placebo in VIALE-A and AGILE, and low-dose cytarabine plus placebo in VIALE-C. AGILE enrolled only IDH1-mutated disease. Eras, age and eligibility rules, marrow blast thresholds and follow-up differ. The planned primary overall-survival analysis was not statistically significant in AZA-AML-001, DACO-016 and VIALE-C. The azacitidine subgroup analysis of a myelodysplastic syndrome trial and the single-arm venetoclax and IDH studies are not plotted. Compare only each trial's own two arms.
  • FLT3-mutated AML, relapsed or refractory. This lane holds one randomized trial. ADMIRAL randomized adults 2:1 to gilteritinib or salvage chemotherapy, and the analysis population is described in the abstract as randomized patients (371 eligible), not as intention to treat. Compare only the trial's own two arms.
  • Maintenance after first remission. This lane holds one randomized trial. Oral azacitidine (CC-486) is not bioequivalent to injectable azacitidine. The abstract gives no hazard ratio and does not use the term intention to treat. Compare only the trial's own two arms.
  • Any setting: other randomised trials. These trials are not part of a curated evidence chain; populations, eras and definitions differ widely within this group.
  • Relapsed / refractory: other randomised trials. These trials are not part of a curated evidence chain; populations, eras and definitions differ widely within this group.
  • Maintenance: other randomised trials. These trials are not part of a curated evidence chain; populations, eras and definitions differ widely within this group.
Data table
TrialYearSubtypeEndpointControl (months)Experimental (months)
AZA-AML-0012009Newly diagnosed AML in older adults or those unsuited to intensive chemotherapyOS (median)6.5 months10.4 months
AZA-AML-0012009Any setting: other randomised trialsOS at 12 months (not charted: different measure or time point)34.2%46.5%
AZA-AML-0012009Any setting: other randomised trialsOS (median)6.9 months12.1 months
DACO-0162009Newly diagnosed AML in older adults or those unsuited to intensive chemotherapyOS (median)5 months7.7 months
ALFA-07012010Gemtuzumab ozogamicin added to intensive inductionOS at 24 months (not charted: different measure or time point)41.9%53.2%
S01062010Gemtuzumab ozogamicin added to intensive inductionOS at 60 months (not charted: different measure or time point)50%46%
APL04062013Any setting: other randomised trialsOS at 50 months (not charted: different measure or time point)92.6%99.2%
aml17-aml19-mrd-monitoring2016Any setting: other randomised trialsOS at 36 months (not charted: different measure or time point)73%70%
aml17-aml19-mrd-monitoring2016Any setting: other randomised trialsOS at 36 months (not charted: different measure or time point)58%69%
aml17-aml19-mrd-monitoring2016Any setting: other randomised trialsOS at 36 months (not charted: different measure or time point)78%69%
aml17-aml19-mrd-monitoring2016Any setting: other randomised trialsOS at 36 months (not charted: different measure or time point)77%72%
3012017Any setting: other randomised trialsOS (median)5.95 months9.56 months
3012017Any setting: other randomised trialsOS (median)5.95 months9.33 months
AGILE2017Newly diagnosed AML in older adults or those unsuited to intensive chemotherapyOS (median)7.9 months24 months
RATIFY2017FLT3-mutated AML, newly diagnosed: added to intensive chemotherapyOS (median)25.6 months74.7 months
ADMIRAL2018FLT3-mutated AML, relapsed or refractoryOS (median)5.6 months9.3 months
VIALE-A2018Newly diagnosed AML in older adults or those unsuited to intensive chemotherapyOS (median)9.6 months14.7 months
VIALE-C2018Newly diagnosed AML in older adults or those unsuited to intensive chemotherapyOS (median)4.1 months7.2 months
QUAZAR AML-0012020Maintenance after first remissionOS (median)14.8 months24.7 months
QuANTUM-First2023FLT3-mutated AML, newly diagnosed: added to intensive chemotherapyOS (median)15.1 months31.9 months

Evidence chains

A lane follows one defined population and setting, where each new trial tests against what was then standard. Compare only within a lane, and read every comparison as illustrative.

Newly diagnosed APL: induction

Adults with newly diagnosed acute promyelocytic leukaemia receiving induction treatment · Any setting · endpoint: EFS

Results are not comparable across trials.

The two event-free survival values are not on the same footing: the European APL91 estimate is an interim 12-month figure from 101 randomized patients, with chemotherapy alone as control, whereas APL0406 reports a 24-month figure in lower-risk disease (white-cell count 10 x 10^9 per litre or less), with ATRA plus chemotherapy as control. The abstracts do not define an event or give the analysis population for these figures, so the entries sit at different landmarks and are not drawn on one scale. The North American Intergroup trial reported overall and disease-free survival, not event-free survival, so it is listed in the lane without an entry. Compare only each trial's own arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100EFS at 12 months (%)1993european-apl-91
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
EFS at 12 months, rate, across 1 trial in this lane. european-apl-91 (1993): experimental 79 % vs control 50 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
european-apl-911993EFS50 %79 %
APL04062013EFS (not plotted: different measure or time point)86 %97 %

Gemtuzumab ozogamicin added to intensive induction

Adults with previously untreated AML receiving intensive induction chemotherapy with or without gemtuzumab ozogamicin · Any setting · endpoint: OS

Results are not comparable across trials.

ALFA-0701 and SWOG S0106 differ in age range, gemtuzumab ozogamicin dose and schedule (3 mg/m2 on days 1, 4 and 7 versus 6 mg/m2 once on day 4) and, in S0106, in the daunorubicin dose (45 mg/m2 with gemtuzumab ozogamicin against 60 mg/m2 in the comparison arm). ALFA-0701 reports overall survival at 2 years and S0106 at 5 years, so the values sit at different landmarks and are not drawn on one scale. Compare only each trial's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100OS at 24 months (%)2010ALFA-0701
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS at 24 months, rate, across 1 trial in this lane. ALFA-0701 (2010): experimental 53.2 % vs control 41.9 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
ALFA-07012010OS41.9 %53.2 %
S01062010OS (not plotted: different measure or time point)50 %46 %

FLT3-mutated AML, newly diagnosed: added to intensive chemotherapy

Adults with newly diagnosed FLT3-mutated AML receiving intensive induction and consolidation chemotherapy plus a FLT3 inhibitor or placebo · Any setting · endpoint: OS

Results are not comparable across trials.

RATIFY enrolled adults aged 18 to 59 with any FLT3 mutation (ITD or TKD), whereas QuANTUM-First enrolled FLT3-ITD-positive disease only, aged 18 to 75. The drugs, chemotherapy details, rates of allogeneic transplant and follow-up differ. The midostaurin median has a confidence interval that reaches not reached and the quizartinib median an upper limit that is not estimable, so both medians are imprecise. P values in RATIFY are one-sided as the design specified. Compare only each trial's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080OS, median (months)20172023RATIFYQuANTUM-First
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 2 trials in this lane. RATIFY (2017): experimental 74.7 months vs control 25.6 months; QuANTUM-First (2023): experimental 31.9 months vs control 15.1 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
RATIFY2017OS25.6 months74.7 months
QuANTUM-First2023OS15.1 months31.9 months

Newly diagnosed AML in older adults or those unsuited to intensive chemotherapy

Older adults, or adults with comorbidities, with newly diagnosed AML treated with lower-intensity regimens (azacitidine, decitabine, venetoclax combinations, ivosidenib plus azacitidine) against conventional care, placebo or treatment choice · Any setting · endpoint: OS

Results are not comparable across trials.

The control arms differ: conventional care regimens preselected per patient (supportive care, low-dose cytarabine or intensive chemotherapy) in AZA-AML-001, treatment choice (supportive care or low-dose cytarabine) in DACO-016, azacitidine plus placebo in VIALE-A and AGILE, and low-dose cytarabine plus placebo in VIALE-C. AGILE enrolled only IDH1-mutated disease. Eras, age and eligibility rules, marrow blast thresholds and follow-up differ. The planned primary overall-survival analysis was not statistically significant in AZA-AML-001, DACO-016 and VIALE-C. The azacitidine subgroup analysis of a myelodysplastic syndrome trial and the single-arm venetoclax and IDH studies are not plotted. Compare only each trial's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

051015202530OS, median (months)20092009201720182018AZA-AML-001DACO-016AGILEVIALE-AVIALE-C
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 5 trials in this lane. AZA-AML-001 (2009): experimental 10.4 months vs control 6.5 months; DACO-016 (2009): experimental 7.7 months vs control 5 months; AGILE (2017): experimental 24 months vs control 7.9 months; VIALE-A (2018): experimental 14.7 months vs control 9.6 months; VIALE-C (2018): experimental 7.2 months vs control 4.1 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
AZA-AML-0012009OS6.5 months10.4 months
DACO-0162009OS5 months7.7 months
AGILE2017OS7.9 months24 months
VIALE-A2018OS9.6 months14.7 months
VIALE-C2018OS4.1 months7.2 months

FLT3-mutated AML, relapsed or refractory

Adults with relapsed or refractory FLT3-mutated AML · Relapsed / refractory · endpoint: OS

Results are not comparable across trials.

This lane holds one randomized trial. ADMIRAL randomized adults 2:1 to gilteritinib or salvage chemotherapy, and the analysis population is described in the abstract as randomized patients (371 eligible), not as intention to treat. Compare only the trial's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

02468101214OS, median (months)2018ADMIRAL
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 1 trial in this lane. ADMIRAL (2018): experimental 9.3 months vs control 5.6 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
ADMIRAL2018OS5.6 months9.3 months

Maintenance after first remission

Adults aged 55 or older in first remission after intensive chemotherapy who were not candidates for stem-cell transplantation, receiving oral azacitidine or placebo as maintenance · Maintenance · endpoint: OS

Results are not comparable across trials.

This lane holds one randomized trial. Oral azacitidine (CC-486) is not bioequivalent to injectable azacitidine. The abstract gives no hazard ratio and does not use the term intention to treat. Compare only the trial's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

051015202530OS, median (months)2020QUAZAR AML-001
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 1 trial in this lane. QUAZAR AML-001 (2020): experimental 24.7 months vs control 14.8 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
QUAZAR AML-0012020OS14.8 months24.7 months

Standard of care, then and now

How the usual sequence of treatment changed for one setting. Choose a point in time; each step links to the milestone behind it.

Newly diagnosed APL: induction treatment

Adults with newly diagnosed acute promyelocytic leukaemia · Any setting

  1. Targeted1Low-risk: ATRA + arsenic trioxideAPL0406: ATRA plus arsenic trioxide versus ATRA plus chemotherapy (2013)
2018. Based only on the trials cited in each step. The EU extended the indication in 2016 and the FDA approved the combination for low-risk APL in January 2018, as listed in the milestone.
Text version
  • 1990: Induction: ATRA (small series)
  • 1997: Induction: ATRA + chemotherapy
  • 2013: Lower-risk: ATRA + arsenic trioxide
  • 2018: Low-risk: ATRA + arsenic trioxide

Newly diagnosed AML other than APL: intensive treatment

Adults with newly diagnosed AML, other than APL, receiving intensive chemotherapy · Any setting

  1. Chemotherapy1Induction: cytarabine + anthracyclineCytarabine with an anthracycline: CALGB trials in AML, 1981-82 (1981)
  2. Chemotherapy2Consolidation: cytarabine dose studiedHigh-dose cytarabine consolidation: CALGB randomised trial, 1994 (1994)
  3. Targeted3FLT3 mutation: + midostaurinRATIFY: midostaurin added to chemotherapy in FLT3-mutated AML (2017)
  4. Targeted4FLT3-ITD: + quizartinibQuANTUM-First: quizartinib with chemotherapy in FLT3-ITD AML (2023)
2023. Based only on the trials cited in each step. QuANTUM-First added quizartinib in FLT3-ITD-positive disease; the two FLT3 trials are separate comparisons.
Text version
  • 1982: Induction: cytarabine + anthracycline
  • 1994: Induction: cytarabine + anthracycline → Consolidation: cytarabine dose studied
  • 2017: Induction: cytarabine + anthracycline → Consolidation: cytarabine dose studied → FLT3 mutation: + midostaurin
  • 2023: Induction: cytarabine + anthracycline → Consolidation: cytarabine dose studied → FLT3 mutation: + midostaurin → FLT3-ITD: + quizartinib

Reversals and dead ends

Approaches that were abandoned or withdrawn, and trials whose results were later questioned, are part of the story.

  • 2010

    Gemtuzumab ozogamicin: approved 2000, withdrawn 2010, back 2017

    Gemtuzumab ozogamicin joins an antibody to a cell-killing drug. It was approved in the United States in 2000 for older people with relapsed AML. In 2010, at the FDA's request, the sponsor moved to withdraw it after a required trial found no survival gain when added to chemotherapy. A new licence followed in 2017 for newly diagnosed and relapsed disease.

    FDA accelerated approval 17 May 2000 (first-relapse CD33-positive AML, age 60 or older; three phase 2 trials, 142 patients). SWOG S0106 (637 registered; daunorubicin-cytarabine with or without GO): CR 69% v 70% (P=.59); 5-year OS 46% v 50% (P=.85). FDA asked for withdrawal 21 May 2010. ALFA-0701 (n=280, age 50-70, fractionated 3 mg/m2): 2-year EFS 40.8% v 17.1% (HR 0.58, 95% CI 0.43-0.78). New BLA approved 1 Sep 2017.

Today and tomorrow

Standard of care today

Any setting

Newly diagnosed acute promyelocytic leukemia (APL), regulatory documents read. FDA, 12 January 2018:

  • a supplement (NDA 021248/S-015) approved arsenic trioxide (Trisenox) in combination with tretinoin for adults with newly diagnosed low-risk APL characterized by the t(15;17) translocation or PML/RAR-alpha gene expression. EMA: the Trisenox procedural-steps document lists an extension of indication (variation II/0058) to induction of remission and consolidation in adults with newly diagnosed low-to-intermediate-risk APL (white blood cell count of 10 x 10^3 per microlitre or lower) in combination with all-trans retinoic acid, with a CHMP opinion on 13 October 2016 and a Commission decision on 14 November 2016
  • the current EPAR page lists the same indication. No regulatory document read here covers high-risk APL. The 2019 European LeukemiaNet expert-panel recommendations for APL (Blood, PMID 30803991) are cited by identifier: only the abstract was read, which describes risk-adapted frontline therapy and names early death and the treatment of high-risk disease as remaining challenges
  • no recommendation is attributed to the panel and whether it is the current edition was not checked
US Food and Drug Administration 2018European Medicines Agency 2026European Medicines Agency 2026Sanz MA, Fenaux P, Tallman MS, Estey EH, Löwenberg B, Naoe T, et al 2019
Relapsed / refractory
Relapsed or refractory APL, regulatory documents read. FDA, 25 September 2000: arsenic trioxide (Trisenox, NDA 21-248) for induction of remission and consolidation in patients with APL who are refractory to, or have relapsed from, retinoid and anthracycline chemotherapy and whose APL is characterized by t(15;17) or PML/RAR-alpha expression. EMA: marketing authorisation issued on 5 March 2002; the EPAR page says it was originally granted under exceptional circumstances, which ended on 10 August 2010, and its current indication includes relapsed or refractory APL (previous treatment should have included a retinoid and chemotherapy). The milestone timeline covers the trials behind these actions.
US Food and Drug Administration 2000European Medicines Agency 2026
Any setting
Newly diagnosed non-APL AML with a FLT3 mutation, intensive chemotherapy, regulatory documents read. FDA, 28 April 2017: midostaurin (Rydapt, NDA 207997 Original 2) for adults with newly diagnosed FLT3 mutation-positive AML as detected by an FDA-approved test, with standard cytarabine and daunorubicin induction and cytarabine consolidation. FDA, 20 July 2023: quizartinib (Vanflyta, NDA 216993) for adults with newly diagnosed FLT3-ITD-positive AML detected by an FDA-approved test, with standard cytarabine and anthracycline induction and cytarabine consolidation and then as maintenance monotherapy; the letter's limitation of use says it is not indicated as maintenance monotherapy after allogeneic hematopoietic stem cell transplantation because improvement in overall survival has not been demonstrated in that setting. EMA: Rydapt marketing authorisation issued 18 September 2017 (the EPAR overview describes use with chemotherapy and then alone once chemotherapy is completed, in responders) and Vanflyta marketing authorisation issued 6 November 2023 (FLT3-ITD-positive disease, with induction and consolidation chemotherapy followed by single-agent maintenance). Two 2026 FDA letters for Vanflyta are not new indications: 13 April 2026 (S-002) approved updates to the drug-interaction and pharmacokinetics sections and records fulfilment of a postmarketing commitment from the 2023 approval letter, and 29 April 2026 (S-003) approved modifications to the REMS.
US Food and Drug Administration 2017US Food and Drug Administration 2023European Medicines Agency 2017European Medicines Agency 2023US Food and Drug Administration 2026US Food and Drug Administration 2026
Any setting

Newly diagnosed AML treated with intensive chemotherapy, antibody-drug conjugate and liposomal chemotherapy, regulatory documents read. FDA, 1 September 2017:

  • gemtuzumab ozogamicin (BLA 761060) for newly diagnosed CD33-positive AML in adults and, as a separate indication, relapsed or refractory CD33-positive AML in adults and pediatric patients 2 years and older (earlier US history is in the hooks). FDA, 3 August 2017: CPX-351 (Vyxeos, NDA 209401) for adults with newly diagnosed therapy-related AML (t-AML) or AML with myelodysplasia-related changes (AML-MRC)
  • on 30 March 2021 a supplement (S-006) expanded the indication to pediatric patients aged 1 year and older. A letter of 27 January 2026 (S-018) approved updates to the Section 8.4 pediatric-use text of the Vyxeos label pursuant to the final report of Study AAML1831
  • it is a labeling update, not a new indication. EMA: Mylotarg marketing authorisation issued 19 April 2018 (patients aged 15 years and above with previously untreated, de novo CD33-positive AML, with daunorubicin and cytarabine, except APL) and Vyxeos liposomal 23 August 2018 (adults with newly diagnosed t-AML or AML-MRC)
US Food and Drug Administration 2017US Food and Drug Administration 2017US Food and Drug Administration 2021US Food and Drug Administration 2026European Medicines Agency 2018European Medicines Agency 2018
Any setting

Guidelines and classification documents for AML, cited by identifier only. The 2022 European LeukemiaNet recommendations (Blood, PMID 35797463; the abstract describes a revised ELN genetic risk classification, revised response criteria and treatment recommendations), the ESMO Clinical Practice Guidelines for AML in adult patients (Ann Oncol 2020, PMID 32171751; PubMed holds no abstract) and NCCN Guidelines Insights:

  • Acute Myeloid Leukemia, Version 3.2026 (J Natl Compr Canc Netw, PMID 42297017
  • the abstract says it summarizes the panel's most recent recommendations on monitoring and management of measurable residual disease). The texts of these guidelines were not opened, so no recommendation is attributed to them
  • whether the ELN and ESMO editions are the current ones was not checked, and the NCCN live guideline may be newer than version 3.2026. The 2022 WHO 5th-edition classification (PMID 35732831) and the 2022 International Consensus Classification (PMID 35767897) are cited by identifier and covered in the milestone timeline
Döhner H, Wei AH, Appelbaum FR, Craddock C, DiNardo CD, Dombret H, et al 2022Heuser M, Ofran Y, Boissel N, Brunet Mauri S, Craddock C, Janssen J, et al 2020Pollyea DA, Altman JK, Assi R, Bachiashvili K, Bhatt VR, Fathi AT, et al 2026Khoury JD, Solary E, Abla O, Akkari Y, Alaggio R, Apperley JF, et al 2022Arber DA, Orazi A, Hasserjian RP, Borowitz MJ, Calvo KR, Kvasnicka HM, et al 2022
Any setting

Newly diagnosed AML in people not eligible for intensive chemotherapy, regulatory documents read. FDA, 21 November 2018:

  • venetoclax (Venclexta, NDA 208573/S-009) with azacitidine, decitabine or low-dose cytarabine for adults aged 75 years or older or with comorbidities that preclude intensive induction chemotherapy, under the accelerated approval regulations
  • FDA, 16 October 2020: supplements S-020 and S-021 gave traditional approval of the same indication on the basis of VIALE-A and VIALE-C, and the letter records under 'SUBPART H FULFILLED' that they fulfil postmarketing requirements 3545-1 and 3545-2 (the letter does not use the word 'converted'). FDA, 21 November 2018: glasdegib (Daurismo) with low-dose cytarabine for newly diagnosed AML in adults older than 75 years or with comorbidities that preclude intensive induction chemotherapy. Ivosidenib (Tibsovo): newly diagnosed AML in adults 75 years or older or with such comorbidities was first added on 2 May 2019 (S-001), and on 25 May 2022 (S-009) a new indication was approved: ivosidenib in combination with azacitidine or as monotherapy for newly diagnosed AML with a susceptible IDH1 mutation detected by an FDA-approved test, in adults 75 years or older or who have comorbidities that preclude intensive induction chemotherapy. The US Vidaza label as returned by the openFDA record (effective 29 January 2024) lists myelodysplastic syndrome subtypes and juvenile myelomonocytic leukemia and no AML indication
  • the US decitabine label was not used here. 2026 FDA letters for venetoclax are not AML indications: 19 February 2026 (S-031) approved the combination with acalabrutinib in previously untreated CLL/SLL and 20 May 2026 (S-032) a tablet-crushing administration option. EMA pages read: Dacogen marketing authorisation issued 20 September 2012 (newly diagnosed de novo or secondary AML, per WHO classification, in adults who are not candidates for standard induction chemotherapy)
  • Daurismo 26 June 2020 (with low-dose cytarabine, same population)
  • Tibsovo 4 May 2023 (with azacitidine, IDH1 R132 mutation, not eligible for standard induction chemotherapy). Venclyxto: the EPAR procedural-steps document lists an extension of indication (variation II/0030) for venetoclax in combination with hypomethylating agents for adults with newly diagnosed AML who are ineligible for intensive chemotherapy, with a CHMP opinion on 22 April 2021 and a Commission decision on 19 May 2021. The Vidaza EPAR page lists AML indications (20-30% blasts with multi-lineage dysplasia
  • more than 30% marrow blasts in adults aged 65 or older not eligible for HSCT) but dates only the original authorisation (17 December 2008), so no date is given for the AML indications
US Food and Drug Administration 2018US Food and Drug Administration 2020US Food and Drug Administration 2018US Food and Drug Administration 2019US Food and Drug Administration 2022US Food and Drug Administration 2024US Food and Drug Administration 2026US Food and Drug Administration 2026European Medicines Agency 2012European Medicines Agency 2020European Medicines Agency 2023European Medicines Agency 2026European Medicines Agency 2021European Medicines Agency 2024
Maintenance
Maintenance after first remission, regulatory documents read. FDA, 1 September 2020: oral azacitidine (Onureg, NDA 214120) for continued treatment of adults with AML who achieved first complete remission or complete remission with incomplete blood count recovery after intensive induction chemotherapy and are not able to complete intensive curative therapy. EMA: Onureg marketing authorisation issued 17 June 2021 for maintenance therapy in adults with AML in complete remission or complete remission with incomplete blood count recovery after induction therapy with or without consolidation who are not candidates for, including those who choose not to proceed to, hematopoietic stem cell transplantation. A letter of 9 April 2026 (S-003) approved updates to the infection and hepatic-impairment text of the US label; it is not a new indication. Quizartinib maintenance is part of the FLT3-ITD indication described above.
US Food and Drug Administration 2020European Medicines Agency 2021US Food and Drug Administration 2026
Relapsed / refractory

Relapsed or refractory non-APL AML, regulatory documents read. FDA letters:

  • enasidenib (Idhifa) for IDH2-mutated disease, 1 August 2017
  • ivosidenib (Tibsovo) for susceptible IDH1-mutated disease, 20 July 2018
  • gilteritinib (Xospata) for FLT3-mutated disease, 28 November 2018 (replacement approval letter)
  • olutasidenib (Rezlidhia) for susceptible IDH1-mutated disease, 1 December 2022 (each in adults, with the mutation detected by an FDA-approved test). Menin inhibitors: revumenib (Revuforj) was approved on 15 November 2024 for relapsed or refractory acute leukemia with a KMT2A translocation (acute leukemia, not AML only) in adult and pediatric patients 1 year and older, and on 24 October 2025 for relapsed or refractory AML with a susceptible NPM1 mutation in patients 1 year and older who have no satisfactory alternative treatment options
  • ziftomenib (Komzifti) was approved on 13 November 2025 for adults with relapsed or refractory AML with a susceptible NPM1 mutation who have no satisfactory alternative treatment options. EMA: Xospata marketing authorisation issued 24 October 2019 (monotherapy for adults with relapsed or refractory AML with a FLT3 mutation)
  • the EPAR page for Idhifa records that the application was withdrawn by the company on 6 December 2019. The EPAR pages for Rezlidhia, Revuforj and Komzifti were not found at the addresses tried, so no EMA status is given for those three
US Food and Drug Administration 2017US Food and Drug Administration 2018US Food and Drug Administration 2018US Food and Drug Administration 2022US Food and Drug Administration 2024US Food and Drug Administration 2025US Food and Drug Administration 2025European Medicines Agency 2019European Medicines Agency 2020

Trials to watch

  • Ziftomenib with venetoclax-azacitidine or 7+3 NCT07007312

    In untreated NPM1-mutated or KMT2A-rearranged AML, does adding ziftomenib to venetoclax plus azacitidine (the registry's nonintensive-therapy study; primary outcomes overall survival and, for US sites, complete remission) or to cytarabine plus daunorubicin 7+3 induction, consolidation and maintenance (the registry's intensive-therapy study; primary outcomes event-free survival and, for US sites, complete remission with MRD negativity in NPM1-mutated disease) improve outcomes compared with placebo? Registry record: two randomised, double-blind, placebo-controlled phase 3 studies under one NCT number, 1,300 estimated, sponsor Kura Oncology. Registry status: recruiting (status verified 2026-08); registry primary completion date 2031-11 (estimated); no results posted.

  • cAMeLot-2 NCT06852222

    In newly diagnosed AML with a KMT2A rearrangement or NPM1 mutation in adults ineligible for intensive chemotherapy, does bleximenib added to venetoclax plus azacitidine improve outcomes compared with placebo plus venetoclax and azacitidine, in complete remission rate and overall survival (the registry's primary outcomes)? Registry record: randomised, double-blind, placebo-controlled phase 3, 600 estimated, sponsor Janssen Research and Development. Registry status: recruiting (status verified 2026-09); registry primary completion date 2029-06-22 (estimated); no results posted.

  • Revumenib with azacitidine and venetoclax (HOVON) NCT06652438

    In newly diagnosed NPM1-mutated or KMT2A-rearranged AML in adults ineligible for intensive chemotherapy, does revumenib added to azacitidine plus venetoclax improve outcomes compared with placebo, in overall survival and complete remission rate (the registry's primary outcomes, both worded for NPM1-mutated AML)? Registry record: randomised, triple-masked phase 3, 448 estimated, sponsor Stichting Hemato-Oncologie voor Volwassenen Nederland. Registry status: recruiting (status verified 2026-08); registry primary completion date 2029-12-13 (estimated); no results posted.

  • REVEAL-ND NPM1 NCT07211958

    In newly diagnosed NPM1-mutated AML in people who are candidates for intensive chemotherapy (age 12 years or older), does revumenib added to an intensive chemotherapy regimen improve event-free survival and the measurable residual disease complete remission rate compared with placebo plus the same regimen (the registry's primary outcomes, worded that way in the record)? Registry record: randomised, double-blind, placebo-controlled phase 3, 468 estimated, sponsor Syndax Pharmaceuticals. Registry status: recruiting (status verified 2026-09); registry primary completion date 2029-06 (estimated); no results posted.

  • QuANTUM-WILD NCT06578247

    In newly diagnosed FLT3-ITD-negative AML in adults aged 18 to 70 years who are candidates for standard 7+3 induction chemotherapy, does quizartinib given with induction and consolidation chemotherapy and as maintenance improve overall survival compared with placebo (the registry's primary outcome, arm A versus arm B; a third arm gives quizartinib with chemotherapy and then placebo maintenance)? Registry record: randomised, double-blind, placebo-controlled phase 3, 700 estimated, sponsor Daiichi Sankyo. Registry status: active, not recruiting (status verified 2026-06); registry primary completion date 2030-06-26 (estimated); no results posted.

  • VAG-3+7-G NCT07407140

    In newly diagnosed FLT3-mutated AML (FLT3-ITD or FLT3-TKD; core-binding-factor AML and APL excluded; the registry also admits MDS/AML with 10-20% marrow blasts) in people aged 14 to under 75 years who are eligible for intensive chemotherapy, does the VAG regimen (venetoclax plus azacitidine plus gilteritinib in induction, consolidation and maintenance, as the registry describes it) differ from cytarabine plus daunorubicin or idarubicin (the registry's 3+7) plus gilteritinib in event-free survival (the registry's primary outcome)? Registry record: randomised, open-label phase 3, 300 estimated, sponsor Institute of Hematology and Blood Diseases Hospital, China. Registry status: not yet recruiting (status verified 2026-03; the registry lists the start date as estimated 2026-04-30); registry primary completion date 2029-02-01 (estimated); no results posted.

References

Show all 134 references

Loading references… Open the list as a page

How to cite this page

Bending the Curve. Acute myeloid leukaemia (including APL): an interactive history. Evidence current to October 2026. https://trialbytrial.com/cancers/aml. Accessed [date].

Every figure was checked against its primary source to the extent possible. Items that could not be fully confirmed are marked “needs review”. Educational history, not medical advice.

Report an error

Links open the primary source. A flag marked ⚑ needs review means a value or claim has not yet been independently confirmed.