Liu J Hematol Oncol (2021) 14:49
`https://doi.org/10.1186/s13045-021-01062-w
`
`REVIEW
`Emerging agents and regimens for AML
`
`Open Access
`
`Hongtao Liu*
`
`
`
`Abstract
`Until recently, acute myeloid leukemia (AML) patients used to have limited treatment options, depending solely
`on cytarabine + anthracycline (7 + 3) intensive chemotherapy and hypomethylating agents. Allogeneic stem cell
`transplantation (Allo-SCT) played an important role to improve the survival of eligible AML patients in the past several
`decades. The exploration of the genomic and molecular landscape of AML, identification of mutations associated with
`the pathogenesis of AML, and the understanding of the mechanisms of resistance to treatment from excellent transla-
`tional research helped to expand the treatment options of AML quickly in the past few years, resulting in noteworthy
`breakthroughs and FDA approvals of new therapeutic treatments in AML patients. Targeted therapies and combina-
`tions of different classes of therapeutic agents to overcome treatment resistance further expanded the treatment
`options and improved survival. Immunotherapy, including antibody-based treatment, inhibition of immune negative
`regulators, and possible CAR T cells might further expand the therapeutic armamentarium for AML. This review is
`intended to summarize the recent developments in the treatment of AML.
`Keywords: AML, Targeted therapy, Novel treatment
`
`Introduction
`AML is a heterogeneous disease, defined by a broad
`spectrum of genomic changes and molecular mutations
`that influence clinical outcomes and provide potential
`targets for drug development. The updated 2017 Euro-
`pean LeukemiaNet (ELN) risk stratification guidelines
`combining cytogenetic abnormalities and genetic muta-
`tions have been widely used to predict the prognosis of
`AML patients [1], while others have been exploring to
`incorporate additional prognostic factors into ELN-2017
`guidelines to improve the risk stratification models [2].
`Advanced by basic and translational research, espe-
`cially through large scale genomic analysis to understand
`the molecular landscape of AML, the development of tar-
`geted therapies, such as targeting fms-like tyrosine kinase
`3 (FLT3) and isocitrate dehydrogenase 1 and 2 (IDH1
`and IDH2) mutations, the treatment of AML landscape
`changed significantly with FDA approvals for several
`
`*Correspondence: hliu2@medicine.bsd.uchicago.edu
`Section of Hematology/Oncology, Department of Medicine, The
`University of Chicago Medical Center, 5841 S. Maryland Ave, MC 2115,
`Chicago, IL 60637-1470, USA
`
`new drugs in the past several years. Even with all these
`improvements, primary resistance to initial treatment
`and disease relapse remain huge unmet need in the treat-
`ment of AML. The majority of AML patients still even-
`tually succumb to the disease. We still have a long way
`to further improve the survival of the AML patients, thus
`many investigational drugs have been explored to target
`the primary and secondary treatment resistance in AML
`patients.
`This review will provide updates of the emerging thera-
`peutic approaches for the treatment of AML, including
`combinations with mutation driven targeted treatments,
`novel immunotherapies in the myeloid disease.
`
`Targeted therapies: alone or combination
`BCL-2 inhibitor: venetoclax
`BCL-2 is a member of the BCL-2 family of anti- and pro-
`apoptotic proteins. BCL-2 protects cells against apopto-
`sis. BCL-2 expression in AML has been associated with
`decreased sensitivity to cytotoxic chemotherapy and a
`higher rate of relapse [3]. Venetoclax is an orally bioavail-
`able selective inhibitor of BCL-2, promoting intrinsic
`apoptotic pathway activation resulting in mitochondrial
`
`© The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
`permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
`original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
`other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
`to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
`regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
`licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco
`mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
`
`CELGENE 2095
`APOTEX v. CELGENE
`IPR2023-00512
`
`

`

`Liu J Hematol Oncol (2021) 14:49
`
`Page 2 of 20
`
`outer membrane permeability through dissociation of
`BCL-2 mediated sequestration of BH3 proteins BIM and
`BID and effector proteins BAX and BAK. Venetoclax was
`initially approved by U. S. Food and Drug Administration
`(FDA) in 2016 to treat individuals with chronic lympho-
`cytic leukemia (CLL) with deletion (17p).
`Venetoclax + hypomethylating agents or low dose cytarabine
`Early studies using venetoclax as monotherapy in AML
`demonstrated only modest efficacy in high-risk relapsed/
`refractory (R/R) AML patients with an overall response
`rate (ORR) of 38% and complete remission/complete
`remission with incomplete hematologic recovery (CR/
`CRi) of 19%. The responses were short lived, with overall
`survival (OS) of only 4.7 months [4]. Based on promis-
`ing results from two large Phase 1b/II trials using com-
`bination of a hypomethylating agent (HMA) or low-dose
`cytarabine (LDAC) with venetoclax in untreated older
`AML patients [5, 6], FDA granted accelerated approval
`to venetoclax in combination with azacitidine (AZA) or
`decitabine (DEC) or LDAC for the treatment of newly-
`diagnosed (ND) AML in adults who are age 75 years or
`older, or who have comorbidities that preclude use of
`intensive induction chemotherapies in 2018.
`Recently published Phase III randomized studies con-
`firmed the results from these early single arm trials, and
`demonstrated a significant survival benefit from add-
`ing venetoclax to azacitidine and to LDAC [7, 8]. The
`major findings from the VIALE-A and VIALE-C trials
`are summarized in Table  1. In summary, the VIALE-A
`trial included 431 patients without history of exposure
`to azacitidine. At a median follow-up of 20.5  months,
`
`the median OS was 14.7 months in the azacitidine-vene-
`toclax group and 9.6  months in the control group. The
`incidence of CR and composite complete remission rate
`(cCR) (CR + CRi) were significantly higher with azaciti-
`dine-venetoclax than with the control regimen. How-
`ever, there were higher rates in key adverse events in the
`azacitidine-venetoclax group than those in the control
`group, but they were manageable [7]. The VIALE-C study
`assigned 211 patients to either venetoclax (n = 143) or
`placebo (n = 68) in 28-day cycles, plus LDAC on days 1
`to 10. In contrast to VIALE-A trial, 20% enrolled patients
`had received prior HMA treatments. The planned pri-
`mary analysis showed a 25% reduction in risk of death
`with venetoclax plus LDAC vs LDAC alone, although
`this was not statistically significant. Median OS was 7.2
`vs 4.1 months, respectively. Unplanned analyses with an
`additional 6-months follow-up demonstrated median
`OS of 8.4 months for the venetoclax arm (HR, 0.70; 95%
`CI, 0.50–0.98; P = 0.04). CR/CRi rates were 48% and 13%
`for venetoclax plus LDAC and LDAC alone, respectively.
`Thus, venetoclax plus LDAC demonstrated clinically
`meaningful improvement in remission rate and OS vs
`LDAC alone, with a manageable safety profile [8]. Based
`on these confirmatory data, FDA granted full approval to
`these venetoclax combinations for treating newly diag-
`nosed AML patients. Both trials established new stand-
`ard of care for unfit newly diagnosed AML patients. Since
`VIALE-A trial excluded patients with previous exposure
`to azacitidine, and 20% patients enrolled on the VIALE-C
`trial had exposure to HMA, venetoclax plus LDAC might
`be a preferred consideration for patients who received
`HMAs in the past.
`
`Table 1 Comparison of randomized prospective studies on venetoclax-based combinations
`LDAC + venetoclax
`Regimen
`
`AZA + venetoclax
`
`in AML: AZA + venetoclax vs
`
`LDAC + venetoclax
`
`Phase
`Population
`Control arm
`h/o HMA
`Patient number
`
`Median age
`(range), years
`30-day mortality, %
`cCR (CR) rate, %
`MRD negativity, %
`Time to CR (response)
`Median DOR, months
`Median OS, months
`Reference
`
`III VIALE-A trial
`Age > 75 years or unfit for chemotherapy
`AZA
`No
`431
`(286 in AZA + venetoclax)
`76 (49–91)
`
`III VIALE-C trial
`
`LDAC
`Yes, allowed (20%)
`211
`(143 in LDAC + venetoclax)
`76 (36–93)
`
`7%
`66.4% (36.7%)
`N/A
`1.3 months (0.6–9.9)
`17.5 (13.6 to NR)
`14.7 (11.9–18.7)
`[7]
`
`13%
`48% (27%)
`6%
`N/A most response at the end of cycle 2
`NA
`8.4 (5.9–10.1)
`[8]
`
`

`

`Liu J Hematol Oncol (2021) 14:49
`
`
`Page 3 of 20
`
`Both trials also identified that patients with NPM1
`and IDH1/2 mutations had high CR rates of 91%, and
`71%, respectively with HMA + venetoclax [5] and high
`CR/CRi rates (89% and 72%), respectively, when treated
`with venetoclax + LDAC [6]. Patients with FLT3 muta-
`tions (Internal tandem duplication (ITD) and/or tyrosine
`kinase domain (TKD) also demonstrated high CR rate
`of 72% [5]. On the other hand, inhibitors to these muta-
`tions have been developed and will be discussed in the
`following sections. It would be continued debate on how
`to choose the first line treatment for AML with these
`mutations: hypomethylating agents with IDH1/2 inhibi-
`tors vs venetoclax-based combination; how to sequence
`the treatment options: venetoclax-based combinations
`first followed by IDH1/2 inhibitors at disease relapse/
`progression or the other way around; or use three drugs
`combination with HMA + venetoclax + IDH1/2 inhibi-
`tor to get deeper remission. Only randomized clinical
`trials could eventually answer these important clinical
`questions.
`Venetoclax + intensive chemotherapy
`Not surprisingly, venetoclax has been studied in com-
`binations with intensive chemotherapy as well (summa-
`rized in Table  2). A retrospective report of 13 patients
`treated with FLAVIDA salvage therapy (fludarabine,
`cytarabine, and idarubicin in combination with veneto-
`clax 100 mg daily for 7 days; dose reduced due to con-
`current azole administration) compared to a control
`cohort received FLA-Ida (fludarabine, cytarabine, and
`idarubicin) reported a higher but not statistically sig-
`nificant CR/CRi rate of 69% compared to 47% in the
`control cohort [9]. A phase 1b/II trial of medically fit
`patients with R/R AML receiving FLAG-Ida induction
`
`and consolidation in combination with a 14 days course
`of venetoclax was conducted at MD Anderson. Early
`results were promising with CRc of 74% in all the patients
`and an impressive CRc of 91% in newly diagnosed (ND)
`patients. Consistent with known venetoclax resist-
`ance mechanisms, high levels of MCL-1 expression
`were found in patients who relapsed following FLAG-
`Ida + venetoclax [10]. The updated data of 62 patients
`(27 with ND AML and 35 with R/R AML) from the trial
`was recently presented. The ORR was 84%, with 89% of
`ND AML and 66% of R/R AML patients achieving a CRc.
`83% of patients achieved minimal residual disease (MRD)
`negative (MRD-) status assessed by flow cytometry. After
`a median follow up of 11  months, median OS was not
`reached. The addition of venetoclax to FLAG-ida demon-
`strated robust efficacy with acceptable safety profile [11].
`The CAVEAT study reported data on 51 newly diag-
`nosed patients with AML, either de novo or second-
`ary, who were treated in five venetoclax dose-escalation
`cohorts (50–600 mg; venetoclax was given over 14 days,
`day -6 to 7 with induction chemotherapy (cytarabine
`100  mg/m2 days 1–5 and idarubicin 12  mg/m2 intrave-
`nously days 2–3)). The same venetoclax dose and sched-
`ule was given for four cycles of consolidation (cytarabine,
`days 1–2, and idarubicin, day 1), and as maintenance
`(up to seven 28-day cycles). The overall CR/CRi rate was
`72%, but was 97% in the 28 patients with de novo AML
`and only 43% in secondary AML. [12]. In our center, we
`have used HiDAC + mitoxantrone + venetoclax for sev-
`eral heavily pretreated patients with R/R acute leukemia
`to control the disease prior to allogeneic stem cell trans-
`plantation (allo-SCT) (personal experience). This combi-
`nation warrants further study in both newly diagnosed
`and R/R AML setting.
`
`Table 2 Summary of venetoclax-based combinations in AML
`
`Combination
`
`Phase
`
`Disease status
`
`Patient number
`
`CR/CRi rate, %
`
`References
`
`FLA-Ida
`FLAG-ida
`CAVEAT (5 + 2)
`
`DEC10
`
`CLIA
`CLAD/LDAC, alternating
`with AZA
`CPX-351
`
`CPX-351 LIT
`GO
`
`Retrospective
`Ib/II
`
`Ib
`
`II
`
`II
`II
`
`II
`
`Ib
`Ib
`
`R/R AML
`ND AML
`R/R AML
`ND AML
`
`ND AML
`R/R AML
`ND AML
`ND AML
`
`R/R AML
`ND AML
`ND AML
`R/R AML
`
`13
`27
`35
`51
`
`70
`55
`18
`48
`
`69%
`89% in ND AML
`66% in R/R AML
`72% in all
`97% in de novo AML
`43% secondary AML
`86% in ND AML
`42% in R/R AML
`88%
`94%
`
`17
`1
`44 planned
`24 planned
`
`37%
`
`NA
`NA
`
`[9]
`[10, 11]
`
`[12]
`
`[13]
`
`[14]
`[15]
`
`[16]
`
`[17]
`[18]
`
`

`

`Liu J Hematol Oncol (2021) 14:49
`
`Page 4 of 20
`
`The results of ten-days of decitabine (DEC10) with
`venetoclax (DEC10-VEN) in AML and high-risk MDS
`were reported. DEC10-VEN is safe and highly effec-
`tive in newly diagnosed AML and can serve as an
`effective bridge to SCT. Median OS in treatment naïve
`AML patients who subsequently underwent SCT
`was not reached (1  year OS of 100%). For previously
`treated AML patients, OS was 22.1  months [13]. In
`addition, propensity score matched analysis (PSMA)
`was employed to compare outcomes of 54 younger
`adult patients with R/R AML treated on the prospec-
`tive phase 2 trial of 10-day decitabine and venetoclax
`(DEC10-VEN) with a historical cohort of patients
`treated with intensive chemotherapy. The analysis
`demonstrated that DEC10-VEN provided comparable
`response of CR/CRi, OS, and rate of patient to proceed
`SCT to non-venetoclax based intensive chemotherapy.
`Thus, DEC10-VEN represents an appropriate salvage
`therapy, and provides an appropriate backbone for add-
`ing novel therapies in R/R AML patients [19].
`The addition of venetoclax to cladribine, idarubicin,
`and Ara C (CLIA) was safe and effective in ND patients
`with AML. The combination was not associated with
`early mortality or prolonged myelosuppression, but
`did result in high rates of durable MRD negative remis-
`sions (NCT02115295) [14]. Addition of venetoclax to a
`low-intensity backbone of cladribine + LDAC (CLAD/
`LDAC) alternating with HMA for older patients with
`newly diagnosed AML provided a CR/CRi rate of 94%;
`and among the subset of patients who had CR with
`complete count recovery, the MRD negative rate was
`92%. The regimen was well tolerated, with 4-week mor-
`tality rates of 0%. With a median follow-up of more
`than 11 months, the median OS has not been reached
`(NR), with 12-month OS rates of 70% [15]. Full dose
`CPX-351 plus 7  days of VEN (300  mg on D2-8) was
`demonstrated to be tolerable with acceptable toxici-
`ties in patients with R/R AML with an ORR of 44%; and
`ORR was high at 60% in patient without prior VEN
`exposure, compared to just 17% among those who had
`prior VEN. 86% of responding patients proceeded to
`SCT. The median OS overall was 6.4 months; and the
`median OS was not reached among the responders
`[16].
`Other ongoing trials include open-label, multicenter,
`2-part, phase 1b study (NCT04038437) to determine
`the maximum tolerated dose and evaluate the safety,
`efficacy, and pharmacokinetics of CPX-351 lower-inten-
`sity therapy (LIT) plus venetoclax [17]. Another single
`arm, open-label, multi-center, dose-escalation phase Ib
`study is evaluating the combination of venetoclax and
`gemtuzumab ozogamicin in R/R CD33 + AML patients
`(NCT04070768) [18].
`
`Venetoclax + experimental drugs or targeted inhibitors
`Given the proven synergies of BCL-2 inhibition, mul-
`tiple combinations with targeted agents, and veneto-
`clax are under investigation. There are many ongoing
`combinations of therapies targeting BCL-2 and other
`pathways, including FLT3 inhibitors (gilteritinib) and
`IDH1 and 2 inhibitors (Ivosidenib and enasidenib)
`(will be discussed in the later sections), MCL-1 inhibi-
`tors (VU661013, A-1210477); MEK1/2 inhibitor (cobi-
`metinib), and MDM2 inhibitor (idasanutlin) (reviewed in
`[20]), combination with TKI in Ph + acute leukemia [21]
`and other emerging pre-clinical combinations includ-
`ing small-molecule inhibitors of CDK9 (the orally active
`A-1592668 and the related analog A-1467729) leading
`to down-expression of MCL-1 [22]; the Exportin inhibi-
`tor, Selinexor, [23]; BET inhibitors, ABBV-075, [24]; SRC
`family kinases (SFK) and Bruton’s tyrosine kinase (BTK)
`inhibitor, ArQule 531 (ARQ 531), [25]; and it is expecting
`much more novel combinations to come.
`
`Resistance mechanisms
`HMA + venetoclax or LDAC + venetoclax have clearly
`advanced the treatment of AML for older or unfit AML
`patients. Unfortunately, these regimens are unlikely
`to provide cure as most patients have relapsed at the
`median of 7 cycles of treatment. A retrospective study
`demonstrated that the outcome of 41 patients who failed
`to respond to HMA + venetoclax was very poor with the
`median OS of only 2.4  months despite salvage therapy
`[26]. To understand the resistance mechanisms, DiNardo
`CD et al. analyzed 81 patients receiving these venetoclax-
`based combinations to identify molecular correlates of
`durable remission, initial response followed by relapse
`(adaptive resistance), or refractory disease (primary
`resistance). Acquisition or enrichment of clones with
`activation of the signaling pathways such as FLT3 or RAS
`or bi-allelic mutations perturbing TP53 were most com-
`monly identified among primary and adaptive resistance
`to venetoclax-based combinations. Single-cell studies
`identified heterogeneous and sometimes divergent inter-
`val changes in leukemic clones within a single cycle of
`therapy, highlighting the dynamic and rapid occurrence
`of therapeutic selection in AML. In functional studies,
`gain of FLT3-ITD mutation or loss of TP53 conferred
`cross-resistance to both venetoclax and cytotoxic-based
`therapies [27]. These data confirmed the previous find-
`ings that TP53 apoptotic network is the primary media-
`tor of resistance to BCL-2 inhibition in AML cells [28].
`Interestingly, recent study demonstrated that monocytic
`AML is intrinsically resistant to venetoclax + AZA due to
`loss of expression of the venetoclax target of BCL-2, but
`instead preferentially reliant on MCL-1 for the survival.
`
`

`

`Liu J Hematol Oncol (2021) 14:49
`
`
`Page 5 of 20
`
`Thus, venetoclax + AZA treatment selects monocytic
`disease at disease relapse, which is derived from pre-
`existing monocytic subclones [29]. AML patients with
`monocytic disease or TP53 mutation might have high
`risk to be resistant to venetoclax-based combinations,
`and clinical trials targeting TP53 mutation or trials spe-
`cifically targeting monocytic AML might be considered
`over venetoclax-based combinations.
`Future clinical research will focus on deepening the
`responses provided by HMA + venetoclax with addi-
`tional targeted agents, like ivosidenib in IDH1 mutated
`AML (to be discussed in next section), FLT3 inhibitors,
`and novel pathways inhibitors to eventually cure a greater
`fraction of newly diagnosed AML, and to explore new
`strategies to deal with relapses after venetoclax-based
`therapies.
`
`IDH1/2 inhibitors
`IDH1 and IDH2 are critical enzymes for the oxidative car-
`boxylation of isocitrate. A mutation in one of these genes
`results in increased concentration of 2-hydroxyglutarate
`(2-HG). 2-HG causes DNA and histone hypermethyla-
`tion, leading to blocked cellular differentiation and tumo-
`rigenesis. Mutations in IDH1 or IDH2 are present in 5%
`to 15% and 10% to 15% of patients with newly diagnosed
`AML, respectively [30]. Oral, small-molecule inhibitors
`have been developed for both mutant IDH1 (ivosidenib)
`and IDH2 (enasidenib). In R/R AML, ivosidenib and
`enasidenib as single agent produced promising responses
`for the corresponding mutations with ORR of 41.6% (CR:
`21.6%) with median OS of 8.8 months [31] and ORR of
`40.3% (CR 20.6%) with median OS of 9.3  months [32]
`respectively. FDA approved ivosidenib and enasidenib
`for patients with relapsed or refractory IDH1 and IDH2
`mutated AML, respectively, in 2018. In the front line
`setting, both inhibitors have also demonstrated clini-
`cal effectiveness [33, 34], leading to FDA approval of
`ivosidenib for patients with newly diagnosed IDH1
`mutated AML based on an ORR of 42% (CR: 30%) with
`median OS of 12.6 months in older patients not eligible
`for intensive therapy [34].
`The Phase 3 IDHENTIFY study evaluating enasidenib
`plus best supportive care (BSC) versus conventional
`care regimens, which included BSC only, azacitidine
`plus BSC, low-dose cytarabine plus BSC, or intermedi-
`ate-dose cytarabine plus BSC, did not meet the primary
`endpoint of OS in patients with R/R AML with an IDH2
`mutation. The safety profile of enasidenib was consistent
`with previously reported findings. IDH inhibitors alone
`are unlikely to provide cure or durable remission for R/R
`AML, but they might provide excellent disease control
`with low toxicity and a bridge to allo-SCT.
`
`IDH inhibitors work in part through induction of
`differentiation of malignant cells, leading to differen-
`tiation syndrome in 10% to 20% of patients. Clinical
`features are similar to those seen in patients with acute
`promyelocytic leukemia (APL) treated with ATRA-
`based regimens [35, 36]. Early studies established a firm
`association between IDH mutations and serum 2-HG
`concentration in AML, and confirmed that serum
`oncometabolite measurements provide useful diag-
`nostic and prognostic information that can improve
`patient selection for IDH-targeted therapies [37]. How-
`ever, 2-HG level reduction and clearance of IDH muta-
`tion by next generation sequencing (NGS) assay does
`not correlate with the clinical response. These inhibi-
`tors are unlikely to provide cure of the AML due to pri-
`mary resistance from co-mutations in other pathways
`especially the NRAS/KRAS, and MAPK pathway effec-
`tors PTPN11, NF1, FLT3 and others [38] and secondary
`resistance from development of second-site IDH2 mis-
`sense mutations or isoform switching [39, 40].
`Since IDH1/2 mutations lead to DNA and histone
`hypermethylation, HMAs might have synergistic effects
`in combination of IDH inhibitors. Combination of HMAs
`with IDH inhibitors has been studied. The combination
`of ivosidenib and azacitidine was studied in 23 patients
`with IDH1 mutated AML as front line treatment. The
`ORR was 78% with CR/CRh rate of 70%, and median
`time to response of 1.8 months; median response dura-
`tion was not yet reached. The ivosidenib and azacitidine
`combination was well tolerated with a safety profile con-
`sistent with ivosidenib or AZA monotherapy and with
`17% incidence of IDH differentiation syndrome. Clear-
`ance of mutated IDH1 was seen in 63% patients with CR/
`CRh. CR and ORR rates exceeded those expected from
`AZA alone [41]; 83% CR/CRh patients achieved MRD
`negativity by flow cytometry [42]. AGILE, a global, dou-
`ble-blind, randomized, placebo-controlled, phase III trial
`for patients with previously untreated IDH1 mutated
`AML who are not candidates for intensive therapy
`(NCT03173248) is actively enrolling patients from 172
`study centers across the world [43]. Patients are ran-
`domly assigned to AZA + ivosidenib or AZA + placebo.
`As for the IDH2 inhibitor of enasidenib, the phase II
`portion of an open-label, randomized phase I/II study
`of enasidenib (E) + AZA (“E + A”) vs AZA monother-
`apy (“A”) in patients with mutated IDH2 (mIDH2) ND
`AML (NCT02677922) was recently reported [44]. 101
`patients with intermediate- or poor-risk cytogenet-
`ics were randomized 2:1 to E + A or A in 28-day cycles.
`ORR (71% vs 42%) and CR (53% vs 12%) rates were sig-
`nificantly improved with E + A with greater clearance of
`mIDH2 allele frequency. Time to first response was about
`2 months in each arm and the time to CR was 5.5 months
`
`

`

`Liu J Hematol Oncol (2021) 14:49
`
`Page 6 of 20
`
`(range, 0.7–19.5). There was no difference in median PFS
`and OS so far [44].
`As discussed in the section of Azacitidine and vene-
`toclax, this combination is very effective in patients
`with IDH1/2 mutation. In a pooled retrospective study,
`79 patients with IDH1/2 mutation were identified and
`treated with VEN + AZA on either the Phase Ib or the
`randomized Phase III (VIALE-A) trials. CR/CRh was
`72% (95% CI: 61%-82%) in the whole population. In
`patients with IDH1, CR/CRh was 59%, median time
`to first CR/CRh response was 2.3  months, and median
`duration of response (DOR) and OS were 21.9 (7.8–29.5)
`months and NR. In patients with IDH2, CR/CRh rates
`were 80%, median time to first CR/CRh response was
`1.0  month. Median DOR and median OS (mOS) were
`NR. Thus, VEN + AZA provided high response rates,
`long DOR, and mOS among treatment-naïve patients
`with IDH1/2 mutation ineligible for intensive chemo-
`therapy with acceptable safety profile [45]. As mentioned
`previously, it will be a continued debate to optimize the
`front line treatment for unfit AML patients with IDH1/2
`mutations.
`There is also a rationale for combining IDH inhibitors
`with BCL-2 inhibitors, since the accumulation of 2-HG
`caused by IDH mutations could decrease the mitochon-
`drial threshold for induction of apoptosis induced by
`BCL-2 inhibition with venetoclax [46]. The combina-
`tion therapy of ivosidenib (IVO) plus venetoclax (VEN)
`with or without azacitidine was found to be effective
`against AML harboring an IDH1 mutation in a phase
`Ib/II trial [47]. Patients with AML or high-risk MDS
`were assigned to one of three cohorts, either receiving
`IVO + VEN 400 mg, IVO + VEN 800 mg, or IVO + VEN
`400  mg + AZA. The median time to best response was
`2  months. In 18 evaluable patients, cCR rate was 78%
`overall (treatment naive: 100%; R/R: 75%), and 67%, 100%,
`and 67% by cohort with median time to best response of
`
`2 months. IVO + VEN + AZA therapy was well tolerated
`and highly effective for patients with IDH1 mutated AML
`[47]. It is reasonable to expect that combination of the
`IDH2 inhibitor enasidenib with venetoclax and azaciti-
`dine might also provide better outcomes than enasidenib
`alone or enasidenib with azacitidine, since enasidenib
`plus venetoclax demonstrated superior anti-leukemic
`activity against IDH2 mutated AML in patient-derived
`xenograft models [48]. Table 3 summarizes the trials of
`combinational therapies for newly diagnosed unfit AML
`patients with IDH1/2 mutation.
`
`Targeting FLT3 mutations
`FLT3 Mutations occurs in approximately 30% patients
`with newly diagnosed AML (20% to 25% with FLT3-ITD
`mutation, 5% to 10% with FLT3-TKD), and associates
`with more proliferative disease, increased risk of relapse,
`and inferior survival. Randomized phase III RATIFY
`study led to approval of 7 + 3 + midostaurin as front
`line for young fit patients [51], and randomized phase
`III ADMIRAL study with single-agent gilteritinib estab-
`lished the approval of gilteritinib for the treatment of R/R
`FLT3-mutated AML [52]. The phase III randomized trial
`using quizartinib vs investigator choice salvage chemo-
`therapy in patients with R/R FLT3-ITD mutated AML
`met the primary objective of OS improvement [53], lead-
`ing its approval in Japan, but not in the US. Many studies
`to combine these FLT3 inhibitors are under investigation.
`An open-label, phase 1 study (NCT02236013) assessed
`the safety/tolerability and anti-leukemic effects of gilteri-
`tinib plus 7 + 3 induction, consolidation, and main-
`tenance therapy in fit adults with newly diagnosed
`FLT3-mutated AML. 80 patients with median age of
`59  years were allocated to treatment. The maximum
`tolerated dose of gilteritinib was 120 mg daily. CRc was
`achieved by 81.8% of patients across all dose groups with
`mutational clearance (FLT3 ITD signal ratio of ≤ 10–4
`
`Table 3 Summary of combination of targeted-therapy trials in IDH1/2-mutant newly diagnosed AML
`
`Regimens
`
`Phase
`
`Patient
`Number
`
`CR/CRi rate, %
`
`Time to CR or response
`(median), months
`
`OS (median),
`months
`
`References
`
`HMA + venetoclax
`AZA + venetoclax
`AZA + venetoclax
`LDAC + venetoclax
`(HMA naïve)
`AZA + ivosidenib
`AZA + enasidenib
`Venetoclax + ivosidenib
`AZA + venetoclax + ivosidenib
`
`Ib
`III
`Pooled data from
`two trials
`Ib/II
`III
`Ib
`II
`Ib/II
`Ib/II
`
`35
`46
`79
`
`18
`
`23
`68
`12
`6
`
`71
`75.4
`72
`
`72
`
`69.6
`68
`83
`67
`
`2.1
`N/A
`1.0
`
`1.4
`
`3.7
`5
`NA
`NA
`
`24.4
`N/A
`24.5
`
`19.4
`
`N/A
`22.0
`NA
`NA
`
`[5]
`[7]
`[45]
`
`[6, 8]
`
`[49]
`[50]
`[47]
`[47]
`
`

`

`Liu J Hematol Oncol (2021) 14:49
`
`
`Page 7 of 20
`
`after induction or consolidation) was achieved by 70% of
`patients with FLT-ITD mutation receiving a gilteritinib
`dose of ≥ 120 mg [54]. Two large randomized clinical tri-
`als of induction and consolidation chemotherapy plus
`gilteritinib vs midostaurin in FLT3 mutated AML patients
`are ongoing in the US (PrECOG trial) (NCT03836209))
`and in Europe (HOVON 156 AML / AMLSG 28–18 trial
`(NCT04027309)).
`The LACEWING study is a phase 3 trial to randomize
`FLT3 mutated ND AML patients ineligible for intensive
`induction chemotherapy to get gilteritinib plus azaciti-
`dine vs azacitidine alone. The safety cohort enrolled 15
`patients and established dose of gilteritinib of 120 mg to
`be used in combination with azacitidine. Overall, a CRc
`of 67% was observed with median duration of remission
`of 10.4 months for the CRc responders. The combination
`treatment was well tolerated with no unexpected adverse
`effect [55]. While the data provides a promising option
`of gilteritinib plus azacitidine for newly diagnosed FLT3-
`mutated unfit AML patients, the company announced
`that Phase 3 LACEWING trial failed to meet primary
`end point of OS at a planned interim analysis and the
`study was terminated for futility in December 2020.
`Many lessons have been learned in the AML field that
`high response rate in AML will not necessary transform
`into survival benefit.
`In the R/R setting, a phase 1b study tested the safety
`and efficacy of combining venetoclax at 400  mg with
`gilteritinib at 120  mg daily. 39 patients were enrolled,
`and among them 64% had previous history of FLT3
`TKI exposure. 37 patients were evaluable for response,
`31 (84%) achieved CRc. This data compares favorably
`to the CRc of 54% with single agent gilteritinib in the
`ADMIRAL study; suggesting gilteritinib plus veneto-
`clax might be better option for R/R FLT3-mutated AML,
`while longer follow-up with OS data is awaited [56]. A
`Phase Ib/II trial explored the combination of quizartinib
`(Quiz) with decitabine (10  days) ± venetoclax mostly in
`patients with R/R AML. CRc of 90% was achieved in the
`DEC10 + VEN + Quiz cohort, and CRc rate of 40% was
`achieved in DEC10 + quiz cohort. In addition, CyTOF
`(single-cell mass cytometry) analysis could be used to
`select patients with the best response based on pre- and
`on-therapy apoptotic and signaling pathway profiles [57].
`
`Targeting TP53 mutation
`The TP53 gene, located on chromosome 17p13.1, is com-
`monly mutated in tumors making it one of the most
`widely mutated genes in human malignancies. TP53
`mutations are detected in 5% to 20% of patients with
`newly diagnosed AML and MDS, with higher incidences
`in older patients and in those with secondary AML or
`therapy-related myeloid neoplasms. TP53 mutation is
`
`enriched in patients with complex karyotype and mon-
`osomal karyotypes and also in patients with relapse or
`refractory disease. TP53 mutation has been associated
`with a poor prognosis in both AML and MDS [58, 59].
`A recent study analyzed 3,324 patients with MDS for
`TP53 mutations and allelic imbalances, and deline-
`ated two subsets of patients with distinct phenotypes
`and outcomes. One-third of TP53-mutated patients had
`monoallelic mutations whereas two-thirds had multi-
`ple hits consistent with biallelic targeting. Established
`associations with complex karyotype, few co-occurring
`mutations, high-risk presentation and poor outcomes
`were specific to multi-hit patients only. The TP53 multi-
`hit state predicted a high risk of death and leukemic
`transformation independently of the revised interna-
`tional prognostic scoring system (IPSS-R). Importantly,
`monoallelic patients did not differ from