Sage Journals: Discover world-class research

Abstract

While the past decade has seen a revolution in understanding of the genetic and molecular etiology of the disease, in clinical practice, initial therapy for acute myeloid leukemia (AML) patients has been a relatively straightforward choice between intensive combination cytotoxic induction therapy as used for decades or less-intensive hypomethylating therapy. The year 2017, however, witnessed US Food and Drug Administration approvals of midostaurin, enasidenib, gemtuzumab ozogamicin and CPX-351 for AML patients, with many other promising agents currently in clinical trials. This review discusses these options, highlights unanswered questions regarding optimal combinations and proposes some suggested approaches for the personalization of initial therapy for AML patients.

Keywords

acute myeloid leukemia novel agents personalized therapy

Introduction

A wealth of knowledge on the genetic and molecular underpinnings of acute myeloid leukemia (AML) has provided the opportunity to develop personalized, targeted treatment approaches. Until recently, there were two mainstays of initial therapy for AML patients. A combination of cytarabine and anthracycline (referred to as ‘7+3’ in the most commonly used dosing schedule in the US), sometimes in combination with a third drug, has been used as an intensive induction regimen for many decades.¹ Alternatively, hypomethylating agents (HMA), such as decitabine and azacitidine, offer a less intensive option often considered for elderly patients or those otherwise unfit for aggressive induction chemotherapy.^1–3 Following advances in our understanding of AML biology, 2017 has set a milestone in finally broadening the therapeutic arsenal for AML. Four FDA drug approvals this year include midostaurin, enasidenib, gemtuzumab ozogamicin (GO) and CPX-351 (liposomal daunorubicin and cytarabine). In addition, US Food and Drug Administration (FDA) breakthrough and fast-track designations were granted to venetoclax, GMI-1271, and gilteritinib.

The hematology community is now faced with the challenge of how to optimally incorporate these novel treatments into everyday clinical practice. It is unknown how to treat patients who may be eligible for more than one approved novel agent, such as in the setting of initial treatment of patients with CD33 positive and fms-like tyrosine kinase 3 (FLT3)-mutated AML. Clinical trials assessing combinatorial and sequential treatment approaches utilizing both novel and traditional therapies are needed to improve the depths and durations of treatment responses, overcome drug resistance, and bridge more patients to an allotransplant where appropriate. Long-term sequelae of treatment toxicities, as in the case of GO-induced veno-occlusive disease (VOD), need be considered in patients who are allotransplant candidates. In this review, we detail recently approved therapeutics and agents in ongoing clinical trials with potential for future use as front-line therapeutics. Proposed approaches for incorporating novel agents into clinical practice are discussed (Figure 1).

Figure 1.

Proposed algorithm for personalizing initial therapy in acute myeloid leukemia.

Newly approved therapeutics

Midostaurin

Mutations in the FLT3 internal tandem duplication (ITD) juxtamembrane or tyrosine kinase domain (TKD) are late or secondary events in leukemogenesis, resulting in constitutive activation of downstream mitogenic signaling pathways, thus producing a proliferative leukemia.⁴ In the multicenter international phase III CALGB 10603 (RATIFY) double-blind randomized controlled trial, patients with newly diagnosed nontherapy-related FLT3 mutated AML received standard 7+3 with consolidative high-dose cytarabine alongside either midostaurin, an oral multitargeted kinase inhibitor, or placebo starting at day 8 of induction. Patients who remained in remission after consolidation received a full year of maintenance therapy with midostaurin or placebo. Maintenance therapy was not part of the protocol for the quarter of patients who underwent an allotransplant while in a CR1. After a median follow up of 59 months, overall survival, the study’s primary endpoint, was significantly longer in the midostaurin cohort than placebo [hazard ratio (HR) 0.78, p = 0.009]. Although patients harboring FLT3–TKD mutations did not derive an overall survival benefit from midostaurin, an event-free survival benefit was derived across all FLT3 mutation types encompassing ITD with low and high allelic ratios and TKD. A total of 57% of the study population went on to receive an allotransplant, more in the midostaurin arm than placebo arm, and the median overall survival among this population was not reached. When stratified by sex, females did not have a survival benefit with midostaurin, but males did; both sexes derived an event-free survival benefit from midostaurin.⁵ The midostaurin arm experienced more nausea, vomiting, skin rash and anemia, but overall the toxicities were manageable. As a result, midostaurin was granted full FDA approval for untreated FLT3 mutated AML in combination with standard induction and consolidation therapy. Use of prophylactic antiemetics are recommended. Midostaurin was not approved for single-agent induction or for maintenance therapy. The companion Invivoscribe LeukoStrat® FLT3 polymerase chain reaction (PCR) and gel electrophoresis-based mutation assay was simultaneously approved to detect ITD mutations of varying sizes, as well as TKD D835 or I836 mutations, to aid in patient selection. Rapid result turnaround time of 2–3 business days should allow midostaurin to be initiated by day 8 of induction. It is noteworthy that the daunorubicin dose in the RATIFY study was 60 mg/m² in both arms as several reports in the literature have supported the notion that 90 mg/m² is superior to both 60 mg/m² and idarubicin 12 mg/m² for patients harboring FLT3-ITD mutations.^6,7 The benefit of midostaurin in the context of high-dose daunorubicin is unknown.

Enasidenib

Mutations in isocitrate dehydrogenase 2 (IDH2) at hotspot residues R140 or R172 are acquired early in leukemogenesis, and by catalyzing alpha-ketoglutarate to 2-hydroxyglutarate they facilitate deoxyribonucleic acid (DNA) hypermethylation and subsequent silencing of genes responsible for cell differentiation.⁸ The prevalence of these IDH2 mutations in AML is around 9–10%, and approaches 20% among cytogenetically normal cases.^9,10 Enasidenib, an oral selective inhibitor of mutant IDH2 enzymes, is FDA approved for use in relapsed/refractory IDH2 mutated AML. In the landmark phase I/II study that resulted in its approval, enasidenib produced response rates of about 40% that were unfortunately short lived, lasting about 6 months. Median overall survival was 9.3 months among all comers but was markedly prolonged at 19.7 months among patients who achieved a complete response. Patients with R172 mutations achieved better treatment responses than those with R140 mutations (overall responses 53.3% versus 35.4%; complete responses 24.4% versus 17.7%). Meanwhile, 11% of the cohort could proceed to curative allotransplant. The most common grade 3–4 toxicities included differentiation syndrome (7%), non-infectious leukocytosis (6%), tumor lysis syndrome (6%) and diarrhea (8%).¹¹ There is an FDA black box warning for differentiation syndrome, which can present with rapid proliferation of differentiated granulocytes and can be accompanied by acute respiratory distress, hypoxia, pleural effusions, fever, peripheral edema, and multiorgan failure, and can be observed from 10 days to 5 months after initiation of enasidenib. Management consists of corticosteroids (typically dexamethasone 10 mg every 12 h) and close hemodynamic monitoring. Differentiation syndrome can be associated with hyperleukocytosis; per enasidenib approval guidelines, non-infectious leukocytosis can be treated with hydroxyurea per institutional practices. An accompanying Abbott RealTime IDH2 PCR assay for single nucleotide variants responsible for nine different nonsynonymous IDH2 mutations, with a turnaround time of about 1 week, was contemporaneously approved to identify patients eligible for enasidenib.

CPX-351 (liposomal daunorubicin and cytarabine)

Secondary AMLs, encompassing therapy-related myeloid neoplasms and AMLs arising from underlying myelodysplasia or myelofibrosis, render a poor prognosis with an estimated survival of <6 months using traditional therapies.¹² A patient’s total cumulative anthracycline dose, generalized organ reserve for intensive therapy, candidacy for a definitive allotransplant, and possibility of a germline predisposition are all important considerations in management of these patients. Liposomal daunorubicin and cytarabine (CPX-351) is a nanoliposomal combination of cytarabine and daunorubicin in a 5:1 molar ratio that is associated with prolonged in vivo drug exposure when compared with standard 7+3.¹³ It was granted full FDA approval for the treatment of secondary AML, encompassing therapy-related AML and AML with myelodysplasia-related changes, based on a multicenter phase III trial comparing CPX-351 with standard intensive induction and consolidation among patients aged 60–75 with secondary AMLs [ClinicalTrials.gov identifier: NCT01696084]. Enrollment was stratified by AML subtype: treatment-related AML, AML with preceding myelodysplastic syndrome (MDS) with prior HMA treatment, AML with preceding chronic myelomonocytic leukemia, and de novo AML with a karyotype characteristic for MDS. The primary endpoint, overall survival, was significantly improved in the CPX-351 arm (9.6 months versus 5.9 months, p = 0.005) when compared with standard 7+3. Response rates were higher (57% versus 40%) as well.¹⁴ Patients harboring FLT3 ITD or TKD mutations were more likely to achieve a complete response than those who were FLT3 wildtype.¹⁵ More patients in the CPX-351 cohort could undergo definitive allotransplant than in the 7+3 cohort (34% versus 25%), and had lower post-transplant mortality, making it an attractive bridge-to-transplant option.

Gemtuzumab ozogamicin

The CD33 antigen is a transmembrane receptor and myeloid differentiation marker that is expressed on most AMLs at varying degrees and on all acute promyelocytic leukemias in a high and homogenous fashion.¹⁶ Its extracellular immunoglobulin-like V domain is the immune dominant epitope that can be exploited with anti-CD33-directed therapy such as the antibody–drug conjugate GO. Importantly, this epitope is not present on all CD33+ AMLs due to alternative splicing,¹⁷ which may explain differential treatment effects. Granted accelerated approval in 2000 at a dose of 9 mg/m² for two doses 14 days apart for older adults with CD33+ AML in first relapse, GO was the first anticancer antibody–drug conjugate approved in the United States.¹⁸ It was later withdrawn from the commercial market in 2010 after the phase III SWOG S0106 study revealed lack of efficacy and increased deaths when used in combination with induction and consolidation therapy.¹⁹ Notably, patients allocated to the GO arm received a suboptimal dose of daunorubicin compared with standard treatment arm (45 mg/m² versus 60 mg/m²). After a long hiatus, it recently gained full FDA approval based on its efficacy using a hypofractionated schedule for CD33+ newly diagnosed or relapsed/refractory AML, and this approval is based on three studies. The multicenter French ALFA-0701 phase III study, randomized patients age 50–70 with de novo AML to receive standard induction and consolidation with or without five doses of hypofractionated GO administered at 3 mg/m². Patients with therapy-related AML or an antecedent bone marrow failure were excluded. In the final analysis, rates of complete responses did not differ between the GO and control arms, (81% versus 74%, p = 0.25), but event-free survival did (31% versus 19%, HR 0.66, confidence interval 0.50–0.87, p < 0.001), as did relapse-free survival (38% versus 25%, p = 0.006), favoring the GO arm. The benefit of GO was more pronounced in patients with favorable or intermediate cytogenetics and among those harboring FLT3-ITD mutations.²⁰ The AML-19 phase III study compared hypofractionated GO with best supportive care among elderly patients with de novo or secondary AML deemed ineligible for intensive chemotherapy, and GO was associated with 27% complete responses and a favorable 1-year overall survival of 24.3% versus 9.7%.²¹ Notably, myeloblast CD33 expression was not an eligibility requirement for the aforementioned two studies. Lastly, the MyloFrance1 phase II study assessed hypofractionated GO for adults with CD33+ AML in a first relapse, and demonstrated efficacy with 26% complete responses and an 11-month recurrence-free survival.²² A subsequent meta-analysis involving 3325 patients from five different studies revealed that GO reduced the 5-year risk of relapse and improved overall survival, and the benefits were most pronounced among those with favorable or intermediate-risk cytogenetics but not adverse-risk cytogenetics, and unlike the ALFA-0701 study, not patients harboring FLT3-ITD mutations.²³ Importantly, GO has an FDA black box warning for hepatotoxicity and VOD. The prevalence of VOD in the ALFA-0701 study was 5%, and half of the cases were fatal. The risk of VOD is heightened by higher doses of GO, pre-existing chronic liver disease, or when GO is used pre- or postallotransplant and hence the recommendation is for the hypofractionated schedule, and to wait at least 2 months between the GO administration and allotransplant.

Novel hypomethylating agents

Guadecitabine is a next-generation HMA associated with a long drug half-life due to inherent resistance to cytidine deaminase degradation. It has been shown to be efficacious in early-phase studies, especially for patients who are treatment naïve or have low-burden disease (blasts < 30%).^24,25 It is being studied in a phase III trial and compared with physician’s choice among those with relapsed/refractory AML [ClinicalTrials.gov identifier: NCT02920008] and in combination with the PD-L1 inhibitor atezolizumab [ClinicalTrials.gov identifier: NCT02892318]. Oral HMA therapy is being studied in the form of oral azacitidine (CC-486). Ongoing investigations assessing the efficacy of CC-486 as maintenance therapy after allotransplant [ClinicalTrials.gov identifier: NCT01835587], as maintenance therapy after induction and consolidation [ClinicalTrials.gov identifier: NCT01757535], and as monotherapy in the treatment-naïve or relapsed/refractory setting [ClinicalTrials.gov identifier: NCT00528983], are underway. If deemed efficacious, CC-486 has the very attractive potential of offering a completely oral AML regimen, especially if combined with other promising oral agents such as venetoclax, FLT3 inhibitors or poly adenosine diphosphate (ADP) ribose polymerase (PARP) inhibitors.

BCL-2 directed therapies

BCL-2 is an antiapoptotic mitochondrial protein that is commonly overexpressed in AML. Its overexpression increases leukemia fitness, renders intrinsic chemoresistance, and contributes to the survival of minimal residual quiescent leukemia stem cells responsible for disease relapse.^26,27 Venetoclax is an oral BCL-2 inhibitor with single-agent activity in relapsed/refractory AML.²⁸ It works synergistically with other agents by blocking antiapoptotic pathways, which enhances the antileukemia activity of partnered drugs. When combined with low-dose cytarabine among treatment-naïve elderly patients unfit for intensive therapy, it was associated with complete response rates of 62% with durability lasting over a year in a phase Ib/II study.²⁹ When partnered with decitabine or azacitidine, it produced similar efficacy with 66% complete responses and durability of about 1 year, and demonstrated activity among those with poor-risk cytogenetics.³⁰ Because of the promising activity demonstrated, larger phase III trials are underway [ClinicalTrials.gov identifiers: NCT03069352, NCT02993523]. Notably, venetoclax is also being studied in combination with FLAG-IDA in the upfront and relapsed/refractory setting [ClinicalTrials.gov identifier: NCT03214562]. This agent was granted FDA breakthrough designation in 2017 and will hopefully soon become part of the standard AML therapeutic armamentarium.

Novel FLT3 inhibitors

Quizartinib is an oral potent inhibitor of FLT3-ITD that has demonstrated single-agent efficacy in relapsed/refractory AML, irrespective of FLT3 mutation status.³¹ It is being studied in the placebo-controlled phase III QuANTUM trial assessing drug efficacy in combination with upfront induction and consolidation therapy [ClinicalTrials.gov identifier: NCT02668653]. Unfortunately, acquired resistance mechanisms to FLT3 inhibitors are common, resulting in short-lived responses. Secondary FLT3-TKD mutations at the D835 residue can confer resistance, as can activation of the receptor tyrosine kinase AXL.^32,33 Crenolanib is a potent FLT3 inhibitor that can overcome acquired resistance through D835 mutations.³⁴ It has single-agent activity in heavily pretreated AML,³⁵ and is being studied in combination with 7+3 in a phase II study [ClinicalTrials.gov identifier: NCT02283177], and in a phase III study comparing it with midostaurin when used in combination with intensive induction [ClinicalTrials.gov identifier: NCT03258931]. Gilteritinib is a highly selective and potent dual FLT3/AXL inhibitor that appears more attractive than quizartinib and crenolanib in that it can circumvent acquired FLT3 inhibitor resistance through both AXL and D835 pathways. It was recently granted fast-track designation for AML. It has single-agent efficacy in relapsed/refractory AML irrespective of FLT3 mutational status.³⁶ As with crenolanib, it’s also being studied in combination with intensive induction and consolidation therapy [ClinicalTrials.gov identifier: NCT02236013] and with azacitidine [ClinicalTrials.gov identifier: NCT02752035].

IDH 1/2 directed therapies

Enasidenib is being studied as upfront monotherapy for older patients with IDH2-mutated AML [ClinicalTrials.gov identifiers: NCT01915498, NCT03013998]. A preliminary report in this setting revealed a response rate of 38%, including 19% complete responses and a median duration of response of about 1 year.³⁷ Among patients who are eligible for intensive therapy, ivosidenib and enasidenib are being studied in combination with intensive induction and consolidation therapy among newly diagnosed AML patients harboring IDH1 and IDH2 mutations, respectively [ClinicalTrials.gov identifier: NCT02632708]. Preliminary data from this phase I study are promising, revealing complete response rates (which included morphological complete remission (CR) and CR with incomplete blood count recovery (CRi)) of 86% and 67% among those with IDH1 and IDH2 mutated de novo AML, respectively.³⁸ High rates of complete responses were also appreciated among those with secondary AML. Analyses of depths of responses in the form of minimal residual disease assessments and duration of responses are underway. Inhibitors of IDH have been shown to produce synergy with azacitidine, as both drugs reduce DNA methylation and circumvent blocks in cell differentiation.³⁹ A phase III study evaluating combination azacitidine with ivosidenib for patients unfit for intensive therapy harboring IDH1 mutations is underway [ClinicalTrials.gov identifier: NCT03173248].

Novel CD33-directed therapies

Novel CD33-directed therapies are being avidly investigated in early-phase studies, including a novel CD33 antibody–drug conjugate, IMGN779, for relapsed/refractory CD33+ AML [ClinicalTrials.gov identifier: NCT02674763], allogeneic CD33 CAR-T cells [ClinicalTrials.gov identifier: NCT03126864], and CD33-bispecific T-cell engagers [ClinicalTrials.gov identifier: NCT03144245]. Vadastuximab talirine (SGN-33A) is an antibody–drug complex with an engineered antibody linked to a more potent cytotoxic agent with the ability to bypass leukemia drug efflux pumps,^40,41 however, development of this agent does not appear to be ongoing at this time, exemplified by the recent discontinuation of phase III CASCADE trial for frontline older AML patients due to increased deaths compared with the control arm.

Poly adenosine diphosphate (ADP) ribose polymerase inhibitors

Leukemia cells can acquire deficits in double strand DNA repair, namely in homologous recombination and nonhomologous end joining, which may be exploited by PARP inhibitors.³⁵ Such deficits can be intrinsic to an AML with a hypermethylated genome due to diminished expression of DNA damage response genes, such as in the setting of IDH, TET2 or DNMT3A mutations.^42–44 PARP inhibitors are being studied in combination with decitabine for untreated and relapsed/refractory AML [ClinicalTrials.gov identifier: NCT02878785], as well as in combination with conventional chemotherapy [ClinicalTrials.gov identifier: NCT03289910] and temozolomide [ClinicalTrials.gov identifier: NCT01139970].

CD123 directed therapy

CD123 is an IL-3 receptor present on myeloblasts that when bound by its ligand, induces cell differentiation and proliferation via downstream signaling of the JAK/STAT and PI3K/MAPK pathways. Unlike hematopoietic stem cells which have low to absent CD123 expression, AML cells, including quiescent leukemia stem cells, have high expression and thus lend opportunity to study CD123 as a therapeutic target.⁴⁵ Anti-CD123-directed therapies are being studied in multiple settings, including relapsed/refractory AML, de novo AML, and for eradication of minimal residual disease. Flotetuzumab is a novel CD123 × CD3 bispecific dual-affinity retargeting (DART) molecule that has demonstrated antileukemia activity in the relapsed/refractory setting in a phase I study, with an overall response rate of 43% among 14 patients and with a tolerable toxicity profile consisting of mainly cytokine release syndrome [ClinicalTrials.gov identifier: NCT02152956].⁴⁶ Ongoing studies are evaluating the safety and efficacy of other CD123-directed agents, to include antibody–drug conjugates [ClinicalTrials.gov identifier: NCT02848248], chimeric antigen receptor T cells [ClinicalTrials.gov identifiers: NCT03222674, NCT02159495], and bispecific T-cell engagers [ClinicalTrials.gov identifier: NCT02730312].

E-selectin-directed therapy

The interaction between the adhesion molecule E-selectin with its receptor in the bone marrow vascular niche represents one mechanism where the marrow microenvironment can facilitate leukemia stem cell fitness, chemoresistance, and quiescence.⁴⁷ GMI-1271 is a novel E-selectin antagonist that has been granted FDA breakthrough designation for use in AML. In a phase I study, the combination of GMI-1271 with 7+3 or intensive salvage chemotherapy produced synergistic efficacy, producing overall response rates of 80% (68% CR/CRi) and 50% (41% CR/CRi) for treatment-naïve and relapsed/refractory AML, respectively [ClinicalTrials.gov identifier: NCT02306291].⁴⁸

Vosaroxin

Vosaroxin is a quinolone-derived nonanthracycline topoisomerase II inhibitor that can circumvent drug efflux pumps and induce apoptosis, and has shown single-agent activity in relapsed/refractory AML.⁴⁹ In a phase II study combining it with decitabine for treatment-naïve older patients, it produced an overall response rate of 74%, 48% of which were complete responses, and of these, 54% were measurable residual disease (MRD) negative [ClinicalTrials.gov identifier: NCT03338348]. When used at a lower dose of of 70 mg/m² there was less mucositis with similar efficacy. Vosaroxin alongside decitabine demonstrated efficacy in patients with prior HMA therapy and P53 mutations.⁵⁰ Vosaroxin is being studied in combination with infusional cytarabine in lieu of an anthracycline for initial intensive induction therapy [ClinicalTrials.gov identifier: NCT02658487], preliminarily with good response rates of 71%, including among those with P53 mutations.⁵¹ Unfortunately, despite being previously granted orphan drug designation in the US and EU, the phase III VALOR trial of this agent in combination with cytarabine in the relapsed or refractory AML setting was unsuccessful in demonstrating any overall survival benefit over cytarabine plus placebo in this patient population, despite a statistically significant difference in CR (30% versus 16%, p < 0.0001) and CR/CRi (37% versus 18%, p < 0.0001) responses.

Immune-checkpoint inhibition

Despite having a relatively low mutational burden in comparison with their solid tumor counterparts, AML cells have high PD-L1 expression, which increases further in the relapsed/refractory setting because of immune exhaustion.⁵² This, in conjunction with the fact that immunotherapy, in the form of allotransplant, can be effective at disease eradication, provides rationale for the study of checkpoint inhibitors in AML. An early-phase study assessed the efficacy of nivolumab with azacitidine in the relapsed or refractory setting, and revealed CR/CRi rates of 22% and a median overall survival of 15 months among responders.⁵³ A study assessing the efficacy of pembrolizumab with 10-day decitabine for relapsed/refractory AML has completed accrual [ClinicalTrials.gov identifier: NCT02996474], and trials of pembrolizumab in combination with other agents (for example [ClinicalTrials.gov identifier: NCT02845297]) are underway. In the upfront setting, feasibility of nivolumab alongside 7+3 for newly diagnosed AML patients was assessed in a phase II study, and revealed a CR/CRi rate of 72%, the median duration of which was not reached after 8 months of follow up. Patients proceeding to allotransplant did not have an increase in graft versus host disease.⁵⁴

Incorporating novel agents into clinical practice

Providers will inevitably face challenges when incorporating novel agents into daily clinical practice. A proposed algorithm for frontline treatment of newly diagnosed AML is presented in the Figure 1.^55–62 Clinical classification as de novo or secondary AML, expression of CD33 on AML blasts, and the mutational profile must be included in the initial assessment of newly diagnosed AML. Although a leukemia’s morphological features and immunophenotype can be determined relatively quickly, testing for genetic aberrations can often take days to weeks to process, which has encouraged the codevelopment of companion diagnostics that can be performed rapidly for many of the newly approved and currently tested drugs. While next-generation sequencing offers many advantages, until more rapid and standardized sequencing methods become available, approved companion diagnostics such as the LeukoStrat® test, with a processing time of as little as 2 days, will remain essential for treatment selection. While ‘myeloid panels’ sequencing dozens of recurrently mutated genes or gene regions in AML are now widely available within institutions and commercially, they can be fraught with limitations, including lengthy turnaround time, nonstandardized variant calling pipelines, and an inability to detect large FLT3-ITD duplications.

The presence of MRD after treatment for AML, representing selection of a chemoresistant or quiescent malignant clone, is increasingly recognized as a clinically important factor.⁶³ Successful approaches to eradicate MRD have not been established. Detectable MRD before allotransplant portends inferior outcomes and increased death.^64–66 RELAZA2 was a single-arm study assessing risk-adapted azacitidine to prevent hematologic relapse in patients with MDS or AML in a complete remission after completion of standard therapy who developed MRD. Of the 53 patients who developed MRD and were treated with empiric azacitidine, 40% achieved a decline of MRD to below a predefined threshold, while 19% experienced stabilization of their MRD. It was concluded that risk-adaptive azacitidine might prevent or delay a hematologic relapse; however, definitive conclusions are difficult to be drawn without a comparator arm.⁶⁷ In addition to randomized studies to determine utility of treating at MRD state,⁶⁸ future studies may also include drug sensitivity and resistance testing to guide therapy choice for such MRD eradication.

It is unknown how to best treat newly diagnosed patients who may be eligible for more than one approved novel therapeutic approach. For example, FLT3 mutations are often associated with high CD33 expression,⁶⁹ and a provider may have difficulty choosing between a GO or midostaurin-based regimen. Both midostaurin and GO improve survival, and studies have demonstrated that FLT3-mutated patients fare better treatment responses to GO than FLT3-wildtype patients.⁷⁰ Knowledge of a FLT3 mutation may be received after a GO-based intensive induction regimen has already been initiated. If knowledge of a FLT3 mutation is discovered after initiation of sole 7+3, if high-dose daunorubicin is used (90 mg/m²) then the safety and efficacy of adding midostaurin on day 8 is not known. In the absence of clinical trials comparing midostaurin with GO head to head, there is limited evidence of best practice in this situation. GO is more efficacious among patients with favorable or intermediate-risk karyotypes and with a higher expression of CD33, and the risk of VOD might be heightened by underlying chronic liver disease or a myeloablative conditioning regimen within 2–3 months of GO administration. A similar clinical conundrum might involve an elderly patient with secondary AML, whose candidacy for intensive induction therapy is borderline, and whose AML harbors a p53 mutation; there are no data to suggest preference of CPX-351 over 10-day decitabine.⁷¹ Similar uncertainties may arise when and if IDH inhibitors or venetoclax become approved in the upfront setting.

It is also not known how to combine or sequence novel agents and traditional therapies. Optimal combinations and sequences have the potential to deepen response depths and durations, to overcome drug resistance, to bridge patients to an allotransplant, to offer completely oral AML treatments, and to improve overall survival. For example, secondary AMLs are associated with mutated FLT3 or IDH up to 10–20% of the time, as well as CD33 overexpression,^72,73 and it is possible that combinatorial or sequential CPX-351 with FLT3, IDH, or CD33-directed therapies may have cytotoxic synergy as demonstrated in preclinical studies. Oral drugs being studied in clinical trials such as venetoclax, oral azacitidine, second-generation FLT3 inhibitors, and PARP inhibitors have the very unique potential of being combined to offer a completely oral regimen for AML, which may offer clinically meaningful improvements in quality of life for many patients.

There is no adequate standard of care for newly diagnosed non-acute promyelocytic leukemia (APL) AML patients, hence clinical trial enrollment should be considered for all patients. The BeatAML master protocol [ClinicalTrials.gov identifier: NCT03013998] is an adaptive trial design utilizing biomarker-driven therapies, with the primary endpoint of assessing the proportion of patients for whom molecular, immunophenotypic and biochemical studies can be completed within 7 calendar days. This important study, as well as others, are essential to improve initial treatment strategies for AML.

Conclusion

Key points and ongoing unanswered questions are presented in Table 1. An improved understanding of the genetic etiology and clonal architecture in AML has laid important groundwork to study personalized therapy for this heterogenous disease. The past year alone has witnessed approvals of four drugs for AML, with more approved therapeutics hopefully coming soon. Incorporating these novel agents into daily clinical practice will pose challenges requiring patient enrollment onto clinical trials as well as experience as a community to learn how to maximize these exciting new opportunities to improve patient outcomes.

Table 1.

Keypoints and unknowns in the initial therapy for acute myeloid leukemia.

Keypoints
(1) Increased knowledge on the biology of AML has laid the framework for personalizing AML therapy
(2) Approved companion genetic tests can provide time-sensitive information in just several days to guide patient selection for novel therapies
(3) Combinations of oral agents such as oral azacitidine, venetoclax, FLT3 inhibitors, and PARP inhibitors offer the potential for infusion-free AML treatments
(4) Due to ongoing uncertainties on how to best combine or sequence therapies and how to incorporate biomarkers into treatment algorithms, clinical trial enrollment should be considered for all patients
Unknowns
(1) It is not known how to standardize genetic testing in clinical practice
(2) It is not known how to optimally treat a patient who may be a candidate for more than one novel therapy, such as a patient with CD33+- and FLT3-mutated AML
(3) It is not known how to best combine or sequence novel agents with traditional therapies

AML, acute myeloid leukemia; PARP, poly adenosine diphosphate (ADP) ribose polymerase.

Footnotes

Acknowledgements

The authors thank Karolyn Oetjen MD PhD for comments.

Funding

This work was supported by the Intramural Research Program of the National Heart, Lung, and Blood Institute (NHLBI) of the National Institutes of Health.

Conflict of interest statement

CSH receives research funding from Merck Sharpe and Dohme and SELLAS Life Sciences Group AG.

References

Yates

Wallace

Ellison

et al . Cytosine arabinoside (NSC-63878) and daunorubicin (NSC-83142) therapy in acute nonlymphocytic leukemia. Cancer Chemother Rep 1973; 57: 485–488.

Celgene Corporation. Vidaza (azacitidine) (package insert). Bedford, OH: Celgene Corporation, https://www.accessdata.fda.gov/drugsatfda_docs/label/2008/050794s011lbl.pdf (2017, accessed August 2008).

Pharmachemie B.V. Haarlem. Dacogen® (decitabine) (package insert). Woodcliff Lake, NJ: Pharmachemie B.V. Haarlem, https://www.accessdata.fda.gov/drugsatfda_docs/label/2010/021790s006lbl.pdf (2017, accessed March 2010).

Welch

Ley

Link

et al . The origin and evolution of mutations in acute myeloid leukemia. Cell 2012; 150: 264–278.

Stone

Mandrekar

Sanford

et al . Midostaurin plus chemotherapy for acute myeloid leukemia with a FLT3 mutation. N Engl J Med 2017; 377: 454–464.

Burnett

Russell

Hills

et al . A randomized comparison of daunorubicin 90 mg/m² vs 60 mg/m² in AML induction: results from the UK NCRI AML 17 trial in 1206 patients. Blood 2015; 125: 3878–3885.

Lee

Kim

Joo

et al . Prospective randomized comparison of idarubicin and high-dose daunorubicin in induction chemotherapy for newly diagnosed acute myeloid leukemia. J Clin Oncol 2017; 35: 2754–2763.

Medeiros

Fathi

DiNardo

et al . Isocitrate dehydrogenase mutations in myeloid malignancies. Leukemia 2017; 31: 272–281.

Abbas

Lugthart

Kavelaars

et al . Acquired mutations in the genes encoding IDH1 and IDH2 both are recurrent aberrations in acute myeloid leukemia: prevalence and prognostic value. Blood 2010; 116: 2122–2126.

10.

Marcucci

Maharry

et al . IDH1 and IDH2 gene mutations identify novel molecular subsets within de novo cytogenetically normal acute myeloid leukemia: a Cancer and Leukemia Group B study. J Clin Oncol 2010; 28: 2348–2355.

11.

Stein

DiNardo

Pollyea

et al . Enasidenib in mutant-IDH2 relapsed or refractory acute myeloid leukemia. Blood 2017; 130: 722–731.

12.

Fianchi

Pagano

Piciocchi

et al . Characteristics and outcome of therapy-related myeloid neoplasms: report from the Italian network on secondary leukemias. Am J Hematol 2015; 90: E80–E85.

13.

Nikanjam

Capparelli

Lancet

et al . Persistent cytarabine and daunorubicin exposure after administration of novel liposomal formulation of CPX-351: population pharmacokinetic assessment. Cancer Chemother Pharmacol 2018; 81: 171–178.

14.

Lancet

Hoering

et al . Survival following allogeneic hematopoietic cell transplantation in older high-risk acute myeloid leukemia patients initially treated with CPX-351 liposome injection versus standard cytarabine and daunorubicin: subgroup analysis of a large phase III trial. In: Proceedings of the American Society of Hematology conference , San Diego, CA, 5 December 2016. Washington, DC: Blood, http://www.bloodjournal.org/content/128/22/906.

15.

Lancet

Cortes

et al . CPX-351 treatment of previously untreated older AML patients with high risk AML markedly increases the response rate over 7+3 in patients with FLT3 mutations. In: Proceedings of the European Hematology Association meeting, Copenhagen, Denmark, 11 June 2016. Pavia, Italy: Haematologica, https://learningcenter.ehaweb.org/eha/2016/21st/135258/jeffrey.lancet.cpx-351.treatment.of.previously.untreated.older.aml.patients.html?f=m3.

16.

Godwin

Gale

Walter

. Gemtuzumab ozogamicin in acute myeloid leukemia. Leukemia 2017; 31: 1855–1868.

17.

Appelbaum

Bernstein

. Gemtuzumab ozogamicin for acute myeloid leukemia. Blood. Epub ahead of print 11 October 2017. DOI: 10.1182/blood-2017-09-797712.

18.

Bross

Beitz

Chen

et al . Approval summary: gemtuzumab ozogamicin in relapsed acute myeloid leukemia. Clin Cancer Res 2001; 7: 1490–1496.

19.

Petersdorf

Kopecky

Slovak

et al . A phase 3 study of gemtuzumab ozogamicin during induction and postconsolidation therapy in younger patients with acute myeloid leukemia. Blood 2013; 121: 4854–4860.

20.

Castaigne

Pautas

Terre

et al . Final analysis of the ALFA 0701 study. In: Proceedings of the American Society of Hematology conference, San Francisco, CA, 4 December 2014. Washington DC: Blood, http://www.bloodjournal.org/content/124/21/376.

21.

Amadori

Suciu

Selleslag

et al . Gemtuzumab ozogamicin versus best supportive care in older patients with newly diagnosed acute myeloid leukemia unsuitable for intensive chemotherapy: results of the randomized phase III EORTC-GIMENA AML-19 trial. J Clin Oncol 2016; 34: 972–979.

22.

Taksin

Legrand

Raffoux

et al . High efficacy and safety profile of fractionated doses of Mylotarg as induction therapy in patients with relapsed acute myeloblastic leukemia: a prospective study of the alfa group. Leukemia 2007; 21: 66–71.

23.

Hills

Castaigne

Appelbaum

et al . Addition of gemtuzumab ozogamicin to induction chemotherapy in adult patients with acute myeloid leukemia: a meta-analysis of individual patient data from randomized controlled trials. Lancet Oncol 2014; 15: 986–996.

24.

Kantarjian

Roboz

Kropf

et al . Predictors of response and survival in 206 AML patients treated with guadecitabine in a phase 2 study. In: Proceedings of the American Society of Hematology conference, Atlanta, GA, 11 December 2017. Washington DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper104053.html.

25.

Issa

Roboz

Rizzieri

et al . Safety and tolerability of guadecitabine (SGI-110) in patients with myelodysplastic syndrome and acute myeloid leukemia: a multicenter, randomized, dose-escalation phase 1 study. Lancet Oncol 2015; 16: 1099–1110.

26.

Lagadinou

Sach

Callahan

et al . BCL-2 inhibition targets oxidative phosphorylation and selectively eradicates quiescent human leukemia stem cells. Cell 2013; 12: 329–341.

27.

Campos

Rouault

Sabido

et al . High expression of bcl-2 protein in acute myeloid leukemia cells is associated with poor response to chemotherapy. Blood 1993; 81: 3091–3096.

28.

Konopleva

Pollyea

Potluri

et al . Efficacy and biological correlates of response in a phase II study of venetoclax monotherapy in patients with acute myelogenous leukemia. Cancer Discov 2016; 6: 1106–1117.

29.

Wei

Strickland

Roboz

et al . Phase 1/2 study of venetoclax with low-dose cytarabine in treatment-naïve, elderly patients with acute myeloid leukemia unfit for intensive chemotherapy: 1-year outcomes. In: American Society of Hematology conference, Atlanta, GA, December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper100832.html.

30.

DiNardo

Pollyea

Jones

et al . Updated safety and efficacy of venetoclax with decitabine or azacitidine in treatment-naïve, elderly patients with acute myeloid leukemia. In: Proceedings from the American Society of Hematology conference, Atlanta, GA, 10 December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper100714.html.

31.

Cortes

Kantarjian

Foram

et al . Phase I study of quizartinib administered daily to patients with relapsed or refractory acute myeloid leukemia irrespective of FMS-like tyrosine kinase 3-internal tandem duplication status. J Clin Oncol 2013; 31: 3681–3687.

32.

Smith

Lin

Stecula

et al . FLT3 D835 mutations confer differential resistance to type II FLT3 inhibitors. Leukemia 2015; 29: 2390–2392.

33.

Park

Mundy-Bosse

Wihtman

et al . Receptor tyrosine kinase Axl is required for resistance of leukemic cells to FLT3-targeted therapy in acute myeloid leukemia. Leukemia 2015; 29: 2382–2389.

34.

Galanis

Rajkhowa

et al . Crenolanib is a potent inhibitor of FLT3 with activity against resistance-conferring point mutations. Blood 2014; 123: 94–100.

35.

Randhawa

Kantarjian

Borthakur

et al . Results of a phase II study of crenolanib in relapsed/refractory acute myeloid leukemia patients (pts) with activating FLT3 mutations. In: Proceedings from the American Society of Hematology conference, San Francisco, CA, 6 December 2014. Washington, DC: Blood, http://www.bloodjournal.org/crossref-forward-links/bloodjournal/124/21/389?sso-checked=true.

36.

Perl

Altman

Cortes

et al . Selective inhibition of FLT3 by gilteritinib in relapsed or refractory acute myeloid leukaemia: a multicenter, first-in-human, open-label, phase 1–2 study. Lancet Oncol 2017; 18: 1061–1075.

37.

Pollyea

Tallman

De Botton

et al . Enasidenib monotherapy is effective and well-tolerated in patients with previously untreated mutant-IDH2 (mIDH2) acute myeloid leukemia. In: Proceedings from the American Society of Hematology conference, Atlanta, GA, 11 December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper104650.html.

38.

Stein

DiNardo

Mims

et al . Ivosidenib or enasidenib combined with standard induction chemotherapy is well tolerated and active in patients with newly diagnosed AML with an IDH1 or IDH2 mutation: initial results from a phase 1 trial. In: Proceedings from the American Society of Hematology conference, Atlanta, GA, 11 December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper102244.html.

39.

Fujii

Khawaja

DiNardo

et al . Targeting isocitrate dehydrogenase (IDH) in cancer. Discov Med 2016; 21: 373–380.

40.

Godwin

Gale

Walter

. Gemtuzumab ozogamicin in acute myeloid leukemia. Leukemia 2017; 31: 1855–1868.

41.

Godwin

McDonald

Walter

. Sinusoidal obstruction syndrome following CD33-targeted therapy in acute myeloid leukemia. Blood 2017; 129: 2330–2332.

42.

Nieborowska-Skorska

Sullivan

Dasgupta

et al . Gene expression and mutation-guided synthetic lethality eradicates proliferating and quiescent leukemia cells. J Clin Invest 2017; 127: 2392–2406.

43.

Gafencu

Tomuleasa

Ghiaur

. PARP inhibitors in acute myeloid leukaemia therapy: how a synthetic lethality approach can be a valid therapeutic alternative. Med Hypotheses 2017; 104: 30–34.

44.

Sulkowski

Corso

Robinson

et al . 2-Hydroxyglutarate produced by neomorphic IDH mutations suppresses homologous recombination and induces PARP inhibitor sensitivity. Sci Transl Med 2017; 9: 1–49.

45.

Al-Hussaini

Rettig

Ritchey

et al . Targeting CD123 in acute myeloid leukemia using a T-cell-directed dual-affinity retargeting platform. Blood 2016; 127: 122–131.

46.

Godwin

Rettig

et al . Preliminary results of a phase 1 study of flotetuzumab, a CD123 × CD3 bispecific dart protein, in patients with relapsed/refractory acute myeloid leukemia and myelodysplastic syndrome. In: Proceedings from the American Society of Hematology conference, Atlanta, GA, 9 December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper101886.html.

47.

Winkler

Barbier

Nowlan

et al . Vascular niche E-selectin regulates hematopoietic stem cell dormancy, self-renewal and chemoresistance. Nat Med 2012; 18: 1651–1657.

48.

DeAngelo

Jonas

Liesveld

et al . GMI-1271 improves efficacy and safety of chemotherapy in R/R and newly diagnosed older patients with AML: results of a phase 1/2 study. In: Proceedings from the American Society of Hematology conference, Atlanta, GA, 11 December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper100360.html.

49.

Stuart

Cripe

Maris

et al . REVEAL-1, a phase 2 dose regimen optimization study of vosaroxin in older poor-risk patients with previously untreated acute myeloid leukemia. Br J Haematol 2015; 168: 796–805.

50.

Daver

Kantarjian

Garcia-Manero

et al . Vosaroxin in combination with decitabine in newly diagnosed older patients with acute myeloid leukemia or high-risk myelodysplastic syndrome. Haematologica 2017; 102: 1709–1717.

51.

Strickland

Podoltsev

Mohan

et al . Phase II trial of vosaroxin and infusional cytarabine for frontline treatment of acute myeloid leukemia (VITAL) patients eligible for intensive chemotherapy: a planned interim analysis. In: Proceedings from the American Society of Hematology conference, Atlanta, GA, 11 December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper108600.html.

52.

Masarova

Kantarjian

Garcia-Mannero

et al . Harnessing the immune system against leukemia: monoclonal antibodies and checkpoint strategies for AML. Adv Exp Med Biol 2017; 996: 73–95.

53.

Daver

Garcia-Manero

Basu

et al . Nivolumab (Nivo) with azacytidine (Aza) in patients (pts) with relapsed acute myeloid leukemia (AML) or frontline elderly AML. In: Proceedings from the American Society of Hematology conference, Atlanta, GA, 9 December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper108402.html.

54.

Ravandi

Daver

Garcia-Manero

et al . Phase 2 study of combination of cytarabine, idarubicin, and nivolumab for initial therapy of patients with newly diagnosed acute myeloid leukemia. In: Proceedings from the American Society of Hematology conference, 11 December 2017, Atlanta, GA. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper101563.html.

55.

Arber

George

Sojitra

. ASH-CAP guidelines for the diagnosis of acute leukemia, http://www.hematology.org/Thehematologist/Mini-Review/7120.aspx (2017, accessed 30 November 2017).

56.

Altman

Rademaker

Cull

et al . Administration of ATRA to newly diagnosed patients with acute promyelocytic leukemia is delayed contributing to early hemorrhagic death. Leuk Res 2013; 37: 1004–1009.

57.

Krug

Rollig

Koschmieder

et al . Complete remission and early death after intensive chemotherapy in patients aged 60 years or older with acute myeloid leukaemia: a web-based application for prediction of outcomes. Lancet 2010; 376: 2000–2008.

58.

Klepin

. Definition of unfit for standard acute myeloid leukemia therapy. Curr Hematol Malig Rep 2016; 11: 537–544.

59.

Dombret

Seymour

Butrym

et al . International phase 3 study of azacitidine vs conventional care regimens in older patients with newly diagnosed AML with >30% blasts. Blood 2015; 126: 291–299.

60.

Short

Kantarjian

Garcia- Manero

et al . A randomized phase II trial of 5-day versus 10-day schedules of decitabine for older patients with previously untreated acute myeloid leukemia. In: Proceedings from the American Society of Hematology conference, Atlanta, GA, 10 December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper101298.html.

61.

Rollig

Middeke

Stasik

et al . Real world data on decitabine treatment in 296 patients with acute myeloid leukemia: outcome and impact of TP53 mutations. In: Proceedings from the American Society of Hematology conference, Atlanta, GA, 10 December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper106924.html.

62.

Dohner

Estey

Grimwade

et al . Diagnosis and management of AML in adults: 2017 ELN recommendations from an international expert panel. Blood 2017; 129: 424–447.

63.

Hourigan

Gale

Gormley

et al . Measurable residual disease testing in acute myeloid leukaemia. Leukemia 2017; 31: 1482–1490.

64.

Buckley

Wood

Othus

et al . Minimal residual disease prior to allogeneic hematopoietic cell transplantation in acute myeloid leukemia: a meta-analysis. Haematologica 2017; 102: 865–873.

65.

Araki

Wood

Othus

et al . Allogeneic hematopoietic cell transplantation for acute myeloid leukemia: time to move toward a minimal residual disease-based definition of complete remission? J Clin Oncol 2016; 34: 328–336.

66.

Hourigan

Goswami

Battiwalla

et al . When the minimal becomes measurable. J Clin Oncol 2016; 34: 2557.

67.

Platzbecker

Middeke

Sockel

et al . Minimal-residual disease guided treatment with azacitidine in MDS/AML patients at imminent risk of relapse: results of the prospective RELAZA2 trial. In: Proceedings from the American Society of Hematology conference, Atlanta, GA, 11 December 2017. Washington, DC: Blood, https://ash.confex.com/ash/2017/webprogram/Paper101383.html.

68.

Ehninger

Kramer

Rollig

et al . Distribution and levels of cell surface expression of CD33 and CD123 in acute myeloid leukemia. Blood Cancer J 2014; 4: e218.

69.

Olombel

Guerin

Guy

et al . The level of blast CD33 expression positively impacts the effect of gemtuzumab ozogamicin in patients with acute myeloid leukemia. Blood 2016; 127: 2157–2160.

70.

Daver

Strati

Jabbour

et al . FLT3 mutations in myelodysplastic syndrome and chronic myelomonocytic leukemia. Am J Hematol 2013; 88: 56–59.

71.

Welch

Petti

Miller

et al . TP53 and decitabine in acute myeloid leukemia and myelodysplastic syndromes. N Engl J Med 2016; 375: 2023–2036.

72.

McNerney

Godley

Le Beau

. Therapy-related myeloid neoplasms: when genetics and environment collide. Nat Rev Cancer 2017; 17: 513–527.

73.

McKenney

Levine

. Isocitrate dehydrations mutations in leukemia. J Clin Invest 2013; 123: 3672–3677.

Personalizing initial therapy in acute myeloid leukemia: incorporating novel agents into clinical practice

Abstract

Keywords

Introduction

Newly approved therapeutics

Midostaurin

Enasidenib

CPX-351 (liposomal daunorubicin and cytarabine)

Gemtuzumab ozogamicin

Novel hypomethylating agents

BCL-2 directed therapies

Novel FLT3 inhibitors

IDH 1/2 directed therapies

Novel CD33-directed therapies

Poly adenosine diphosphate (ADP) ribose polymerase inhibitors

CD123 directed therapy

E-selectin-directed therapy

Vosaroxin

Immune-checkpoint inhibition

Incorporating novel agents into clinical practice

Conclusion

Footnotes

Acknowledgements

Funding

Conflict of interest statement

References