Advances in treatment formulations for acute myeloid leukemia

(1)

Teaser This review presents current treatments generally used in acute myeloid leukemia therapy, and highlights innovative formulations designed to overcome drug resistance or adverse effects observed with current treatments.

Advances in treatment formulations for acute myeloid leukemia

Thomas Briot

^1,2

, Emilie Roger

¹

, Sylvain Thépot

^3,4

and Frederic Lagarce

^1,2

1Micro&NanomédecinesTranslationelles-MINT,UNIVAngers,INSERM1066,CNRS6021,UniversitéBretagne Loire,MINTIBS-CHU,4rueLarrey,49933Angers,France

2UniversityHospitalofAngers,PharmacyDepartment,4rueLarrey,49933Angers,France

3UniversityHospitalofAngers,ClinicalHematologyDepartment,4rueLarrey,49933Angers,France

4INSERMCRCINA,UniversityofAngers,4rueLarrey,49933Angers,France

Acute myeloid leukemia (AML) is the most common cause of leukemia- related mortality. The combination of cytarabine and anthracycline has been the gold standard of treatment over the past 40 years, but the distribution of the drugs in the body leads to severe adverse effects. Poor prognosis of older patients with AML is the consequence not only of comorbidities, but also of chemoresistance resulting from frequent secondary AML. Numerous strategies using nanotechnologies are in development to improve drug targeting, pharmacokinetics,

administration route, chemoresistance, and adverse effects generally observed. Among the four new drugs approved for AML by the US Food and Drug Administration (FDA) in 2017, Vyxeos

¹

is a novel liposomal

formulation of historical AML drugs. Here, we review current AML treatments and discuss how the development of new formulations will change the therapeutic armamentarium.

Introduction

AML is the most common acute leukemia in adults, and its incidence increases with age.

Although the disease is usually of unknown etiology, it can develop following either an underlying hematological disorder (e.g., myelodysplastic syndromes, MDS) or exposure to genotoxicagents(e.g.,topoisomerasesII,alkylatingagents,orradiation).Thepathogenesisof AMLinvolves successive genomicalterationsin multipotential hematopoieticstem cellsand leadstotheabnormalclonalproliferationanddifferentiationofmyeloidcells.

AMLmainlyaffectsadultpatients,withamedianageof64years[1].Prognosisremainspoor, especiallyinolderpatients,withfewerthan30%ofpatientssurvivingwithinayearofdiagnosis[2].

Althoughourunderstandingofthediseaseaswellasitsbiologicalcharacteristicshasadvanced inrecentdecades,inductiontherapyhaschangedlittlesince1973.However,the therapeutic

ReviewsKEYNOTEREVIEW

ThomasBriotisa hospitalpharmacistandis alsocurrentlystudyingfor hisPhDinpharmaceutical technologyintheINSERM U1066/CNRS6021Unitat theUniversityofAngers.

Hismaininterestsaredrug formulations,analytical

qualitycontrol,newdrugdeliverysystems (nanomedicines),andoraldeliverysystems.

FredericLagarce receivedhisPhDin2004, andhasbeenafull professorof

pharmaceuticaltechnology andbiopharmaceuticsat theUniversityofAngersin Francesince2012.Heis alsoahospitalpharmacist.

Hisresearchfocusesontranslationalproducts(from benchtobedside)andismainlyfocusedoncancer therapy,especiallyonbioavailabilityenhancementby exploitingtheinteractionsbetweendrugproducts (mainlynanosystems)andlivingtissues.

Correspondingauthor:Lagarce,F. (frederic.lagarce@univ-angers.fr)

1359-6446/ã2018ElsevierLtd.Allrightsreserved.

1

(2)

landscapeofAMLdramaticallychangedin2017withfournovel drugs approvedby the FDA.Among them,Vyxeos¹(Jazz Phar- maceuticals)isanovelliposomalformulationofaco-encapsula- tionofdaunorubicinandcytarabine.

TheprincipaladverseeffectgenerallyobservedinAMLtreat- mentismyelosuppressionwith neutropenia,whichcanleadto severeinfections,anemiasrequiringredbloodcelltransfusions, andthrombocytopeniainvolvinghemorrhagicevents.Support- ive care with anti-infectious treatments, such as antifungals, antibiotics, andantivirals [3,4] and growth factors [5] has led to improved overallsurvival rates for patients with AML over recentdecades[6,7].

Here,wehighlightinnovativeformulationsasapotentialway tooptimizeAMLchemotherapies.Specifically,wefocusonhow nanoparticles (NPs) might be designed to counteract adverse effects,enhancetreatmentcompliance,avoidresistance,orpro- longexposuretochemotherapy.

Current treatments

Inyoungerpatients,AMLrequirestheinstaurationoftherapywith a minimal delay after diagnosis [8]. Bycontrast, delayingAML treatmentdoesnotaffectprognosisinolderpatients,exceptthose withveryproliferativedisease[9].Intensivechemotherapyisthe goldstandardforyoungerpatientswithAML,whereasepigenetic therapy is an option for older patients, especially those with comorbiditiesorbelongingtohigh-riskpopulations(i.e.,second- aryAMLorpoorriskcytogenetics)[10].

Standardchemotherapy

Induction therapy for younger patients with AML in 2018 remains largely as it was during the 1970s [11]. The most commontherapyusedinAMListhe‘7+3regimen’,a combi- nationofcytarabine(anucleosideanalog)combinedwithdau- norubicinoridarubicin(ananthracycline)[12].Patientseligible for intensive chemotherapy usually undergo this treatment, which comprises a continuous infusion of cytarabine (100–

200mg/m²) for 7 consecutive days and daunorubicin at 60mg/m² or idarubicin at 12mg/m² on Days 1–3. Reducing daunorubicin to 45mg/m² per day is considered suboptimal [13],whereasitisthoughtthatidarubicinisassociatedwitha highercurerateinthelongterm[14].Afterinduction,complete remissionisachievedin60–80%ofyoungerpatientswithAML, compared with approximately 45–65% in older patients [13,15,16]. This induction is then followed by consolidation therapyduringthepostremissionphase.Thistherapycomprises twotofourcyclesofhigh-dosecytarabine(2000–3000mg/m²/ daygivenonDays1,3,and5)andcanbefollowedbyhemato- poietic stemcell transplantation dependingon age,AMLrisk group,anddonoravailability.

Specific therapy

Theintroductionofall-transretinoicacid(ATRA)intothethera- pyofacutepromyelocyticleukemia(APL),whichrepresents10–

15%ofallcasesofAML,completelyrevolutionizedthemanage- ment andoutcomesof thisdisease[17].ATRAcombined with anthracyclinesresultsinaremissionrateof>90%,andprovides thefirstexampleofmolecularlytargetedtreatmentinAML[18].

ATRA,administeredorallyat45mg/m²twicedaily,ispresented

as10-mgsoftcapsulesinwhichATRAissolubilizedinamixture ofyellowbeeswax,hydrogenatedsoyabeanoil,partiallyhydro- genated soya beanoil, andsoya bean oil. ATRAhas also been combinedwitharsenictrioxide(ATO),asaresultofsynergyin inducing differentiation and apoptosis. The combination of ATRAwithATOresultedinthefirstchemotherapy-freeregimen for AML andhas become thenew standard for APL [19] with inferior toxicity compared with ATRA combined with chemotherapy[20,21].

Relevantmoleculartargetshavebeen discoveredinAML.For example,isocitratedehydrogenases(IDHs)canbemutated,lead- ing to the accumulation of R-2-hydroxyglutarate in the cell.

MutationsoftwoisoformsofIDHs(IDH1andIDH2)havebeen linkedtoepigeneticchangesresultinginglobaldysregulationof geneexpression. Thosemutationshave beenfoundin approxi- mately20%ofpatientswithnewlydiagnosedAML[22].FMS-like tyrosinekinase3(FLT3)isagenethatencodesatyrosine-kinase andismutatedinsomepatientswithAML.MutantformsofFLT3 canbetargetedbykinaseinhibitors[23].

MidostaurinwasthefirstkinaseinhibitorapprovedforAMLby theFDAandtheEuropeanMedicinesAgency(EMA),in2017.This drugisgivenorallyat50mgtwicedailyinassociationwiththe conventional‘7+3regimen’,inAMLwithFLT3internaltandem duplicationortyrosinekinaseactivation.IntheRATIFYtrial,the medianoverallsurvivalwasprolongedfrom25.6to74.7months withmidostaurininpatientsaged18–59years[24].

TheFDAalsoprovidedacceleratedapprovalfortheoralIDH2 inhibitorenasidenibinAugust2017,basedonapromisingPhase 2studyinrelapsedorrefractoryAML[25].Givenorallyat100mg perday,alone,thistreatmentproducedanoverallresponserate of 37%. Based on these results, this treatment is likely to be combinedin thenear future withconventional chemotherapy ordemethylatingagentsinpatientswithfirst-lineAMLandIDH2 mutations.

Othertargetedagentsinducingmutationsareunderdevelop- ment,suchasIDH1inhibitors(ivosidenib)[26]andBCL2antago- nism (venetoclax) either alone or in combination [27–29].

Interesting reviews of new active pharmaceutical ingredients (API)usedinAMLtreatmentareavailableelsewhere[30,31].

Antibody–drugconjugates

Gemtuzumabozogamicin(GO)wasthefirstantibody-drugcon- jugateapprovedby theFDA forCD33-positiveAML.CD33isa membranereceptor expressedby myeloid progenitors but not presentin normalhematopoietic stemcells.Thisdrugisa hu- manized anti-CD33 antibody conjugated to calicheamicin (a DNA-intercalating antibiotic). Accelerated approval was given byFDAin2000butwithdrawn10yearslaterbecauseofincreased treatment-related mortality [32]. Final approval was given in 2017basedonnewGOdoseregimens[33].Inadditiontonew CD33-targetingagents,otherclinicaltrialstargetingCD123are ongoing[34].

Epigenetictherapy

Forpatientswhoarenotcandidatesforintensivechemotherapy, suchasolderpatientswithcomorbiditiesorrelapsedorrefractory AML,lowerintensitytherapy isrequired. Forseveral years,the only solutions available were best-supportive care or low-dose 2 www.drugdiscoverytoday.com

(3)

cytarabine. Lower-intensity AML treatment is currently repre- sented by hypomethylating agents, such as decitabine (DEC) and azacytidine (AZA). The role of epigenetic deregulation in the pathogenesisofAML andMDS hasbeen a focusofintense researchoverthe past10years. Initiallydevelopedforpatients withhigher-riskMDS,DECandAZAhaveshowntheirabilityto improvemedianoverallsurvivalcomparedwithlow-dosecytar- abineorsupportivecareonly[35,36].InAML,bothhypomethy- lating agents have been compared with conventional care regimensintwo Phase3trials[10,37],inwhichthey displayed animprovementinmedianoverallsurvivalratesaswellaslower toxicity.WhereasDECinjectedintravenouslyat20mg/m²every dayover5days,repeatedevery4weeks,wasapprovedin2012by the EMA for older patients with de novo AML who were not candidatesforstandardinductiontherapy[38],itisonlyapproved bythe FDAforpatients withMDS. AZA(75mg/m²) subcutane- ouslyfor7 daysevery28days isalsoapprovedin theUSAand EuropeforMDS,butnotforAML.However,patientswith20–30%

bonemarrowblasts(refractoryanemiawithexcessblastintrans- formation according to French-American-British subtypes) are nowconsideredpatientswithAMLaccordingtotheWHO’sclas- sification[39],meaningthattheuseofDECandAZAisalabeled option.

Other novel agents currently under investigation in clinical trialswererecentlyreviewedbyMontalbanetal.[40].

Questions and challenges for AML treatment

Chemotherapiesusedin AMLaremostlypresentedassolutions and are administered intravenously. A general distribution is mostly observed without drug targeting towards specific cells.

Therefore, these chemotherapies can lead to general adverse effects, such as pancytopenia or cardiac toxicities related to anthracyclines. Treatment-related mortality is high, especially in the older population. However, specific AML therapies are mostly oral treatments,although the newschedules developed forpatientswithAMLarealsoaffectedbyadverseeffectsanddrug resistance.Tyrosinekinaseinhibitorsareassociatedwithdigestive adverseeffects(vomiting,nausea,diarrhea,andheartburn),vari- oustypesofskinrash,headaches,musclecramps,heartdysfunc- tion, hepatic impairments, pulmonary toxicities, or cardiotoxicities[40–42].

Despitenewtreatments,personalizedtherapiesandmanyclin- icaltrials,overall survival remainspoor. Complete remission is reached in 70–80%of youngerpatients with AML (<60 years), comparedwithonly45–65%ofolderpatientswithAMLbecauseof comorbiditiesanddrugresistance[13,43].Moreover,accordingto EuropeanLeukemiaNet,medianoverallsurvivalfallsfrom15.5to 5.5yearsinfavorableAMLgeneticgroupsamongpatientsoverthe ageof60[44].Relapseoccurswhendrugsareunabletoeradicate leukemicstemcells(LSCs)protectedinthehematopoieticniche.

Oneofthemajorcausesofdrugresistanceisthealteredfunctionof transmembranetransporters.SeveralmembersoftheATP-binding cassettetransporters familyareupregulatedin tumorcells[45], andcouldbeatargetinAML[46].

To bypass these phenomena, pharmaceutical technologies, such as nanotechnologies, are being developed, and this field representsapromisingapproach[47].Inparticular,inAMLthera- py, nanocarrierscould prolong the half-life of drugs generally

administeredviaanintravenousadministration,thusenhancing thecontactofthedrugwithcells.Specificcells,suchasLSCs,could be targeted to reduce relapses [48,49]. Finally, modifying the administration route can help to improve patient compliance, especiallyofolderpatients.

Challenges for innovative formulations

According to the definition given by the FDA and the EMA, nanomaterials involved in drug delivery should have at least oneexternal,internal,orsurfacestructureinthenanoscalerange (1–100nm). Nanomaterials involved in drug development can includeliposomes,lipidNPs,polymericNPs,micelles,dendrimers, and metallic NPs, depending ontheir formulationprocess and their composition. Since the first nanocarrierswere developed, severalapproacheshaveemergedtoenhancetheireffectiveness, depending on the purpose of the nanocarrier. Typically, four generationsofNPshavebeendistinguishedovertime.Inthefirst one,particleswere‘nude’andrapidopsonizationswereobserved.

Inthesecondgeneration,hydrophilicpolymersweregraftedonto the surfaces ofNPsto formabrush-like barrier.Slowed protein adsorption onto NPs was achieved, and mean blood residence timeswereenhanced.Second-generationparticlesdiffusetosolid tumorsbecauseoftheenhancedpermeabilityandretention(EPR) effect. However, thiseffect doesnotappearto work in human therapeutics[50]andhasnoimplicationsfortherapeuticsinAML, anonsolidtumorcancer.Third-generationparticlescompriseNPs with specificligands on their surfaces, designed specifically to target cells of interest. The fourth generation corresponds to theranosticstrategies(multifunctionalnanocarriersthatcombine a diagnostic agent with a therapeutic agent) or targeting cell organelles[51].

Severalnanocarrierformulationshavebeenapprovedandare availableontheUS,European,Indian,Japanese,orChinesemar- kets.Theyaremostlyliposomalformulationsandarelargelyused forsolidtumors.Nanocarriersaregenerallydesignedforfourmain reasons:(i)drugtargetingwiththeaimofreducingadverseeffects resulting fromthegeneral distributionofthe APIor fromexci- pients;modifydrugpharmacokinetics;(3)bypassdrugresistance;

and(4)modifytheroutebywhichthedrugisadministeredorto improve the oral bioavailability of APIs [52]. This can lead to improvedqualityoflifeforpatientifintravenousadministration isreplacedbyoraladministration.

Otherhematologicmalignancieshavealsobeentreatedusing liposomalformulations.Forexample,liposomalpegylateddoxo- rubicinformulations,indicatedinmultiplemyeloma,decreasethe cardiac toxicity associated with anthracycline. However, new adverseeffectshavebeenobservedwiththeseformulations,such asacutehypersensitivityreactions.Liposomalformulationsmodi- fy doxorubicin pharmacokinetics by increasingits half-life and plasmaconcentration(areaundercurveincreased)anddecreasing the volume of distribution [53]. Moreover, doxorubicin-loaded liposomes are less subject than free doxorubicin to multidrug resistancemechanismsdrivenbyp-glycoprotein,andarecapable ofreversingmultidrugresistanceeffluxmechanisms[54].

There are a few NP formulations on the market, and one, Vyxeos¹,wasapprovedforAMLtreatmentbythe FDAin2017.

This liposomal formulationis a co-encapsulation of cytarabine anddaunorubicin,andisdiscussedinmoredetailbelow.

www.drugdiscoverytoday.com 3 ReviewsKEYNOTEREVIEW

(4)

Forty-twoclinicaltrialscorrespondingtoacombinationofthe search terms ‘acute myeloid leukemia’ and ‘liposomes’,

‘nanoparticles’, ‘nanocarriers’, ‘micelles’ or ‘nanocapsules’ were foundontheclinicaltrials.govwebsiteinDecember2017(Table1).

Among the 42clinical trials, eight corresponded to antifungal drugs, and six did notused nanocarrier formulations but were foundbecauseoftheiruseofassociatedkeywords(MESHterms).

Finally;28clinicaltrialswereidentifiedthatusedliposomalfor- mulations;butonlyonewasbasedonpolymericNPs.NootherNP formulationiscurrentlyunderclinicalinvestigationforthetreat- mentofAML.

Innovative nanocarrier formulations

Liposomes

Liposomes arevesicles generally comprisingphospholipids and cholesterol ina bilayer arrangement(Fig. 1). Liposomescanbe loadedwithoneormoremoleculesintheaqueouscompartment orinthelipidicbilayer.Liposomalformulationsarealreadyused in clinicalsettings fornon-AMLindications,suchasanti-infec- tiousdrug encapsulation(amphotericinB,Ambisome¹,Astellas Pharma), analgesic drugs [morphine sulfate, Depodur¹, Pacira Pharmaceuticals (commercializationdiscontinued)] or cytotoxic drugs.Intermsofcytotoxicdrugs,onlytwodrugclasseshavebeen encapsulated intoliposomes; anthracycline and antimetabolite agents. Two anthracycline molecules were encapsulated and reachedthemarket,namelydoxorubicin(Caelyx¹,Janssen-Cilag InternationalN.V,Doxil¹,Janssen-CilagInternationalN.V,and Myocet¹, Teva B.V.) and daunorubicin (Daunoxome¹, Gilead Sciences) forwhichcommercializationwasdiscontinued.Cytar- abine(DepoCyt¹,PaciraLtd)istheonlyrepresentativedrugofthe antimetabolite class encapsulated in a liposome-like structure (nonconcentric vesicles). The commercialization of Depocyt¹ was discontinued in 2017. The main advantages of liposomes

arethattheyhaveshownagoodabilityforlarge-scaleproduction (i.e.,thepresenceofdrugproductsonthemarket)andalsotheir toleranceisnowwellstudied.

LiposomesapprovedasAMLtherapy

In2017,the FDA approved Vyxeos¹(previously denotedCPX- 351),aliposomal formulationwith aco-encapsulationofcytar- abineand daunorubicin in a 5:1 molar ratio. To obtain theses liposomes, an emulsion was formed comprising distearoylpho- sphatidylglycerol,distearoylphosphatidylcholine,and cholesterol, solubilized in chloroform and dispersed in a solution comprisingcopper gluconate and triethanolamine.The solvent (chloroform)isevaporatedandthefilmisthenrehydratedandthe liposomes extruded on a polycarbonate filter [55]. Vyxeos¹ is presentedinalyophilizedformandshouldbereconstitutedwith waterfor injection; itshould then be diluted in 0.9%sodium chloride injection or 5%dextrose injection before intravenous injection.Themeansizeofliposomesis107nm,withapolydis- persityindex(PI)of0.243andanegativesurfacecharge(potential zetaof33mV)[55].InaPhase2randomizedtrial,CPX-351was comparedwitha ‘7+3regimen’in olderpatientswithfirst-line AML[56].This clinicaltrial enrolled127patients, ofwhom86 wereallocatedtotheCPX-351 armand41tothegoldstandard treatmentarm.Twocoursesofinductionandconsolidationwere allowed in each arm. CPX-351 was associated with a higher complete remission rate and a lower 30-day treatment-related mortalityrate(3.5%versus7.3%).Itwasdemonstratedthatlipo- somescanenterleukemiccellsintact,and thismechanismwas suggestedasameansofovercomingdrugresistance[57].Indeed, anthracyclinesaresusceptibletoeffluxpump(becauseofthep- glycoprotein) and, once encapsulated into liposomes, cellular internalization is mediated by liposome properties rather than byanthracycline properties.Thus,daunorubicinwasableto es- capetothep-glycoprotein.Cytarabineisrapidlymetabolizedby TABLE1

Chemotherapeuticdrug-loadednanocarriersinvolvedinclinicaltrialsinDecember2017inanAMLindication^a

Nanocarriercategory Drug Clinicaltrialphase ClinicaltrialID

PolymericNPs AZD2811 1–2 NCT03217838

Liposomes Vincristine 1 NCT00933985

2 NCT02337478

Cytarabine–daunorubicin Notprovided NCT01804101

1 NCT00875693

1–2 NCT02642965

2 NCT02286726/NCT00822094/NCT02019069/

NCT00788892/NCT02269579/NCT02238925/NCT03335267

3 NCT01696084

Grb2antisenseoligonucleotide 1 NCT01159028

2 NCT02781883

Doxorubicin 1 NCT00003243

1–2 NCT00237627

3 NCT01736943

Annamycin 1 NCT00430443

1–2 NCT03315039

Daunorubicin 1 NCT02722135

1–2 NCT00005942

2 NCT00005962

2–3 NCT00111345

3 NCT02724163/NCT00186966

Cytarabine 1 NCT00003073

2 NCT00002547

aFromclinicaltrials.gov,December2017.

4 www.drugdiscoverytoday.com

(5)

cytidine deaminase and its encapsulation into liposomes can protectitfromthisdegradationpathway.CPX-351wasalsoeval- uatedinfirstrelapseinyoungerpatientswithAMLaged18to65 [58].ThisPhase2trialwasbasedon2:1randomization[i.e.,CPX- 351oranalternativechosenbytheinvestigatorandbasedmainly on cytarabine (97.3%) and anthracycline (77.3%)]. As in the previous clinicaltrial, a highernumberof completeremissions wasobservedintheCPX-351arm,alongwithalowernumberof deathsafter90days.APhase3clinicaltrialwasthenconductedin patientsaged60to75withuntreatedhigh-riskorsecondaryAML (NCT01696084).Thistrialrecruited309patientsandconfirmed the higher complete remission rate (47.7% versus 33.3%) and overallsurvival(median,9.56 versus5.95 months)forCPX-351 overthe‘7+3regimen’[31,59,60].

Vyxeos¹isnowlabeledforthetreatmentofadultswithnewly diagnosedtherapy-relatedAMLor AMLwithmyelodysplasia-re- latedchanges.Accordingtocliniclatrials.gov,inDecember2017, 11clinicaltrialswereongoingusingtheliposomalformulationof

daunorubicinandcytarabine(Table1),withtheaimofexpanding Vyxeos¹indications.

LiposomesunderclinicaltrialasAMLtherapy

Other liposomal formulations are under clinical evaluationfor AMLindications(Table1).Thesearedrugformulationsalreadyon the market and are used in clinical trials as drug support to investigateothernewagents,todevelopnewschedulestrategies, ortoinvestigatenewchemotherapeuticagentassociations.Vin- cristine-sulfate liposome, annamycin liposomes, or BP1001 (a liposomal growthfactor receptor-boundprotein-2antisenseoli- godeoxynucleotide loadedin a liposome formulation) arenew drugsthatarenotyetapproved.TheyarecurrentlyatthePhase1 and2evaluationstagestodetermineeitherthemaximumtoler- ated dose (annamycin liposomes and BP1001) or their efficacy (vincristinesulfateliposomes)inpatientswithrelapsedorrefrac- toryAML.

Liposomescanbeused forco-encapsulation,and Vyxeos¹is proofthatco-deliverycouldbemoreeffectivethanacombination Aqueous core

Terminal function

Hydrophobic chain Hydrophilic group

Polymer chains Core

Dendrimer

Liposome Metallic nanoparticle

Micelle

Lipid nanoparticle

Polymeric nanoparticle Interior branching

Hydrophilic API Hydrophobic API Surfactant-containing surface

Lipid core (Solid or liquid) Phospholipids and cholesterol bi-layer

Drug Discovery Today

FIGURE1

Overviewofthesixmaintypesofnanocarriersinvolvedinacutemyeloidleukemia(AML)therapyatdifferentstagesofpreclinicalorclinicaldevelopment.

NanocarriersaregenerallyloadedwithdrugsalreadyusedinAMLtherapytoreducesevereadverseeffectsandcouldbefunctionalizedwithspecificligandsto enhancetargetingproperties.Abbreviation:API,activepharmaceuticalingredient.

(6)

ofdrugsadministeredseparately,probablybecausethemolarratio is preserved throughout drug release [61]. Moreover, Vyxeos¹ confirmsthatusingnanotechnologiescouldovercomedrugresis- tance,suchasp-glycoprotein(anthracycline),and protectdrugs from enzymaticdegradation pathways,suchascytidinedeami- nase(cytarabine).

Liposomesevaluatedininvitroorinvivomodels OtherliposomeformulationshavebeendevelopedforAMLtreat- ment, and are designed to avoid adverse effects (Fig. 2). For example, a method has been developed for the co-delivery of safingol and C2-ceramide [62]. Safingol is a sphingolipid with antileukemicactivity,butisnotusedintheclinicalsettingbecause ofitspoorwatersolubilityandhemolyticproperties[63].Tanetal.

designed liposomes forsafingol encapsulation. These werepre- pared by mixing safingol, distearoylphosphatidylcholine, and cholesterolinchloroformbeforeevaporatingthelatter.Thelipid filmwasthenrehydratedinaHEPESbuffer,whichwasfollowedby extrusiontechniquesonpolycarbonatefilters.Finally,liposomes withapositivezetapotentialwereobtained.Theliposomeswere able to reduce significantly the in vitro hemolytic toxicity of safingolandsafingolplasmaeliminationinamouse-basedmodel, and to improve survival in U937-inoculated mice (histiocytic lymphoma cells) [63].The same authors then developed a co- encapsulationofsafingolandC2-ceramide(1:1molarratio),an- othersphingolipid[62].GiventhatC2-ceramideandsafingolact independently,withdifferentcellularpathways,theauthorspos- tulatedthat co-encapsulationof thesetwosphingolipids would enablethemtoactinsynergistically.Aliposomalformulationwas obtainedafterthedissolutionofthelipids(safingol,C2-ceramide, cholesterol, and distearoylphosphatidylcholine) in chloroform.

The solvent was then evaporated, and the film obtained was rehydrated with piperazineethanesulfonicacid buffer.The lipo- someswerethenextrudedonpolycarbonatefilters.Liposomesof 135nmwithaco-encapsulationofsafingolandC2-ceramidewere obtainedthatweresimilartoliposomescontainingonlyC2-ceramide (119nm).Liposomessignificantlyreducedthe hemolysis properties ofsphingolipids comparedwithfree drugs.The con- centrationsrequiredfor liposomalC2-ceramideto achieve50%

celldeathweresignificantlyhigherthanthoseforliposomalco- encapsulationofsafingolandC2-ceramideonfiveleukemiacell lines (KG-1, MV-4–11, HL-60, U937, and HEL). Finally, in vivo experimentswerecarriedoutinvolvingintravenousinoculation ofU937usingamouse-basedmodel.Despiteareductioninthe drugdosesadministered,co-encapsulationextendedthemedian survivaltimefrom24to37dayscomparedwithliposomeswith C2-ceramide encapsulation alone (P<0.005), demonstrating onceagainthepotentialofco-encapsulationintoliposomes.

Myhrenetal.alsodevelopedaliposomeformulationfortheco- administrationofananthracycline(daunorubicin)andaprotein synthesisinhibitor,emetine[64].Liposomesweregraftedontothe liposomesurfacewithpolyethyleneglycol(PEG)toconferstealth properties,andfolicacidwasalsograftedontotargetAMLcells.

The liposomescomprised phosphatidylcholine,cholesterol,and distearoylphosphatidylethanolamineordistearoyl-glycero-phos- phoethanolamine-N-PEG5000-folate.Afterthelipidsweresolubi- lized in chloroform, the solvent was removed, and the film obtained wasrehydratedin ammoniumsulfatebuffer andthen extruded.Liposomeswerethenloadedwithdrugsbyapplyingthe

acid-precipitationconcept.Finally,liposomeswithasizeof121–

127nm (depending on the folic acid grafting),and a negative chargeofapproximately10mVwereobtained.Afteraninvitro studyonanAMLcellline(MOLM-13),theauthorsshowedthat liposomeswithaco-encapsulationweremoreeffectivethanaco- incubationofdaunorubicin-loadedliposomesandemetine-loaded liposomes.Apoptosiswasenhancedbyafactorofsix.However, theauthorsalsoshowedthatincubationofdaunorubicin-loaded liposomesfollowed30minlaterbyincubationofemetine-loaded liposomes was more effectivethan the administration of lipo- somescontainingthetwodrugs.

Liposomesandbiopharmaceuticals

Biopharmaceuticalsarecurrentlyaresearchfocusandliposomes could help in administering these APIs. Biotechnology is an emerging field with great potential in cancer treatment, and AMLisaffectedbythisdevelopment.Biotechnologyencompasses severaltreatmentstrategies, suchasgenetherapy,antisenseoligonucleotide therapy, and chimeric antigen receptor T cells, whichwererecentlyapprovedforacutelymphoblasticleukemia treatmentby theFDA.Antisenseoligonucleotidesrefertosmall interfering(si)RNA,miRNA,orshorthairpin(sh)RNA.Thereare fewantisenseoligonucleotidesonthemarket,probablybecauseof a lack of stability in blood circulation, inefficient intracellular delivery,difficultiesintargetingcellsortissues,andpooraffinity withthetargetsequence[65].Therefore,aspartofcancertreat- mentstrategies,differentNPsweredesignedtoimproveeitherthe stabilityinthebloodstreamortheirabilitytotargetcancercells [66]. Hundreds of Phase 1, 2, and 3 clinical trials for cancer indications are ongoing [67]. However, only two clinical trials combining liposomes and antisense oligonucleotide for AML treatment were found on clinicaltrials.gov in December 2017 (Table1).

BP1001isaliposomalformulationofanantisenseoligonucleo- tide (growth factor receptor-bound protein-2) comprising dio- leoylphosphatidylcholine,witha meansize<900nm,andhigh encapsulationefficiency(>95%)[68].Twoclinicaltrialsareongo- ing.APhase1trialisunderway(NCT01159028)todeterminethe highest safe dose for patients with Philadelphia chromosome- positive chronic myeloid leukemia, AML, acute lymphoblastic leukemia,andMDS(activebutnotrecruitingpatients).ThePhase 2trial(NCT02781883)isdesignedforpatientswithnewlydiag- nosedAML whoarenotsuitablefor standardor high-intensity chemotherapyregimens(currentlyrecruitingpatients).Thestudy isdesignedtoassesswhetherthecombinationofBP1001andlow- dosecytarabineismoreeffectivethanlow-dosecytarabinealone.

Otherformulationsareunderinvitroorinvivoevaluation.miR- 126isoverexpressedinLSC[69]andthecellsappeartobeengaged inAMLtreatmentresistance;thus,targetingLSCsappearstobea promising therapeutic approach [70]. Dorrance et al. designed lipoplextargetingLSCs,andmorespecificallymiR-126,withthe objectivetoreduce quiescentLSCand leukemiagrowth [49].A negativelychargedmiR-126-antagonistwasmixedwithpositively chargedpolyethylamine,producingapolyplexcorethatwasthen addedtoananionicliposomalformulation,finallyresultingina lipopolyplex. To target LSC, transferrin or CD45 antibody was thengraftedontothe surfaceofthe lipoplex.The formulations weretestedonmiceengraftedwithhumanAMLprimaryblasts.

Themiceweretreatedwithlipopolyplexloadedwithanti-miR-126 6 www.drugdiscoverytoday.com

(7)

orwithanti-miR-scramble(SCR)asacontrol.After48hoftreat- ment, bone marrowcells were collected and transplanted into secondarymice.Whereasa significantimprovementinsurvival rateofsecondaryrecipientswasobserved,therewasnoimpactfor primaryrecipientmicetransplantedwithcellsharvestedafteran anti-miR-126-lipopolyplex formulation compared with mice transplantedwithcellsharvestedafterananti-miR-SCR-lipopolyplex formulation. Thus,this combinationof anti-miR-126 and chemotherapycouldbeaneffectivewayto targetLSCsandde- creaseleukemiarelapses.

LiposomeswerealsodevelopedtoencapsulateGTI-2040(also known as LOR-2040) and to target ribonucleotide reductase, which is overexpressed in cytarabine-resistant AML cells [71].

GTI-2040, targeting the R2 ribonucleotide reductase subunit, was evaluated in a Phase 1 clinical trial in combination with cytarabine. R2 downregulation was observed but no complete remissionwasnoted,indicatingthattheeffectofGTI-2040needed tobeenhanced[72].GTI-2040-loadedimmunoliposomes,grafted withananti-CD33totargetAMLcells,weredesignedandevalu- atedbyLietal.[71].Theseliposomeswerepreviouslyformulated

using an emulsion comprising lipids (2,3-dioleyloxy-propyl-tri- methylammoniumchloride,dioleoylphosphatidylethanolamine,

a

-tocopherylPEG-1000succinate,andcholesterol)solubilizedin ethanolandtheninjectedinanHEPESbuffer.Meanparticlesize was 9318nm, and the zeta potential was relatively neutral (4.30.2mV).After cytotoxicitytesting onKasumi-1 (an AML cellline)andK562(chronicmyelogenousleukemiacellline),the authors developedan invivo modelof a Kasumi-1 xenografted tumor.FollowingdailyadministrationofGTI-2040-loadedimmu- noliposomesfor3days,significantlydownregulatedR2expression wasobserved comparedwith freeGTI-2040.Tumor volumede- creasedbyafactorofapproximatelytwo.Whencombinedwith cytarabinetreatment,thesurvivalofmicewasalsoprolongedwith the use of GTI-2040-loaded immunoliposomes compared with cytarabinealone,cytarabineplusfreeGTI-2040,orcytarabineplus GTI-2040-loadedliposomeswithoutanti-CD33grafting.

Finally,severalliposomalformulationsusingchemotherapeu- tic drugsor antisenseoligonucleotides arein various stages of development(Table 2).Combinationtherapy withdrugco-encapsulation ordrugco-administrationisnow offeredona sys- O

O

OH

OH OH

OH

HO

HO HO

OH OH

OH

N

H H

H

H H H H

H H

OH N

N N

O N

O O

O

OH

OH OH

O

O O

O

O O O

O

O O

NH

H₂N CH₃

CH₃

CH₃ H₃C CH₃

CH₃

CH₃ OH

OH HO

OH OH

OH

HO

OH OH

H₂N

Doxorubicin Daunorubicin

C2-ceramide

Fingolimod

Cytarabine Decitabine

Etoposide Safingol

AZD2811

All-trans retinoic acid

HO H₃C

H₃C

CH₃

CH₃ H₃C

H₃C

H₂N

NH₂

NH₂ NH₂

H₃C

O O O

O

O O O

N N

N

N NH

HN

F HN

O O O

Drug Discovery Today

FIGURE2

Examplesofmoleculesloadedininnovativeformulationsdevelopedtotreatacutemyeloidleukemia(AML)toimprovepharmacokineticparametersaswellas reducingadverseeffectsormodifyingtheadministrationroute.

(8)

tematic basis. Liposomes have beendesigned to enhance drug targeting and to overcomedrug resistance, andtheir development iscontinuing, with severalformulationscurrently under clinicalevaluation.

Nanoparticlesbasedonpolymers

Those particles in the nano-size range contain supramolecular associationsofpolymers(micellesanddendrimers)andaggregated densepolymericparticles.The stabilityoftheseformulationsin thebodycanbeverydifferent,ascantheirtoxicityandscale-up ability.Thus,thefutureoftheseformulationsintheclinicislinked notonlytotheirperformance,butalsototheirfateinthehuman

body,resultingfromtheirphysicochemicalproperties.Thediffi- culties in producing and accurately characterizing large-scale batchescouldhindertheirfuturedevelopment.

PolymericNPs

Polymeric NPs (Fig. 1) are NPs formulated from polymers and designedtoencapsulatehydrophilicorlipophilicdrugsdepending ontheformulationprocessused.Alongwithliposomes,polymeric NPsrepresentthe mostsignificant nanomedicinecategoriesap- provedbytheFDA[73].

ForAMLindications,oneclinicaltrialusingpolymericNPshas beenregistered(Table1),andothersarecurrentlybeinginvesti- gatedinpreclinicalmodels.

TABLE2

ExamplesofnanocarriersincurrentdevelopmenttoimproveAMLtreatment

APIclass API Drugformulation Drugassociation Developmentstage Objectives Refs

Anthracyclin Daunorubicin Liposomes Yes(cytarabine) FDAapprovedand clinicaltrial investigations

Bypassdrugresistance [55]

Yes(emetine) Invitroevaluation Modify

pharmacokinetics;target AMLcells

[64]

Doxorubicin PolymericNPs No Invivoevaluation Modify

pharmacokinetics;target AMLcells;bypassdrug resistance

[77]

Polymericmicelles No Invitroevaluation TargetAMLcells [85]

Antipyrimidic Cytarabine Liposomes Yes(daunorubicin) FDAapproved Bypassdrugresistance [55]

Dendrimers No Invitroevaluation Modify

pharmacokinetics;

bypassdrugresistance;

targetAMLcells

[88,89]

IronNPs No Invitroevaluation [118]

Hypomethylatingagent Decitabine LipidNPs No Exvivo,invitro,andin vitroevaluations

Modifyadministration route(improvepatients’ qualityoflife)

[112–114]

ProteinkinaseCinhibitors Safingol Liposomes No Reduceadverseeffects;

modify

pharmacokinetics

[63]

Liposomes Yes(C2-ceramide) Invivoevaluation Reduceadverseeffects [62]

C2-ceramide Liposomes Yes(Safingol) Invivoevaluation Reduceadverseeffects [62]

Oligonucleotideantisense miR-126 antagonist

Liposomes No Invivoevaluation TargetLSCs [49]

GTI-2040 Liposomes No Invivoevaluation TargetAMLcells [71]

CD44siRNA PolymericNPs No Invitroevaluation TargetAMLcells;modify pharmacokinetics

[78,80]

siRNA Cyclodextrin-basedNPNo Invitroandexvivo

evaluation

TargetAMLLSCs [48]

Differentiatingagents All-transretinoic acid

NLCs No Invitroevaluation Modifyadministration

route

[103–105]

SLNs No Invitroevaluation;

physicochemical characterization

Modifyadministration route

[101,102]

PolymericNPs No Invitroevaluation Modifyadministration route

[76]

TopoisomeraseIIinhibitors Etoposide NLCs No Invitroevaluation Targetcancerouscells;

modifycurrent formulationtoreduce adverseeffects

[100]

Sphingosine1-phosphate receptormodulators

Fingolimod SLNs No Invivoevaluation Modify

pharmacokinetics;target AMLcells

[109]

AuroraBkinaseinhibitors AZD2811 PolymericNPs No Invivoevaluation Modify

pharmacokinetics

[74,75]

8 www.drugdiscoverytoday.com

(9)

PolymericNPsencapsulatingAZD2811,amolecularlytargeted agent(auroraBkinaseinhibitor),havebeendevelopedandwillbe evaluatedinanapprovedclinicaltrial(NCT03217838).NPscom- priseblock copolymersof poly-D,L-lacticacid (PLA)and PEG. A nanoemulsionofbenzylalcoholandanaqueousphasecontaining anorganicacidhavepreviouslybeenobtained.Theorganicacid ledto a counterion interactionwith AZD2811,increasingdrug encapsulationefficiencyanddecreasingthereleaserateofthedrug [74].Aninvivo pharmacokineticstudy using arat-based model confirmedthe slowreleaseof the drug, extendingits exposure profile.Pharmacodynamicstudiesinmultiplepreclinicalmodels havedemonstratedthatAZD2811-loadedNPsweremoreeffective ininhibitingtumorgrowthwhenadministeredathalfofthedose ofAZD1152.AZD1152 correspondstotheAZD2811phosphory- latedprodrug (alsoknownasbarasertib) [74].InaHL-60-xeno- graftedmouse-basedmodel,AZD2811NPsshowedthepotentialto improveAZD2811efficacyinregressingtumorsizeatthevarious doses investigated, as well as being well tolerated [75]. When combinedwithcytarabineinthesamemodel,tumorsizeregres- sionandthe sustainabilityoftheresponse wereincreased compared with cytarabine alone or AZD2811 NPs alone. When AZD1152(used at a highertotaldosecomparedwithAZD2811 NPs)wascombinedwithcytarabine,thedurationoftheresponse wasreduced,confirmingthepotentialofAZD2811-PLANPs[75].

Simonetal.developedNPstoencapsulateATRAinpolymeric NPs basedonpoly-D,L-lactide-co-glycolide (PLGA)polymer [76].

PLGANPswereobtainedusingthesolventevaporationtechnique.

After a lyophilization process, the mean particle size was 150–

200nm.TheNPsdidnotaffecttheantiproliferativeanddifferen- tiationcapabilitiesofATRAinHL-60cellswhenentrapped.Given thatPLGANPsarealreadyusedforparenteraladministration,and PLGAisaFDA-approvedpolymerforparenteraluse,theauthors concludedthatATRApolymericNPscouldbeusefulforintrave- nousadministration,particularlybecausethereleaseofthedrug wasprolonged.Currently,ATRAisonlyavailablefororaladmin- istration,and a parenteral administration method could bere- quiredforpatientswhoareunabletoswallow.

Multifunctionallipid-polymericNPsloadedwithdoxorubicin and obtained by the nanoprecipitation methodhave alsobeen developed[77].Inthisstrategy,doxorubicinwasencapsulatedina polymericcorecomprisingPLGAandPluronic¹85(co-polymers basedon ethyleneoxide and propyleneoxide).Transferrin was thengraftedontoalipidicshellcomprisingstearicacid,afterbeing conjugatedtoPEGandoxalicacid.Thegraftingledtothetargeting oftransferrinreceptorsexpressedinleukemiccells.Particleshada mean size of 160nm and a negative charge. The formulation showed the ability to prolong doxorubicin plasma circulation timeinarat-basedmodel.Themeanresidencetimewasenhanced byafactorofatleastfive.Tumorreductionvolumeinamouse- basedmodelxenograftedwithHL-60cellsordoxorubicin-resistant HL-60cellswasalsoenhanced.Doxorubicinmultidrugresistance wasalsoovercome,probablybecauseofthepresenceofPluronic¹ 85inthecoreoftheNPs,whichisabletoinhibitp-glycoprotein.

AspecificCD44siRNAtotargetLSCsloadedinpolymericNPs has also been developed[78]. CD44, an adhesion molecule, is associatedwithpoorAMLprognosisandhighlevelofAMLrelapse [79]. Lipid-modified polyethyleneimine dissolved in water was added to CD44-specific siRNA, solubilized in a saline medium

[80]. The mean size of the complexes obtained was <100nm [80]. Gul-Uludag et al. loaded specific CD44 siRNA into these polymeric NPs in a 8:1 polymer:siRNA weight ratio [78]. They investigatedthesilencing efficiencyofCD44ina AMLcell line modelknowntoexpressahighlevelofCD44(KG-1cells).After3 days,CD44silencingwassignificantlysuperiorwhenCD44siRNA- loaded polymeric NPs were applied compared with scramble siRNA-loaded polymeric NPs. The cytotoxic activity of CD44 siRNA-loadedpolymericNpswasalsomoreimportant thanthe cytotoxicityofthepolymer–SCR–siRNAcomplexes.Experiments wereonlyperformedininvitromodels,andneedtobecontinued toconfirmthepotentialofthispolymericformulationcompared with the toxicity of CD44-neutralizing antibodies previously reportedinclinicaltrials[81].

Micelles

Polymericmicellesarelargelyusedincancertherapiesinvolving lipophilicdrugs[82].Theirpharmacokineticbehaviorisdrivenby the hydrophilicstructureoftheir shell[83].Polymeric micelles wererecentlyapprovedby differentmedicinesagenciesandare goodcandidatesforsolubilizinghydrophobicdrugs(notablypac- litaxel)withouttheneedforatoxicsurfactant.Polymericmicelle formulationsofpaclitaxelalreadyexistandareusedintheclinic (e.g.,Genexol-PM¹,Samyang-Biopham;Nanoxel-M¹,Samyang- Biopham;andPaclical¹,OasmiaPharmaceuticalAB)[84].

Doxorubicin-loaded polymericmicelles havebeen developed forAMLindications[85].Dextranwasusedasacopolymer,and folicacidandretinoicacidwerethengraftedontotheshellofthe micellestotargetAMLcells.Themicelleshadameansizeof80–

100nmandaencapsulationefficiencyofatleast90%.Theinvitro cytotoxic activity of doxorubicin-loaded polymeric micelles graftedwithfolicacidwasenhancedinKG-1cellscomparedwith the free drug. The concentration required to provide 50% cell deathwasapproximatelytwiceashighusingdoxorubicinalone compared with doxorubicin-loaded polymeric micelles grafted with folic acid.When folicacid was notgrafted, nodifference between the free drug and drug-loaded micelles was observed, demonstratingtheimportanceofthefolicacidligand.

Moreadvancedassessmentsarerequiredtoconsidermicellesfor clinicaltrialsinAMLtreatments.

Dendrimers

Dendrimers are3D branchedstructures with a nano-size range (Fig.1).Theyareusefulforimprovingthesolubilityofhydropho- bicdrugs.Moreover,arangeofterminalfunctionalgroupsexists and, thus, dendrimers are ideal for drug delivery systems and targetingstrategies[86].Drugscanbeconjugatedorencapsulated intothecoreofthedendrimer.Incancertherapy,dendrimershave beenusedtocontroldrugreleaseandspecificdrugtargeting[87].

Nodendrimerparticles arecurrentlyinvolvedin anyclinical trialforAMLtreatment.However,theliteratureincludesseveral studiesassessingtheirpotentialtoimprovetreatment.

Dendrimersforcytarabineencapsulationhavebeendeveloped by Szulc etal.[88]. First,maltose residues wereattachedat the amino terminalgroupofpoly(propyleneimine).Cytarabinetri- phosphatewasthenmixedwithdendrimerinphosphatebuffer saline, and a complexwas finally obtained asa resultof ionic interactionsbetweenthenegativelychargedcytarabineand the cationicdendrimer[89].Cytarabinetriphosphate-complexedden- drimers did notsignificantlydecrease HL-60 cell viabilitycom- www.drugdiscoverytoday.com 9

(10)

pared with cytarabine in solution. By contrast, cytarabine tri- phosphate-complexeddendrimerssignificantlyincreasedthecy- totoxic potency of cytarabine on the 1301 cell line (a T cell leukemiacellline).Accordingtotheauthors,thisenhancement resultedfromanadditionalentryortoxicitymechanism[88].

There are no dendrimer-bound chemotherapy formulations currentlyonthemarket.ForAMLindications,fewinvitrostudies are underway, and theseneed to bedevelopedthroughin vivo studies. These formulations also need to be scaled up before clinicalapplicationscanbeconsidered.

Lipid-basednanoparticles

Thereisarangeoflipophilicdrugsavailablebuttheycannotbe solubilized in aqueous media in great quantity. Liposomes, as mentionedabove,wereoriginallydesignedforhydrophilicdrugs.

Lipophilicdrugscanbeencapsulatedintoliposomesaspreviously described and are then located in the phospholipidbilayer or grafted onto the surface.Inthe latter case,rapidreleaseof the drugcanbeobserved,andthedrugloadingislimited.LipidNPs can be usedto prevent this burst release and to enhance drug loadinginNPs.

Two mainkinds oflipid-basedNP arecurrently being developed;solidlipidNPs(SLNs),withasolidlipidmatrix,andnano- structuredlipidcarriers(NLCs),comprisingsolidlipidsandliquid lipids(Fig.1).Intheidealscenario,lipid-basedNPscanbeprepared using a low energy process, do not require the use of organic solvent,andcanbescaledup[90–92].Thus,numerouslipophilic cytotoxicdrugshavebeenencapsulatedinlipid-basedNPs,suchas paclitaxel, etoposide, docetaxel, SN38, and doxorubicin; some drugshavebeenmodifiedtobemorelipophilic,suchascarbon- yl-graftedgemcitabine[93–97].However, therearecurrentlyno lipid-basedNPseitheronthemarketorundergoingclinicaltrials.

EtoposidehasbeenfoundtobeactiveindenovorelapsedAML [98],andcanbeusedincombinationwithcytarabineinpatients whoareeitherolderwithAMLorhaverefractoryAML[99].Given itspoorwatersolubility,etoposideinitscommercialformulation isonlyavailablecombinedwithalcoholandsurfactants,suchas polysorbate 80or CremophorEL, which display cell toxicities.

Therefore, etoposide lipid-based nanocarrier formulations were developedforAMLindications.Khajaviniaetal.developedetopo- side-loadedlipidnanocarriersconjugatedwithtransferrin[100].

Particles were prepared by an emulsion–solvent diffusion and evaporation method followed by sonication. The lipid phase, containingetoposide,wasdissolvedinamixtureofacetoneand ethanol, and comprised cetylpalmitate, octyldodecanol, soy lecithin, and a fatty amine. A solution of poloxamer 188 was mixedandtheemulsionwasthenultrasonicated.Finally,trans- ferrinwasconjugatedtolipidnanocarriers.The sizeofparticles rangedfrom158nmto247nm,dependingonthefattyamineand the transferrin conjugationmethod used.When etoposide was loadedintolipidnanocarriers,itscytotoxicactivityonK562cells wassystematicallyenhancedcomparedwithfreeetoposide.

Givenitslipophilicstructure,ATRAisagoodcandidateforlipid- basedNPs,andtherearenumerousexamplesintheliteratureof lipid-basedNPs forATRAencapsulation.Oneexampleis ATRA- loaded SLNs obtained by an emulsification-ultrasound process [101,102].Depending onthecomponentsandtheoil:surfactant ratiooftheSLNs,particlesrangedinsizefrom75nmto>200nm

[101,102].Invitrocytotoxicityassessmentshavebeencarriedout ondifferentacuteleukemiacelllines(HL-60,Jurkat,andTHP-1) [101]. On the HL-60 cell line, ATRA-loaded SLNs significantly decreased cell viability compared with ATRA in solution. On theJurkatandTHP-1celllines,ATRAwasinactive,evenathigh doses.WhenATRAwasloadedinSLNs,asignificant,butmodest, decreaseincellviabilitywasobserved.SeveralATRA-loadedNLCs havealsobeendevelopedtoincreasedrugloadingcomparedwith SLNs.Sabzichietal.useda high-pressurehomogenizationtech- niquetoproduceATRA-loadedNLCs[103].Amixtureofsolidand liquidlipidswasproduced,followedbytheadditionofasurfactant (poloxamer407)underhighpressurehomogenization.Ahotoil- in-water emulsion wasproduced before itwas cooled downto ambient temperature, resulting in particles from 65–115nm.

Chinsriwongkuletal.andOuriqueetal.developedNLCsforATRA encapsulation using an homogenization technique [104,105].

Particlesof200nmwithanarrowPDIandnegativezetapotential wereobtained,and the cytotoxicityof theATRA wasincreased comparedwith free solution in HL-60 cells. The concentration requiredtoachieve50%cellgrowthinhibitionforATRA-loaded LNCswashalforless(dependingontheoilusedtoproducethe NLCs)that requiredforanATRAsolution.[104].Ouriqueetal.

developedNLCsof213nmwithanincreasedlipid-coreviscosity toretainATRAandtoreduceATRAleakagefromNLCs.ATRAwas dispersedinthelipidcore,comprisingsorbitanmonostearateand acaprylic/caprictriglyceridemixture,anddemonstratedasimilar antiproliferativeeffectonanAMLcellline(HL-60)tothatofthe freedrug[105].

Aspreviouslydiscussed,newtargetsarebeinginvestigated,for example,proteinphosphatase2A(PP2A), whichis functionally inactivatedinnumerouscancersandleukemias[106,107].Phar- macologicalrestorationofPP2Aantagonizescancerdevelopment andprogression.Fingolimod,asphingosineanalog(alsoknownas FTY720)isonemoleculeabletoactivatePP2A[106].Itisawater- solubledrugwithahighoralbioavailabilityandisFDAapproved for patients with multiple sclerosis. Fingolimod has also been evaluatedforAMLtreatmentand,accordingtoasystemicevalua- tionofitssafetyandtoxicity,itappearstobeapotentialtreatment optionforAML,becauseofanacceptablebenefit:riskratioanda lack of bone marrow toxicity [108]. Fingolimod has also been encapsulatedinSLNspreparedbyhothomogenizationfollowed by sheer homogenization and ultrasonication [109]. NPs of 207nmwereobtained.Cytotoxicactivity inAMLcells(HL-60), pharmacokineticsfollowingoraladministration(inamice-based model), and antitumor activity of fingolimod-loaded NPs were evaluatedandcomparedwithafingolimodsolution.Fingolimod- loadedSLNsinducedahigherpercentageofapoptosis.Moreover, oralbioavailability wasenhancedbyafactorofnine(themaxi- mumdrugconcentrationwasenhancedby a factorof40),and significantly decreased tumor burden in a HL-60 xenografted mouse-basedmodel.

Therefore,NPscouldbedevelopedtoincreasetheoralbioavail- abilityofdrugs,allowingoraladministrationofdrugsonlyavail- able by intravenous infusion because of their physicochemical structures.Suchstrategieshavebeenproposedforhypomethylat- ing agents, suchas AZA or DEC, which are only available by subcutaneousorintravenousinfusion,respectively.Oralformula- tionsmightbemoreconvenientforpatients,eliminateinjection 10 www.drugdiscoverytoday.com

(11)

sitereactions,andallownewschedulesforimprovingdemethyla- tion[110].OralAZA(CC-486)hasbeenassessedinaPhase1trial [111]forAMLand MDS,and aPhase3trial inlow-riskMDS is ongoing(NCT01566695),showingtheinterestinthepossibleoral administrationofsuchagents.DECencapsulationinlipid-based NPsfororaladministrationiscurrentlyunderinvestigation[112–

114].Forexample,afterDECsolubilizationinTranscutol¹HPand Precirol¹ATO5,followedbyhomogenizationunderhighpressure (500–1000bar) in cold conditions, particles of 116nm were obtained by Neupane etal. [113].An exvivo permeation study (onanileumsegment)wasperformedandafourfoldincreasein DEC in vitro permeability was observed compared with a DEC solution[112].WhenDECwasloadedintoanNLCformulation of30nmobtainedusingalow-energyphaseinversionprocess,its cytotoxicityontwoAMLcelllines(HELandHL-60)wasenhanced byafactorof1.5onHELcellsand1.8onHL-60cellscompared withfreeDEC[114].Moreover,theapparentpermeabilityofan NCLformulationacrossaCaco-2cellsmonolayerwasenhancedby afactorofovernine.Themajorissuefortheclinicaldevelopment oforalnucleotideanalogsistheirrapidclearanceinthegutand liverbycytidinedeaminase.Adoseescalationtrialofconcomitant oraladministration ofDECand a cytidine deaminaseinhibitor (E7727)hasbeenconducted,withpromisingresults[115].

Lipid-based NPs are currently undergoing in vitro or in vivo assessmentswithregardtoAMLtreatment. Promisingstrategies havebeensuggestedtoenhancedrugtargeting,tomodifyphar- macokineticsofdrugsalreadyusedinAMLtreatment,ortomodify the administration routeto facilitate drug administration orto improvepatients’qualityoflife.Newtherapeuticstrategies,such asfingolimodusedforAMLindications,arealsobeingdeveloped.

Asaresultofstrategiesdevelopedtoenhancedrugloading,and afterqualitycontrolimprovementsinthefieldoflipidNPs,such strategiescanbetestedinclinicaltrials.

Metallicnanoparticles

MetallicNPshavemostly beendevelopedaspartoftheranostic strategies,becauseofacombinationofsuperparamagneticprop- ertiesandreactiveoxygenspeciesproductioncapabilitiesusedto destroycancercells.Theycanbeusedasdrugcarriersbecauseofa covalentconjugationwithdrugsorimagingagents.Chemothera- peuticdrugsgenerallyattachedtoironNPs arecharacterizedby lowmolecularweights,rapidbiodistribution,andrapidclearance.

OnceattachedtoironNPs,bloodcirculationtimeisincreasedas wellasEPReffects,enhancingtumorretention[116,117].Metallic NPs include iron, gold, silver, zinc, and titanium, and can be produced ona largescale, although they have to demonstrate theirgoodtoleranceandthelackofbodyaccumulationovertime ifchronicadministrationistoberealized.

NometallicNPformulationiscurrentlyunderclinicalinvesti- gation for AML indications. A strategy designed to conjugate cytarabinewith ironNPs coatedwith silicahasbeen presented

[118].Meanparticlesizevariedfrom23nmto57nmdepending onthemethodusedtomeasureit(microscopyordynamiclight scattering). The cytotoxic capacity of these negatively charged particles (11.5mV) was evaluated in AML cell lines (HL-60 and KG-1). An improvement of the antiproliferative effect of cytarabinewasobservedwhenitwasconjugatedwithironNPs.

The concentration ofNPs required to achieve50% cell growth inhibitionwashalfofthatrequiredforfreecytarabine.

Iron NPsarepreferred fortheirsuperparamagnetic properties andimagingcapability.UsingsuchNPstoconjugatechemothera- piescouldproveapromisingapproachforcombiningtreatment anddiagnostics.However,thereiscurrentlylittleworkassessing thisstrategyforAMLindications.

Concluding remarks

Here, wehave discussedcurrenttreatments andinnovativefor- mulations(Table1)thathaveeitherbeenrecentlyapprovedorare indifferentstagesofdevelopment(clinicaltrials,invivoorinvitro evaluations) for use against AML. These new formulations are designedtoovercomedrugresistanceortobypassadverseeffects ofAPIscurrentlyusedinchemotherapy(Fig.1).Asisthecasein otherpathologieswhereNPsaredeveloping,liposomesremainthe mostadvancedarea,withongoingclinicaltrialsandarecentFDA- approved formulation (Vyxeos¹). The second NP category for which future trialinvestigations areplanned is polymericNPs.

Clinical transfer fromlaboratory investigations using other NP strategiesmustaddressinadequatedrugloading,toxicityevalua- tions, andscaling-upissues.Aspreviouslyreported,thetransfer from benchto bedsideisa major challengefornanomedicines [119,120].

Finally,fourmainresearchapproachesaredevelopinginAML formulationresearch,namelyAMLcelllinetargeting,overcoming drug resistance, modifyingthe pharmacokinetics of drugs, and modifyingadministrationroutesforclinicalreasons.Ifpathology outcomescannotbemodifiedintheneartermforrefractoryAML, NPscouldhelptoimprovepatients’qualityoflifebyminimizing pain,inconvenience causedbythe administrationroute,or the adverse effects of treatment. Over the longer term, there is a requirement to ensure the best possible AML outcomes, and combininginnovativeformulationsandnewtargetsrecentlydis- coveredcouldcontributetosignificantimprovements.Innovative formulationsopenthefieldofpersonalizedtherapyincombina- tionwithnewlyidentifiedbiomarkers.

Conflicting interest statement

S.T.thatheisaconsultanttoAstellasandCelgene.

Acknowledgments

TheauthorsareverygratefultotheLiguecontreleCancer,and particularlytoitsMaineetLoireCommittee,foritsfinancial support.

References

1Rodriguez-Abreu,D.etal.(2007)Epidemiologyofhematologicalmalignancies.Ann.

Oncol.18(Suppl.1),i3–i8

2Shah,A.etal.(2013)SurvivalandcureofacutemyeloidleukaemiainEngland,1971–

2006:apopulation-basedstudy.Br.J.Haematol.162,509–516

3Segal,B.H.etal.(2007)Preventionandearlytreatmentofinvasivefungalinfectionin patientswithcancerandneutropeniaandinstemcelltransplantrecipientsintheera ofnewerbroad-spectrumantifungalagentsanddiagnosticadjuncts.Clin.Infect.Dis.

44,402–409