Liposomal amphotericin B(AmBisomeÒ)efficacy in confirmed invasive aspergillosis and otherfilamentous fungal infections in immunocompromised hosts:a pooled analysis
C.Cordonnier,1M.Bresnik2and R.Ebrahimi2
1Service d’He´matologie Clinique,Hoˆpital Henri Mondor,Creteil,France and2Gilead Sciences,Foster City,CA,USA
Summary A pooled efficacy analysis applying current diagnostic standards for case selection was
performed on previously published trials of liposomal amphotericin B for invasive
filamentous fungal infections(IFFI).Favourable responses were observed in51%of
microbiologically confirmed cases of proven or probable IFFI.Despite the limitations
inherent in a retrospective analysis of pooled studies,the response rates observed in this
analysis were consistent with previous reports for antifungal therapy with
amphotericin B deoxycholate or voriconazole in the treatment of invasive aspergillosis. Key words:AmBis
ome,amphotericin B,aspergillosis.
Introduction
Invasive aspergillosis(IA)and other invasivefilamen-tous fungal infections(IFFI)continue to be associated with significant morbidity and mortality,particularly in immunocompromised patients.Efficacy and survival rates for antifungal therapies for the treatment of IFFI have varied considerably depending on the patient population underlying disease,allogeneic stem cell transplantation(SCT),presence and severity of neutropenia,and receipt of immunosuppressive therap-ies].The liposomal formulation of amphotericin B(L-AMB)exhibits a broad spectrum of activity against yeast and moulds comparable with amphotericin B deoxych-olate1but with a more favourable safety profile that significantly reduces dose-limiting toxicities associated with amphotericin B deoxycholate.2,3The efficacy and safety of L-AMB have been reported in a number of randomised comparative trials,3,4as well as in open-label trials and retrospective analyses.5–9Subsequent to the publication of these studies,diagnostic criteria for defining and classifying proven and probable cases of IFFI in immunocompromised patients for inclusion in antifungal efficacy trials have been made uniform.10To further define the efficacy of L-AMB in the treatment of IA and other IFFI for patients treated in clinical trials before2002,a pooled analysis of four prospective, clinical trials was performed,a
pplying the currently accepted diagnostic standards for case selection and efficacy analyses.
Materials and methods
Patient population
Original case report forms(CRF)for patients treated with1–15mg kg)1day)1L-AMB[AmBisome;Gilead Sciences,Inc.,Foster City,CA(formerly Nexstar)and Fujisawa Healthcare,Inc.,Deerfield,IL,USA]collected from four clinical studies were reviewed individually. The CRF were obtained from three published,prospect-ive,multicentre clinical trials.3–5These studies included: L-AMB5mg kg)1day)1vs.amphotericin B deoxycho-late for the treatment of documented and suspected neutropenia-associated invasive fungal infections,3 L-AMB1mg kg)1day)1vs.L-AMB4mg kg)1day)1 for the treatment of IA4and a safety,tolerance and pharmacokinetic trial of high-dose L-AMB(7.5–15mg kg)1day)1).5The fourth study was a compas-sionate-use,multicentre study,from which cases have been previously reported in four publications.6–9
Correspondence:Catherine Cordonnier,Service d’He´matologie Clinique,
Hoˆpital Henri Mondor,51Avenue du Marechal de Lattre de Tassigny,Creteil
Cedex94010,France.
Tel.:+33(0)149812059.Fax:+33(0)149812067.
E-mail:carlcord@club-internet.fr
Accepted for publication27December2006
Original article
Journal CompilationÓ2007Blackwell Publishing Ltd•Mycoses(2007),50,205–209
Methodology and analysis
The diagnosis of proven or probable IFFI was deter-mined using criteria defined by the European Organisa-tion for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group(EORTC/MSG).10Cases were confirmed by histopathology or were microbiologically confirmed by culture or Aspergillus antigen.This was required so as to increase the level of certainty of IFFI and to avoid the inclusion of misleading cases only documented through halo sign or air-cresc
ent sign).Chest computerised tomography scan abnormalities were assumed to be non-halo infiinor EORTC/ MSG criterion)unless otherwise stated.Additional clinical signs and symptoms of infections reported in the CRF were required to fulfil clinical criterion for IFFI. All cases with the protocol-defined IFFI that met EORTC/MSG criteria for proven or probable IFFI and in which greater than or equal to one dose of L-AMB was administered were included in the pooled analysis.A favourable response(i.eplete or partial response) was based on the investigator-determined assessments, in which criteria for complete and partial responses were previously defined per each protocol.3–9Death or discontinuation from the study before response assess-ment could be made was considered a failure of therapy for this analysis.Survival was determined from the last follow-up visit reported in the CRF.
Results
Patients
A total of212cases were screened,of which69(33%) cases met the EORTC/MSG criteria for proven(n¼20) or probable(n¼49)IFFI.The proven and probable cases were identified infive of26cases reported in Leenders et al.[3],25of115cases reported in Ellis et al.
[4],21of44cases reported in Walsh et al.[5]and18of 27cases reported in the compassionate-use trial.6–9 Rejected cases included137cases that lacked a micro-biologically documented fungal pathogen and six cases that lacked appropriate host factors,including malig-nancy,organ transplant or infection with human immunodeficiency virus.Of the69confirmed proven or probable cases,microbiologic data used to confirm the diagnosis of IFFI included positive culture(n¼59), positive Aspergillus antigen assay(n¼3)and histopa-thology(n¼13).Six cases had both a positive culture and documented histopathology.The most common fungal isolate identified was Aspergillus species(n¼61), with six Zygomycetes and four Fusarium isolates were also documented.
The majority of patients(n¼61;88%)had haema-tological diseases,with38patients diagnosed with acute leukaemia.Of these61patients,30%(n¼18) had received an allogeneic SCT.Eight of the69patients had undergone a solid organ transplant,had a solid tumour or were infected with human immunodeficiency virus.There were51(74%)males included in the analysis,and the median age of the patient population was43years(range,14–72years).The median L-AMB dose administered was4mg kg)1day)1(range,1–15mg kg)1day)1),with48cases treated with L-AMB 1–5mg kg)1day)1and21cases treated with L-AMB 7.5–15mg kg)1day)1.L-AMB was administered as thefirst-line therapy in44patients and as the second-line therapy in25patients(20because of failure of the first-line treatment andfive because of amphoterici
n B-induced nephrotoxicity).None of the patients treated with L-AMB as the second-line therapy had received voriconazole or an echinocandin before entry into the trials.
Response
A favourable response(complete response,n¼15; partial response,n¼20)with L-AM
B was observed in 35of69(51%)cases:11of20(55%)cases of proven IFFI and24of49(49%)cases of probable IFFI(Table1). Treatment with L-AMB as thefirst-line therapy showed a higher favourable response(61%)compared with the administration of the second-line therapy(32%).In addition,patients with severe neutropenia(absolute neutrophil count<500cells mm)3)at baseline showed a response similar to that of patients without severe neutropenia,with16of34(47%)patients and19of35 (54%)patients achieving a favourable response respect-ively(Table1).In patients with haematological disease, a favourable response was observed in31of61(51%) patients(Table1).Of these61patients,eight of18 (44%)who received allogeneic SCT and four of seven (57%)who received autologous SCT showed a favour-able response with L-AMB.
Favourable response rates varied by the site of infection,ranging from44%(21of48cases)for pulmonary infections,64%(seven of11cases)for sinus/nasal infections,57%(four of seven cases)for disseminated i
nfections and one of one case each for subcutaneous abscess,pericarditis,and mastoiditis (Table1).Favourable response rates by pathogen indi-cated that L-AMB was effective against Aspergillus,
C.Cordonnier et al.
Journal CompilationÓ2007Blackwell Publishing Ltd•Mycoses(2007),50,205–209
Zygomycetes and Fusarium pathogens.A favourable response was observed in28of59(47%)patients with IA,five of six(83%)patients with zygomycosis and two of four patients with fusariosis.There were no apparent trends in response rates based upon the dose of L-AMB received(Table2).
Survival
Of the69patients with probable or proven IFFI,35 (51%)patients treated with L-AMB survived to the last follow-up visit.Three of the patients who survived were from the clinical trial reported by Leenders et al.[3],15 patients were from Ellis et al.[4],nine patients were from Walsh et al.[5],and eight patients were from the compassionate-use trial.6–9Of these surviving patients, 23of35patients had survival documented to ‡12weeks after the initiation of treatment.For the remaining12patients whose last study
visit was <12weeks following the initiation of L-AMB treatment, the mean follow-up time was65days(range,31–82days).
Discussion
The consensus guidelines defining the criteria for the diagnosis of invasive fungal infections10were developed to standardise the inclusion of patients in clinical trials. Cases of IFFI reported from studies before the latest standardisation of the diagnostic criteria by EORTC/ MSG are difficult to assess because of the inclusion of ÔsuspectedÕorÔprobableÕcases that lacked microbiologic confirmation of a fungal pathogen.Many of these cases were presumed to be aspergillosis based on fevers unresponsive to broad-spectrum antibacterial therapy, new and/or persisting pulmonary infiltrates and lack of isolation of specific pathogens.These would be charac-terised as possible cases by the current EORTC/MSG definitions.Furthermore,this has made it difficult to compare more recently conducted antifungal clinical studies to historic controls.Applying these newer diagnostic standards for identifying proven or probable IFFI to previously reported cases treated with L-AMB demonstrated a favourable response rate(complete or partial response)of51%in a population of69severely immunocompromised patients with microbiologically confirmed IFFI.Because of the small number of patients treated at each dose level(1–15mg kg)1day)1)and the different protocols followed,we cannot draw any conclusions on
the dose-related efficacy of L-AMB in proven or probable infections.However,these data help reaffirm the efficacy of L-AMB,using currently accepted standards for documenting IFFI.Additionally,these response rates are consistent with the results that were recently reported in a prospective,randomised trial of two dose regimens of L-AMB(3mg kg)1day)1vs. 10mg kg)1day)1)in confirmed mould infections(Am-BiLoad trial).In this study,favourable response rates were found in50%and46%of patients in the standard-and high-dose groups respectively.11
Applying these newer diagnostic standards also allowed for more precise determination of IA cases and for consideration of the efficacy results in the context of prospective studies recently reported.12,13
Table1Response rates according to presentation,treatment
indication,underlying disease and site of infection.
Overall response,n/n(%)1
Total352/69(51;95%CI39–63)
Diagnosis category
Proven11/20(55)
Probable24/49(49)
Treatment indication
First-line therapy27/44(61)
Second-line therapy8/25(32)
After failure offirst-line treatment6/20(30)
Because of nephrotoxicity2/5(40)
Neutropenia at baseline
ANC<500cells mm)316/34(47)
ANC‡500cells mm)319/35(54)
Status of underlying haematological disease3
Uncontrolled419/40(48)
Controlled412/21(57)
Underlying disease
Haematological disease31/61(51)
Acute leukaemia25/38(66)
Chronic myeloproliferative disorders1/4(25)
Lymphoproliferative disorders3/8(38)
Aplastic anaemia1/5(20)
Myeloma1/6(17)
Allogeneic stem cell transplantation8/18(44)
Solid organ transplant0/3(0)
Solid tumour2/2(100)
HIV infection2/3(67)
Site
Pulmonary21/48(44)
reportedSinus/nasal7/11(64)
Disseminated4/7(57)
Other53/3(100)
ANC,absolute neutrophil count;HIV,human immunodeficiency
virus.
1Complete or partial response.
2Complete response¼15;partial response¼20.
3Eight patients had non-haematological conditions.
4Uncontrolled disease¼evidence of active disease present at study
entry.
5One site each of subcutaneous abscess,pericardium or mastoid
process.
Liposomal amphotericin B efficacy Journal CompilationÓ2007Blackwell Publishing Ltd•Mycoses(2007),50,205–209
In an open-label,non-comparative trial of116patients with acute IA,13a complete or partial response was reported in35of60(58%)patients treated with voriconazole as afirst-line therapy and21of56 (38%)patients treated with voriconazole as a second-line therapy[6mg kg)1bid for2days,3mg kg)1IV twice daily(bid)for6–27days followed by oral 200mg day)1for up to24weeks].
In the larger,randomised comparative trial of IA published by Herbrecht et al.[12],patients were treated with voriconazole(6mg kg)1IV bid on day1;4mg kg)1 bid for‡7days followed by oral voriconazole200m
g bid) (n¼144)or with amphotericin B deoxycholate1.0–1.5mg kg)1day)1(n¼133).A complete or partial response was observed in32%and53%of patients treated with amphotericin B deoxycholate and voricon-azole respectively.Approximately one-third of these cases were included based on clinical and imaging criteria but without histologic or microbiologic confirmation of aspergillosis.For these cases that lacked microbiological confirmation,higher response rates were seen for both the voriconazole and amphotericin B deoxycholate groups(67.4%and42.9%respectively).This could represent a more favourable treatment effect associated with earlier intervention and/or lower fungal burdens. Recently,an open-label study of caspofungin as the second-line therapy was reported for83patients with proven or probable aspergillosis.14An overall(complete or partial)response was reported in45%of the83 patients.However,the overall response rate was only 14%in21allogeneic SCT recipients and26%in19 neutropenic patients.In the current analysis,an overall response was reported in32%of25patients treated with L-AMB as the second-line therapy.
In the current pooled analysis,the underlying con-ditions resulting in immunosuppression were generally similar to those in the previously reported trials.12,13For example,in Herbrecht et al.[12]and Denning et al.[13]the number of patients who had undergone allogeneic SCT(20–26%),had acute leukaemia(35–45%),had solid organ transplantation(3–6%)or were infected with human immunodeficie
ncy virus(4–5%)was gen-erally similar to the numbers in the current study.In addition,34(49%)patients were neutropenic at base-line in the current analysis,similar to that reported (45%)in the randomised study comparing voriconazole with amphotericin B deoxycholate12and substantially higher than the24%reported in the non-comparative trial of voriconazole.13Based on the patient populations and similar diagnostic criteria applied to document IFFI, the clinical response rate in this pooled analysis for L-AMB(51%)in proven or probable infection was higher than that reported for amphotericin B deoxych-olate(32%)and within the range of clinical response reported for voriconazole(48–53%)in the treatment of IA.It is important to note that L-AMB is active against Zygomycetes infections;however,voriconazole and caspofungin are not active against this family of organisms.Zygomycetes infections may be difficult to distinguish from Aspergillus infections clinically or via imaging;therefore,this limitation with voriconazole should be taken into consideration when treating patients with IFFI if microbiological confirmation of the fungal pathogen is lacking.
In the current study,survival to the last follow-up visit in patients with proven or probable IFFI was51%, with66%of these patients surviving for‡12weeks. Because this analysis was a retrospective case review of patients from multiple clinical trials that followed different protocols,it is difficult to draw further conclu-sions from the survival data or compare the favourable results in this analysis to other more rec
ently published trials.However,it is encouraging that more than half of the cases of proven or probable IFFI in a population of severely immunocompromised patients survived to the last follow-up visit.
Table2Response rates according to dose
and regimen.
Dose
(mg kg)1day)1)Patients(n)Overall response,
n/n(%)1Response by dose regimen
1169/16(56)Standard dose(1–5mg kg)1day)1)
26/48(54%)
2105/10(50)
342/4(50)
41327/13(54)
553/5(60)
7.542/4(50)High dose(>5mg kg)1day)1)
9/21(43%)
1052/5(40)
12.553/5(60)
1572/7(29)
1Complete or partial response.
2Median.
C.Cordonnier et al.
Journal CompilationÓ2007Blackwell Publishing Ltd•Mycoses(2007),50,205–209
In conclusion,L-AMB has documented efficacy in the treatment of proven or probable IFFI in cases me
eting current diagnostic criteria.Data from this retrospective analysis suggest an overall response rate(51%)that is comparable with voriconazole and higher than that of amphotericin B deoxycholate;however,these observa-tions have not been confirmed in randomised compa-rator trials.Furthermore,conclusions drawn from this study are limited in that it was a pooled analysis of selected studies that differed in drug dosage and diagnostic criteria.Despite these limitations,mounting evidence supports the role of AmBisome as an effective antifungal therapy.
Acknowledgments
This work was supported by Gilead Sciences.
References
1Ellis D.Amphotericin B:spectrum and resistance.J Anti-microb Chemother2002;49(Suppl.1):7–10.
2Walsh TJ,Finberg RW,Arndt C et al.Liposomal ampho-tericin B for empirical therapy in patients with persistent fever and neutropenia.National Institute of Allergy and Infectious Diseases Mycoses Study Group.N Engl J Med
1999;340:764–71.
3Leenders AC,Daenen S,Jansen RL et al.Liposomal amphotericin B compared with amphotericin B deoxych-olate in the treatment of documented and suspected
neutropenia-associated invasive fungal infections.Br J
Haematol1998;103:205–12.
4Ellis M,Spence D,de Pauw B et al.An EORTC interna-tional multicenter randomized trial(EORTC number
19923)comparing two dosages of liposomal amphotericin
B for treatment of invasive aspergillosis.Clin Infect Dis
1998;27:1406–12.
5Walsh TJ,Goodman JL,Pappas P et al.Safety,tolerance, and pharmacokinetics of high-dose liposomal amphoteri-cin B(AmBisome)in patients infected with Aspergillus
species and other filamentous fungi:maximum tolerated
dose study.Antimicrob Agents Chemother2001;45:3487–
96.
6Ringde´n O,Meunier F,Tollemar J et al.Efficacy of amphotericin B encapsulated in liposomes(AmBisome)in the treatment of invasive fungal infections in immuno-
compromised patients.J Antimicrob Chemother1991;28: 73–82.
7Tollemar J,Ringde´n O.Early pharmacokinetic and clinical results from a noncomparative multicentre trial of
amphotericin B encapsulated in a small unilamellar lipo-some(AmBisomeÒ).Drug Invest1992;4:232–8.
8Mills W,Chopra R,Linch DC,Goldstone AH.Liposomal amphotericin B in the treatment of fungal infections in
neutropenic patients:a single-centre experience of133
episodes in116patients.Br J Haematol1994;86:754–
60.
9Ng TT,Denning DW.Liposomal amphotericin B(AmB-isome)therapy in invasive fungal infections.Evaluation of United Kingdom compassionate use data.Arch Intern Med 1995;155:1093–8.
10Ascioglu S,Rex JH,de Pauw B et al.Defining opportunistic invasive fungal infections in immunocompromised pa-
tients with cancer and hematopoietic stem cell trans-
plants:an international consensus.Clin Infect Dis2002;
34:7–14.
11Cornely OA,Maertens J,Bresnik M,Herbrecht R.Lipo-somal amphotericin B(L-AMB)as initial therapy for
invasive filamentous fungal infections(IFFI):a random-ized,prospective trial of a high loading regimen vs.
standard dosing(AmBiLoad Trial).Blood2005;106:
900a(abstract3222).
12Herbrecht R,Denning DW,Patterson TF et al.Voricon-azole versus amphotericin B for primary therapy of inva-sive aspergillosis.N Engl J Med2002;347:408–15.
13Denning DW,Ribaud P,Milpied N et al.Efficacy and safety of voriconazole in the treatment of acute invasive asper-gillosis.Clin Infect Dis2002;34:563–71.
14Maertens J,Raad I,Petrikkos G et al.Efficacy and safety of caspofungin for treatment of invasive aspergillosis in pa-tients refractory to or intolerant of conventional antifun-gal therapy.Clin Infect Dis2004;39:1563–71.
Liposomal amphotericin B efficacy
Journal CompilationÓ2007Blackwell Publishing Ltd•Mycoses(2007),50,205–209
发表评论