Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, December 2004, p. 4808-4812, Vol. 48, No. 12
0066-4804/04/$08.00+0 DOI: 10.1128/AAC.48.12.4808-4812.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
UZ Gent, Ghent,1 UZ Gasthuisberg, Leuven,3 Clinique Universitaire Mont Godinne, Yvoir,4 EORTC Data Center, Brussels, Belgium,14 St. Elisabeth Cancer Institute, Bratislava, Slovakia,2 Hadassah University Hospital, Jerusalem, Israel,5 Hôpital Saint Louis,6 Hotel-Dieu, Paris,10 Hôpital Henri Mondor, Créteil,7 CHR Hotel-Dieu, Nantes,11 Hôpital de Hautepierre, Strasbourg, France,15 University Medical Centre, Nijmegen,8 Leiden University Medical Center, Leiden, The Netherlands,12 Royal Brisbane Hospital, Brisbane, Australia,9 University General Hospital, Heraklion, Greece,13
Received 31 December 2003/ Returned for modification 12 February 2004/ Accepted 25 June 2004
|
|
|---|
|
|
|---|
Until the recent approval of voriconazole, standard treatment of IA consisted of intravenous amphotericin B given at doses of at least 1 mg/kg of body weight/day (11, 24). Amphotericin B is associated with considerable toxicity and frequent treatment failure, particularly in the case of advanced disease and prolonged neutropenia. The liposomal formulation of amphotericin B has allowed the administration of more than fivefold doses of the drug with considerable improvement in the safety profile (15, 25).
The liposomal encapsulation technique can, however, be extended to other antifungal agents that were previously excluded for treatment of systemic fungal infections due to the lack of a well-tolerated intravenous formulation. Nystatin is a naturally occurring product of Streptomyces noursei that has been used as a topical fungicidal drug since the 1950s. It is active against Candida, Cryptococcus, Histoplasma, Blastomyces, and Aspergillus spp. and has a spectrum of antifungal activity similar to that of amphotericin B (1, 4, 13, 22).
Early attempts at free drug administration resulted in dose-limiting toxicity. Recent studies have demonstrated that nystatin can be incorporated into liposomes with significant reductions in toxicity while preserving antifungal activity (4, 8, 10, 13, 18, 19, 22).
This approach has resulted in extensive animal testing, with good results in short- and long-term toxicity studies with healthy rodents and dogs (9, 19). Efficacy has been shown for experimental infections including aspergillosis: liposomal nystatin at doses of 2 and 4 mg/kg/day, but not at 1 mg/kg/day, prolonged survival for neutropenic rabbits and reduced tissue fungal burden (8). In an experimental murine aspergillosis study, liposomal nystatin at a dose of 5 mg/kg given four times within the first week of infection was as effective as liposomal amphotericin B and more effective than amphotericin B deoxycholate and amphotericin B lipid complex (7). Clinical studies are very limited and have not yet been published. They include a phase I maximal-tolerated-dose study, a compassionate-use study, and phase II and III trials for candidemia, for empirical therapy of persistent febrile neutropenia, and for cryptococcal meningitis (2, 14, 20). Doses used in clinical trials ranged from 2 to 4 mg/kg for empirical therapy of persistent febrile neutropenia and for candidemia (23, 26). To test the activity and safety of liposomal nystatin in documented or probable IA, the European Organisation for Research and Treatment of Cancer (EORTC) Invasive Fungal Infections Group undertook a clinical phase II study of liposomal nystatin for patients who had previously been treated with amphotericin B and found to be resistant or intolerant.
(This work was presented in part at the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Ontario, Canada, 17 to 21 September 2000.)
|
|
|---|
Patient population.
Patients aged 6 years or older were included if they were shown to have definite or probable Aspergillus infections, had either failed to respond to or were intolerant of conventional amphotericin B and/or a lipid formulation of amphotericin, and required continued antifungal therapy. Reasons for noninclusion were evidence of hepatic function impairment (baseline bilirubin level of 4 mg/dl or higher or baseline transaminase levels more than 5 times the upper limit of the normal value), severe renal dysfunction (
4.5 mg of creatinine/dl or on dialysis), life expectancy of less than 14 days due to underlying disease, or pregnancy or lactation.
Definitions. In the absence of consensus definition criteria at the time the study was designed, the following criteria, which later proved to be close to the EORTC/MSG criteria published in 2002 (3), were used. Definite aspergillosis was defined as tissue histopathology showing septate acute branching hyphae with a positive culture for an Aspergillus sp. from the same site or, in the absence of histopathology, a positive culture from tissue obtained by an invasive procedure, such as transbronchial biopsy or percutaneous needle aspiration, from a consolidated tissue or closed body fluid.
Probable aspergillosis was defined in the context of neutropenia (<500 neutrophils/µl), administration of cytotoxic agents for a malignant or immunologic disease, a corticosteroid dosage of more than 10 mg of prednisone or the equivalent daily, or a congenital or acquired immunodeficiency. Patients with these preexisting conditions had probable aspergillosis if either of the following radiological signs was demonstrated in the presence of clinical signs and symptoms: (i) for pulmonary infection, chest radiography or computed tomography (CT) scan showing new nodular or cavitary lesions, in addition to two sputum cultures or one bronchoalveolar lavage (BAL) or brushing culture testing positive for Aspergillus species or cytologic examination of BAL fluid showing characteristic septate hyphae; (ii) for central nervous system infection, CT scan or magnetic resonance imaging demonstrating symptomatic lesions compatible with a previous diagnosis of definite aspergillosis elsewhere.
Patients were considered refractory to conventional treatment if they had received the equivalent of at least 5 days of amphotericin B (
0.8 mg/kg/day), amphotericin B lipid complex (
5 mg/kg/day), liposomal amphotericin B (
3 mg/kg/day), or amphotericin B colloidal dispersion (
3 mg/kg/day) and nevertheless showed no improvement upon radiographic assessment and/or showed persistence of fever (
38.3°C) in the absence of any other cause.
Patients were considered intolerant of amphotericin B if they developed or had preexisting nephrotoxicity, indicated by a serum creatinine level of
2.5 mg/dl (>1.5 mg/dl for pediatric patients), or if they developed severe infusion-related adverse events not controlled by premedication.
Antifungal therapy.
Liposomal nystatin was provided by Aronex Pharmaceuticals, Inc. (The Woodlands, Tex.). Patients were started on a 4-mg/kg/day dose of liposomal nystatin. Liposomal nystatin was reconstituted with saline to produce a 1-mg/ml solution and was delivered at an infusion rate of 2 mg/min once daily. For patients with preexisting renal impairment (baseline creatinine level,
2.5 mg/dl), the initial dosage level was 2 mg/kg/day.
The option to increase the daily dose to 6 mg/kg after at least 5 days of therapy at the initial dosage, or to decrease to 2 mg/kg, was left to the investigator's clinical assessment of response and tolerance. Patients continued treatment until failure or to a maximum of 40 days.
The initial dose was given without any premedication. In case of infusion-related side effects, the infusion of liposomal nystatin was temporarily interrupted and administration of paracetamol or meperidine was allowed. If shivering did not subside, the use of chlorpheniramine or steroids was left to the discretion of the investigator. Hypotension was treated with intravenous fluid replacement.
Systemic use of additional antifungal agents for treatment of aspergillosis or another fungal infection disqualified patients from this study.
Follow-up and evaluation. Patients were followed up clinically on a daily basis. Routine laboratory tests, creatinine clearance, and chest X rays were performed on the day of inclusion, days 3 and 5, and weekly thereafter. Safety was assessed by using modified National Cancer Institute (NCI) Common Toxicity Criteria.
Activity was assessed at the end of treatment, and survival was evaluated up to 1 month later. Case report forms and imaging were collectively reviewed by the two coordinators of the study and two site investigators in order to reach a consensus on eligibility and response evaluation. Response evaluation was as follows: complete response was defined as the resolution of all attributable symptoms and clinical and radiographic signs; partial response was defined as an improvement in attributable symptoms and clinical signs and a decrease of at least 50% in radiographic signs; a patient was considered to have stable disease if, at the end of the study, he or she was alive and had only minor or no improvement; failure was defined as deterioration in attributable clinical and/or radiographic abnormalities.
Statistical methods.
The main end point in this trial was the response rate. A two-stage Simon design was followed such that if the true underlying response rate with liposomal nystatin was 20% or more, the probability of rejecting the drug for further study should be
0.05 (ß = 0.05), whereas if the true underlying response rate was
5%, the probability of rejecting the drug for further study should be
80% (
= 0.20).
Under these assumptions, 16 patients were initially entered into the study. If at least 1 success was observed among these 16, an additional 17 patients would be entered, for a total of 33. The drug could be rejected as being inactive if two or fewer responses were observed and should be considered of potential interest if three or more responses were observed.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Baseline demographics and characteristics of the 26 eligible patients
|
|
View this table: [in a new window] |
TABLE 2. Characteristics of invasive aspergillosis at initiation of liposomal nystatin for the 26 eligible patients
|
Among the 25 evaluable patients, 1 achieved a complete response. He was a 67-year-old man with colorectal cancer who was on a ventilator for pulmonary failure secondary to anaerobic sepsis following a colostomy. He developed fungal pneumonia, with cultures of the tracheal aspirate positive for both Aspergillus fumigatus and Aspergillus flavus. He received 70 mg of amphotericin B/day for 14 days while on dialysis and was switched to liposomal nystatin because of absence of clinical and radiological improvement and persistent positive cultures. He became asymptomatic and totally cleared CT scan abnormalities. There were six partial remissions and four patients with stable disease, three of whom progressed after discontinuation of liposomal nystatin. Fourteen patients failed despite 1 to 48 days of treatment. Thus, the overall favorable response rate was 7 of 25 (28%; 95% confidence interval, 12 to 49%).
Two patients underwent a lobectomy after achieving a partial response, one because of hemoptysis and one at the discretion of the investigator. Both remained disease free at 1 month posttreatment.
Treatment responses for the 25 evaluable patients with respect to definite or probable IA and intolerance or refractoriness to amphotericin B are shown in Table 3.
|
View this table: [in a new window] |
TABLE 3. Responses of 25 evaluable patients to liposomal nystatin according to the degree of certainty and the reason for inclusion
|
|
View this table: [in a new window] |
TABLE 4. Adverse events reported for the 33 enrolled patients
|
For two patients, infusion-related toxicity was the reason for discontinuation of drug treatment; one experienced severe shivering, chills, and hypertension on the first infusion, and the other decided to discontinue in the middle of the second infusion because of respiratory distress and chills.
Non-infusion-related side effects included a decline in renal function for 10 patients (all with grade 1 toxicity), electrolyte disturbances (hypokalemia for 13 patients [8 with grade 1 and 5 with grade 2 toxicity] and grade 1 hypocalcemia for 4), liver toxicity for 8 patients (6 with grade 1 and 2 with grade 2 toxicity), and diarrhea for 4 patients. Arrhythmias were not noted in this study; neither were influences on glucose or fat metabolism. Hematological toxicity was not observed for 15 patients starting with normal blood counts. For the other patients, anemia, neutropenia, and thrombocytopenia were present at the start of treatment and were not noticeably influenced by liposomal nystatin. In a population with underlying hematological malignancies that are not in remission, any drug for refractory Aspergillus infection is difficult to score for hematological toxicity.
Survival and causes of death. Of the 25 eligible patients, 17 died and 8 remained alive at the end of treatment. All eight patients who were alive at the end of treatment remained alive 1 month thereafter. For nine patients the primary cause of death was IA, while eight patients died of underlying malignancy in spite of improvement of the fungal infection.
|
|
|---|
The severity of fungal disease in our patients needs to be stressed. All patients but two were refractory to previous conventional amphotericin B therapy and/or a lipid formulation of amphotericin B. Although patients could be included after only 5 days of amphotericin B therapy according to the design of the study, the median duration of previous therapy was much longer. This long duration and the high median cumulative dose of amphotericin B that was given prior to inclusion suggest true refractoriness and not an early and inappropriate conclusion of failure. In addition, the median Apache II score of 21 also confirms the severity of the underlying condition and/or of the fungal infection.
Interestingly, all 7 responses occurred in the subgroup of 23 patients refractory to previous amphotericin B therapy, leading to a 30.4% favorable response rate in this population of patients with severe disease. Only a few studies on salvage therapy with recently developed antifungal drugs have been specifically devoted to IA. The response rate for patients failing previous therapy is not always precisely indicated. The response rate we observed in this population is in the range of previously published results for caspofungin (39.4% in a group of 71 refractory patients), voriconazole (37.5% of 56 patients refractory to or intolerant of previous therapy), liposomal amphotericin B (45.5% of 11 patients in a single-center study), and a colloidal dispersion of amphotericin B (29.2% of 48 patients included in five phase I/II studies) (6, 12, 17, 21).
Infusion-related events, including chills, shivering, fever, hypo- or hypertension, and tachycardia are, as for amphotericin B, the most common adverse events during liposomal nystatin therapy. They remained mild or moderate in all but three cases. Liposomal nystatin was associated with nephrotoxicity, although limited in severity, in 10 patients. The investigator believed it was appropriate to reduce the daily dosage to 2 mg/kg in three cases. Hypokalemia was more common but never severe and did not limit the dose or duration of therapy. Anemia, thrombocytopenia, and leukopenia, including grade 3 and 4 toxicity, were frequently reported, but the relationship with the study drug cannot be ascertained, because all the patients affected by these events suffered from hematological malignancies and had abnormal blood cell counts at baseline. Such hematological adverse events are more likely due to the underlying condition than to the antifungal therapy.
This study demonstrates that liposomal nystatin at a daily dose of 4 mg/kg can be effective for salvage therapy of invasive aspergillosis. However, its administration is associated with frequent infusion-related adverse events and mild renal toxicity, which would, in the context of other currently available drugs, reduce its level of utility in the antifungal armamentarium.
We thank Ann Marinus and Stéphane Lejeune for assistance in analyzing the data and preparing the manuscript.
Members of the EORTC Invasive Fungal Infections Group are as follows: Liliana Baila (coordinating physician), Hamdi Akan, Mickael Aoun, Sibel Ascioglu, Daniel Benhammou, Zwi Berneman, Hartmut Bertz, Jacques Bille, J. Magnus Bjorkholm, Igor Wolfgang Blau, Angelika Boehme, Aida Botelho de Sousa, Patrick Boutard, Stephane Bretagne, Denis Caillot, Thierry Calandra, Bernadeta Ceglarek, Francois Chapuis, Jose Miguel Cisneros, Alain Cometta, Oliver Cornely, Robrecht de Bock, Siem de Marie, David W. Denning, Anna Dmoszynska, Peter Donnelly, C. Duhamel, Hermann Einsele, Michael Ellis, Zoran Erjavec, Peter Ernst, Edgar Faber, Bertrand Gachot, Jorge Garbino, E. Gautier, Mareva Giacchino, Raffaella Giacchino, Axel Glasmacher, Renee Grillot, Andrease Groll, Jan Haber, Petr Hamal, Ian Hann, Ulrich Jehn, Elizabeth Johnson, Alain Kentos, Winfried Kern, Chris C. Kibbler, Paul Kotoucek, Bart Jan Kullberg, Jean-Paul Latge, Bernadette Lebeau, Jean-Claude Legrand, R. Lindblad, Per Ljungman, Olivier Lortholary, Johan Maertens, Rodrigo Martino, Georg Maschmeyer, Jacques Meis, Francoise Meunier, Mauricette Michallet, Marco Montillo, Frank Michael Mueller, D. Nemet, Nicole Nolard-Tintigner, Karoly Pecze, George Petrikkos, Ray L. Powles, Elisabeth Presterl, C. Rayon, Jorg Ritter, Tom Rogers, Emmanuel Roilides, Montserrat Rovira, Markus Ruhnke, Stefan Schwartz, Dominique Selleslag, David G. Spence, Anne Thiebaut, Eckhard Thiel, Jan Tollemar, Janez Tomazic, Andrew J. Ullmann, Omrum Uzun, Catalina Vadell-Nadal, Bernard Vandercam, Paul Verweij, Claudio Viscoli, and Marianna Viviani.
Participants in the EORTC Invasive Fungal Infections Group are listed in Acknowledgments. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»