Fax: (713) 745-6839
Itraconazole added to a lipid formulation of amphotericin B does not improve outcome of primary treatment of invasive aspergillosis†
Version of Record online: 20 APR 2005
Copyright © 2005 American Cancer Society
Volume 103, Issue 11, pages 2334–2337, 1 June 2005
How to Cite
Kontoyiannis, D. P., Boktour, M., Hanna, H., Torres, H. A., Hachem, R. and Raad, I. I. (2005), Itraconazole added to a lipid formulation of amphotericin B does not improve outcome of primary treatment of invasive aspergillosis. Cancer, 103: 2334–2337. doi: 10.1002/cncr.21057
Presented, in part, at the 44th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, DC, October 29–November 2, 2004.
- Issue online: 18 MAY 2005
- Version of Record online: 20 APR 2005
- Manuscript Accepted: 18 JAN 2005
- Manuscript Received: 11 JAN 2005
- lipid amphotericin B formulations
Invasive aspergillosis (IA) is associated with poor outcome in patients with hematologic malignancy treated with amphotericin B (AMB)-based therapy. Itraconazole (ITC), a triazole with activity against Aspergillus, has been used in combination with AMB or lipid formulations of AMB (LipoAMB) in the treatment of IA, although the efficacy of this strategy is uncertain.
To determine whether the addition of ITC to LipoAMB improves outcome of IA, the authors retrospectively studied 179 consecutive patients with hematologic malignancies and definite or probable IA who received primary antifungal therapy with either LipoAMB (n = 146), or lipoAMB plus ITC (n = 33) between June 1993 and June 2003. In view of the erratic absorption of ITC tablets, only patients who received either intravenous or liquid ITC were analyzed. Patients who received < 1 week of treatment were excluded.
Evaluable patients in both groups (LipoAMB: n =101; ITC and LipoAMB: n = 11) had comparable distribution of risk factors of poor outcome such as neutropenia at onset of IA, persistent neutropenia, systemic steroids, previous antifungal prophylaxis, admission to the intensive care unit, disseminated IA, previous bone marrow transplant, and IA due to infection by a non-fumigatus Aspergillus species. Response to primary antifungal therapy was equally poor in both groups (LipoAMB group: 10%; ITC and LipoAMB group: 0%; P = not significant).
In the authors' 10-year study of patients with hematologic malignancy and IA, the response rate to LipoAMB given as primary therapy was very poor. In a comparable group of patients, the addition of ITC did not result in a therapeutic benefit. Cancer 2005. © 2005 American Cancer Society.
Invasive aspergillosis (IA) is a leading cause of death due to infections in patients with hematologic malignancies.1 Pneumonia is the most frequent clinical manifestation of this opportunistic mycosis.1 The efficacy of current antifungal therapies against IA is suboptimal, and the disease has a mortality rate approaching 60–80% in patients with acute leukemia and in bone marrow transplant (BMT) recipients.2 The availability of newer amphotericin B formulations with an improved therapeutic index and of itraconazole (ITC), a triazole with activity against Aspergillus species, has renewed interest in combination treatments that, when given either concomitantly or sequentially, could result in additive or synergistic effects.3 The clinical validity of such combinations is unknown, as there is little human data.3 Most of these data are derived from salvage treatment strategies where the confounders are very hard to assess.3 In particular, there are virtually no data about antifungal combinations given as primary therapy in IA.3 Furthermore, the existing preclinical studies suggest that the combined or staggered administration of amphotericin B (AMB) with ITC or other triazoles might not be beneficial in patients with IA.3–11 We retrospectively reviewed our experience using the combination of ITC with lipid formulations of amphotericin B (LipoAMB) compared with LipoAMB monotherapy, given as primary treatment in patients with hematologic malignancies who had documented IA.
We identified patients who received LipoAMB formulations such as liposomal AMB (ambisome), AMB lipid complex, and AMB colloidal dispersion with or without ITC at The University of Texas M. D. Anderson Cancer Center (Houston, TX) between June 1993 and June 2003. We analyzed only patients who received primary treatment for documented (proven or probable) IA. Patients with possible aspergillosis were excluded. Patients receiving LipoAMB with or without ITC as salvage therapy were also excluded from the study. We extracted the following information from the medical records: age, gender, underlying malignancy, bone marrow transplantation (BMT) type (autologous or allogeneic), presence of graft-versus-host disease, presence of pulmonary or extrapulmonary diseases, severity of the underlying immunosuppression (e.g., duration and degree of neutropenia, steroid use within 1 month before the initiation of the combination), and recovery from neutropenia. Patients were divided into two groups, i.e., those who received combination therapy (ITC and LipoAMB) and those who received LipoAMB alone. Response was assessed at the end of primary antifungal therapy. If the patient continued to receive long-term antifungal therapy, the global response assessment was made immediately before the time at which the patient's physician considered the goal of further therapy to be secondary prophylaxis. Because of the erratic absorption of ITC capsules, we included in our analysis only patients treated with oral ITC liquid solutions or intravenous ITC. Efficacy was considered evaluable if the patients received the ITC and LipoAMB combination for ≥ 7 days. All patients who received the combination for ≥ 1 day were evaluated for toxicity.
IA was defined as proven, or probable, according to recently published guidelines.12 The end of primary antifungal therapy was defined as the point of discontinuation of successful LipoAMB treatment with or without ITC or as treatment withdrawal by the primary team because of evidence of progressive infection, toxicity, or death. Response was defined as resolution or major improvement of symptoms and signs of IA (combined with improvement of radiologic changes on chest X-rays or computed tomography scans by 75%) and the requirement of no further systemic antifungal treatment in the treating physician's judgment. Failure was defined as the deterioration or lack of significant improvement of these same variables (including death of the patient or drug withdrawal with evidence of infection still present) after ≥ 7 days of therapy or as the need to discontinue drugs due to toxicity.
The combination of ITC and LipoAMB was defined as concomitant administration of both agents (started within 24 hours of each other) for ≥ 7 days. ITC was given as a 200-mg loading dose (orally or intravenously) twice daily for 2 days, followed by a daily dose of 200 mg, whereas LipoAMB was given at a daily dose of 5 mg/kg as initial therapy. Patients with a serum creatinine level of ≥ 2.5 mg/dL were considered to have mild to moderate renal insufficiency. Neutropenia was defined as an absolute polymorphonuclear cell count < 500 cells per microliter.
Descriptive and inferential statistics were obtained with attention to parameters associated with success or failure. For categorical data, we used the chi-square or Fisher exact text for analysis. For numerical data, we used the Student t test or the Wilcoxon rank-sum test if appropriate. All testing was 2 sided and a significance level of 0.05 was used. No adjustments were made for multiple comparisons. All analyses were performed using the SPSS software program (version 12.0 for Windows; SPSS Inc., Chicago, IL). Of the 179 consecutive patients who received ITC and LipoAMB (n = 33) and LipoAMB alone (n = 146) as primary therapy for IA during the study period, 112 (63%) were evaluable (ITC and LipoAMB in 11 patients, LipoAMB alone in 101 patients). The reasons for exclusion of 22 patients in the ITC and LipoAMB group were ITC capsules in 19 patients and administration of the combination for < 7 days in 3 patients. The reasons for exclusion of 45 patients in the LipoAMB group was administration of LipoAMB for < 7 days. Table 1 shows the characteristics of these 112 patients. In both groups, the underlying malignancy in most patients was leukemia, and median age and gender were comparable. Similarly, underlying allogeneic BMT, neutropenia and its duration, percentage of neutrophil recovery during the course of treatment, receipt of corticosteroids within 1 month from the diagnosis of IA, and rates of infection by Aspergillus fumigatus or non-fumigatus Aspergillus species were comparable in the 2 cohorts (Table 1). Finally, both groups had comparable frequencies of extrapulmonary IA and use of immunomodulators (γ-interferon, granulocyte-macrophage–colony-stimulating factor) or white blood cell transfusions (Table 1).
|Characteristics||LipoAMB + ITCa||LipoAMBa|
|(n = 11) (%)||(n = 101) (%)|
|Male gender||6 (55)||61 (60)|
|Mean age (yrs) ± SD||54 ± 12||46 ± 16|
|Definite IA||4 (36)||39 (39)|
|Probable IA||7 (64)||62 (61)|
|Site of IA|
|Invasive pulmonary aspergillosis||9 (82)||80 (79)|
|Disseminated aspergillosis||1 (9)||11 (11)|
|Other||1 (9)||10 (10)|
|Acute leukemia||6 (55)||53 (52)|
|Chronic leukemia and other hematologic malignancies||3 (27)||23 (23)|
|Lymphoma/myeloma||2 (18)||25 (25)|
|History of BMT within previous year||3 (27)||41 (41)|
|Neutropenia (< 500 ANC) during previous month||6 (55)||64 (63)|
|Mean duration (days) ± SD||17 ± 7||18 ± 11|
|Median (range)||16 (10–31)||17 (1–31)|
|Neutropenia (< 500 ANC) at onset of IA||6 (55)||51 (50)|
|Mean duration (days) ± SD||25 ± 18||18 ± 16|
|Median (range)||26 (2–45)||14 (1–70)|
|Persistent neutropenia||2/6 (33)||22/51 (43)|
|Corticosteroid use during previous month||2 (18)||30 (30)|
|Corticosteroid use during infection||9 (82)||76 (75)|
|Immunotherapy during infection|
|Leukocyte transfusion||4 (36)||17 (17)|
|Growth factors||8 (73)||81 (80)|
|γ-Interferon||2 (18)||8 (8)|
|ICU during the infection||5 (46)||41 (41)|
|Mean duration (days) ± SD||12 ± 6||14 ± 11|
|Median (range)||9 (5–21)||10 (1–42)|
|Prophylactic antifungal before infection||7 (64)||77 (76)|
|Prophylactic fluconazole||3 (27)||46 (46)|
|Aspergillus-active prophylaxis||5 (46)||55 (54)|
|Aspergillus fumigatus||5 (46)||22 (22)|
|Aspergillus terreus||1 (9)||28 (28)|
|Aspergillus flavus||2 (18)||20 (20)|
|Other||3 (27)||31 (31)|
|Response at end of primary therapy||0||10 (10)|
|Death during hospitalization||10 (91)||77 (76)|
|Death with aspergillosis||5 (46)||45 (45)|
|2-week mortality||2 (18)||13 (13)|
|4-week mortality||4 (36)||34 (34)|
|Duration of therapy (days)|
|Mean ± SD||19 ± 11||20 ± 14|
|Median (range)||17 (8–45)||15 (7–85)|
End-of-treatment responses in the two groups were equally poor. The median durations of administration of ITC and LipoAMB and of LipoAMB were 17 days (range, 8–45 days) and 15 days (range, 7–85 days), respectively. The 2-week and 4-week mortality rates in the 2 groups were 18% and 36% and 13% and 34%, respectively (P = not significant). Previous receipt of fluconazole or ITC prophylaxis did not influence the response rates to LipoAMB. Finally, both treatments (LipoAMB, LipoAMB and ITC) were well tolerated as primary antifungal therapy. Mild to moderate renal insufficiency occurred in 5 (5%) of the patients receiving LipoAMB monotherapy.
The efficacy of AMB-based antifungal therapy for documented IA in high-risk patients with hematologic malignancy is poor.13 For example, in recent randomized trials, the response rate of IA to AMB-d14 or liposomal AMB15 was as low as 10%. Therefore, administration of combination antifungal regimens for documented IA has been suggested as a strategy to improve the outcome for this relatively refractory mycosis. However, no prospective clinical trials to date have examined the use of these regimens against IA in humans.3 This is not surprising, because such studies are logistically difficult to perform and because complex host factors can confound the true efficacy of antifungal combinations. Most importantly, the optimal timing of these interventions is unclear.
Only one retrospective study has tried to address the value of the AMB/ITC combination in IA.16 In that study, Popp et al. reviewed outcomes in 21 patients with definite or probable IA at Memorial Sloan-Kettering Cancer Center (New York, NY). They noted that of the 11 patients who received combination therapy with ITC and AMB, 9 (82%) experienced a clinical cure compared with 5 (50%) of the 10 patients who received only AMB, suggesting that simultaneous administration of ITC and AMB may be beneficial.16 However, this small study lacked important information on confounders of outcome such as neutropenia and did not separate primary from salvage therapy.
The current study had several limitations, mainly its retrospective nature and small sample size. Therefore, because the decision to initiate combination antifungal therapy was left to the discretion of the treating physician, one should always consider the possibility of undocumented prognostic variables that were not equally distributed between the two groups. Despite these limitations, however, our experience in a relatively homogeneous patient population (all of the patients had hematologic malignancies, and all had documented IA) indicates that the addition of ITC to LipoAMB as primary therapy does not improve the response rate in aspergillosis. The poor responses to LipoAMB with or without ITC in our patient population is, perhaps, due to the finding that documented IA typically carries a worse prognosis because the fungal burden in documented IA is higher.
- 7Antagonistic interaction between liposomal amphotericin B and ravuconazole in experimental invasive pulmonary aspergillosis: implications for combination therapy [abstract M-238]. In: Proceedings of the 44th Interscience Conference on Antimicrobial Agents and Chemotherapy. Washington, DC: American Society for Microbiology, 2004: 403., , , et al.