The trusted source for
healthcare information and
abstract & commentary
Source: Walsh TJ, et al. Liposomal amphotericin B for empirical therapy in patients with persistent fever and neutropenia. N Engl J Med 1999; 340:764-771.
The niaid mycoses study group compared the safety and efficacy of conventional amphotericin B desoxycholate (AB) to AmBisome, a unilamellar liposomal amphotericin B (L-AB) product in the empirical therapy of persistent fever and chemotherapy induced neutropenia. Patients studied were ages 2-80 and were being treated for malignancy and who had an ANC less than 500/mm3 and continuing fever despite having received empirically administered antibacterial therapy for at least five days. Three hundred forty-four patients received AB (0.6 mg/kg/d) and 343 received L-AB (3.0 mg/kg/d) for mean durations of, respectively, 10.8 and 10.3 days. The dose could be increased according to a set of guidelines or decreased in response to toxicity. Among the reasons for exclusion from study entry were a serum creatinine concentration greater than twice the upper limit of normal. The patient groups were similar at baseline, with approximately one-third in each being categorized as high risk and approximately 45% in each undergoing bone marrow transplantation. Study drug was continued until recovery from neutropenia.
In a modified intent-to-treat analysis (all patients who received at least 1 dose of study drug were included), there was no significant difference in outcome by any of the following individual end points examined: resolution of fever, survival for seven days after study initiation, lack of study drug discontinuation due to toxicity or lack of efficacy, and cure of baseline fungal infection. The last mentioned of these outcomes was the result of 11 patients in each group having positive blood cultures for Candida at baseline, prior to receipt of the first dose of study drug. There was also no difference between groups when a composite score was used. There were, however, significantly fewer proven breakthrough fungal infections among the patients given L-AB (3.2%) than in those given AB (97.8%; P = 0.009). There were three breakthrough candidemic episodes in the former group and 12 in the latter (P = 0.03). A blinded assessment found that fungal infection contributed to or was the primary cause of death in four L-AB recipients and 11 AB recipients (P = 0.11).
Infusion-related adverse effects, including fever, chills, hypertension, tachycardia, hypotension, and hypoxia, were significantly more common in the AB recipients. Decreased arterial oxygen saturation was seen in 22 AB recipients and only one L-AB recipient (P < 0.001). Premedication was more commonly felt to be required in the AB recipients. An increase in serum creatinine concentration of either two-fold or greater or three-fold or greater was significantly more common in AB than L-AB recipients (P < 0.001); a peak serum creatinine of more than 3.0 mg/dL was observed, respectively, in 26% and 12% (P < 0.001).
L-AB is one of three lipid-associated amphotericin B preparations currently available in the United States and is the only one of the three that contains true liposomes.1,2 (See Table.) Its phospholipid bilayer, stabilized with cholesterol, provides an intravesicular hydrophilic and external hydrophobic environment. The amphipathic nature of amphotericin B makes it ideal for entrapment in these structures. Transfer of amphotericin B from liposomes to fungal cells may be assisted by the production and secretion of phospholipases by some pathogens; such enzymes are also produced by a variety of mammalian cell types, including phagocytic cells.
|Lipid-associated Amphotericin B Products________________________________|
|Abelcet (ABLC; amphotericin B lipid complex)||1600-11,000 nm ribbonlike bilayers of 7:3 molar ratio of dimyristoyl phosphatidylcholine & dimyristoyl phosphatidylglycerol|
|Amphocil (ABCD; amphotericin B colloidal dispersion)||120-140 nm disks of cholesteryl sulfate|
|AmBisome (liposomal amphotericin B)||80 nm unilamellar bilayered vesicles of hydrogenated soy phosphatidylcholine and distear- oylphosphatidylglycerol with cholesterol|
Clearance of liposomal vesicles from the bloodstream varies with a number of factors. Smaller vesicles are cleared more slowly than large ones and positively or neutrally charged ones (the charge depending upon the phospholipids used) are also cleared more slowly than positively charged ones. Stabilization of the liposome with cholesterol also decreases its rate of clearance. Liposomes are predominantly removed from the circulation by cells of the reticuloendothelial system; L-AB (and ABLC) concentrate primarily in the spleen and liver, with limited distribution to the kidneys, heart, and lungs. The correlation between this pattern of drug distribution and both organ-specific toxicity and treatment efficacy have been speculated upon, but remain unproven. The observation that liposomal encapsulation of amphotericin reduces production of proinflammatory cytokines by phagocytic cells may account for a reduction in infusion-related toxicity. Despite its clearance by cells of the reticuloendothelial system, administration of L-AB does not interfere with clearance of bacteria from the bloodstream in a rodent model.3
Most in vitro and animal studies have demonstrated that, compared to AB, lipid-associated amphotericin B products cause less nephrotoxicity but at the cost of reduced (on a weight basis) antifungal activity. The critical question, thus, becomes whether the lipid products have an improved therapeutic toxic ratio in human infection. We are getting closer to an affirmative answer to that question as a result of studies such as the one reviewed above.
In another recent report, 338 adults and children with neutropenia (ANC < 500/mm3) and fever persisting despite empirical antibacterial therapy for 96 hours were randomized to receive AB at 1 mg/kg/d, L-AB at 1 mg/kg/d, or L-AB at 3 mg/kg/d.4 The rates of success (no development of fungal infection and no addition of other systemic antifungal therapy) were, respectively, 49%, 58%, and 64%. High-dose L-AB was superior to AB (P = 0.03). There was, however, no difference in time to defervescence. Toxicity was significantly less frequent in the L-AB recipients. Separately, a randomized comparison of amphotericin B colloidal dispersion and AB in a similar group of patients found no difference in efficacy but did find evidence of reduced toxicity with the lipid preparation.5
In a smaller study, 66 patients who could be assessed with regard to efficacy of treatment of proven (9 with fungemia, 17 invasive mold infections) or suspected (40 with possible pulmonary aspergillosis) infection were randomized to receive either AB (1 mg/kg/d) or L-AB (5 mg/kg/d).7 The complete response rate was 44% in the L-AB recipients and 18% (P = 0.03) in the AB recipients. The proportions of partial responders in the two groups were, respectively, 22% and 38%; the comparable failure rates were 34% and 44%. When the entire range of responses was considered, the difference did not reach statistical significance (P = 0.09). Mortality rates, adjusted for malignancy status, were lower in patients given L-AB (P = 0.03). AB administration was more frequently associated with a rise in serum creatinine concentration (P < 0.001).
One important randomized comparison of two doses of L-AB in the treatment of invasive aspergillosis in 87 assessable patients found no difference in outcomes between those who received 1 mg/kg/d and those who received 4 mg/kg/d.6 There were nine deaths due to aspergillosis infection in each arm.
L-AB (4 mg/kg/d for 3 weeks) has also been compared to AB (0.7 mg/kg/d for 3 weeks) in the initial treatment of cryptococcal meningitis in a small number of AIDS patients.8 Patients subsequently received fluconazole in each arm. L-AB was significantly less nephrotoxic and was associated with earlier achievement of a negative cerebrospinal fluid culture. There was, however, no apparent difference in clinical efficacy.
As we accumulate evidence to validate the proposed superiority of lipid-associated amphotericin B products such as L-AB over conventional AB, we face another very difficult question—is the benefit sufficient to justify the cost of these expensive products? This question cannot be addressed until the issue of the optimal dose of drug is settled. Until that information is forthcoming and a convincing pharmacoeconomic justification is available, most institutions will likely restrict their use to particular circumstances in which conventional AB therapy is failing or cannot be administered because of toxicity.
1. Hiemenz JW, Walsh TJ. Clin Infect Dis 1996;22 (Suppl 2):S133-144.
2. Wong-Beringer A, et al. Clin Infect Dis 1998;27: 603-618.
3. van Etten EW, et al. Antimicrob Agents Chemother 1998;42:1677-1681.
4. Prentice HG, et al. Br J Haematol 1997;98:711-718.
5. White MH, et al. Clin Infect Dis 1998;27:296-302.
6. Ellis M, et al. Clin Infect Dis 1998;27:1406-1412.
7. Leenders AC, et al. Br J Haematol 1998;103:205-212.
8. Leenders AC, et al. AIDS 1997;11:1463-1471.
a. L-AB was associated with a greater frequency of decreased arterial oxygen saturation during infusion.
b. Proven breakthrough fungemia occurred more frequently in the L-AB recipients.
c. An increase in serum creatinine occurred more frequently in the AB recipients.
d. Fever resolved more rapidly in the L-AB recipients.
a. Liposomes are predominantly removed from the bloodstream by cells of the reticuloendothelial system.
b. Amphocil is the only true liposomal amphotericin B available in the United States.
c. Liposomal encapsulation of amphotericin B enhances production of proinflammatory cytokines by phagocytic cells in vitro.
d. Administration of liposomal amphotericin B interferes with the clearance of bacteria from the bloodstream in a rodent model.