Editorial Response: Estimating the True Cost of Amphotericin B
- 1Division of Infectious Diseases, Department of Internal Medicine, Center for the Study of Emerging and Re-emerging Pathogens, University of Texas Medical School, Houston, Texas
- 2Immunocompromised Host Section, Pediatric Oncology Branch, National Cancer Institute, Bethesda, Maryland
- Reprints or correspondence: Dr. John H. Rex, 6431 Fannin, 1728 JFB, Houston, TX 77030 (jrex{at}heart.med.uth.tmc.edu).
Like all other cost centers in the hospital, pharmacies remain under pressure to closely watch and control their direct expenditures. However, drugs have many associated costs. Of primary importance to the pharmacy budget is the cost of simply buying the drug. Then there is the cost of preparing and administering the drug. There are also the costs of monitoring for and treating side effects of the drug. Finally, there are the costs due to any lack of or delay in therapeutic response. Further complicating matters, not all of these costs are monetary or immediately obvious. Patients may pay a very high physiological price indeed if a critical organ is damaged by a drug's toxicity or if a slow response puts them at risk for developing additional complications.
Thus it is increasingly recognized that a pharmacy's focus on acquisition and administration costs alone may be shortsighted and could both increase a hospital's overall monetary cost and produce worse outcomes for the patients. Such global cost analyses have now been provided for many drugs (e.g., the recent analyses of low-molecular-weight heparins [1–3]), and taking steps toward such an analysis is especially important to our understanding of the proper use of amphotericin B.
Amphotericin B, available for 40 years now in its traditional deoxycholate preparation (dAmB; Fungizone' Bristol-Meyers Squibb, Princeton, NJ), remains unchallenged as the antifungal agent with the broadest spectrum. However, its formidable administration-related and renal toxic effects have motivated the development of the 3 now-licensed lipid-associated formulations of amphotericin B (LFABs): amphotericin B lipid complex (Abelcet; Liposome, Princeton, NJ), amphotericin B colloidal dispersion (Amphotec; The Liposome Company, Princeton, NJ; Amphocil; Sequus Pharmaceuticals, Menlo Park, CA), and liposomal amphotericin B (AmBisome; Vestar, San Dimas, CA) [4, 5].
Although differing in composition and pharmacokinetics, the LFABs share 4 key characteristics. First, they reduce (but do not eliminate!) the toxic effects associated with dAmB. Second, they have been shown to be active in animal models of fungal infections. Although they sometimes appear less potent on a weight basis than dAmB [6], their improved safety profile permits administration of higher doses that more than overcome any such differences [7, 8]. Third, the results from the animal studies have now been supported by randomized trials with humans, which have found these agents at suitable doses to be consistently less toxic than but at least as efficacious as amphotericin B [9–13]. Although equivalent clinical data are not available for all drugs versus all fungi, there is little reason to doubt that these agents are generally efficacious against the same range of fungi and fungal infections as dAmB.
Were it not for the fourth common characteristic of these agents, their cost, these agents would thus be widely adopted and not further debated. Unfortunately, the LFABs are significantly more expensive than dAmB. Whereas the 50-mg vial of dAmB required to administer a typical 0.7-mg/kg dose of dAmB to a 70-kg adult costs ∼$20 (US currency), the licensed doses of the LFABs are 3–6 mg/kg, and the acquisition cost of a single dose for the same 70-kg adult is ∼10–50 times the cost of dAmB.
In addition to the cost factor, the studies that led to licensing of these agents focused primarily on their use for salvage therapy. With the exception of the licensing of liposomal amphotericin B as first-line therapy for febrile neutropenia in patients at high risk of invasive fungal infection, the current US Food and Drug Administration (FDA) guidelines for these agents specify that they are to be used for patients who are intolerant of dAmB or whose conditions are refractory to it. Taken with the cost data, the licensed indications have led many hospitals to adopt internal usage guidelines that mirror the FDA's licensing guidelines.
However, the reduced administration-related and renal toxic effects of the LFABs have attracted much interest in earlier use of these agents, and physicians are asking (and being asked) increasingly difficult questions about how much toxicity is required before it is considered “intolerable.” Moreover, there is increasing demand for use of LFAB as initial therapy of invasive fungal infections in patients who are known to have a high risk of developing nephrotoxicity. Such patients include those who are receiving or have recently received concomitant nephrotoxic agents or those with preexisting intrinsic renal impairment.
This line of reasoning also inevitably leads to questions about the price we assign to these toxic effects. As the acute administration-related toxicity can usually be controlled and also tends to abate with time, it is the cost of the renal toxicity that is of primary interest. Medically induced renal failure has been associated with increased mortality and morbidity [14, 15], but what, specifically, is the cost of dAmB-induced renal dysfunction, and how much of such toxicity should we tolerate before we switch to an LFAB?
In an effort to answer at least some portion of this question, Wingard et al. report in this issue of Clinical Infectious Diseases the results of a review of 239 cancer patients who received dAmB as therapy for suspected or proven invasive aspergillosis [16]. Nephrotoxicity was significant in this group, with 29% experiencing a rise in creatinine level to >2.5 mg/dL and 15% requiring dialysis. These are indeed outcomes with a high cost, at least as far as the physiology of the patient is concerned.
Not all groups were at equal risk for these negative results. When controlled for use of other nephrotoxic agents, baseline creatinine level and duration of amphotericin B therapy predicted the rise in creatinine level to >2.5 mg/dL. Experiencing such a rise and being a bone marrow-transplant recipient increased the likelihood of dialysis, and dialysis increased the likelihood of death. The authors conclude that preexisting renal dysfunction and/or even small increases in serum creatinine level should be taken very seriously, especially in bone marrow-transplantation patients, and that early conversion to an LFAB might be beneficial.
These results are instructive but must be placed carefully into context. First, the authors provide little information (beyond underlying disease) on the severity of illness of the patients under study. Analyses that control for known concomitant nephrotoxic agents are provided, but other authors have not noted such frequent nephrotoxicity, and the possibility of a systematic bias remains.
For example, Luber et al. analyzed 178 patients given dAmB for an average of 17 days and found that only 1 of these patients developed renal insufficiency requiring hemodialysis [17]. However, 87% of these patients lacked a documented fungal infection, and the mean daily dose of dAmB was ∼0.5 mg/kg—presumably much lower than the 0.8- to 1.2-mg/kg dose of dAmB that was likely used for the patients described by Wingard et al. [16]. Use of other nephrotoxic agents and prior renal injury may have also differed between the groups. Second, the dAmB dosing data are limited to duration information; no dose-intensity data are provided. Third, more data on the rate of change of the serum creatinine level might have been of value. Arbitrary levels of serum creatinine may not be as meaningful as the rate of change of these values.
These problems aside, the authors are to be commended for their efforts in this area. The demonstration of potential differences in risk of nephrotoxicity in relation to patient group is intriguing and will doubtless fuel needed work in this area. For example, the authors did not attempt to go beyond the implicit physiological cost of renal dysfunction to a monetary analysis. Others, however, have begun to address the problem from this standpoint.
For example, a pharmacoeconomic analysis of the recently published [9] National Institute of Allergy and Infectious Diseases Mycoses Study Group trial, which compared a 3-mg/kg dose of liposomal amphotericin B with a 0.6-mg/kg dose of dAmB in empirical antifungal therapy for persistently febrile neutropenic patients, found that nephrotoxicity (defined as greater than twice the baseline serum creatinine value) increased hospital costs, most notably those related to duration of hospitalization and use of the pharmacy and blood-bank [18]. When modeled in a sensitivity analysis, the higher acquisition cost of liposomal amphotericin B counterbalanced the added cost of managing nephrotoxicity once the cost of liposomal amphotericin B was no higher than $85 per 50-mg vial, or $340 for a 3-mg/kg dose for a 70-kg adult.
From a personal standpoint, these data suggest to us that LFAB guidelines should permit their use by those patients whose therapy with dAmB has failed, who are intolerant of dAmB, or who are highly likely to be intolerant of dAmB. Although this last clause is subjective and not part of the current labeling for these compounds, both personal experience and the data of Wingard et al. suggest that pushing forward with dAmB until the creatinine level rises significantly (for example, in a diabetic patient with known nephropathy and a baseline creatinine clearance of 25 mL/h) is unlikely to be cost-effective.
We currently think that LFABs could be considered for first-line therapy for life-threatening invasive fungal infections in patients who have preexisting renal dysfunction (calculated creatinine clearance <50 mL/min) or a high probability of developing severe renal impairment. Patients at high risk of severe renal impairment would include those who are receiving (concomitantly) ≥1 highly nephrotoxic agents (e.g., cyclosporine, tacrolimus, foscarnet, aminoglycoside, cis-platinum, or ifos-famide), as well as those who have underlying primary or intrinsic renal disease (e.g., renal dysfunction due to diabetes mellitus).
Additional work is clearly needed to refine the concepts “intolerant of” and “highly likely to be intolerant of.” Such work should include surveys of a broader range of dAmB-treated patients so that we could better correlate toxicity with the multiple preexisting physiological factors that contribute to severity of illness. The data presented by Wingard et al. are a good step in this direction.
- Received June 29, 1999.
- Revision received August 9, 1999.
- © 1999 by the Infectious Diseases Society of America
References
This Article
- ExtractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
