Companion Drugs for Amphotericin B in Cryptococcal Meningitis: Flucytosine, Fluconazole, or…Nothing?
Year: 2011
Type: editorial
Abstract: (See the article by Loyse et al, on pages 121–8.) In their latest article, Dr Harrison and colleagues continue to lead the way in efforts to improve care of South African AIDS patients with cryptococcal meningitis, in this case retroviral-naive patients with their first episode of the disease [1]. The article argues convincingly that 2 weeks of 0.7–1.0 mg/kg amphotericin B can provide the same benefit whether flucytosine or fluconazole is used during this period. This question is highly relevant in sub-Saharan Africa, where flucytosine is unavailable or very expensive. In this randomized but unblinded 4-armed study, all patients received amphotericin B 0.7–1.0 mg/kg once daily for 2 weeks, along with a companion drug: flucytosine 25mg/kg four times per day (21 patients), fluconazole 800 milligrams once daily (22 patients), fluconazole 1200 milligrams once daily (23 patients), or voriconazole 300 milligrams twice daily (13 patients). The authors avoided any firm conclusions about replacing flucytosine with voriconazole but concluded that fluconazole at 800 milligrams could replace flucytosine in the treatment of cryptococcal meningitis. Although this reviewer would have preferred to see fewer arms and more patients, this study remains informative. However, a few questions remain. First, if flucytosine contributes little or nothing to an initial 2 weeks of amphotericin B, then fluconazole, being equivalent, may have contributed little or nothing to amphotericin B during that 2 weeks. Although flucytosine is recommended in the Infectious Diseases Society of America guidelines for addition to amphotericin B for 2 weeks in the treatment of cryptococcal meningitis [2], the evidence supporting this practice is underwhelming. The pivotal study in AIDS patients with cryptococcal meningitis compared 202 patients who received flucytosine plus amphotericin B 0.7 mg/kg with 179 who received amphotericin B alone [3]. There was no clinical or culture benefit of flucytosine after 2 weeks or after an additional 8 weeks of itraconazole or fluconazole. Although the incidence of negative cerebrospinal fluid (CSF) cultures at 2 weeks was not significantly higher in flucytosine recipients, the power of this observation was weakened by the 16.7% incidence of missing culture data. A multivariate analysis found a difference in CSF cultures negative for cryptococcal meningitis favoring flucytosine at 2 weeks but not after 10 weeks. The second study to address this issue compared maintenance therapy with fluconazole versus itraconazole [4]. In a retrospective secondary analysis, absence of initial flucytosine appeared as a statistically significant factor that favored relapse during maintenance. Insufficient attention has been given to the fact that the benefit of flucytosine was not found in patients who received amphotericin B followed by fluconazole, which is now the standard regimen, but the benefit was observed only in patients given itraconazole after amphotericin B or in the few given no initial amphotericin B, (ie, flucytosine may have provided benefit only if followed by regimens that are no longer recommended). A small unblinded study in Thailand comparing amphotericin B alone or with flucytosine found a faster rate of CSF-culture-colony count fall with flucytosine but found no difference in mortality at 2 or 10 weeks [5]. Another question is whether the endpoint in the recent trial used a valid surrogate marker. The endpoint was the rate at which the CSF-culture-colony count fell over 2 weeks. The best answer to this question is a careful retrospective analysis of 4 studies in undeveloped countries, spanning 12 different regimens [6]. Compelling evidence was presented that the rate of fall, particularly a very slow fall, correlated with death at 10 weeks. As the authors pointed out, inclusion of 3 regimens with no initial amphotericin B may have influenced results. Initial therapy with fluconazole has been observed to cause a very slow fall in CSF-culture-colony counts and higher mortality than regimens with initial amphotericin B [7]. Because of the extent that the validation study differed from the current study in which initial fluconazole was not used alone, the use of this surrogate marker may be open to question. The final question concerns the authors’ conclusion that 800 milligrams fluconazole should be substituted for flucytosine during the first 2 weeks of therapy. This conclusion is based on the surrogate endpoint and supported by the similar 10-week mortality rate: 6 of 20 patients (30%) treated with flucytosine and 7 of 21 patients (33%) treated with fluconazole (3% difference, 95% confidence interval, −25% to 32%). As the authors point out, their results differ from their small Thai study with a similar design, though with 400 milligrams of fluconazole rather than 800 or 1200 milligrams [5]. In the Thai study, the 2-week mortality rate was greater when amphotericin B was combined with fluconazole (5 of 16 patients, 31%) than when amphotericin was combined with flucytosine (1 of 15 patients, 7%). This discrepancy shows the problem in extrapolating data from one population to another, particularly when patient numbers are small. None of the above questions should take away from the authors’ conclusion that in South Africa the difference in efficacy between flucytosine and fluconazole 800 milligrams or 1200 milligrams was too small to be detected when added daily to the first 2 weeks of amphotericin B 0.7–1 mg/kg. As reported in other studies, toxicity of either fluconazole or renally dosed flucytosine is similarly low over the 2 weeks. This conclusion provides an important option for initial therapy of cryptococcal meningitis in South African retroviral-naive AIDS patients. This work was supported by the Division of Intramural Research, National Institute of Allergy and Infectious Diseases. Author certifies no potential conflicts of interest. The author has submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
Source: Clinical Infectious Diseases
DOI: 10.1093/cid/cir758
Author: John Bennett
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
Funder: National Institutes of Health
OpenAlex ID: W2011742965