Correspondence
Anidulafungin and Fluconazole for Candidiasis
N Engl J Med 2007; 357:1347-1348September 27, 2007
- Article
To the Editor:
The study of anidulafungin as compared with fluconazole for invasive candidiasis by Reboli et al. (June 14 issue)1 shows the challenges involved in the design and reporting of noninferiority trials.2,3 First, the dose of fluconazole may have been inadequate.4,5 Second, no justification is provided for the conspicuously large prestated margin of noninferiority used in this trial, even though justification is recommended.6 This prestated margin of noninferiority implies that a difference in treatment effect of less than 20 percentage points is clinically negligible, a contention that few reasonable clinicians would support, especially given the lack of other advantages of anidulafungin (e.g., with regard to the side-effect profile or cost). The authors themselves appear to believe that a treatment effect of less than 20 percentage points is not negligible, since they are willing to declare anidulafungin superior to fluconazole on the basis of a difference of 15.4 percentage points. But they cannot have it both ways — a difference of less than 20 percentage points is either negligible or it is not.
6 ReferencesScott K. Aberegg, M.D., M.P.H.
James M. O'Brien, Jr., M.D.
Ohio State University, Columbus, OH 43210
scottaberegg@hotmail.com 1
Reboli AC ,Rotstein C ,Pappas RG , et al. Anidulafungin versus fluconazole for invasive candidiasis. N Engl J Med 2007;356:2472-2482
Full Text | Web of Science | Medline2
Le Henanff A ,Giraudeau B ,Baron G ,Ravaud P . Quality of reporting of noninferiority and equivalence randomized trials. JAMA 2006;295:1147-1151
CrossRef | Web of Science | Medline3
Powers JH ,Dixon CA ,Goldberger MJ . Voriconazole versus liposomal amphotericin B in patients with neutropenia and persistent fever. N Engl J Med 2002;346:289-290
Full Text | Web of Science | Medline4
Pappas PG ,Rex JH ,Sobel JD , et al. Guidelines for treatment of candidiasis. Clin Infect Dis 2004;38:161-189
CrossRef | Web of Science | Medline5
Brown AL ,Greig JR ,Kartsonis NA ,Perfect J ,Walsh TJ . Caspofungin versus amphotericin B for invasive candidiasis. N Engl J Med 2003;348:1287-1288
Full Text | Web of Science | Medline6
Piaggio G ,Elbourne DR ,Altman DG ,Pocock SJ ,Evans SJW . Reporting of noninferiority and equivalence randomized trials: an extension of the CONSORT statement. JAMA 2006;295:1152-1160[Erratum, JAMA 2006;296:1842.]
CrossRef | Web of Science | Medline
To the Editor:
The study by Reboli et al. suggests that there is now an additional tool against invasive candidiasis. A concern may arise that the results were biased in favor of anidulafungin, for several reasons: patients receiving anidulafungin received longer intravenous treatment than those receiving fluconazole (median, 14 vs. 11 days) and shorter oral treatment (1 vs. 3 days), and more patients in the fluconazole group either did not receive the study drug or did not have candidemia (11, vs. 5 in the anidulafungin group). Moreover, since microbiologic failure was measured at the end of intravenous therapy, a longer duration of intravenous anidulafungin therapy could have improved its success rates. Since most patients (78%) had central catheters, it would be valuable to mention the study policy for catheter removal, especially since the catheter was not removed in a larger number of patients in the fluconazole group (11, vs. 4 in the anidulafungin group). Thus, for better appreciation of the data, it would be beneficial to present the full results of univariate and multivariate analyses for the influence of such confounders. Finally, for an improved understanding of the results, the statistics that were applied for comparison of continuous variables should be described.
Ran Nir-Paz, M.D.
University of California at Berkeley, Berkeley, CA 94720
ran.nirpaz@gmail. com Allon E. Moses, M.D.
Hadassah–Hebrew University Medical Center, 91120 Jerusalem, IsraelAuthor/Editor Response
The 400-mg dose of fluconazole is the dose approved by the Food and Drug Administration and that specified in the Infectious Diseases Society of America guidelines.1 Higher doses have been suggested for isolates with minimum inhibitory concentrations of 16 to 32 μg per milliliter. The response rate for fluconazole in our trial was similar to that in a recently reported study using 800 mg of fluconazole.2 As we note, only five isolates from both groups had a minimum inhibitory concentration of fluconazole of 16 μg per milliliter or more.
Our study was designed as a noninferiority trial with a prespecified two-step comparison. In the first step, noninferiority was concluded if the lower limit of the two-sided 95% confidence interval was greater than −20 percentage points. In the second step, if the lower limit was greater than 0, then anidulafungin was considered to be superior to fluconazole. The noninferiority margin was chosen on the basis of the severity of disease and the expected response rates for fluconazole. We do agree that there is some debate among statisticians about whether a lower limit of the 95% confidence interval that is greater than 0 or greater than the prespecified margin of noninferiority indicates superiority. Regardless, the claim of superiority was not based on the point estimate (15.4 percentage points); it was based on a value above 0 for the lower limit of the 95% confidence interval.
Since this was a double-blind trial, the duration of exposure to intravenous treatment was not subject to bias. There was no significant difference between the anidulafungin group and the fluconazole group with respect to the duration of intravenous treatment, number of patients switched to oral fluconazole, or duration of oral fluconazole therapy. Median treatment differences can probably be attributed to more frequent treatment failure in the fluconazole group.
Removal of central venous catheters was recommended for patients with candidemia. Although the numbers are small, among patients who did not have their catheter removed, 3 of 4 (75.0%) in the anidulafungin group were successfully treated, as compared with 3 of 11 (27.3%) in the fluconazole group. Analysis of variance was the method used to analyze continuous variables.
2 ReferencesAnnette C. Reboli, M.D.
Cooper University Hospital, Camden, NJ 08103
reboli-annette@cooperhealth.edu Michele Wible, M.S.
Wyeth Research, Collegeville, PA 19426Thomas J. Walsh, M.D.
National Cancer Institute, Bethesda, MD 208921
Pappas PG ,Rex JH ,Sobel JD , et al. Guidelines for treatment of candidiasis. Clin Infect Dis 2004;38:161-189
CrossRef | Web of Science | Medline2
Rex JH ,Pappas PG ,Karchmer AW , et al. A randomized and blinded multicenter trial of high-dose fluconazole plus placebo versus fluconazole plus amphotericin B as therapy for candidemia and its consequences in nonneutropenic subjects. Clin Infect Dis 2003;36:1221-1228
CrossRef | Web of Science | Medline
- Citing Articles (2)
Citing Articles
1
Suganthini Krishnan-Natesan, Elias K. Manavathu, Jessica L. Cutright, Pranatharthi H. Chandrasekar. (2010) Efficacy of anidulafungin, caspofungin and fluconazole in the early phase of infection in a neutropenic murine invasive candidiasis model. International Journal of Antimicrobial Agents 36:1, 33-36
CrossRef2
A.M. Bal. (2010) The echinocandins: three useful choices or three too many?. International Journal of Antimicrobial Agents 35:1, 13-18
CrossRef
