|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 23, No 36 (December 20), 2005: pp. 9428-9429 © 2005 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.04.1608
Misleading Statements in Industry-Sponsored Meta-Analysis of ItraconazoleNordic Cochrane Centre, Copenhagen, Denmark To the Editor: Glasmacher et al justify their meta-analysis of itraconazole by saying that the "efficacy of antifungal prophylaxis has not been convincingly proven."1 They discard our meta-analysis of antifungal agents versus placebo or no treatment2,3 by saying that it is "seriously flawed in assuming an equivalent range of antifungal activity and bio-availability across very different drugs and by different modes of delivery (eg, pooling oral polyenes with azoles)" and that it was also incomplete "as only three of thirteen relevant randomized controlled trials of itraconazole were included."1 All of these statements are wrong. We showed in 19972 that amphotericin B and fluconazole reduce the incidence of invasive fungal infections. We also found that amphotericin B decreased all-cause mortality significantly, but interpreted this finding cautiously as the trials were small. We distinguished between the drugs of course. In addition to the overall estimates for the effect of the drugs combined, we also showed and discussed the results for each individual drug. Also, our analysis is not incomplete as we compared drugs versus placebo or no treatment, and included all the trials that were available to us when we wrote our review (Glasmacher et al1 also included active comparisons). Glasmacher et al’s unreserved conclusion that immunosuppressed cancer patients "should receive an antifungal prophylaxis with itraconazole"1 is not supported by their data, but is undoubtedly useful for the marketing department of Janssen, the manufacturer of itraconazole, which funded their review. First, they claim that there was an effect on mortality from invasive fungal infections. However, this has also been shown for other drugs; further, this outcome can be biased.2 All-cause mortality is without bias and it is therefore interesting that they failed to find an effect on this outcome (207 v 206 deaths). Second, they fail to mention that, like itraconazole, amphotericin B and voriconazole are also effective against Aspergillus.4 Third, they did not include trials where amphotericin B was given intravenously, although such trials exist, and although this is the appropriate mode of delivery for this drug when given to patients who are at risk of dying from a fungal infection.5 Fourth, although they received an unpublished report from the manufacturer, there seems to exist at least one other unpublished report, of a large trial.6 Fifth, azoles have important adverse effects. Some are suspected of increasing deaths because of an increase in bacterial infections,2,3 and itraconazole may cause congestive heart failure.7 It is therefore prudent to base one’s conclusions on all-cause mortality and not on mortality from invasive fungal infections as Glasmacher et al1 do. We have previously shown serious problems with both the design and the analysis of trials comparing fluconazole with amphotericin B sponsored by Pfizer.5 We wonder whether this is the reason we seem to be so unpopular with authors like Glasmacher et al who benefit from industry sponsorship of their reviews? Authors’ Disclosures of Potential Conflicts of Interest The authors indicated no potential conflicts of interest. REFERENCES
1. Glasmacher A, Prentice A, Gorschluter M, et al: Itraconazole prevents invasive fungal infections in neutropenic patients treated for hematologic malignancies: Evidence from a meta-analysis of 3,597 patients. J Clin Oncol 21:4615-4626, 2003 2. Gøtzsche PC, Johansen HK: Meta-analysis of prophylactic or empirical antifungal treatment versus placebo or no treatment in patients with cancer complicated by neutropenia. BMJ 314:1238-1244, 1997 3. Gøtzsche PC, Johansen HK: Routine versus selective antifungal administration for control of fungal infections in patients with cancer. Cochrane Database Syst Rev 2:CD000026, 2002 4. Herbrecht R, Denning DW, Patterson TF, et al: Voriconazole versus amphotericin B for primary therapy of invasive aspergillosis. N Engl J Med 347:408-415, 2002 5. Johansen HK, Gøtzsche PC: Problems in the design and reporting of trials of antifungal agents encountered during meta-analysis. JAMA 282:1752-1759, 1999 6. Prentice AG, Bradford GR: Prophylaxis of fungal infections with itraconazole during remission-induction therapy. Mycoses 32:96-102, 1989 (suppl 1) 7. Ahmad SR, Singer SJ, Leissa BG: Congestive heart failure associated with itraconazole. Lancet 357:1766-1767, 2001[CrossRef][Medline]
Related Reply
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
