Reply

Abstract

Thank you for the opportunity to respond to these letters regarding our trial of antimycobacterial therapy in Crohn’s disease.1Chamberlin W. Importance of the Australian Crohn’s disease antibiotic study.Gastroenterology. 2007; 133: 1744-1745Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar, 2Kuenstner J.T. The Australian antibiotic trial in Crohn’s disease: alternative conclusions from the same study.Gastroenterology. 2007; 133: 1742-1743Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar, 3Gitlin L. Biesecker J. Australian Crohn’s antibiotic study opens new horizons.Gastroenterology. 2007; 133: 1743-1744Abstract Full Text Full Text PDF PubMed Scopus (16) Google Scholar, 4Lipton J.E. Barash D.P. Flawed Australian CD study does not end MAP controversy.Gastroenterology. 2007; 133: 1742Abstract Full Text Full Text PDF PubMed Scopus (16) Google Scholar Most of the comments and criticisms raised by your correspondents were addressed in our original article.5Selby W. Pavli P. Crotty B. et al.Two-year combination antibiotic therapy with clarithromycin, rifabutin, and clofazimine for Crohn’s disease.Gastroenterology. 2007; 132: 2313-2319Abstract Full Text Full Text PDF PubMed Scopus (281) Google ScholarThe higher doses proposed by Drs Gitlin, Biesecker, Lipton, and Barash are derived from regimens used for treatment of Mycobacterium avium complex infections, not specifically for Mycobacterium avium paratuberculosis (MAP). We chose the same doses of clarithromycin and rifabutin that were reported to be effective in Crohn’s disease in the open-label study performed by Gui et al.6Gui G.P. Thomas P.R. Tizard M.L. et al.Two-year-outcomes analysis of Crohn’s disease treated with rifabutin and macrolide antibiotics.J Antimicrob Chemother. 1997; 39: 393-400Crossref PubMed Scopus (145) Google Scholar Our placebo-controlled, double-blind study added clofazimine to this regimen, but did not replicate their results. To our knowledge, there are no published reports of the use of higher doses of these antibiotics in Crohn’s disease.We were also concerned about the nondissolution of clofazimine in vitro (which may or may not have occurred in vivo). We performed a supplementary analysis, excluding patients who were exposed to the defective capsules. This demonstrated a higher relapse rate in the active treatment group, which is inconsistent with a loss of efficacy owing to failure to dissolve.It is argued that the results of the trial are invalid because we did not identify MAP by culture or polymerase chain reaction testing. This is discussed in our paper and does not influence the findings that the antibiotic regime was not effective in the long term.Dr Chamberlin states that “the antibiotic limb did significantly better than the comparison limb as long as antibiotics were being administered.” This is incorrect—the only statistically significant difference was at 16 weeks. It is not clear why there was a difference at 16 weeks, but it seems unlikely that it was due to elimination of MAP because there was no statistically significant difference at 12, 24, or 36 months despite continuing the antibiotics for 24 months. We have speculated that it may be due to a nonspecific antibacterial effect—the data do not permit any firm conclusions.Dr Chamberlin cites recently published data suggesting that mesalamine and 6-mercaptopurine have an antibiotic effect against MAP in vitro.7Greenstein R.J. Liya S. Haroutunian V. et al.On the action of methotrexate and 6-mercaptopurine on M. avium subspecies paratuberculosis.PLos One. 2007; 2: e161Crossref PubMed Scopus (64) Google Scholar The extrapolation of in vitro data to the clinical arena is speculative and is not consistent with the well-established activity of other immunomodulators, such as corticosteroids and infliximab.Dr Kuenstner suggests that the study required an unrealistically high response rate for a positive outcome. We disagree; the power calculation was based on demonstration of a difference of 40% in relapse rates between the active treatment and placebo arms. Dr Kuenstner also suggests that reexposure may have occurred. This should not have affected patients who continued to take MAP therapy. Dr Kuenstner comments that the study may not have had adequate power to detect a difference if only a proportion of patients were infected with MAP. This is true for any trial of any therapy for any infection. However, the results of the study do not support a significant pathogenic role for MAP in the majority of patients with Crohn’s disease. Dr Kuenstner’s advocacy of public health measures and development of better MAP therapies for Crohn’s disease is not supported by current evidence.Drs Gitlin and Biesecker are critical of endpoints chosen for the study (proportion of patients experiencing at least one relapse at 12, 24, and 36 months), suggest that the proportion of patients achieving remission would be more appropriate, and compare the 66% remission rate at 16 weeks with a 39% response rate with infliximab. We believe the choice of endpoints is appropriate (the aim of the trial was to assess long-term anti-MAP therapy) and that it is misleading to perform such cross-trial comparisons. Their criticisms are contradictory; if the response to MAP therapy “ranks among the highest remissions generally achieved,” the dosage must have been adequate and the patients continuing in the active arm should have stayed in remission.We conducted a randomized, double-blind, controlled trial comparing steroids plus 3 anti-Mycobacterial antibiotics with steroids alone. To our knowledge, this is the only randomized controlled trial to assess the role of therapy directed against MAP. There were no statistically significant differences in outcome for any of our predetermined primary endpoints. We see no reason to alter our conclusion that the results of this trial do not support a pathogenic role for MAP in the majority of patients with Crohn’s disease. The only way this conclusion could be challenged is by conducting further randomized controlled clinical trials. Thank you for the opportunity to respond to these letters regarding our trial of antimycobacterial therapy in Crohn’s disease.1Chamberlin W. Importance of the Australian Crohn’s disease antibiotic study.Gastroenterology. 2007; 133: 1744-1745Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar, 2Kuenstner J.T. The Australian antibiotic trial in Crohn’s disease: alternative conclusions from the same study.Gastroenterology. 2007; 133: 1742-1743Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar, 3Gitlin L. Biesecker J. Australian Crohn’s antibiotic study opens new horizons.Gastroenterology. 2007; 133: 1743-1744Abstract Full Text Full Text PDF PubMed Scopus (16) Google Scholar, 4Lipton J.E. Barash D.P. Flawed Australian CD study does not end MAP controversy.Gastroenterology. 2007; 133: 1742Abstract Full Text Full Text PDF PubMed Scopus (16) Google Scholar Most of the comments and criticisms raised by your correspondents were addressed in our original article.5Selby W. Pavli P. Crotty B. et al.Two-year combination antibiotic therapy with clarithromycin, rifabutin, and clofazimine for Crohn’s disease.Gastroenterology. 2007; 132: 2313-2319Abstract Full Text Full Text PDF PubMed Scopus (281) Google Scholar The higher doses proposed by Drs Gitlin, Biesecker, Lipton, and Barash are derived from regimens used for treatment of Mycobacterium avium complex infections, not specifically for Mycobacterium avium paratuberculosis (MAP). We chose the same doses of clarithromycin and rifabutin that were reported to be effective in Crohn’s disease in the open-label study performed by Gui et al.6Gui G.P. Thomas P.R. Tizard M.L. et al.Two-year-outcomes analysis of Crohn’s disease treated with rifabutin and macrolide antibiotics.J Antimicrob Chemother. 1997; 39: 393-400Crossref PubMed Scopus (145) Google Scholar Our placebo-controlled, double-blind study added clofazimine to this regimen, but did not replicate their results. To our knowledge, there are no published reports of the use of higher doses of these antibiotics in Crohn’s disease. We were also concerned about the nondissolution of clofazimine in vitro (which may or may not have occurred in vivo). We performed a supplementary analysis, excluding patients who were exposed to the defective capsules. This demonstrated a higher relapse rate in the active treatment group, which is inconsistent with a loss of efficacy owing to failure to dissolve. It is argued that the results of the trial are invalid because we did not identify MAP by culture or polymerase chain reaction testing. This is discussed in our paper and does not influence the findings that the antibiotic regime was not effective in the long term. Dr Chamberlin states that “the antibiotic limb did significantly better than the comparison limb as long as antibiotics were being administered.” This is incorrect—the only statistically significant difference was at 16 weeks. It is not clear why there was a difference at 16 weeks, but it seems unlikely that it was due to elimination of MAP because there was no statistically significant difference at 12, 24, or 36 months despite continuing the antibiotics for 24 months. We have speculated that it may be due to a nonspecific antibacterial effect—the data do not permit any firm conclusions. Dr Chamberlin cites recently published data suggesting that mesalamine and 6-mercaptopurine have an antibiotic effect against MAP in vitro.7Greenstein R.J. Liya S. Haroutunian V. et al.On the action of methotrexate and 6-mercaptopurine on M. avium subspecies paratuberculosis.PLos One. 2007; 2: e161Crossref PubMed Scopus (64) Google Scholar The extrapolation of in vitro data to the clinical arena is speculative and is not consistent with the well-established activity of other immunomodulators, such as corticosteroids and infliximab. Dr Kuenstner suggests that the study required an unrealistically high response rate for a positive outcome. We disagree; the power calculation was based on demonstration of a difference of 40% in relapse rates between the active treatment and placebo arms. Dr Kuenstner also suggests that reexposure may have occurred. This should not have affected patients who continued to take MAP therapy. Dr Kuenstner comments that the study may not have had adequate power to detect a difference if only a proportion of patients were infected with MAP. This is true for any trial of any therapy for any infection. However, the results of the study do not support a significant pathogenic role for MAP in the majority of patients with Crohn’s disease. Dr Kuenstner’s advocacy of public health measures and development of better MAP therapies for Crohn’s disease is not supported by current evidence. Drs Gitlin and Biesecker are critical of endpoints chosen for the study (proportion of patients experiencing at least one relapse at 12, 24, and 36 months), suggest that the proportion of patients achieving remission would be more appropriate, and compare the 66% remission rate at 16 weeks with a 39% response rate with infliximab. We believe the choice of endpoints is appropriate (the aim of the trial was to assess long-term anti-MAP therapy) and that it is misleading to perform such cross-trial comparisons. Their criticisms are contradictory; if the response to MAP therapy “ranks among the highest remissions generally achieved,” the dosage must have been adequate and the patients continuing in the active arm should have stayed in remission. We conducted a randomized, double-blind, controlled trial comparing steroids plus 3 anti-Mycobacterial antibiotics with steroids alone. To our knowledge, this is the only randomized controlled trial to assess the role of therapy directed against MAP. There were no statistically significant differences in outcome for any of our predetermined primary endpoints. We see no reason to alter our conclusion that the results of this trial do not support a pathogenic role for MAP in the majority of patients with Crohn’s disease. The only way this conclusion could be challenged is by conducting further randomized controlled clinical trials. The Australian Antibiotic Trial in Crohn’s Disease: Alternative Conclusions From the Same StudyGastroenterologyVol. 133Issue 5PreviewI wish to comment on the recent article by Warwick Selby and others entitled, “Two-year combination antibiotic therapy with clarithromycin, rifabutin, and clofazamine for Crohn’s disease.” In the conclusion of the abstract, the authors claim that the findings of their study do not support a significant role for Mycobacterium avium paratuberculosis (MAP). Had the authors identified which patients were infected with MAP, provided evidence of eradication of MAP infection, and shown that patients were still symptomatic, they could more convincingly claim that their study does not support a significant role for MAP in Crohn’s disease. Full-Text PDF Importance of the Australian Crohn’s Disease Antibiotic StudyGastroenterologyVol. 133Issue 5PreviewThe article by Warwick Selby, “Two-year combination antibiotic therapy with clarithromycin, rifabutin, and clofazimine for Crohn’s disease”1 and the accompanying commentary2 are well written and worth reading carefully. They contain important data and commentary. Unfortunately, despite data supporting the positive impact of properly chosen antibiotics in treating Crohn’s disease (CD) patients, the authors emphasize the negative. The authors report the data accurately, but their interpretations need to be challenged. Full-Text PDF Australian Crohn’s Antibiotic Study Opens New HorizonsGastroenterologyVol. 133Issue 5PreviewThe recently reported trial by Selby et al, “Two year combination antibiotic therapy with Clarithromycin, Rifabutin and Clofazimine for Crohn’s Disease”1 and the accompanying editorial refer to the controversial topic of Crohn’s disease causation by an infectious agent. In this letter, we address 3 important aspects that may have given careful readers concern. Full-Text PDF