OUP user menu

Journal of Crohn's and Colitis: 10 (12)

Editor-in-Chief

Laurence J. Egan, Ireland

Associate Editors

Shomron Ben-Horin, IsraelSilvio Danese, ItalyPeter Lakatos, HungaryMiles Parkes, UKJesús Rivera-Nieves, USABritta Siegmund, GermanyGijs van den Brink, NLSéverine Vermeire, Belgium

6.585
5.586

Published on behalf of

On the updated ECCO consensus guidelines for medical management of Crohn's disease

Dirk Esser, Freddy Cornillie, Robert H. Diamond, Robert J. Spiegel
DOI: http://dx.doi.org/10.1016/j.crohns.2010.02.002 165-166 First published online: 1 April 2011

Dear Sir,

During the 2009 annual meeting of the European Crohn's and Colitis Organisation (ECCO), the updated consensus guidelines for medical management of Crohn's disease (CD) were presented. ECCO deserves recognition for using a rigorous, evidence-based approach in developing these guidelines. However, whilst we acknowledge our own interests as marketing authorisation holder and distributor of Remicade, we would like to respectfully point out our objections to an assumption of a class effect between anti-TNF agents, as the updated guidelines and in particular section 5I suggest:

All currently available anti-TNF therapies appear to have similar efficacy and adverse-event profiles, so the choice depends on availability, route of delivery, patient preference, cost and national guidance [EL5, RG D].

Firstly, the different biologics targeting TNF-α have different molecular structures and seem to have different clinical effects. For example, the anti-TNF drug etanercept, although currently indicated for use in other inflammatory diseases, was not efficacious in CD clinical trials, and is not indicated for treatment of CD.1 Likewise, development of the monoclonal anti-TNF antibody CDP571 in CD was stopped after a Phase III trial demonstrated only modest, short-term efficacy of unclear clinical relevance.2 More recently, the European Medicines Agency (EMEA) rejected a CD indication for the anti-TNF agent certolizumab pegol, because the therapeutic effect was “not considered clinically relevant”.3 EMEA also recognises differences between the two anti-TNF drugs indicated for use in CD, Remicade and adalimumab. In their respective Summaries of Product Characteristics, no effect of corticosteroids on the pharmacodynamics of Remicade is described, whereas adalimumab needs to be co-administered with corticosteroids during the induction phase unless these are contraindicated.4,5 The route of administration may also have an impact on pharmacokinetics and pharmacodynamics of a drug. We believe that these examples suggest that caution is warranted before concluding a class effect in the absence of appropriate studies.

Secondly, the impact on mucosal healing and long-term outcomes may also be different. In the SONIC trial, 30–44% of patients treated with Remicade achieved complete mucosal healing (complete absence of ulcerations) at week 26,6 confirming previous experiences.79 In contrast, the efficacy for subcutaneous agents to induce complete mucosal healing remains to be proven conclusively in placebo-controlled clinical trials. In the MUSIC trial, an open-label CD trial with certolizumab pegol, only 4% of the patients at week 10 showed complete mucosal healing.10 In the EXTEND trial, which used an open-label induction with adalimumab, the primary endpoint of complete mucosal healing at week 12 was not met in the intention-to-treat analysis.11 Mucosal healing has a demonstrated effect on long-term outcomes such as hospitalisations and surgeries — major cost drivers in CD.8,9,12,13 We submit that for a lifelong disease, it should be not only immediate concerns, but also long-term implications which need to be considered when choosing a biologic, and that complete mucosal healing should be a recognised treatment goal in CD.

In addition to these general remarks, we are also concerned about specific sections of the updated guidelines, although we realise that we have not yet seen them in full detail. Several sections specifically acknowledge that Remicade clinical trials are the only source of evidence in a particular patient population such as that from the SONIC trial (“…[EL1A, RG B for infliximab]”). The rapid inclusion of the SONIC data into the guidelines is remarkable and clearly underscores ECCO's commitment to current scientific evidence. It is therefore surprising to see a general recommendation for use of all anti-TNF agents in the same sections (“…anti-TNF therapy… is an appropriate option…”, “…should be treated with anti-TNF therapy…”). In SONIC, Remicade, in combination with azathioprine or as monotherapy, was superior to azathioprine monotherapy in achieving steroid-free clinical remission and complete mucosal healing in patients failing aminosalicylate and/or corticosteroid therapy.6 Other biologics in this patient population have not been compared to azathioprine monotherapy prospectively in controlled clinical trials. As the examples of etanercept, CDP571 and certolizumab pegol mentioned above have shown, it cannot be assumed that all anti-TNFs work equally well in CD, for whatever reason.

Finally, the updated ECCO guidelines also cite evidence for Remicade and adalimumab in the use of perianal disease: “Infliximab [EL1B, RG A] or adalimumab [EL1B, RG B] should be used as a second line medical treatment [EL1B, RG B].” We would like to emphasise that the only prospective trials with a biologic aimed specifically at fistulising CD patients were conducted with Remicade.14,15 Evidence for adalimumab is limited to subanalyses from clinical trials that were neither powered nor prospectively designed to show efficacy in this particular patient population.

In conclusion, we applaud ECCO for issuing timely, evidence-based guidelines for inflammatory bowel diseases on a European level, but feel that the general assumption of an anti-TNF class effect does not reflect the existing clinical evidence. Although we recognise that existing data support adalimumab's efficacy in reducing signs and symptoms of CD, its effect in immunosuppressive-naïve patients and in fistulising disease remains to be determined, as does its ability to induce complete mucosal healing, and we believe the guidelines should reflect this. We further suggest that you consider including long-term aspects and the ability of a drug to induce complete mucosal healing as important considerations when making a treatment decision.

References

  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
View Abstract