A significant proportion of UC patients remain in long-term remission with thiopurines (azathioprine, mercaptopurine). However, therapeutic efficacy of thiopurines has been poorly studied in UC. Therefore, many patients are switched to expensive treatment with anti-TNF agents early on or even undergo colectomy.
There have been performed several studies reporting benefit of thiopurines in UC. The only two placebo-controlled trials that have been conducted in UC suffered from lack of power and poor endpoint definition. Jewell et al compared azathioprine to placebo and failed to demonstrate superiority of azathioprine (BMJ.1974). There was no additional benefit of adding azathioprine to a standard regimen of corticosteroids. Hawthorne et al performed a withdrawal trial in UC patients who were in clinical remission (BMJ.1992). They investigated if maintenance treatment with azathioprine could prevent disease relapse. This small trial was positive since one year relapse rates were significantly lower for patients continuing azathioprine compared to placebo (36% vs 59%). Another trial compared azathioprine, infliximab and combination therapy with these agents in UC (Panaccione et al. Gastroenterol.2014). Week 16 clinical remission rates were 22% and 24% in the infliximab and azathioprine arm resp., and 40% in the infliximab-azathioprine combination group. However, more than 20% of patients already failed on previous therapy with azathioprine prior to inclusion and this trial was prematurely discontinued because of slow patient recruitment. Moreover, no therapeutic drug monitoring (TDM) was applied, despite present evidence supporting this strategy (Haines et al. IBD 2011). Therefore, we can conclude that the efficacy of thiopurines in UC has not been appropriately assessed with a modern adequately powered design using objective endpoints. Indeed, none of the trials have made use of TDM and endoscopic endpoints, which now have become standard practice in UC trials.