Background Aims: Approximately 2-5% of adult-onset coeliac disease patients fails to respond to a gluten free diet and develops refractory celiac disease (RCD). Two types of RCD have been recognized: type I with a phenotypically normal and type II with an aberrant intraepithelial Tcell population. In contrast to RCDI, RCDII seems to be unresponsive to common immunosuppressive treatment. Transition into Enteropathy Associated T-cell Lymphoma (EATL) is seen in 50-60%. This study reports on the effect of cladribine (2-CDA), a purine analogue inducing T-cell depletion, regarding clinical, histolopathologic and immunologic parameters. Design and methods: An analysis was performed in a tertiary referral centre for coeliac disease between 2001 and 2008. Overall, 29 patients diagnosed with RCDII (16 men, 13 women) were treated with 2-CDA (0,1mg/kg/day) intravenously for 5 days, in 1-3 courses every 6 months depending on the response. Symptoms of malabsorption, weight, albumin, haemoglobin, Marsh classification and percentage of aberrant intraepithelial T-cells were evaluated during follow-up. Results: At the time of 2-CDA treatment, the mean age of the 29 patients included was 62.6years (SD±7.6). All patients tolerated 2-CDA without serious side effects. The mean follow-up period was 30.6 months (SD±23.6). Sixteen patients (55%) showed a clinical improvement, nineteen (66%) a histopathological improvement, ten (35%) a significant decrease in aberrant T-cells and fourteen (48%) a complete remission (Marsh 0-1) during follow-up. Four patients had a partial small bowel resection due to UJ before 2-CDA treatment, complete remission was seen in three of them. Four non-responders have had bone marrow transplantation (BMT), among them three eventually showed a complete histological remission and one developed an EATL. Sixteen of 29 (55%) RCDII patients presented as ulcerative jejunitis (UJ). Overall, twelve patients (41%) died. Five out of these 12 patients developed an EATL and died within one year after diagnosing EATL. Seven patients died because of progressive refractory state. The 2-, 3-, 4-year survival after 2-CDA therapy was 75%, 62% and 44%, respectively. Conclusion: Although treatment with 2-CDA of patients with RCDII does not prevent EATL development in all RCDII patients, it seems feasible, well tolerated and induces clinical and histological improvement in 50% of the patients, and complete histological remission in 48%. BMT was only an alternative for nonresponders up to 70 years of age.