Abstract: Introduction: We reported last year, during the ACR annual meeting, the result of different treatment modalities for ocular lesions of BD in 856 patients. We showed that pulse cyclophosphamide (PCP), low dose pulse cyclophosphamide (LDP), oral cyclophosphamide (OCP), weakly methotrexate (MTX), chloambucil (CHL), and cyclosparine A (CYA) had all the same efficacy. The percentage of the eyes aggravated (visual acuity) despite the treatment was: For PCP: 21%, LDP: 22%, OCP: 17%, MTX: 20%, CHL: 19%, CYA: 23%. The difference was not statistically significant. We proposed for these non responder patients, to switch their treatment protocol to another one. The aim of this study was to look at these patients, after they were switched to another treatment, to find their outcome. The main question was will they respond to another treatment, and if yes to what extend. Materials & Methods: The Total Inflammatory Activity Index (TIDI) was chosen to evaluate the result. TIDI was calculated for each patient on the inflammatory state of the anterior chamber, the uvea, and the retina. An improvement of more than 20% of the TIAI was accepted as a Good result, an aggravation of more than 20% as a Bad result, and between the two the result was considered to be Null. A confidence interval at 95% was calculated for each percentage. Results: From 856 patients treated with one of the above methods, 125 were non responder and were switched to a second method (14.6%). The result was: Good 31,2% ± 8.1, Null 56% ± 8.7, and Bad 12.8%±5.9. Looking at each method separately, the difference between them was not statistically significant, From 125 patients, 18 were switched to a third method. The result was: Good 27.8% ± 21.2, Null 50%±23.6, and Bad 22.2%±19.5. Two patients were switched to a forth method. For one the result was Good and for the other Null. These were the results calculated for each new Treatment. Looking at the baseline data of each patient, from the beginning of the first treatment to the end of the last one, the results were: Good 26,4%±7.7, Null 46.4%±8.7, and Bad 27.2%±7.8. Conclusion: A non responder patient to a given cytotoxic therapy may respond well to another one. However the change must be done enough soon before the occurrence of any sequela.