chemotherapy (eight), usually pixantrone. One patient received high-dose chemotherapy and ASCT. None of these patients were among the group who were still alive and under follow-up. Factors related to OS were as expected. The IPI at the time of DECC chemotherapy was predictive, with the median survival being not reached in the good-risk group, 8Á6 months in the intermediate-1-risk group, 3Á8 months in the intermediate-2-risk group and 3Á5 months in the poor-risk group. The number of prior lines of therapy did not affect OS; however, the time since completion of the last line of treatment was predictive, with a median survival of 3Á6 months if this was <6 months compared to 8Á3 months if >6 months. The radiological response to treatment was also predictive were also predictive of OS. Median survival was not reached in the group achieving a CR or CMR, 36Á0 months in those achieving a PR and 8Á1 months in those with progressive disease. These results are better than expected given the combined median OS of only 5Á6 months. The reason for this is that patients who progressed or died shortly after commencing DECC often did not receive any further imaging so are not included in the data above (median survival 1Á6 months). In conclusion, DECC chemotherapy is an effective palliative regimen for patients with DLBCL, with a small number of patients achieving durable remissions without additional treatment. The combination has the advantage of being an oral, outpatient-based treatment that can be tolerated even in elderly or frail patients. Given the baseline patient characteristics and often refractory disease, the response rates and OS are encouraging and comparable to other treatment options in this setting. With cellular therapies now available, the role of DECC for bridging treatment should be assessed.