OBJECTIVES: Health utilities, measured by the EQ-5D, have not been reported for MCL, an aggressive form of non-Hodgkin's lymphoma. This research aimed to explore the validity of the EQ-5D in patients with RR MCL and to estimate minimally important differences (MIDs). METHODS: Data were collected in a phase II, single-arm, open-label study of enzastaurin in RR MCL patients (Morschhauser et al, Ann Oncol 2008). Criterion validity was assessed by grouping patients by International Prognostic Index (IPI) and by Eastern Cooperative Oncology Group performance status (PS). Responsiveness to change was assessed by grouping patients according to changes in PS (improved/stable or worse) and disease status. Distribution- and anchor-based methods were used to estimate MIDs. RESULTS: EQ-5D data were available for 58 of 60 enrolled patients at baseline and for 38 patients at discontinuation. Median age was 66 years, 70% were male, 88% had a baseline PS of 0 or 1, and 52% had an IPI score ≤2. There was a statistically significant (p<0.05) difference in baseline EQ-5D index (0.81, 0.63, 0.58) and visual analogue scale (VAS) (66.9, 56.4, 44.7) by PS group (0, 1, 2, respectively) and at discontinuation by PS group for the index (0.77, 0.61, 0.28). There were no significant differences for the index or VAS scores by IPI group. Those patients with worsened PS (n=12) had significantly worse index and VAS scores than those patients with improved/stable PS (n=18). The estimated MID ranges across the varying methodologies were 0.05-0.10 for the index and 5.2-10.0 for the VAS. CONCLUSIONS: This retrospective study provides evidence that the EQ-5D is a valid instrument to assess health status in patients with RR MCL, based on differentiation of patients by clinical status. The estimated MIDs are consistent with those reported for other cancers and will be helpful in designing future studies.