Baseline Quality of Life Assessed with the EORTC QLQ-C30 Predicts the Treatment Outcome of Patients with Diffuse Large B-Cell Lymphoma: Results of a Prospective Cohort Study

Abstract

Abstract 1595 Background:The self-reported Quality of life (QoL) assessment has been widely used as an endpoint for clinical trials along with traditional endpoints of response and survival because the QoL can provide the health-related information, and it should be considered another parameter to estimate the efficacy of treatment. However, there is growing evidence that the QoL assessment can be used as a prognostic factor in cancer patients. Thus, the prognostic value of baseline QoL was reported in several different types of cancers independent of other clinical prognostic factors. Considering the performance status of patients is an important prognostic factor in lymphoma, the assessment of QoL at diagnosis might be an informative tool for predicting treatment outcome of lymphoma. However, few data exists regarding the prognostic relevance of baseline QoL in patients with lymphoma. Thus, we evaluated the prognostic impact of baseline QoL on the outcome of patients with diffuse large B-cell lymphoma (DLBCL), the most common subtype of non-Hodgkin lymphoma. Methods:The Samsung Medical Center Lymphoma Cohort Study (SMCLCS) is a prospective cohort study registering patients diagnosed with Hodgkin and non-Hodgkin lymphoma at the Samsung Medical Center (NCT00822731). All patients completed the version 3.0 of the European Organization for Research and Treatment of Cancer (EORTC) Core Questionnaire (QLQ-C30) after they provided written inform consents. Among patients registered in the SMCLCS, we analyzed the baseline QoL and clinical data of 263 patients consecutively diagnosed with DLBCL between September 2008 and February 2011. Primary CNS lymphoma was excluded, thus, all patients were treated with rituximab-CHOP. We made a prognostic model according to the scores of three items: (1) Global health status representing general health condition, (2) Functional scale representing physical function, (3) Symptom scales representing symptoms and signs. Each item was counted as one point if the score of each item was higher than 50. Thus, patients were classified as good (3 points), moderate (2 points) and poor QoL (score =0 or 1) groups. The QoL scores of patients with marginal zone B-cell lymphoma (MZL, n = 110) were used as a control. Results:The median score of Global health status was 50 (range 0–100) while the median scores of Functional and Symptom scales were 78.4 (24.1–100.0) and 79 (25.9–100.0), respectively in patients with DLBCL. These scores were significantly lower than a control group, MZL (P < 0.05). When patients were categorized into three groups, the good and moderate QoL groups accounted for 43.0% (n = 113) and 41.1% (n = 108) of patients, respectively. The proportion of poor QoL group of DLBCL was 16.0% (n = 42), and it was higher than that of MZL (2.7%). When the QoL at diagnosis was compared with other clinical factors, the poor QoL was significantly associated with poor ECOG performance status (≤ grade 2) whereas age older than 60 years was not significantly related with the poor QoL. The clinical factors representing high tumor burden such as elevated serum LDH, the number of extranodal involvement and Ann Arbor stage were significantly associated with the QoL model, thus, patients with high tumor burden showed poor QoL (P < 0.05). In consistent with this, patients with high or high-intermediate International Prognostic Index showed worse QoL scores than low or low-intermediate risk group. The frequency of treatment-related mortality was significantly higher in patients belonging to moderate and poor QoL groups than good QoL group (P < 0.05). As a result, more than 20% of the moderate or poor QoL group failed to complete the planned treatment schedule of rituximab-CHOP while only 10% of patients of the good QoL group could not finish the treatment. As a result, the overall survival of the good QoL group was significantly better than moderate and poor QoL groups (p < 0.05). This risk stratification based on the baseline QoL assessment was independently prognostic for overall survival in the multivariate analysis after controlling for International Prognostic Index. Conclusion:The baseline quality of life assessed with the EORTC QLQ-C30 is significantly associated with high tumor burden and worse overall survival of DLBCL. Thus, this prospective result suggests that the it may become a feasible prognostic factor in patients with DLBCL. Disclosures:No relevant conflicts of interest to declare.