Abstract
Abstract 935
Long term impairment of quality of life (QoL) and elevated fatigue levels in Hodgkin Lymphoma (HL) survivors have been reported. However, few longitudinal data and no conclusive knowledge on components and determinants of QoL exist so far. Therefore, the German Hodgkin Study Group (GHSG) assessed the patients` QoL within the prospectively randomized studies HD10-12 for a detailed longitudinal evaluation of QoL and fatigue.
QoL was assessed with a psychometrically proven questionnaire (QLQ-S) which contains the EORTC QLQ-C30 among other scales and items. Patients answered the questionnaires before, during, and at the end of therapy and at regular follow-up visits. For all QLQ-C30 functional scales and fatigue, longitudinal courses up to 27 months from diagnosis are given with means and 95%-confidence intervals. Reference values from a German control population were used for interpretation of the results. Components and determinants of QoL were analyzed with special modeling software (MPlus) which allows for full information maximum likelihood estimation of multivariate longitudinal models in the presence of missing data. The predictive value of fatigue at baseline for progression free survival and overall survival was tested in Cox proportional hazards analyses together with other known risk factors.
4,160 patients were included in HD10-12, and 3,208 are evaluable for this analysis (total of 15,722 assessments). Before therapy, HL patients had clearly poorer mean scores in each QoL scale when compared to the German reference population. All scales at baseline were negatively influenced by gender (females) and more advanced disease. Before therapy age ≥50 years was negatively related to physical functioning and cognitive functioning, but positively to social functioning. After a decrease of QoL during chemotherapy, all scales showed considerable improvement. However, usually long term QoL remained below normal reference values and this was most pronounced in patients ≥50 years of age and advanced stages. A QoL model with three factors (physical, mental, and social) showed very good fit (RMSEA<.05) and high stability of QoL already 12 months after diagnosis. No relevant effect of the type of treatment could be detected. Overall, 44.7% of patients never experienced severe fatigue ≥50 (relative scale from 0–100), and 17.4% had fatigue only temporarily during treatment, and 15.1% had severe fatigue before therapy which vanished after therapy. In addition, 6.8% of patients developed severe long term fatigue without being severely fatigued before, and 6% had permanently severe fatigue. Cox regression for overall survival revealed that severe fatigue at baseline is a significant, strong and independent risk factor for death from any cause (p<.05, HR= 1.5). Other significant risk factors for OS included age, infradiaphragmatic nodes, and large mediastinal mass. In contrast, gender, high ESR, extranodal involvement, B-symptoms, intermediate stages, advanced stages were not significant.
This is the first detailed QoL and fatigue analysis in HL patients covering all stages of the disease before, during, and after therapy. QoL domains are clearly impaired before the onset of chemotherapy, but improve over time substantially. Baseline QoL is affected in considerable degree by tumor- and patient-specific characteristics. Importantly, type and intensity of HL treatment have no relevant negative impact on long term QoL or fatigue. The strong impact of severe fatigue at baseline on overall survival is currently analyzed in more detail and results will be presented.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
