Fatigue in Hodgkin Lymphoma Patients: Longitudinal Evaluation and Curve Types up to 5 Years

Background: Long-term impairment of quality of life and elevated fatigue levels in Hodgkin Lymphoma (HL) survivors have been frequently reported. However, few longitudinal data and no knowledge on types of fatigue development exist. Therefore, the German Hodgkin Study Group (GHSG) assessed the patients´ fatigue within the prospectively randomized HD13-15 studies (G5) aiming at a detailed longitudinal evaluation and classification of fatigue curves in these studies.

Methods: Fatigue measurements used the respective scale of the EORTC QLQ-C30 with scores between 0 and 100. Patients answered the questionnaires before, during, and at the end of therapy and at regular follow-up visits. Longitudinal development of fatigue up to 5 years after end of therapy is given with means and 95%-confidence intervals. The impact of different risk factors and treatments was estimated with linear multiple regression. Analysis of treatment effects was performed separately within the studies for long term fatigue, 2 and 5 years after end of treatment. These analyses were adjusted for age, gender and baseline scores of fatigue. Growth mixture models using Mplus were applied to identify different types of fatigue trajectories within the clinically best arms of the studies which represent current GHSG standards. The level of significance was set to 0.05 and no correction for multiple testing was made. Full information maximum likelihood techniques (FIML) were used to account for missing data.

Results: 5,306 patients ≤ 60 years qualified, 325 patients from the early stopped arms B and D in HD13 were excluded and 3,619 survivors contributed at least a baseline fatigue score. Before therapy, fatigue scores were increased in all stages with a significant impact of stage (HD13: 30.8, HD14: 39.8, HD15: 49.0). The respective percentages of patients with very severe fatigue ≥ 50 were 22.3% in HD13, 36.8% in HD14 and 48.2% in HD15. Significant risk factors for baseline fatigue were female gender, albumin < 4g/dl, Hb < 10.5 g/dl, stage IV disease and high ESR (all P<0.01). After a large increase of fatigue during chemotherapy, many survivors showed fast and considerable improvement irrespective of stage and treatment intensity. Long term fatigue as opposed to baseline fatigue was remarkably similar in the three studies. The best predictors of long term fatigue were baseline fatigue and age (P<0.0001).

Analyzing treatment effects on long term fatigue we detected only one significant effect: in the 5^th year after end of treatment, arm C in HD15 (8x BEACOPP-14) led to significantly higher fatigue scores than the standard arm A (8x BEACOPPesc). The clinically best arms of the studies (HD13 arm A with 2xABVD+30Gy IF-RT, HD14 arm B with 2xBEACOPPesc+2xABVD+30Gy IF-RT, HD15 arm B with 6x BEACOPPesc+30Gy IF-RT of tumor rests) all compared favorably and without any significant difference with the other arms in consideration.

Growth mixture modeling identified 3 subgroups in arm A of HD13 and 4 subgroups in the arms B of HD14 and HD15. The subgroups resulted primarily from different baseline fatigue levels, which translated to different levels of long-term fatigue. Percentages of survivors with very severe long-term fatigue ≥ 50 were 16.8% in HD13 arm A, 27.2% in HD14 arm B and 22.0% in HD15 arm B. In each study further 30%-40% of survivors reported less extreme but clinically relevant long-term fatigue <50. Noteworthy, all subgroups of survivors showed very stable long-term fatigue levels after end of treatment.

Conclusion: This is the first detailed longitudinal analysis of fatigue in HL patients covering all stages of the disease, starting before therapy, and identifying different types of fatigue development up to 5 years after end of therapy. In all stages around 15% to 25% of survivors suffer chronically from very severe long-term fatigue and further 30% to 40% suffer to a less extreme but clinically relevant degree. The type and intensity of chemotherapy has surprisingly little influence on long-term fatigue. The most important determinants of long-term fatigue are high fatigue levels before therapy and increasing age, which seems to limit the capability of survivors to recover.