In total, 68 patients were eligible. Eight dyads decided not to participate. Reasons not to participate were that participating was perceived too burdensome (n = 4); patient had caregivers, but no principal one (n = 2); dyad was not recruited by accident (n = 1); or reasons were unclear (n = 1). After inclusion, three dyads withdrew from participation for unclear reasons and one patient passed away. The mean recruitment rate was three dyads every month. For analyses, complete data of 56 dyads was available for T0, 49 dyads for T1 (withdrew n =3; incomplete data n = 1; patient deceased n = 1, recurrent disease n =1, reason unclear n = 1), and 45 dyads for T2 (recurrent disease n = 2, logistic mistake (n = 1), incomplete data n = 1). Baseline characteristics are given in Table 1. For age, gender, marital status, level ofeducation, and baseline fatigue, no significant differences were found between caregivers who dropped out (n = 11) and those who completed all assessments at all time points (n = 45), except for baseline burden, which was significantly lower in the group who dropped out (mean rank 16.6) compared to the group who completed all measurements (mean rank 31.4; p = 0.006). Patients who dropped out (n = 11) showed no significant differences compared to the patients who completed all measurements (n = 45), except for baseline HRQoL, which was significantly higher in the group of patients who dropped out (mean rank 37.6) compared to the patients who completed all measurements (mean rank 25.5; p = 0.03).