Caren van Roekel

102 Chapter 3 microspheres (typically several million). 166 Ho- and 90 Y-resin microspheres will therefore have a larger embolic effect and likely also more post-embolic symptoms such as pain, fever and loss of appetite. The study of Cosimelli et al. is most comparable to the HEPAR I and II studies. Cosimelli et al. reported that QoL was not adversely affected in their cohort of patients with metastatic colorectal carcinoma. However, QoL was not tested shortly after treatment, which is an important difference (15). The changes in QoL after RE were also investigated in a first-line setting. In the SIRFLOX, FOXFIRE and FOXFIRE-Global studies, the possible role for RE as a first-line treatment was investigated. QoL was assessed in the patient group receiving systemic therapy alone and in the patient group receiving RE as an addition to systemic therapy. QoL was slightly worse in the combination group at 2-3 months follow-up, but this was not deemed clinically meaningful (24). There are several limitations to this study. First, the total number of patients was limited. Second, there was a large loss to follow-up since patients were excluded from the HEPAR II study after diagnosis of progressive disease. This may also have led to a biased representation of the QoL of our study population and it may explain why response category did not significantly influence QoL in the analyses. Third, the QLQ-LMC21 questionnaire, created for patients with colorectal liver metastases, was used to complement the more general QLQ-C30 questionnaire, although colorectal cancer was not the only tumor type in this study. One of the strengths of this study is its prospective nature and the high compliance rate regarding the QoL questionnaires. QoL was frequently assessed and especially the 1-week post treatment questionnaire offered valuable insight in the short-term effects on QoL and patients’ transient symptoms. Another strength of this study is the use of a longitudinal approach for the data analysis. By using a mixed model with a random intercept per patient, the variation between patients and data clustering were taken into account. More knowledge on the influence of 166 Ho-RE on QoL is important for several reasons. Above all, this information is needed to better inform patients on treatment-related adverse effects and may help them to make a well-informed choice between all the available palliative treatment options. In selected