Presentation Authors: Wout Devlies*, Leuven, Belgium, Greet De Coster, Nancy Van Damme, Thierry Roumeguère, Thierry Quakels, Brussels, Belgium, Ben Van Cleynenbreugel, Leuven, Belgium, Peter Dekuyper, Filip Ameye, Ghent, Belgium, Wouter Everaerts, Steven Joniau, Leuven, Belgium
Introduction: Robotic assisted laparoscopic prostatectomy (RALP) is increasingly used as treatment for high-risk prostate cancer (PCa). However, quality of life (QoL) data in this subgroup is lacking. We evaluated the QoL following RALP in high versus low/intermediate risk PCa patients using the Be-RALP database: a large, prospective, multicentre database.
Methods: A total of 9235 patients treated by RALP between 2009 and 2016 were included in the database. The EORTC QLQ-C30 questionnaire were completed on baseline, 1-, 3-, 12- and 24-months post-surgery._x000D_
Two groups, a high and low-intermediate risk group of 2745 patients each were matched 1:1 using age, year of surgery, nerve sparing and centre size as matching criteria. Longitudinal mixed models quantified the relation of the QoL changes from baseline with follow-up time, risk group and postoperative radiotherapy (RT)/androgen deprivation therapy (ADT). All models included the two- and three-way interactions of time, risk and postoperative treatment). Backward selection was used to remove nonsignificant interaction terms (Î± = 0.01). Covariates (table 1) corrected for case-mix variation.
Results: Risk was not a significant predictor of QoL (all p>0.07). QoL changed over time and was related to hospital size (all p < 0.003). In global, physical and cognitive QoL, surgery year was a significant predictor (all p < 0.003).Postoperative treatment was not significant in predicting QoL after surgery (all p>0.04)
Conclusions: Risk group fails to predict quality of life after RALP. Longitudinal figures curves and modelling were risk independent. (Figure 1)Hospital size was the strongest covariate with better QoL in higher volume centres (all p < 0.0001).Longitudinal graphs present a negative effect of postoperative treatment on QoL. This trend is not significant in the mixed effect modelling (p>0.04).
Source of Funding: De Wever foundation - Prostate cancer