Combination therapies in prostate cancer: proceed with caution
Mené sur 806 patients atteints d'un cancer de la prostate résistant à la castration et présentant des métastases osseuses, cet essai de phase III évalue l'efficacité, du point de vue de la survie sans événement indésirable lié au squelette, et la toxicité de l'ajout du radium-223 à un traitement combinant acétate d'abiratérone et prednisone ou prednisolone (durée médiane de suivi : 21,2 mois)
In The Lancet Oncology, Matthew Smith and colleagues report the results of the phase 3 ERA 223 trial, in which they investigated the effect of combining two therapies approved by the US Food and Drug Administration for the treatment of metastatic castration-resistant prostate cancer: radium-223 and abiraterone acetate plus prednisone or prednisolone. There is a biological rationale for this combination in view of the known cross-talk of DNA repair from ionising radiotherapy and androgen signalling. Despite the rationale for this combination, the ERA 223 trial suggests that the combination of radium-223 with abiraterone acetate plus prednisone or prednisolone causes more harm than good. The trial, which included 806 patients (all of whom received abiraterone acetate plus prednisone or prednisolone), was unblinded prematurely after more fractures and deaths were noted in the radium-223 group than in the placebo group. Median symptomatic skeletal event-free survival did not differ between groups (hazard ratio [HR] 1·122 [95% CI 0·917–1·374]; p=0·2636). In the initial analysis, which was done after 136 deaths occurred in the radium-223 group, overall survival was significantly worse in patients who received radium-223 than in those with received placebo (HR 1·35, 95% CI 1·05–1·73). In the current paper, which includes longer follow-up (155 deaths in the radium-223 group), overall survival no longer differs between groups (HR 1·195, 95% CI 0·950–1·505). Unexpectedly, fractures were more common in the radium-223 group than in the placebo group (112 [29%] of 392 patients vs 45 [11%] of 394 patients).
The Lancet Oncology , commentaire, 2018