The robust outcome of CARD with a hazard ratio (HR) for radiographic progression-free survival (rPFS) 0.54 in favor of cabazitaxel allows meaningful subgroup analyses. Preplanned subgroups of particular interest and that, importantly, retained statistical significance including the time to failure on the first androgen receptor targeted agents (ARTA) (0-6 months, 6-12 months) and the docetaxel- first ARTA sequence.
It was found that the cross-over between ARTA in patients who had experienced some benefit of the prior ARTA (6-12 months until failure) had no better outcome as compared to those who had virtually no benefit or primary resistance. If there was any trend it was in the opposite direction. Therefore, although the study did not test the cross-over in patients with longer benefits to the initial ARTA, it is very unlikely that the outcome would be completely different in someone who had responded for instance 18 months on prior ARTA.
Likewise, there was no difference in the sequence of ARTA followed by docetaxel or vice versa. Here it can be concluded that docetaxel in between the two lines of ARTA obviously does not restore the AR sensitivity.
In a posthoc analysis, solicited by the reviewers, we also examined the sequence of the 2 ARTAs, as some may consider the cross-over to enzalutamide after failure on abiraterone effective treatment; the sequence abiraterone-enzalutamide, however, did not perform any better, the HR for rPFS remained of the same magnitude, 0.57.
We collected information about treatment post-progression. Apparently, the 33% of patients who subsequently received cabazitaxel upon failure of the ARTA, did not confound the equally robust OS benefit, HR 0.64. Moreover, 22% of patients received palliative radiotherapy only, as presumably some of these patients had become too frail to receive systemic treatment. This obviously points to the conclusion that we should not miss the window of opportunity of choosing the right treatment at the right time.
One question that is more difficult to address is the dose of cabazitaxel used; 25 mg/m2 according to the EU label, with primary G-CSF prophylaxis. The question is if the results would have been different if a 20 mg/m2 would have been chosen. PROSELICA1 formally tested and showed non-inferiority of the 20 mg dose. Thus, it is extremely unlikely that CARD would have notably different. Yet, in PROSELICA the 25 mg dose did result in a higher tumor response rate and it is important to note that in our CARD study cabazitaxel 25 mg/m2 was very well tolerated; only 3.2% neutropenic fevers occurred and the total rates of AEs, SAEs, and serious AEs were similar between the 2 groups. Therefore the choice for the use of either the 20, or 25mg dose, will remain an Institutional choice, or personal preference.
Written by: Ronald de Wit, MD, PhD, Department of Medical Oncology, Erasmus University Medical Center, Rotterdam, the Netherlands.
Clinical Trial Information: NCT02485691
References:
1. De Bono et al. “Phase III non-inferiority study of cabazitaxel (C) 20 mg/m2 (C20) versus 25 mg/m2 (C25) in patients (pts) with metastatic castration-resistant prostate cancer (mCRPC) previously treated with docetaxel (D).” Journal of Clinical Oncology. 2016 DOI: 10.1200/JCO.2016.34.15_suppl.5008.