Mario Eisenberger, MD
Phase III results from the PROSELICA trial showed that 20 mg/m2
of cabazitaxel (Jevtana) was as safe and effective as the 25 mg/m2
dose for postdocetaxel patients with metastatic castration-resistant prostate cancer (mCRPC). The findings were were published online in the Journal of Clinical Oncology.
Median overall survival (OS) for patients assigned to the 20 mg/m2
dose (C20) was 13.4 months versus 14.5 months for those assigned to the 25 mg/m2
dose (C25; hazard ratio [HR], 1.024). The one-sided 98.89% upper boundary of the confidence interval (UCI) of the HR was 1.184, which was less than the 1.214 noninferiority margin, thus satisfying the predefined criteria for noninferiority in the intent-to-treat population of patients.
Noninferiority was defined as maintenance of ≥50% of the OS benefit associated with C25 versus mitoxantrone in the TROPIC trial, with 95% CI. The UCI of the HR for C20 versus C25 could not exceed 1.214 under a one-sided 98.89% CI after interim analyses. Secondary endpoints included progression-free survival (PFS), prostate-specific antigen (PSA), tumor and pain responses and progression, health-related quality of life, and safety.
“PROSELICA provides further evidence of the efficacy of cabazitaxel in the treatment of patients with mCRPC who have previously received docetaxel,” first author Mario Eisenberger, MD, The Johns Hopkins University and coinvestigtors wrote. “Data also suggest that patients can achieve a similar clinical benefit from cabazitaxel at both the 20 mg/m2
and 25 mg/m2
starting doses and that dose reductions may be implemented in patients who require them without significant detriment to outcome.”
To determine whether that the 20 mg/m2
dose was noninferior for OS, researchers assigned 598 patients with mCRPC to the C20 group and 602 to the C25 group in a phase III, randomized, open-label trial. Patients enrolled at 172 centers in 22 countries worldwide from April 2011 to December 2013.
More than 85% of patients in both groups had increasing PSA levels at baseline, and most patients in each group had received at least 1 cycle of docetaxel. Nearly half of all patients in each arm had received 3 or more regimens of hormonal therapy. Of the total randomly assigned population, 308 patients (25.7%) had received prior targeted therapy with abiraterone acetate (Zytiga) or enzalutamide (Xtandi). Median time between last dose of docetaxel and random assignment and median time between last dose of docetaxel and disease progression were similar between the 2 groups.
In a secondary analysis of the per-protocol population evaluating only patients who had received at least 3 doses of cabazitaxel, the observed median OS for patients receiving C20 was 15.1 months compared with 15.9 months for C25 (HR, 1.042). The one-sided 98.89% UCI of the HR was 1.224, which exceeded the 1.214 noninferiority margin and thus did not confirm the noninferiority of the C20 dose compared with the C25 dose for OS.
Median PFS was 2.9 months in the C20 arm verse 3.5 months in patients receiving C25 (HR, 1.099; 95% CI, 0.974-1.240). Researchers observed similar rates in each arm for tumor, PSA, and pain progression. The most frequent PFS events were PSA progression (C20, 39.8%; C25, 34.7%) and pain progression (C20, 25.8%; C25, 28.1%).
There was no significant difference in the tumor response rate in evaluable patients receiving C20 and C25 (18.5% vs 23.4%, respectively; nominal P
= .1924). Median time to tumor progression was 9.0 months for patients receiving C20 and 9.3 months for patients receiving C25 (HR, 1.096; 95% CI, 0.902-1.331).
PSA response rates were significantly higher in the C25 arm, with 29.5% of C20 patients and 42.9% of C25 patients demonstrating a ≥50% decline in PSA from baseline (nominal P
<.001). Median time to PSA progression was 6.8 months or longer for patients receiving C25 compared with 5.7 months in the C20 arm (HR, 1.195; 95% CI, 1.025-1.393).
Researchers observed no significant difference in the number of patients who experienced a pain response and the rate of pain progression was similar in the 2 groups. Median time to pain progression was 6.2 months for patients receiving C20 versus 6.4 months for patients assigned to C25, with a similar risk for pain progression in both groups (HR, 1.046; 95% CI, 0.874-1.251).
The safety population consisted of 580 patients assigned to C20 and 595 assigned to C25 who received at least 1 dose of cabazitaxel. Patients in the C20 group received a median of 6 treatment cycles for a median duration of 18 weeks. Patients in the C25 group received a median of 7 treatment cycles for a median duration of 21 weeks.
“The safety profiles of both doses were consistent with previous reports, with a relatively low incidence of manageable grade ≥3 AEs, particularly with the lower cabazitaxel dose,” Eisenberger, et al wrote.