At the European Society of Medical Oncology (ESMO) Congress 2019, the randomized phase 3 CARD trial was presented, with a simultaneous publication in the New England Journal of Medicine.1  This trial randomized 255 men with metastatic castration-resistant prostate cancer in a 1:1 fashion, who previously received docetaxel and an Androgen-Signaling-targeted Inhibitor (ASI), either abiraterone or enzalutamide, to cabazitaxel 25 mg/m2 plus prednisone and granulocyte colony-stimulating factor or the other ASI. 

The trial results were uniformly positive, with the primary endpoint of imaging-based median progression-free survival of 8.0 vs. 3.7 months for cabazitaxel vs. ASTI (HR 0.54, 95% CI 0.40-0.73; p<0.001), respectively.  The median overall survival was 13.6 vs. 11.0 months for cabazitaxel vs. ASI (HR 0.64, 95% CI 0.46-0.89; p<0.008), respectively.  The confirmed PSA >50% decline rate was also in favor of cabazitaxel over ASI with a 35.7% vs. 13.5% (p<0.001), respectively.  Similarly, the RECIST v1.1 response rate was 37% with cabazitaxel and 12% with ASI (p=0.004).  Finally, a clinically significant endpoint of confirmed pain response was 45.0% for cabazitaxel compared with 19.3% with ASI.

Although cabazitaxel was obviously superior in regards to treatment efficacy, the toxicity differences require evaluation.  Adverse events led to treatment discontinuation more frequently with cabazitaxel with at a 19.8% rate compared to 8.9% for ASI.  It should be recognized that the dose of cabazitaxel was 25 mg/m2, as that is the approved dose in Europe, where the trial was performed.  In the United States, the PROSELICA trial, previously demonstrated non-inferior overall survival of cabazitaxel 20 mg/m2 compared with 25 mg/m2 with fewer adverse events with the lower dose.2  This led to the subsequent United States Food and Drug Administration approval of the 20 mg/m2 dose.  In the CARD trial, there were some adverse events of grade 3 or greater that occurred in cabazitaxel more frequently than ASI that are worth considering when applying this data to clinical practice.  This includes fatigue (4.0% vs. 2.4%), diarrhea (3.2% vs. 0.0%), peripheral neuropathy (3.2% vs. 0.0%), and febrile neutropenia (3.2% vs. 0.0%).  These toxicities should be considered when evaluating overall patient fitness, comorbidities, and quality of life goals, to determine the appropriateness of individualized treatment selection in clinical practice.

After evaluating the results of the CARD trial, it appears that there should be pragmatic clinical practice applications.  Reflex interpretations are that patients who previously received an ASI and docetaxel for metastatic castration-resistant prostate cancer should immediately receive cabazitaxel.  Although this principle seems solid, it is not clear that this recommendation should apply to uniform populations, and the details of patient selection in the CARD trial should be reviewed.  The inclusion criteria mandated that patients progress within 1 year of starting on their prior ASI.  Hence, the population in the CARD trial included a more hormonally-insensitive tumor population.  Another key eligibility criterion mandated that patients previously received a minimum of 3 cycles of docetaxel.  That eligibility criterion deselects patients who harbor extremely taxane-resistant metastatic castration-resistant prostate cancer that theoretically could be less likely to respond to subsequent taxane-based cabazitaxel chemotherapy.

When evaluating the demographics of the CARD trial, it becomes clear that this patient population had rather aggressive disease characteristics.  For many trials, the predominant form of progression leading to eligibility is PSA progression only.  In this trial, far less than 10% entered the trial that way, and 66.7% and 71.4% of the populations entered, respectively, into the cabazitaxel and ASI arms for pain progression.  The duration of initial androgen deprivation therapy prior to developing castration-resistant prostate cancer was a median of 13.7 months and 12.6 months, respectively, for the cabazitaxel and ASI arms.  This is a very short time to development of castration-resistance, again hinting at a very hormonally-insensitive tumor population.

In summary, the CARD trial strongly supports the use of cabazitaxel chemotherapy in a patient who previously progressed on an ASI and docetaxel for metastatic castration-resistant prostate cancer.  However, we must always consider the individual patient when making treatment decisions.  After a thorough evaluation of the eligibility criteria and demographic data, it becomes evident that the typical CARD patient may not be the average patient in this advanced disease setting. 

One important implication beyond standard clinical practice extends into the clinical trial arena.  For example, many ongoing clinical trials are randomizing novel agent(s) vs. a second ASI in a previously ASI treated population.  We, as a field, should remain dedicated to completing accrual to these trials, and the eventual findings from these trials should not be hastily dismissed.  If these trials are positive in favor of the novel therapeutic agent(s), it will be important to closely evaluate the control ASI arm demographics.  Additionally, some of the trials that are ongoing include administration of the second ASI in the pre-chemotherapy disease state, hence the results from the CARD trial may not be applicable to this disease state.  Finally, in clinical practice, there are many patients who prefer to avoid transitioning directly from docetaxel to cabazitaxel chemotherapy due to comorbidities, potential treatment toxicity and quality of life considerations.  All of these details should be taken into consideration for patient therapeutic selection, both in standard clinical practice and clinical trial enrollment.

Highlighted randomized trials for metastatic castration-resistant prostate cancer patients who have progressed on prior abiraterone or enzalutamide and now are using the other androgen-signaling-targeted inhibitor (ASI) in the control arm

  • Randomized phase 2 trial of I131-1095 with enzalutamide vs. enzalutamide in chemotherapy naïve patients (NCT03939689)
  • KEYLYNK 010: Randomized phase 3 trial of pembrolizumab plus olaparib vs. abiraterone or enzalutamide (NCT03834519)
  • KEYNOTE 641: Randomized phase 3 trial of enzalutamide +/- pembrolizumab (prior abiraterone allowed) (NCT03834493)
  • Randomized phase 2 trial of enzalutamide +/- radium-223 (prior abiraterone allowed) (NCT03344211)
  • TRITON3: Randomized phase 3 trial of rucaparib vs. physician’s choice (abiraterone, enzalutamide or docetaxel) for those with homologous recombination deficiency (NCT02975934)

Written by: Evan Yu, MD, Professor, Department of Medical Oncology, University of Washington School of Medicine,  Member, Fred Hutchinson Cancer Research Center and Assistant Fellowship Director, Hematology and Oncology Fellowship Training Program, University of Washington and Fred Hutchinson Cancer Research Center


  1. de Wit, Ronald et al. 2019. “Cabazitaxel versus Abiraterone or Enzalutamide in Metastatic Prostate Cancer”. New England Journal of Medicine. Massachusetts Medical Society. doi:10.1056/nejmoa1911206.
  2. Eisenberger, Mario et al. 2017. “Phase III Study Comparing a Reduced Dose of Cabazitaxel (20 mg/m2) and the Currently Approved Dose (25 mg/m2) in Postdocetaxel Patients With Metastatic Castration-Resistant Prostate Cancer—PROSELICA”. Journal of Clinical Oncology 35 (28): 3198-3206. American Society of Clinical Oncology (ASCO). doi:10.1200/jco.2016.72.1076.