Cabazitaxel is a novel microtubule inhibitor, thought by many to supersede the effects of docetaxel. However, at this time, it is currently used in the second line for patients who have failed first-line therapy.
In this study, Kim Chi, MD, and colleagues specifically focused on patients with poor prognosis mCRPC, specifically men with liver metastases, early CRPC (<12 months from ADT start), and/or >4 of 6 poor prognostic criteria (Chi et al, Annals of Oncol, 2016). Consensus opinion for these patients is to use chemotherapy ahead of AR-targeted therapies.
The study protocol is seen below in slides from Dr. Chi’s presentation.
This was a planned cross-over trial. Patients were randomized 1:1 to either cabazitaxel 20-25 mg/m2 or abiraterone+prednisone (AAP) or enzalutamide (Enza). Once they failed on this first line treatment (progression), they would cross over to the other pathway. The primary outcome was “clinical benefit rate” (CBR), defined as PSA decline >= 50%, measurable disease response OR stable disease > 12 weeks. Their secondary objectives are listed above, but include time to progression, PFS and OS. Importantly, they included correlative studies of cfDNA/ctDNA at each step to potentially identify biomarkers for sequencing.
They had planned for accrual of 120 patient (60 per arm) in order to detect a 20% difference in CBR. However, due to changing treatment paradigms and poor accrual, only 95 patients were randomized.
Baseline characteristics were evenly matched between the groups. Key highlights include:
- Median age 67-68
- 86-93% had CRPC within 12 months, 11-24% had liver metastases, 31-38% had visceral metastases
- 91-96% were ECOG status 0-1
- Median PSA was lower in the cabazitaxel group (18.7 vs. 39.4)
- >50% in both groups had prior docetaxel (either for CRPC or CSPC) – 46-47% did not have prior docetaxel
In terms of tolerability, Arm A had a higher rate of Grade 3-5 adverse events (40.9% vs. 6.0%), primarily due to neutropenia and diarrhea.
However, when looking at the first treatment, the CBR was much higher in the cabazitaxel arm (Arm A) – 90% vs. 70% (p = 0.02). This was primarily driven by stable disease at 12 months though, rather than PSA response or measurable disease response. Looking at secondary endpoints, time to PSA progression (HR 0.73, p = 0.26) and PFS (HR 0.72, p = 0.18) favored CAB, but were not statistically significant.
In the patients that moved on to second therapy, there was no difference in CBR (75% arm A, 84% Arm B, p = 0.483). Again, there was no difference in time to PSA progression and OS. In subgroup analyses looking at time to PSA progression and OS benefit (for entire treatment course), while Arm A was favored in all groups, there was no statistically significant difference.
He did briefly review the ctDNA/cfDNA biomarker work. They compared the study group of poor prognosis mCRPC patients to an “all-comers” group, and as expected, found that AR amplification, high levels of ctDNA and DNA repair defects were more common in the study population. Baseline ctDNA fraction was a prognostic indicator of PFS and OS, as was on-treatment change in ctDNA fraction. AR amplification was associated with cabazitaxel PFS benefit (p = 0.055, HR 0.52).
Invited Discussion:
Following Dr. Chi’s presentation, Dr. Stéphane Oudard provided a brief summary and add the following points.
First, he commended the authors on completing this study – the first prospective study to assess sequencing in this disease space.
Here are the current options for management of metastatic prostate cancer – when hormone naïve (mHNPC) and castration-resistant (CRPC):
As he reviewed the study, he had a few points. First, he noted that the cabazitaxel arm had a much higher rate of toxicity, though primarily neutropenia and diarrhea. He did question the lack of presentation of cardiac toxicity. This should be reported. He briefly reviewed the criteria for “poor prognosis” mCRPC – while agreed with all their criteria (and provided the studies to support them), he also pointed out additional markers that should potentially be included:
1) Neutrophil-to-Leukocyte ratio (NLR)
2) Pain related to cancer
Both are important predictors that could have been included.
He also pointed out that while most of the patients saw docetaxel prior to starting on the trial, 46-47% of patients had never been treated with docetaxel. Yet, all major guidelines and approval for cabazitaxel is in patients who have failed docetaxel. So, this is not clinically relevant for patients in most countries, as they would not be eligible to get the drug!
- He also provided some retrospective data to support the fact that patients treated with docetaxelàCABàabi/enza do the best, compared to those treated with docetaxelàabi/enzaàCAB, which supports the current data
1) This is a phase II trial that under-accrued
2) Underpowered for OS (most important endpoint)
3) These patients are likely to die rapidly from prostate cancer
4) Around 50% of patients never even make it to 2nd line therapy
5) Best OS is when patients receive Docetaxel, cabazitaxel and an ART
6) Do NOT miss the opportunity to give chemotherapy!
7) Results must be confirmed in a phase III trial
Presented by: Kim N. Chi, MD, Clinical Associate Professor, Medical Oncologist, BCCA, University of British Columbia, Vancouver, Canada
Invited Discussant: Stéphane Oudard, MD, Ph.D., Georges Pompidou European Hospital, Department of Medical Oncology, Paris, France
Written by: Thenappan Chandrasekar, MD, Clinical Instructor, Thomas Jefferson University, twitter: @tchandra_uromd, @TjuUrology at the 2018 European Society for Medical Oncology Congress (#ESMO18), October 19-23, 2018, Munich Germany