Read the Full Video Transcript

Zach Klaassen: Hello, and thank you for joining us for this UroToday Journal Club discussion. My name is Zach Klaassen. I'm an associate professor in the Department of Urology at the Medical College of Georgia, and with me today is Rashid Sayyid, who is the urologic oncology fellow in the division of urology at the University of Toronto. Today we're discussing neoadjuvant cabazitaxel plus abiraterone plus leuprolide in high risk prostate cancer patients, the ACDC-RP Phase II Trial, and this trial was recently published in Clinical Cancer Research from the team from Toronto, led by Dr. Neil Fleshner. We know that definitive local therapy provides excellent oncological outcomes for most patients with clinically localized prostate cancer. However, patients with high risk localized disease treated with radical prostatectomy experience treatment failure rates as high as 65%. Unfortunately, the 15 year mortality rate remains in excess of 30% despite treatment intensification with adjuvant and salvage and/or chemohormonal therapy, which does suggest the presence of micrometastatic disease at the time of diagnosis.

Early treatment intensification with neoadjuvant therapy has been proposed to improve oncological outcomes for these high risk prostate cancer patients. In this setting previous trials and meta-analyses have shown improved pathological outcomes, specifically with regards to pathological complete response, tumor down staging and decreased positive margin rates. However, these approaches have focused on short-term outcomes and have not demonstrated clinically significant survival benefits. So the hypothesis of the ACDC trial was that the combination of abiraterone and taxane chemotherapy may overcome early resistance mechanisms, characteristics of ADT exposure with this treatment strategy already showing survival benefits in the first line metastatic hormone sensitive prostate cancer setting. So the study objective was to compare the pathological early oncologic and safety outcomes of neoadjuvant leuprolide acetate and abiraterone acetate with or without cabazitaxel in localized high-risk prostate cancer patients prior to planned radical prostatectomy. As we can see here, this was a randomized, open label, multicenter, parallel group phase 2 trial with a planned safety run-in.

The patients were randomized upfront to arm A, which is leuprolide, abiraterone, and six cycles of cabazitaxel versus leuprolide and abiraterone in arm B, and this was for 24 weeks, and then all patients subsequently had a prostatectomy with the outcomes of pathological complete response and minimal residual disease at the time of prostatectomy. Inclusion criteria for this trial were quite straightforward. 18 years or older histologically confirmed prostate adenocarcinoma. As mentioned at the top, D'Amico high risk patients defined as PSA greater than 20, Gleason score greater than or equal to eight with a minimum of 2 core positive or Gleason score 4+3 with a minimum 3 cores positive and PSA greater than 10. These patients all had excellent performance status, ECOG zero to one, life expectancy greater than six months, and no evidence of nodal metastatic disease on conventional imaging defined as non-pathological lymph nodes less than 15 millimeters.

During this trial, PSMA PET became more available and this was allowed in the trial at the discretion of the treating physician. Exclusion criteria included prior chemo, surgery or radiotherapy for prostate cancer, as well as significant abnormalities on screening laboratory exams or active uncontrolled medical comorbidities. Patients were allowed three months of ADT prior to enrollment, but not administered within six months prior to randomization in the trial. Follow-up after radical prostatectomy was quite straightforward. This included PSA and testosterone checks every three months for the first two years, every six months thereafter or as clinically indicated by the physician. With regards to study endpoints, the primary endpoint was pathological complete response or minimal residual disease. In the radical prostatectomy specimen pathological complete response was defined as the absence of morphologically identified carcinoma in the prostatectomy specimen and minimal residual disease was defined as residual cancer burden less than or equal to 0.25 cubic centimeters.

Secondary outcomes included positive surgical margin rate, tumor size, extra prostatic extension, seminal vesicle involvement and nodal involvement. And the authors also performed a post-hoc analysis evaluating demographic and oncological predictors of pathological complete response and minimal residual disease. With regards to safety assessments, a safety run-in after administration of two cycles of cabazitaxel was pre-planned to evaluate the first six participants assigned to arm A for toxicity. Safety and tolerability were determined by frequent assessment of adverse events, physical exams, vital signs and safety lab assessments, and a safety follow-up visit was conducted roughly 30 days after the last dose of abiraterone.

The sample size was calculated on the basis of pathological complete response and minimal residual disease, and a rate of 18% following six months of neoadjuvant leuprolide and abiraterone was anticipated. The study was designed so that a two-sided Z test at 70% to detect a 25% difference between the group, assuming a type one error rate of 5% leading to a calculated sample size of 76 patients with 38 patients per arm in the trial. Statistics were quite straightforward. Chi-square and Fisher exact tests were used to compare categorical variables. Mann Whitney U-test was used for non-parametric comparisons of continuous variables. Survival analysis performed using Kaplan Meier curves with a log rank test used for group comparisons and predictors of pathological complete response and minimal residual disease were evaluated using univariable logistic regression. Now I'm going to pass it off to Rashid who's going to walk us through the results and take home discussion points from this trial.

Rashid Sayyid: Thank you for that excellent introduction, Zach. So we see here we have the consort flow diagram. Initially 78 patients were randomized, 42 to arm A, which is the cabazitaxel arm, and 38 to arm B, which was the no cabazitaxel arm. A number of patients were withdrawn for numerous reasons such as a chemotherapy related adverse event, whether they were ineligible for chemotherapy, consent withdrawal, or different issues such as hepatotoxicity, particularly in arm B. And given that this is a surgical trial and although we try to follow the intent to treat or the intent to treat principle with this analysis, given that it is a surgical trial and patients who don't undergo surgery cannot be analyzed, the per protocol analysis was applied in these patients. And as such, we have 37 patients in arm A and 33 patients in arm B, which is the no cabazi arm. A safety run-in was performed for the first six patients, and we note that there were no dose limiting toxicities for cabazi, so the cabazi dose of 25 milligrams per meter squared was continued as the standard dose.

Next, let's look at the baseline patient characteristics in table one. We see that the median age was 63 and a half, which is typical for patients with this disease. In contrast to other trials, one of the major criticisms of neoadjuvant trials of the prostate cancer space is that they're over enriched with low and intermediate risk patients. And we see here that almost three quarters of patients had high risk disease. The baseline PSA was about 22, and when we look at the biopsy Gleason score, over 90% of patients had grade group three disease or worse. We see that 29% had grade group three and then 37% had Gleason score nine disease on biopsy. We see that not in addition to a high grade, there's a high volume of disease. So when we look at the number of cores taken, the median was 12, which is standard, but of these 12, the median of 10 and a half were positive, again, high volume disease with a maximum percent core involved in 95%.

Next, when we look at the number of chemo cycles received, we see that in arm A the cabazi arm, 81% of patients received six cycles. Again, we want to note that these patients were clinically localized based on conventional imaging. And so a question that's going to arise is, well, what about if these patients had a PSMA PET? So a number of patients had a PSMA PET performed, so seven out of the 70 total about 10%, and this was at the investigator's discretion. And we see that of these seven patients, all seven had evidence of pelvic nodal disease and all of these patients underwent a radical prostatectomy with extended pelvic lymph node dissection. Next, let's look at the pathologic outcomes on the radical prostatectomy specimen. When we look at the best response with complete response, we see that 7% overall had a complete response, but this was a bit lower in arm A, the cabazitaxel arm at 5.4% compared to 9.1% in arm B.

So again, surprisingly, patients who received cabazi had a bit of a worse complete response. Next, when we look at the co-primary endpoint of complete response or minimal residual disease, again, we see that these are somewhat similar, 43 and 45% between the two arms. Next, when we look at tumor volume, we don't see any differences. They may be slightly lower in arm A and when we look at surgical margins, on the other hand, although this was not statistically significant, we see that the rate of positive surgical margins was almost half at 13 and a half versus 30% in favor of the cabazitaxel arm. There were no differences in the pathologic stage, and surprisingly the node positivity was almost twofold higher than cabazitaxel arm of 32 versus 18%. Looking at outcomes following surgery, radiation therapy was given to about 20% of patients in both arms, and the radiation type was for the most part salvage consistent with current recommendations for early salvage radiation therapy. Further chemotherapy was given in one patient cabazitaxel arm and none of the patients in the no cabazitaxel arm.

Next, we looked at time to biochemical failure defined as receipt of salvage radiotherapy or PSA greater than 0.2, and similar to what we saw with the pathologic outcomes, again, there was no difference in the biochemical failure rates between the two arms. And we note here that given that these patients receive testosterone suppression neoadjuvantly, in order to minimize any biases we calculated the time from testosterone recovery, given that if patients had a suppressed testosterone, we expect the PSA has already suppressed. So to minimize the lead time bias, we only assess that in patients who had recovered their testosterone, and we see that within a median followup of 13 months, just over a year, almost all patients had recovered their testosterone low. So this is very important when we're counseling our patients, giving them the testosterone suppression pre-op with ADT plus an ARAT that just over 90% of patients are going to recover their testosterone within the first year.

What about differences in biochemical failure rates by the pCR or MRD status? In other words, if patients had an improved pathologic response, did this correlate with oncologic outcomes? And the answer appears to be yes, while not statistically significant, and this is limited by that relatively small sample size and short-term follow-up, we see that quite clearly from this Kaplan Meier curve that patients who had a pathologic complete response or minimal residual disease on the prostatectomy specimen had evidence of superior biochemical failure rates. Next, we look at the grade three adverse events. We don't see any new signals here, but as we would expect with patients receiving chemotherapy, there was a higher incidence of febrile neutropenia in the cabazitaxel arm at 7.5%. Also, surprisingly, the hepatotoxicity rates, the grade three or worse were higher in the no cabazitaxel arm at 10% compared to 5% in the cabazitaxel arm.

We also note that of the six initial patients who were recruited, two of these six had venous thromboembolic events postoperatively and so the protocol was amended afterward to include 28 days of low molecular weight heparin postoperatively in addition to the standard pre-op sub-Q heparin that's given to all these patients. And following that, none of the patients had any VTE events. Next, how do we put the ACDC trial results in context? We see that overall the pathologic complete response minimal residual disease rate was about 43 to 45%, and we saw a pathologic complete response in 5 to 9%. So let's look at the analysis or the study by Taplin et al. that was published in the Journal of Clinical Oncology in 2013. In this trial, patients were given 12 weeks of an LHRH agonist plus or minus abiraterone acetate. And after the 12 weeks, all patients received both LHRH agonist and abi for the 12 weeks before surgery.

So a total of 24 weeks with the first 12 weeks randomized to LHRH plus or minus abi and the second 12 weeks everybody got both medications and we see that the pathologic complete response rate was 4 to 10%, which is quite similar to the results from our trial. But the pathology complete response or minimal residual disease rate was 48 to 62%. It was a little bit higher. Next, we look at the PUNCH trial, which randomized patients to six months of neoadjuvant ADT plus docetaxel prior to surgery versus surgery alone. So again, an important note here that the control arm was surgery alone. While this was a negative trial, meaning it did not meet its primary endpoint of three-year biochemical progression-free survival improvement, it did demonstrate that the overall biochemical progression-free survival rate was better in the neoadjuvant ADT plus docetaxel arm with the hazard ratio of 0.69.

And we saw an improvement as well in the metastasis free survival rate with a hazard ratio of 0.7. However, we see that none of the patients had evidence of a pathologic complete response in the radical prostatectomy specimen and so this begs the question whether this lack of complete response is related to the lack of androgen receptor signaling inhibitor used in this trial, which is in contrast to the ACDC trial and that by Taplin et al. More recently we had the ARNEO trial published in European Urology 2023, where patients received three months of degarelix plus apalutamide versus degarelix alone. And again, we note this is three months as opposed to six months in the other neoadjuvant trials. We see here that the minimal residual disease with MRD rate was 38% in the combinatory arm versus 9.1% in the control arm degarelix and again, the pathologic complete response rate was zero.

And so given the difference in the neoadjuvant treatment duration, we wonder whether the pCR rate of 0% is related to the short period of neoadjuvant therapy. And so in conclusion, we note from the ACDC trial that neoadjuvant addition of cabazitaxel to abi and ADT does not improve pathologic outcomes in D'Amico high-risk patients undergoing RP and as such, we currently cannot recommend it's routine use in clinical practice. But despite these negative results, there are two important takeaways from this trial. So we argue that ADT and androgen receptor signaling inhibitor should probably be the backbone therapy of future trial design in the neoadjuvant setting and this is based on the results of Taplin with six months of ADT plus abi showing a pCR/MRD rate of 62% with a pathologic complete response of 10%. Again, these are similar to ACDC where we see excellent pathologic outcomes, 45% and 5% for pathologic complete response. And again, with ARNEO as well, we see excellent results, particularly with the pCR/MRD rate, and these all outperform results from historic series.

So we see a consistent signal that when you use an adrenoreceptor signaling inhibitor in the pre-op setting, that we have superior pathologic outcomes. And then the next key takeaway message from this trial is that although we see a movement towards early treatment intensification in the prostate cancer space, where we see triplet therapy emerging in the metastatic hormone sensitive disease space, particularly with ARASENS and PEACE-1 and ENZAMET, these results suggest that there may be a limit to how far upstream triple therapy combinations can be moved. And the rationale for cabazitaxel addition in this trial was targeting the androgen insensitive cells such as those that have PTEN mutations. But we know that the mutational burden in the clinically localized setting is much lower than what we see in the metastatic setting. It's about 6% in the localized setting, whereas it's 12 to 25% in the metastatic setting.

And as such, just giving further treatment when those mutations may not be there, may not quite be the most cost-effective regimen because we need to consider the additional toxicities of this treatment intensification. And as such, any treatment intensification efforts in this space need to be biomarker driven. Which leads us to the GUNS trial or the Genomic Umbrella Neoadjuvant Study that's currently open and enrolling patients, which is a biomarker selected neoadjuvant trial that is including patients with high risk localized prostate cancer who receive LHRH agonist plus apalutamide for eight weeks and afterwards undergo genomic profiling and are further randomized to one of four arms based on the mutation that patients harbor. And as such, any further treatment in addition to the ADT plus the antigen receptor signaling inhibitor is guided by the mutation that a patient has. And so we await the results of this trial that's currently enrolling, which will be very exciting for their patients. We thank you for joining us for this Journal Club today of the ACDC-RP study.