Read the Full Video Transcript

Zachary Klaassen: Hi, my name is Zach Klaassen. I'm a urologic oncologist at the Georgia Cancer Center in Augusta, Georgia. I'm delighted to be joined on UroToday by Dr. Bram De Laere, who's a translational cancer researcher at Ghent University. Bram, thanks very much for joining us today.

Bram De Laere: Hey. Hello, Dr. Klaassen. Thanks for having me. Looking forward to our discussion.

Zachary Klaassen: Yeah, we're going to discuss a really, really fascinating clinical trial platform that you guys have put together and recently published in Nature Medicine, the ProBio trial. Why don't you share your slides and walk us through the trial design and some of the key results you have already?

Bram De Laere: Okay, will do so. First of all, Dr. Klaassen, thanks for having me. It's a privilege to be able to present our latest findings on AR pathway inhibitors versus taxane-based chemotherapy in men with metastatic castration-resistant prostate cancer who have been treated in the ProBio trial. During this presentation, we will touch upon the ProBio workflow, the trial design in which the first results were obtained, the main and secondary findings, and end up with a take-home message of the study.

The ProBio, or simply Prostate Biomarkers trial, was designed to study systemic treatments in men with metastatic prostate cancer. It's an academic study being the first to use an outcome-adaptive platform designed to evaluate treatments across predefined genomic biomarkers. The sponsor of the trial is the Karolinska Institute, with Professor Henrik Grönberg as coordinating investigator, and is currently active in four countries: Sweden, Norway, Belgium, and Switzerland, where Professors Henrik Grönberg, Jan Oldenburg, Piet Ost, and Ashkan Mortezavi are the national principal investigators.

Now, ProBio has a focus on genomic biomarkers and actually uses the analysis of a minimally invasive liquid biopsy. During ProBio screening, blood plasma-derived circulating cell-free DNA and germline DNA from patients are analyzed using a prostate cancer-tailored sequencing panel, which is capable of identifying copy number alterations, point mutations, and structural rearrangements.

Now, upon sequencing and data curation, we basically infer the biomarker subgroup combinations in men with detectable circulating tumor DNA. This is actually defined simply based on the presence or absence of alterations in the androgen receptor, TP53, the TMPRSS2-ERG fusion, and genes associated with DNA repair. Now we actually use this information for the randomization to either one of our investigational treatment arms or a common control arm, which is physician's choice standard of care.

Now, our first results were actually obtained in the CRPC indication, where ProBio enrolled and screened 343 patients who were starting first- or second-line systemic therapy for mCRPC across 25 hospitals in Sweden, Belgium, and Norway. Prior ADT intensification with either docetaxel or an AR pathway inhibitor for hormone-sensitive or nonmetastatic CRPC was allowed and actually occurred in 55% of our patients.

A total of 67% of our patients actually had detectable circulating tumor DNA and were, based on the predefined biomarkers, randomized to an androgen receptor pathway inhibitor—abiraterone acetate or enzalutamide—or a taxane-based chemotherapy, being docetaxel or cabazitaxel. Or, as a control arm, the physician's choice standard of care.

Now, when the primary endpoint, which was here the time to no longer clinically benefiting per PCWG3 criteria, was reached, patients could be re-randomized upon a new circulating tumor DNA analysis. You actually see that feedback loop over there. Now, importantly, men randomized to physician's choice standard of care remained in standard of care upon re-randomization, which allows us to assess overall survival as well.

Now, after 218 randomizations, the statistical threshold for superiority of an AR pathway inhibitor versus taxanes and versus standard of care was reached in the biomarker-unselected all-patients group. So irrespective of the biomarker, we saw that on average an AR pathway inhibitor was significantly extending the time that the patients were clinically benefiting from the treatment, compared to a taxane-based chemo and the control arm. With a median of 11.1 months for the AR pathway inhibitors, versus 6.9 months for the taxanes, and 7.4 for the control arm.

Now, as a secondary endpoint, we also saw that patients treated with the AR pathway inhibitor also had an extended overall survival with a median of 38.7 months, compared to those treated with taxanes at 21.7 months, or a control arm at 21.8 months. Now, these results remained robust in different subgroup sensitivity analyses where we only focused on men starting first-line systemic therapy for mCRPC, first-line mCRPC after ADT monotherapy—so completely naive for either AR pathway inhibitors or taxanes—and also irrespective of the patients who have not previously been treated with that type of drug class to which they have been randomized.

Now, as a secondary finding, ProBio is now also in the unique position to actually have a randomized comparison between AR pathway inhibitors and taxanes, and we can actually assess what the impact is of our predefined genomic biomarkers that were detected in the circulating tumor DNA. Here we are going to measure the differential treatment effect on the time to no longer clinically benefiting between an AR pathway inhibitor in orange and taxanes in blue, within the biomarker-positive and -negative population.

So better outcomes can be seen within specific biomarkers. We actually saw that the effect of an AR pathway inhibitor versus a taxane was 44% higher in patients without mutations or rearrangements in the AR and TP53 genes, resulting in a median time of clinical benefit of 17 months. Now, similarly, we also saw that the effect of an AR pathway inhibitor versus taxanes was 42% higher in patients with a TMPRSS2-ERG fusion, resulting in a median time of clinical benefit of 13.8 months.

But conversely, we also saw that the effect of an AR pathway inhibitor versus taxane decreased by almost 40% in TP53-altered patients compared to wild-type patients. As you can see, no difference in outcomes between an AR pathway inhibitor and taxane-based chemotherapy was observed in patients with TP53-mutant disease, which resulted here on average in a median time of therapeutic benefit of six to seven months. So this clearly warrants the need for alternative treatment options for these men and careful monitoring as these men will rapidly progress.

Now, to conclude, I think we can say that ProBio is acting as this real-time screening platform for multiple treatment-biomarker combinations and has embedded several novel trial design features, such as prospective circulating cell-free DNA profiling, outcome-adaptive randomization, and re-randomization. Our first data actually showed us that on average an AR pathway inhibitor will always outperform a taxane-based chemotherapy in men with, importantly, ctDNA-detectable mCRPC for both our primary endpoint—time to no longer clinically benefiting—and overall survival as a secondary endpoint.

However, the take-home message here is that the genomic markers as detected in the circulating tumor DNA can actually impact the patient outcomes. We're observing better ARPI outcomes in patients without mutations and structural rearrangements in the AR and TP53 genes and those harboring the TMPRSS2-ERG fusion. Importantly, no difference in AR pathway inhibitors or taxanes was observed in TP53-altered mCRPC. So we believe that ProBio is providing a new way to maybe study systemic therapies across genomic biomarkers with several other new treatment-biomarker combinations ongoing as well in different indications.

Finally, and on behalf of the whole ProBio team, we want to express of course our sincerest gratitude to all of our patients, our sponsors, and the whole consortium basically, which is depicted here in the picture, which we take during our biannual international investigator meetings. This one is actually just from three days ago. So if you want to learn more on ProBio, I would like to invite you to visit our website or X handle. Thanks again, Dr. Klaassen, for providing me the opportunity to present our data.

Zachary Klaassen: Bram, beautiful work, a wonderful discussion. I mean, this is really next-generational trial design, so to speak. So maybe just take us under the hood a little bit more about how this unique and really elegant trial design came to be.

Bram De Laere: Yeah. We knew from literature, from retrospective series, that different tumor genotypes can actually yield varying therapeutic benefits, and we know that there are several prognostic molecular biomarkers out there, and that the number of biomarkers actually predictive of therapeutic or clinical benefit remains quite small. I mean, today our best-in-class example that we have is actually PARPi sensitivity in the context of BRCA-altered disease.

And perhaps one reason for this is that the traditional clinical trials have not really been biomarker-driven and have not focused on prospectively investigating which patient subgroups actually are benefiting more from a specific treatment. And the promise of precision oncology has always been that we should select or tailor our treatment based on the molecular characteristics of a patient's cancer. However, given that there are numerous biomarkers, as also therapies, this would actually result in a large number of hypotheses to evaluate.

And if you want to test all of these hypotheses together in a traditional two-armed RCT design, then that actually would be very laborious and expensive. You would require, of course, that prospective molecular profiling. But above all, you would require large sample sizes. And actually, to meet these challenges, we designed ProBio as a platform just for that. The platform format actually provides you an efficient way, we believe, to screen and evaluate multiple biomarker-therapy combinations at the same time. So therefore, one could actually see ProBio as a screening device, which basically aims to elucidate whether certain biomarker signatures can actually help us to identify patients who might derive a greater or a longer clinical benefit from a treatment.

And therefore we actually have used a number of design elements to maximize the information for every patient actually entering into our trial to rapidly see through these multiple hypotheses, which is, one, we have a single common control arm. And these are concurrent controls against which we can contrast the outcomes from the different other investigational treatment arms. We're using outcome-adaptive randomization. This means that we are updating the randomization probabilities over time based on the observed outcomes we have seen from patients that have entered the trial before. So this is actually aiming to allocate our patients more quickly to those investigational biomarker-treatment combinations that have shown to be more effective in others. In other words, this means that fewer patients are being wasted on hypotheses that are unlikely to be true.

And finally, ProBio uses re-randomization of our patients within the investigational arms when the primary endpoint has been reached. There is a plan B for patients that are progressing in the trial. And at the same time, this also increases the amount of information that we can actually gather from each patient enrolled in the trial.

Zachary Klaassen: That's a great explanation. I think we've seen cell-free DNA come on the scene over the last few years. It's made a mark in bladder cancer, and you guys have really put it on the map with this trial. It was two-thirds of your patients had cell-free DNA positive, I believe.

And I think when you look at this, we've been doing somatic testing, we've been doing germline testing. Now we take a cell-free DNA sample, which is certainly easy to do. How do you see cell-free DNA transforming how we treat advanced prostate cancer, maybe over the next five years or so?

Bram De Laere: In the next five years, even if it will take that long, I foresee that the molecular diagnostic laboratories will have embraced these liquid biopsy technologies, which today, I have to be frank here as well, you rather see them in academic or university hospital centers, given their steep learning curve to implement and interpret these results.

At the same time, I'm from Europe and we have, here in Belgium, for instance, a reimbursement-driven medicine system. And so I believe that the reimbursement of these tests, which are typically a bit more expensive compared to the gold standard tissue profiling, will come into play and will come into place. And the treating physician, I think, will be able to order a liquid biopsy ctDNA analysis to first actually prognosticate the patient, based on that example of detectable or undetectable ctDNA. And in those cases, with detectable circulating tumor DNA, we will be able to infer the tumor profile prior to starting a new line of systemic therapy.

Hopefully by then the liquid biopsy and the biomarker community as such will be able to expand the number of prognostic and treatment-predictive biomarkers to assist the treating physician in tailoring the treatment strategy for their patients. Simultaneously, I believe that ctDNA analysis will also have a use or will be implemented rather in the MRD setting—so in the minimal residual disease setting—for monitoring therapy response.

Based on, for instance, ctDNA clearance or a certain dynamic during a new line of systemic therapy, I think we will be defining new therapy response intervention and progression endpoints, which will dictate when a patient needs to switch therapy. This ideally should result in a clinically meaningful lead-on time compared to classical radiographic and clinical progression, which hopefully should also result in an improved survival for our men.

Zachary Klaassen: That's great. You hinted at it on your last slide on the take-home messages about what may be coming down the line for ProBio. Maybe just give us a little more information about—you mentioned mHSPC, some other additional treatments. What's coming over the next little bit for ProBio?

Bram De Laere: Yeah, very good question. I mean, this is an ongoing study. And during ESMO 2023, we actually presented new data on the health-related quality of life on the ARPi versus taxane comparison. So that was actually also a secondary endpoint in that trial. We also reported or presented data on our investigational carboplatin arm in patients with mCRPC, and also reported the outcomes of patients with undetectable ctDNA at ProBio screening. These manuscripts are now being reviewed internally and we will be submitting them soon for publication.

Today, also, it's important to know that there are multiple evaluations with new agents and biomarkers ongoing. For instance, we are evaluating PARP inhibition in DNA repair-deficient mHSPC and mCRPC, and also are evaluating Akt inhibition in, for instance, PI3K-altered mCRPC patients.

As mentioned, besides men with metastatic castration-resistant prostate cancer, ProBio is also running the trial in patients with metastatic hormone-sensitive disease. Now today we are limiting our enrollment to patients with solely de novo synchronous M1 disease. A new protocol amendment is in the making where we also will start enrolling men with metachronous metastatic disease as well.

Simultaneously, ProBio is also thinking about how can we salvage our patients with undetectable ctDNA. As you said as well, Dr. Klaassen, one-third of our patients were unable to detect ctDNA, and today these men are not randomized in the trial. This info is however communicated to the treating physician and where these men are treated per standard of care.

Now, if they were willing, these patients could actually enter an observational arm within ProBio, and actually where we have observed excellent outcomes in these men. So now ProBio is aiming actually to open a new treatment arm to test de-escalation strategies in this population, given their excellent outcomes.

And then finally, we have received interest from new countries. They have actually expressed their interest in the ProBio platform, so now we are also looking into expanding the trial.

Zachary Klaassen: That's excellent. Congratulations on just absolutely phenomenal work and on a really impactful publication and multiple presentations over the last year or so. And to you and the whole ProBio team, congrats. And thanks so much for your expertise and your time discussing this on UroToday.

Bram De Laere: Okay, super. Thank you very much, Dr. Klaassen, for having me.