Read the Full Video Transcript

Zachary Klaassen: Hello, my name is Zach Klaassen. I'm a urological oncologist in Augusta, Georgia. I'm delighted to be joined on UroToday by Doctor Alessio Crippa at the Karolinska Institute. We are going to be discussing their ProBio Platform. We've discussed ProBio on UroToday before. They have some very exciting data recently published in Nature Medicine looking at the prognostication of ctDNA and mCRPC. Alessio, thanks so much for joining us today.

Alessio Crippa: Thank you. Thank you so much for inviting me and for presenting the results of the recently accepted paper in European Urology Oncology entitled, "Prognostic Value of Circulating Tumor DNA Fraction in Metastatic Castration-Resistant Prostate Cancer." And this is part of the results from the ProBio trial.

So the outline for today, I'm going to reintroduce a little bit the ProBio Platform trial which is the basis for this clinical study. And I will move forward to the aim of the study itself and discuss more on the results. And I will conclude a little bit with the take home message that we can take from these results.

So the ProBio trial is an adaptive trial for personalized metastatic prostate cancer treatment. I represented here the figure for the life circle of ProBio which is intended to be, as opposed to traditional clinical trials when you have just one specific research question in a specific time frame, this is intended to be a framework. So it's an infrastructure that is going to stay for a longer time where we can evaluate different research hypotheses simultaneously.

Basically, the idea is that as patients enroll in ProBio and consent to be part of the studies, we are going to take their blood and analyze it in order to characterize their genetic profile of the tumor which drives their disease. So this is a very non-invasive blood draw, often referred to as liquid biopsy.

And from the genetic analysis of the circulating tumor DNA, which I'll refer as ctDNA, we can randomize the patient either to the control group, which is physician standard of care—so the physician chooses the treatment using the information available, not the genetic biomarker—whereas the randomized experimental arm are instead separate arms which are driven by the genetic profiling.

And we continuously monitor the outcome data generated from the accumulating information from the patient. And we continually evaluate the therapies in the trial in order to decide if continuing to randomize new patients or to early stop because of evidence of superiority, which I refer here as graduation, as we have done in the first paper published in Nature Medicine last year, showing the superiority of androgen receptor pathway inhibitors—so abiraterone or enzalutamide—compared to the control group (physician choices alone or the taxanes). Or we can also stop because of futility when we see that there is no evidence of superiority but the opposite. And then, of course, this platform offers a very good framework for adapting and integrating new therapies and new biomarkers as old therapies leave the platform or new ones are integrated. I listed here two references, one of the initial design of the platform trial and one with the first publication of the disclosed data.

This specific clinical study is framed within ProBio. And we are going to focus on a subpopulation together with the first disclosed data which consists of patients with undetectable ctDNA. Basically, this is the subpopulation of patients where we were not able to detect a signal from the liquid biopsy. This often occurs in around 30% of the patients, and we are referring here to first-line metastatic castration-resistant prostate cancer patients.

These are, unfortunately, excluded from the trial because we cannot stratify them according to their genetic biomarkers and instead they are treated outside of ProBio with physician choice. Basically, again, the physician decides which treatment best suits the patient.

The patients, instead, with detectable ctDNA, then—we already mentioned—they are either randomized to the control, where they are treated similarly to the excluded patients by just the physician, or to one of the experimental arms that we have been disclosing, so the ARPI or the taxane. And we are not able to use the other information from the other experimental arms which are still under investigation.

So the specific aim of this study is then to evaluate how this negative outcome in terms of ProBio—not being able to be randomized—can actually translate to a positive outcome for the patient in terms of a better prognosis. It has been shown in different other studies.

But we are not only focusing on detectable versus undetectable; we also tried to assess the prognostic value of the specific ctDNA fraction level, which is measured on a continuous scale between 0 or more, in the range of undetectable to 100%. And again, we are going to do this by contrasting this set of patients against the patients in the control group randomized in ProBio, and then also into the experimental arms.

For the first part regarding the prognostic role of undetectable ctDNA, I'm presenting here modeled survival curves. In green, we have the curves for undetectable ctDNA, so patients who are excluded in the clinical trial. And I'm reporting two survival [INAUDIBLE], no longer clinical benefit, referred to as progression-free survival, and then overall survival.

And then we have the corresponding curves for patients in red with detectable ctDNA in the control group—so treated similarly to those with undetectable ctDNA—and then those in the experimental arm receiving a taxane or an ARPI experimental arm.

We have seen, like I summarized here on the right, that those with undetectable ctDNA have a 2.5 longer time to no longer clinical benefit compared to similar patients in the control, and a similar two-fold increase in the expected overall survival times. This translates to a median progression-free survival time of 21 months approximately, compared to 8.95 months in the control group for progression-free survival. Similarly for overall survival, we have a difference when comparing median overall survival times of 45 months versus 23 months in the control group.

Of course, this group of patients are really different in terms of various prognostic variables. When we take into account those variables in a multivariate adjusted model, the results remain stable, as presented also in the additional material of the paper.

Talking instead about the additional information of the quantitative value of the ctDNA fraction, then I'm presenting here the results of our dose response analysis where we flexibly model the association between increasing value of ctDNA fraction and longer or shorter survival time.

Even though we allow for a very flexible association, it turns out to be pretty linear. Here, I'm reporting different results for different analyses: the green where we don't adjust, the blue where we adjust only by the therapy received, and the fully adjusted in purple or red.

We can see that compared to a low value of ctDNA fraction but detectable—here we chose 2.5% as reference—we still see the observed increase in benefit in terms of no longer clinical benefit and overall survival, around a two-fold increase for progression-free survival and 1.5 increase for overall survival for patients with undetectable ctDNA. Whereas when ctDNA has been detected, we see that a 10% increase in the ctDNA fraction level translates to a 9% reduction in the observed survival endpoint.

I would like to conclude with a few take home messages which can be taken from these clinical studies where we have been evaluating the role of ctDNA as a prognostic tool. In doing this, we have demonstrated that indeed ctDNA fraction—both detectability and the actual value—can be used to predict the prognosis of metastatic castration-resistant prostate cancer patients.

This, of course, can be utilized in the framework of personalized care and clinical application by supporting more tailored treatment decisions. For example, suggesting less intensive treatment regimens or follow-up for patients with undetectable ctDNA, or suggesting ctDNA integration when deciding treatment strategies and possibly monitoring of the patient. Having said that, I would like also to thank you once more again, UroToday, for giving us the opportunity to present our last published results. Thank you.

Zachary Klaassen: Fantastic. Great presentation. I think just laying out the platform and the really exciting results from the ctDNA prognostication. We've seen, in the last couple of years, especially in urothelial cancer guiding adjuvant immunotherapy with IMvigor010, some really elegant work there. And so now, for the first time really, we're seeing in the mCRPC population from your guys' work the prognostication of ctDNA negative versus positive.

So let's take this to the clinic and just sitting with a patient, how are we able to explain what's coming down the road whether somebody has ctDNA positive versus negative? Certainly, it looks like if you're starting off your mCRPC journey with a ctDNA negative, your outcome is quite good, is that correct?

Alessio Crippa: Yeah, exactly. So this information, I think, is essential for deciding which is the best therapy for the patient, especially for prostate cancer where we know that intensive treatment regimens are also associated with severe side effects affecting the quality of life of the patient.

Giving this additional information I think is particularly relevant for both the clinician but also for the patient, for deciding how to better balance treatment efficacy and also quality of life. And I think we also tried to translate our results in terms of longer or shorter expected time, which is a measure that is usually easier for the patient to understand and to translate to their situation.

Zachary Klaassen: Yeah, absolutely. I think if we take this very practically, patients want to know, "How bad is my disease?" And at least with this test, we have an idea of giving them, hopefully, some example of: this is going to really need to be treatment intensified versus ctDNA negative in 30% of the population—this isn't quite as bad as what it could be. And I think that's going to be very, very helpful in terms of validation and everything else.

So the ProBio trial is just getting going. There's a lot of exciting stuff coming down the pipeline. Without spilling all of the secrets, maybe share some of the stuff we can look forward to from the ProBio trial over, let's say, the next year or two?

Alessio Crippa: Yeah, there's a lot going on. So in very different phases—later stages, early stages. I can anticipate that the next piece of work is going to provide more insights on the quality of life of the patient treated with ARPI and taxane. I think it's going to be a very novel piece of work, as this direct comparison for this patient population is actually missing, and it's actually a very fruitful comparison as we are going to analyze very different dimensions of quality of life and adverse events. So this is definitely something to look forward to.

Then we have, of course, other manuscripts in preparation. I can say something about the fact that we are going to present the first preliminary data on treatment with carboplatin, so platinum, in a subset of the initial patients. And of course, we are going to also present a lot of work that is being done on auxiliary research, which is also interesting. We have been talking about ctDNA fraction. We're going to see also if we can predict ctDNA fraction with the available information and how accurately we can predict it. But more importantly, if ctDNA fraction can predict treatment response over time. So I think there's a lot. And this is just the final stages. Of course, I presented the results only for the metastatic castration-resistant setting.

ProBio is open also in the early stage where we have hormone-sensitive patients. And there, we are actually implementing these findings about the positive prognostic for a patient with undetectable ctDNA. So instead of excluding them from ProBio, we are in the process of opening a new subarm where we are going to treat those patients only with darolutamide alone, removing the ADT treatment which has been the basis for the last so many years.

Zachary Klaassen: Sure, that's great. Great summary. We're excited to see additional work from ProBio. Congratulations on this absolutely beautiful work. I know you gave some great take home messages in your slides. Any last-minute thoughts for our listeners?

Alessio Crippa: Well, we are always open to hear any kind of feedback on ProBio. Something that really excites the readers or other reporters is that clinicians or patients would like us to address. And of course, it's a really complex study. But since we have this infrastructure, I think it's a unique environment. And it is a very international collaboration of patients, urologists, oncologists, statisticians, bioinformaticians. So I think it's a perfect framework for producing good knowledge.

Zachary Klaassen: Yeah, absolutely. Very well said. Alessio, thanks so much for joining us on UroToday. Great presentation. We look forward to having you on again soon. Thanks again for your time and expertise.

Alessio Crippa: Thank you.