JCO Precision Oncology Conversations

JCO PO author Dr. Jonathan D. Tward, M.D., Ph.D., FASTRO, at the HCI Genitourinary Cancers Center and the Huntsman Cancer Institute at the University of Utah, shares insights into his JCO PO article, “Using the Cell-Cycle Risk Score to Predict the Benefit of Androgen-Deprivation Therapy Added to Radiation Therapy in Patients With Newly Diagnosed Prostate Cancer.” Host Dr. Rafeh Naqash and Dr. Tward discuss how the cell-cycle risk score predicts the benefit of androgen-deprivation therapy in prostate cancer treatment.

TRANSCRIPT

Dr. Abdul Rafeh Naqash: Hello and welcome to JCO Precision Oncology Conversations, where we bring you engaging conversations with authors of clinically relevant and highly significant JCO PO articles. I'm your host, Dr. Rafeh Naqash, Assistant Professor at the OU Health Stephenson Cancer center. Today, we are excited to be joined by Dr. Jonathan Tward, Leader at the HCI Genitourinary Cancer Center, and Vincent P. and Janet Mancini Presidential Endowed Chair in Genitourinary malignancies at the Huntsman Cancer Institute at the University of Utah. Dr. Tward is also the lead author of the JCO Precision Oncology article titled “Using the Cell-Cycle Risk Score to Predict the Benefit of Androgen-Deprivation Therapy Added to Radiation Therapy in Patients With Newly Diagnosed Prostate Cancer.”

At the time of this recording, our guest's disclosures will be linked in the transcript.

Doctor Tward, welcome to the podcast and thank you for joining us today.

Dr. Jonathan Tward: Thank you so much, Dr. Naqash. I'm excited to share this important research with your audience.

Dr. Abdul Rafeh Naqash: Awesome. For the sake of simplicity, we'll refer to each other using our first names, if that's okay with you.

Dr. Jonathan Tward: That's great.

Dr. Abdul Rafeh Naqash: Okay. So, Jonathan, this complex but very interesting topic revolves around a lot of different subtopics as I understand it. There is genomics, there are implications for treatment, there is machine learning and computational data science research. So, to start off why you started this project or why you did this research, could you, for the sake of our audience, try to help us understand what androgen deprivation therapy is? When is it used in prostate cancer? When is it used in combination with radiation therapy? And that would probably give us a decent background of why you were trying to do what you actually did in this research.

Dr. Jonathan Tward: Yes, thank you very much. So, men who are diagnosed with localized prostate cancer, which is the majority of prostate cancer diagnosis, are faced with a lot of treatment decisions. And those decisions range all the way from, “Should I just go on active surveillance with the idea that it might be safe to treat later?” to “Should I consider surgery or radiation?” And then there's various forms of radiation. Now, as a radiation oncologist, one of the things that I have to consider when I meet a patient with localized prostate cancer who is pondering receiving radiation therapy, is whether or not we want to intensify treatment by doing more than just radiation alone. And androgen deprivation therapy, very specifically also thought of as chemical castration, what that really is is some kind of therapy where you are trying to reduce a man's testosterone levels to nearly zero. And the rationale for using androgen deprivation therapy in prostate cancer and in this case, specifically localized prostate cancer, is that one can think of testosterone almost as the food and growth signal for prostate cancer. There have been numerous prospective randomized trials that have been performed in the past that have clearly demonstrated that adding androgen deprivation therapy to certain contexts of patients with localized prostate cancer receiving radiation improves the outcome, including risk of metastasis and overall survival.

The problem is, we don't want to just intensify therapy for everybody who walks through our doors with localized prostate cancer. Some men have lower risk disease, and some men have higher risk disease. And conventionally, the way we make this decision is by looking at things like NCCN risk groups, which kind of lump patients into a few different boxes, generally speaking, called low risk, intermediate risk, and high risk. And if you think of those risk groups, the patients with the contemporary standard of who to add ADT to are men who are considered high risk localized, or men who are considered unfavorable intermediate risk localized. That being said, I think there's a recognition that we're overtreating some unfavorable intermediate risk men and undertreating them, and the same could be said of high-risk disease. So, I think we're always looking for better tools that make it a little bit more personalized, rather than lumping men into just one of several boxes.

Dr. Abdul Rafeh Naqash: Sure. And this sort of reminds me of the oncotype DX, in a way, trying to connect people with ER/PR, breast cancer, and where chemotherapy, plus anti-estrogen and progesterone therapy may be applicable. So, I think you were trying to do something similar in this research, and as far as I remember, please correct me if I'm wrong, this is knowledge that I remember from my board exams, we classify this high risk, intermediate risk, and low risk based on the Gleason score. Is that correct? Is that still true, or has this changed?

Dr. Jonathan Tward: It's still true. Conventional risk stratification, which is still used, literally only looks at a few parameters. You mentioned one, which is the Gleason score, which is really a human subjective judgment by a pathologist about how deranged cells look under a microscope. That's one parameter. The second parameter is the PSA value at the time of diagnosis. And the third parameter is the cT stage, which is really based on the digital rectal exam. Now, when you ponder that the entirety of our risk classification system is based on two subjective and one objective pieces of information, meaning what a Gleason score looks like, what the T stage is based on human interpretation, and then the only objective piece of data, PSA, it's rather rudimentary way of classifying men. I mean, it's done us well since the late ‘90s, when that particular classification system was derived. But it strikes me as odd that we should take all newly diagnosed localized prostate cancer patients and say you fit into one of three boxes, when we know there's so much more complexity to people and so many different treatment options and choices out there, which we're trying to match to the patient to ensure that we right size the treatment for them.

Dr. Abdul Rafeh Naqash: Understood. Now, as we go into the precision medicine component of this research, there's genomics research in metastatic cancers. But is there any genomics research in early-stage prostate cancer where there have been differences that have been identified between the intermediate low risk, high risk? Is that something that has been explored to date?

Dr. Jonathan Tward: Well, there are certainly somatic mutations that track with certain aggressive features. But I think when I think about the spirit of your question, within the localized prostate cancer space, there's been several molecular signatures that have been developed and, in fact, been commercialized that have been shown quite clearly that if you have a certain array of gene expressions, let's say, that that can correlate with metastasis or risk of recurrence or death. And the work that we're talking about today is one that actually uses one of the commercially available biomarkers, commercially it's known as Prolaris. But very specifically, in the work that I think we're discussing today, what we're looking at is cell cycle progression genes. And these are genes that maybe, to simplify it, are sort of hallmarks of how quickly cells are turning over. And what's interesting about looking at cell cycle progression is it's not certainly particular to prostate cancer. I mean, you could make an argument that cell cycle progression genes are probably relevant measures in any cancers, but there's been much work done over the past 15 to 20 years that have clearly validated that this particular cell cycle progression gene signature, which is now commercially available, clearly correlates with risk of progression, risk of metastasis in localized prostate cancer patients, whether they're receiving surgery or radiation. But what we've done is we've built upon this molecular work and added clinical risk features and added results of prospective randomized trials to use this test to personalize the precise risk reduction of what would happen to a man who is pondering adding ADT to radiation therapy. So, it's a very powerful precision tool.

Dr. Abdul Rafeh Naqash: Sounds very interesting. When you go deeper into this platform, is this genomic testing platform, does it incorporate RNA transcriptome or is it DNA, or is it a composite of both?

Dr. Jonathan Tward: There are various molecular tests that are out there. In this particular case, these are mRNA expression levels of cell cycle progression genes, and they are kind of calibrated against some normal housekeeping genes, which is how the test is run.

Dr. Abdul Rafeh Naqash: Understood. So, from what I understand in the discussion, you very appropriately said, in fact in your first paragraph, the goal here is to match patient level precision medicine approaches and reconcile them with population level therapeutic options. It's a very catchy statement. Can you help explain for our audience how you tried to do that? And this goes back to the question that you were trying to understand, where to use combination therapy in a localized prostate cancer based on risk stratification and deriving that risk stratification from the cell cycle score and then arriving to certain thresholds. So could you go through that in simple terms to help us understand how you tried to do it and what was the outcome and what are the implications of that?

Dr. Jonathan Tward: Sure, there's a lot to unpack there, but I'll do my best to simplify it. So, we'll start with the basic question that faces a patient and their radiation oncologist, which is, if they're going to receive radiation, should you add hormone therapy? And if hormone therapy was completely nontoxic, you'd say, “Sure, just add it to everybody if there's a benefit.” But the problem is, of course, hormone therapy is associated with all kinds of unpleasant side effects and additional risks, so we don't want to utilize it unless we're sure that the benefit is clear. When you think about the way most of oncology decides whether or not adding an intervention should be done in a particular patient context, it's actually been derived originally from prospectively randomized trials, which usually assigned a hazard ratio or some kind of known relative reduction to doing ‘thing B’ versus ‘thing A’ or ‘thing B’ in addition to ‘thing A’.

But what's curious about always looking at hazard ratios and saying that those are the reasons why you should do additional things, discounts a really important fact, which is the baseline risk of something bad happening has to be accounted for first before you decide whether or not it a relative risk reduction matters. So to state more clearly, if I knew a prostate cancer patient sitting in front of me only had a 2% risk of developing metastasis within 10 years, if I just did radiation alone, if I then say adding hormone therapy might cut that in half from 2% to 1%, a patient might say, “You know what? I'm not sure I want to accept the toxicity of many months of hormone therapy to cut my risk of metastasis from 2% to 1%.” But if you had a patient where that risk was 20% risk of metastasis with radiation alone, and you told them I can cut that risk down to 10% or 12%, then that's something they would seriously consider.

And so what this work really does is precisely that. It gives us a tool where, using the molecular signature of the cell cycle progression genes, which afford a patient a certain risk of metastasis, and also taking into account clinical risk factors that we know are prognostic, Gleason score, PSA, their age, how many cores of the biopsy were possible. We use all this information, and I'll use a strange term, multiplex it into a robust risk model that will prognosticate extremely clearly what that patient's precise risk of metastasis will be within the next 10 years, and this is the key point, if they receive radiation alone.

So, think of this work in two phases. Phase one is calibrate the risk in a patient if they get radiation alone, by using both molecular and clinical prognosticators. But then take the power of numerous randomized trials, which have clearly set the hazard ratio reduction for adding the hormone therapy, and then using mathematical principles, applying that hazard ratio risk reduction to the absolute risk. And then what you ultimately do is, at a very individual level, have a patient sitting in front of you where you can say, “Mr. Jones, I've run this test on you, and I can tell you definitively that if you receive radiation therapy for your localized prostate cancer, the risk of metastasis will be 12%. But if you add, let's say, six months of hormone therapy, that could be reduced to 7%, and the absolute risk reduction might only be 5%.” And if you think about that number in a number needed to treat mentality, then you could say, “Listen, I have to give 20 men identical to you, hormone therapy for one to benefit. Is that worth it to you?”

And what it really does is it empowers the patient. Rather than following a guideline that says, “Effectively, thou shalt do this for this risk group,” you really want to engage the patient in the conversation about the risk benefit of what you're going to do. And I think it's uncommon in oncology for physicians to be able to very precisely tell a patient sitting in front of them, if you do ‘thing A’, this is the risk, something bad happen. If you do ‘thing B’, this is how the risk reduces. And I think now we really get into shared decision making, rather than a, “Trust me, I'm a doctor,” paternalistic situation.

Dr. Abdul Rafeh Naqash: That's a very interesting approach. Again, you're basically personalizing the personalized medicine approach, refining it further, and involving the patient in discussions, which helps them understand why something would make sense. And some of this, as you might already know, people have tried to do in some other tumor types, hasn't necessarily led to significant clinical decision-making changes. But I think the way the field is evolving, especially this research that you published on and others are working towards, will hopefully result in more personalized approaches for individual decision making for these patients.

Now, I do understand that simplicity sometimes results in more uptake of some information versus when sometimes things get more complex. So, in your assessment, when you came up with these results, you looked at the genomic score, you took the randomized clinical trial data, you did the absolute risk reduction. From what I understood in the manuscript, it does look like you did come up with a threshold of what would appropriately risk stratify individuals, meaning individuals that are at a higher risk if they cross that threshold, versus individuals that are at a lower risk if they cross that threshold. Is that a fair statement or is this a continuum? So there is no binary, but this is over a scale that this assessment can be made.

Dr. Jonathan Tward: So, there are elements of your summary that are fair, but this is a continuum which allows any individual to accept whatever risk reduction they want. That being said, there is no standard in oncology for what percent risk should you intensify a treatment for? And when you poll physicians and doctors as to how much reduction in death or how much reduction in metastasis, doctors and patients are all over the map at what they consider to be a threshold. But we designed these thresholds actually from prior work, based on surveying both patients themselves, as well as experts who were on cooperative trial group steering committees, and ask them, essentially, “At what level of risk reduction would you want to intensify treatment?” And what's interesting is most people who are asked that question are willing to do more treatment intensity for an important outcome like metastasis if the absolute risk reduction of that event happening is 5%. So as a general principle, that's how it was set.

These thresholds in the current paper we're discussing actually weren't defined in this current work. They were defined in prior works, where we had clearly shown in retrospective data sets that they could discriminate very well who does or doesn't benefit from hormone therapy. What's, I think, novel about this paper, even though we had previously validated those thresholds, is that now that we're using the randomized trial data, it's extremely robust in our risk estimates, and we can say that it's truly a predictive biomarker. Because it's one thing to prognosticate an outcome, but predict a difference in treatment A versus treatment B usually requires randomized trial data so that you get the highest level of evidence and the confidence that it works.

Dr. Abdul Rafeh Naqash: So the next steps for this very, very provocative research, is it something prospective validation or are you going to try to utilize maybe proper group trial data or other pharma trial data, individual patient data to risk stratify these individuals and validate?

Dr. Jonathan Tward: So these thresholds, for example, that you refer to are very well validated. There's multiple prior studies, well over at this point, 1500 patients where there's validation. And yes, we have reached out to cooperative groups to do some additional validation. That being said, this work is already ready for prime time and being used. In fact, this test is the commercially available Prolaris test. The results gleaned from this work are published on the score report that a patient and a physician receives. So the reality is that this is already existing as a clinical tool in the community. And the NCCN guidelines also support the use of this and other tests to move from a stratification to personalized medicine. So it's not like this is so much in the experimental realm as it is effectively a complete tool that is being used today. And effectively, it's available for any patient or physician diagnosed with localized prostate cancer to immediately order on biopsy tissue.

Dr. Abdul Rafeh Naqash: One naive question, Jonathan, I wanted to ask is most prostate cancers tend to be prostatic adenocarcinoma. So if it's a neuroendocrine localized prostate cancer, does the same risk assessment apply? Because neuroendocrine tumors in general seem to be higher replication stress or higher tendency to metastasis. Does it change from your perspective, from the genomic assessment standpoint, the CCR score standpoint?

Dr. Jonathan Tward: That's a very interesting question, because what I will tell you is that there are probably a lot of, well, I wouldn't say a lot, but there are some neuroendocrine cancers mixed in with the adenocarcinomas that no one identified as neuroendocrine, which in a way were baked into the cake of the risk signature. Even though that is so, I dont think we’ve independently looked very specifically at known neuroendocrine cancers and compared them to the adenocarcinomas. What I would actually argue though, is that if you have a neuroendocrine cancer sitting in front of you, the point about whether or not you're adding ADT is relatively moot because neuroendocrine cancers may or may not respond to ADT, and you have to start considering chemotherapeutic-like decisions.

So the question, which is very interesting and academic, is that I would presume the cell cycle progression score should be elevated, although I don't know that in a neuroendocrine cancer, this tool doesn't appear to be useful at this moment for neuroendocrine cancers because we're not making decisions about chemo. That's an interesting and provocative question, and now you make me want to study that. So potentially, the next paper would be neuroendocrine cancers, whether or not it might prognosticate using a topicide or something else like this. But we would have to rely on prospective trial data as well to see whether or not we could use it the same way.

Dr. Abdul Rafeh Naqash: Hopefully, if you do work on it, then you can submit the manuscript again to JCO PO for us to talk again.

Dr. Jonathan Tward: Yeah, and you'll be on the author bar.

Dr. Abdul Rafeh Naqash: Appreciate the inclusion. So thank you so much, Jonathan, for talking to us about the science. And a few quick minutes about yourself. Can you tell us a little bit about your career trajectory, how you ended up doing what you're doing, and maybe some lessons learned and some advice for early career junior investigators that would be helpful for them?

Dr. Jonathan Tward: Yes, that's a happy memory. When I was a young undergrad, I was fortunate to do some volunteer work in a radiation oncology department and had mentors there who guided me into considering a career in medicine and specifically a career as a physician scientist. So I'll start with the best advice is to get mentors early on and throughout your career who are really interested in your career development and who are accomplished that can kind of help you along. But I went to medical school with an open mind and continued to love oncology. I think it has some of the most complex questions that are unanswered. It is very high stakes oncology. There's still a lot of death and disability and consequences of our therapies. And I just love the idea of working in an environment, both clinically and as a researcher, to try to solve some of those questions like, how do I improve outcomes? How do I make therapy less toxic?

And radiation oncology for me, was a nice fit in genitourinary cancer, I guess, specifically because mid GU cancer realm patients are presented with a menu of treatment options. It's kind of interesting. It's a little bit unlike other cancers. But I had fantastic mentors throughout both my medical school as well as residency program who really helped guide me and encourage me along the way. And so without spending too much time, I would say go out of your way to find people who are successful at what they do, are interested in making you better, and really sit at their knee and listen to them when they are trying to guide you because they really have your best interests in mind. And I think as a mentor and a mentee, what makes me most proud is watching people I've trained go out and succeed. I mean, the reward of being a mentor is watching your mentees succeed.

Dr. Abdul Rafeh Naqash: Thank you. Appreciate all those words of wisdom, Jonathan, and excited to see all the subsequent steps and results from the research that you're doing. Thank you again for joining us today and providing a very simple summary of a very complex topic which I think our audience and perhaps some of the trainees listening to this podcast will appreciate. We really appreciate your time.

Dr. Jonathan Tward: Thank you so much, Rafeh.

Dr. Abdul Rafeh Naqash: And thank you for listening to JCO Precision Oncology Conversations. Don't forget to give us a rating or review and be sure to subscribe so you never miss an episode. You can find all ASCO shows at asco.org/podcast.

The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions.

Guests on this podcast express their own opinions, experience, and conclusions. Guest statements on the podcast do not express the opinions of ASCO. The mention of any product, service, organization, activity or therapy should not be construed as an ASCO endorsement.

Dr. Tward Diclosures:

Honoraria
Company name: Bayer

Consulting or Advisory Role
Company name: Myriad Genetics, Blue Earth Diagnostics, Janssen Scientific Affairs, Merck, Bayer, Boston Scientific, Myovant Sciences, Myriad Genetics, Lantheus Medical Imaging

Research Funding
Company name: Bayer, Myriad Genetics

Travel, Accommodations, Expenses
Company name: Myriad Genetics, Bayer