Comparing Cost-Effectiveness of Surgery, Radiation, and Chemo for Seminoma - Daniel Joyce

January 12, 2024

Ruchika Talwar interviews Daniel Joyce about his journey into cost-effectiveness studies, particularly in testicular cancer. Initially skeptical, Dr. Joyce now sees these studies as crucial for interpreting clinical trial data in real-world contexts. He discusses his analysis of the SEMS trial, focusing on small-volume stage two seminoma, where RPLND emerged as the most cost-effective option compared to radiation or chemotherapy. The study's sensitivity analyses reveal the importance of considering patient values, like fertility, and the need for better long-term toxicity data. Dr. Joyce advocates for incorporating cost-effectiveness data into clinical guidelines, emphasizing the balance of cost and quality in healthcare decisions. He calls for more high-quality cost-effectiveness analyses and greater clinician engagement with these studies to inform practice and guideline care.


Daniel Joyce, MD, Urologic Oncology Fellow, Vanderbilt University Medical Center, Nashville, TN

Ruchika Talwar, MD, Urologic Oncology Fellow, Department of Urology, Vanderbilt University Medical Center, Nashville, TN

Read the Full Video Transcript

Ruchika Talwar: Hi everyone, and welcome back to UroToday's Health Policy Center of Excellence. Today, I'm really excited to be joined by my colleague, Dr. Dan Joyce, who's an assistant professor and a urologic oncologist at Vanderbilt University Medical Center. He's a real expert in the space of cost-effectiveness studies and will be discussing a recent analysis that he did with the data from the SEMS trial exploring strategies in the management of seminoma. Thanks, Dr. Joyce, for being here with us today.

Daniel Joyce: Thanks so much for having me, Ruchi. It's a pleasure.

Ruchika Talwar: Dr. Joyce, you have been exploring cost-effectiveness studies in a variety of urologic oncology disease processes, so tell us a little bit about what got you interested in looking at this specific testicular cancer or seminoma space?

Daniel Joyce: Yeah, absolutely. I actually started out looking at cost-effectiveness analyses because I thought they were kind of worthless, truthfully. I would often read these papers, not really know much about them, except that somebody came up with some arbitrary rules, pumped them into a model, and then came to the conclusion that one treatment was of higher value than another. So, I totally get why people might not put a lot of stock in their outcomes. But, through the process of learning how to do them, what I came to find was that it's actually not the primary outcome that means the most in these papers; it's a way of viewing high-quality data, like clinical trials, in a way that contextualizes them, beyond the clinical trial patient, who we all know is not representative of real-world patients.

And so, you can take a lot of the data we find from clinical trials and basically extrapolate findings from them, in order to understand how to treat our patients better, in a more high-value way, and also in a way that helps us view how we should be good stewards of our healthcare resources in general, but also contextualizing that for our patients and providing the best value for them individually, as well. So, I actually started this process with Dr. Borgen, who's been an incredible mentor to me during my fellowship at Mayo, and we first started looking at the VESPER trial and neoadjuvant chemotherapy. Through that process, I really started to see that, okay, there are these clinical trials out here that are maybe changing practice, maybe not changing practice, they're a little controversial. Do we do... They're not perfect. VESPER had six cycles of chemotherapy. We don't really give that. How do we interpret that data? How do we then apply that? Should everybody be getting DDM back, and it sort of helped me see how CEAs can help you understand where the value is in those trials and how to translate it into practice.

And so, SEMS, when it came out, was really the perfect setting for that, because it was the same kind of thing. You have a treatment that was showing a 22% recurrence, when we know we have chemotherapy that cuts that down to less than 1%. So, why not do the more effective treatment, and what does it mean to really pursue surgery upfront in these stage two seminoma patients? Really, the catch is that we're trying to avoid long-term toxicity. But, as you know, in SEMS, there were no long-term toxicity data reported. So, what do you do with that? Is it really saving these patients by pursuing surgery upfront, and through our CEA we're able to show that, yes, indeed. When we look at it, we take published long-term toxicity data and plug that into the model using the SEMS trial oncology outcomes, we see that over time, the impact of those long-term toxicities makes RPLND a very worthwhile endeavor for people who are young and going to live a long life.

Ruchika Talwar: Yeah, absolutely. I think a lot of times, especially in these disease processes, where we have a long period of survivorship after what we consider to be hopefully a cure or at least a period of disease latency, we really have to start considering things outside of even the quality of life space, what makes sense for both these patients from a financial toxicity perspective, but also from our health system as a whole? Because we all pay into this big model of healthcare in one way or another. So, I agree with you. I think these sorts of analyses are really meaningful, especially if they go ahead and change the way we think about these really groundbreaking trial results. So, let's dive into your results a little more. Tell me about what specifically you found.

Daniel Joyce: So, the overarching message was that RPLND was the most cost-effective option compared to radiation or chemotherapy in small volume stage two seminoma. That, to me, is less interesting personally, than the sensitivity analyses, and that's really where the meat of those studies comes from. What that allows you to do is, you can imagine you have a model that you make a bunch of assumptions to put these probabilities and utilities into, and those assumptions are, or can be flawed, because we as clinicians are making those assumptions based on the best data we have, and they're not perfect. A sensitivity analysis allows us to really test the robustness of those assumptions and also helps us extrapolate a little bit from our base case scenario into other patient scenarios. I'll give you an example of what I mean.

So, in a one-way sensitivity analysis, you can vary one of the variables that you put into your model and see how varying that variable changes your outcome, and what we found were some interesting themes. So one, infertility was highly influential on the model, so the utility that you place, so the quality of life that is equated with having infertility, and the probability of that infertility, impacted whether RPLND was really cost-effective. What does that mean? Well, it means we need to understand our patient's values with fertility before we recommend RPLND. But also, if we can spare the nerves, we know we do really well in preserving integrated ejaculation, and we know that fertility is actually preserved in as many as 75% of patients desiring children. So, we can, one, it says, okay, in the right patient RPLND is still worthwhile, and also, maybe we should really prioritize in these young patients, nerve sparing, and we could do it. In low volume seminoma, that's often possible.

Another thing we found was that progression after RPLND didn't have to increase very much for RPLND no longer to be cost-effective. So in SEMS, 22%, anything 23% or higher in our model made radiation the preferred option as opposed to RPLND. And so, that is a narrow margin, but there are a couple of things about that. One, there were, I think, 17 recurrences in the SEMS trial, of those, many had a modified template, and of those that recurred, many were in the retroperitoneum. So, we could potentially avoid roughly around 42% of the recurrences that happened in SEMS, if we did a bilateral template and not a modified template. Now, that's obviously some extrapolation and some assumption that may not be true, but it helps us think a little bit more critically about the way we're doing these surgeries and how we can continue to improve value in an option that we think is worthwhile upfront in seminoma.

And then, the final thing was really identifying a knowledge gap, and that's that a lot of these long-term toxicity data that we have aren't great, and they're not great because you need a long period of time to characterize long-term toxicities, and by the time you've characterized those toxicities, well, you're analyzing treatments that are antiquated. And so, in radiation for instance, much of the data we have comes from radiation templates that involve the mediastinum. One of the factors we put into our model was cardiovascular toxicity, and that mattered in our model. The degree to which cardiovascular toxicity patients experience can change the outcome of what's the highest value option. So, you could say that perhaps there's still a very real role for radiation as well, in patients that don't want to undergo surgery, especially if we assume that the cardiovascular toxicity is lower than what it was for those templates, we don't do that anymore. But again, an important area where we need better data, we need better research to understand the long-term toxicity data of those radiation templates, now that we're using different regimens.

Ruchika Talwar: I think there's a lot of nuance here, that people don't necessarily appreciate when interpreting cost-effectiveness studies, what goes into the model? What are your levers that you can adjust, for example, on specific sensitivity analysis and follow-up? So, great point, and I think that it's something we should all make a more concerted effort to do when we're interpreting studies like yours. Now, with all of this really meaningful data, in your opinion, what's the next step to get this to actually change practice and be incorporated into ways that will affect urologists who treat testicular cancer patients?

Daniel Joyce: Yeah, really good question. It's maybe a charged answer, but I don't think it needs to be. I think we need to start thinking about the value of the interventions when we're creating our guideline statements. It's something that the NCCN has already started to do with their evidence blocks and including an affordability category in that, I think it's done imperfectly, personally. I think the EAU guidelines would love to see some buy-in, at least including some of the data surrounding cost-effectiveness when making those recommendations because... And, what I don't mean, and I think this can often be misconstrued, is that I'm not saying we should make our guideline statement recommendations based on a CEA outcome. That is not what I'm saying at all.

But, what I am saying is that there is this data that shows that in these patients, this might be a higher value intervention than these other treatments, and to at least consider that and discuss that with your patients, so that they can, one, understand the value of what they're getting. But, also allows us as clinicians to think about the healthcare system, as a whole, as we're recommending these treatments, because I think we do have a responsibility, as clinicians, to be judicious with the treatments we recommend, and healthcare costs continue to rise and rise over time, and we make little impact. The only way we're really going to make an impact, I think, is if we as clinicians start thinking about it and changing our practice, in a way that's efficient and effective.

Ruchika Talwar: Absolutely. I just want to remind our audience that value incorporates both cost and quality, because I know in a lot of scenarios people will bring up as a criticism for these sorts of analyses, that we're so focused on cost, we're not thinking about the care. Every patient situation is a little different, everyone's priorities are different, but we are incorporating that in our assessment of value. You alluded to it earlier, depending on what the patient's treatment priorities are, it actually changed the cost-effectiveness, in some ways. So, just want to point that out for our audience because I know it is certainly a point that's brought up often, as a weakness here.

Daniel Joyce: From people who do this research like me, what we need more of is the outlook from a patient perspective. We're really lacking in the evaluation of things like patient out-of-pocket costs and more of a societal perspective, as opposed to just the healthcare system's perspective. And so, we need more efforts to better describe indirect costs for patients, travel time, parking fees, all of these things we need to better understand, in order to inform our cost-effectiveness analyses, so they can be higher quality and give better recommendations.

Ruchika Talwar: Yes, for sure. Also, things like caregiver burden, time away from work, all of those are really important aspects. So, as we wrap up, just tell me what do you think is next in this space? How can we continue to move the needle here?

Daniel Joyce: I think we need more high-quality CEAs, honestly, and also I hope more clinicians don't write them off as useless and actually pay attention to the details that can help inform their practice. I think when we do clinical trials, cost is never an aspect of that, and also, a lot of the downstream effects are not a part of that. Also, those patients are not representative of the patients we see in our clinic. CEA, I think, is a really unique way to then contextualize those studies. So, I think just more effort, more effort from us, more effort from me and people like me doing this kind of research, and then just maybe more attention brought to those types of studies, to help inform guideline care.

Ruchika Talwar: Well, we are so thankful that you are here with us today highlighting the importance of this work and cost-effectiveness studies, in general. Thanks for taking the time.

Daniel Joyce: Thanks so much for having me, Ruchika. I really appreciate it.

Ruchika Talwar: To our audience, we appreciate you joining us, and we'll see you next time.