Read the Full Video Transcript

Zachary Klaassen: My name is Zach Klaassen. I’m a Urologic Oncologist at the Georgia Cancer Center in Augusta, Georgia. I’m delighted to be joined here today by Dr. Tim Daskivich, who is a urologic oncologist at Cedars-Sinai Hospital in Los Angeles, California.Today, we’re going to be discussing Dr. Daskivich, his recent publication in JAMA Internal Medicine, overtreatment of prostate cancer among men with limited longevity in the active surveillance era. Tim, thanks very much for joining us on here today.

Timothy Daskivich: My pleasure to be here. And thanks for the kind introduction. I’ve been working on this topic for 10 years now, maybe even more than 10 years, and tried to make the case for why men with limited life expectancy are at risk for overtreatment. We’ve done this in a number of ways.

The principal reason why they’re at risk for overtreatment is that these men are unlikely to live long enough to benefit from treatment. So we know from randomized trial evidence that benefits of prostatectomy and radiation therapy are delayed for about 10 years after the time of treatment. So if you die before then from another cause, you really haven’t benefited from it. Worse, if you do get treated, men with limited longevity have worse quality of life. And so we also know that from registry data. We also showed that treatment is actually less effective for these men. As you might expect, if many of them are dying before a cancer mortality may occur, the treatment is overall less effective as you get older and sicker. And this all leads to aggressive local treatment being poorly cost-effective for these men.

We actually published that with men with less than 10-year life expectancies, even for intermediate risk disease, active surveillance is a superior option in terms of cost-effectiveness. So for these reasons, life expectancy has been ensconced as the first triage point in guidelines. So these are NCCN guidelines from 2024.

And prostate cancer is unique in that it’s the only solid malignancy, to my knowledge, that life expectancy is hardwired into guidelines as the first triage point. And so for all risk subtypes, life expectancy cutoffs are given. However, what we do know is that men with limited longevity have been historically overtreated even for low and intermediate risk disease, which is unlikely to harm them over the short term.

So this is data from SEER-Medicare that we published in 2014. And this is the rate of aggressive local treatment broken down by age and comorbidity status. And the groups highlighted in yellow are those with less than 10-year life expectancy.

And what you can see is that the probability of aggressive treatment of these men in the early 2000 era, in aggregate, was over 50%. So over half of them were being treated with surgery and radiation, despite a less than 10-year longevity, which is obviously against guidelines. But in comes active surveillance, and this is thought to be going to help with this situation. It certainly has helped with overtreatment of men with low-risk prostate cancers.

And we see from data from the AQUA Registry that rates of active surveillance for low-risk prostate cancer have increased 35% from 2014 to 2021, and these rates are continuing to go up. But many people believe that the emergence of active surveillance has solved the problem of overtreatment for men with limited longevity. But this is still an open question because there are two separate problems related to overtreatment.

So we sought in this study to determine if overtreatment of men with limited life expectancy has persisted in the active surveillance era. And also, if it does exist, to determine if it varied by tumor risk and treatment type. So in this study, it was an observational study of 243,000 men in the VA, a national VA sample with localized prostate cancer.

And principally, we were looking at trends in rates of definitive treatment defined as radical prostatectomy or radiation therapy over time across life expectancy subgroups. Additionally, we wanted to know what types of treatment were used for men who were overtreated.

We used a multivariable linear and log-linear Poisson regression model. And our primary predictor here was life expectancy, as estimated by the validated age-adjusted Prostate Cancer Comorbidity Index, which we validated across 181,000 men in the VA. And you can see how overall mortality pans out right by the PCCI—robust, validated scale.

So these are the sample characteristics. And I think the important things to point out here is it’s a typical prostate cancer population, mean age is 67, had adequate representation across all risk categories. About half were treated with surgery or radiation and half were treated with conservative management. And I think probably the most interesting statistic is at the lower right, which is their life expectancy breakdown.

And according to the PCCI, roughly 20% of men, 21% of men, had life expectancies of less than 10 years, and 4.7% of men had life expectancies less than five years. So men at risk for overtreatment are not a small proportion. It’s about 1 in 5 men diagnosed with prostate cancer.

So when we look at trends in definitive treatment by life expectancy, you can see that there’s been significant reductions in treatment for low-risk disease across all life expectancy categories. And this is what we’d expect. Fifty percent reduction in men with greater than 10-year life expectancies and less, but appreciable, reductions in men with less than 10- or 5-year life expectancy for low-risk disease. And of course, this is due to active surveillance.

But what we were surprised to find is that rates of aggressive local treatment for men with intermediate risk disease first are increasing in men with limited longevity. So you can see here for both favorable and unfavorable intermediate risk, men with greater than 10-year life expectancies—on the leftmost column—the aggressive treatment rates are flat, basically the same over the 19 years of observation.

But looking at men with less than 10-year life expectancies, and even in men with less than five-year life expectancies, we see significant increases over time. And I would say those increases range from 19% to 59%, you can see it right here. And I would also note that all of these men with less than 10- or less than five-year life expectancies are recommended to consider conservative management or outright recommended against aggressive treatment by guidelines.

What we also were surprised to find is that for high-risk disease, there was also an increase in risk of potentially overtreated populations. Now, for high-risk disease, we have a higher bar for potential overtreatment because it’s potentially lethal within even 5 to 10 years. So really, the group that is at risk for overtreatment here is the men with less than five-year life expectancies.

But despite that, there was a 29% increase in treatment of these men over the time horizon of the study, whereas men with long longevities, it’s basically flat again. Another point of interest in this study is we parsed out the types of treatments that were used in men who were overtreated. I won’t go through the figures here, but the high points are that the vast majority of overtreatment that’s happening in men with limited longevity is by radiotherapy. And this makes a lot of sense because most of these men aren’t great operative candidates.

So among men with less than 10-year life expectancies who received a definitive treatment for low or intermediate risk disease, it was 78% receiving radiotherapy. And among those overtreated for less than five-year life expectancies with high-risk disease, it was 85% receiving radiotherapy. So we, as urologists, are not exonerated from this, even though it’s not us doing the treatment, because we refer these patients to radiation oncology.

So in summary, I think that these data clearly show that overtreatment of men with limited longevity for intermediate and high-risk disease has gone up in the active surveillance era, which is surprising. We thought it might be stable, but it’s actually going up. And over three-quarters of these men were actually treated with radiation therapy.

And I think that the take-home point here is that we need to think hard about how to integrate life expectancy into our decision-making paradigms, and the way we communicate to patients about the risks and rewards of therapy. And ultimately, the goal of that would be to help them make better decisions about treatment and avoid unnecessary toxicity. Thanks again for the invitation, and I look forward to our discussion.

Zachary Klaassen: So, Tim, phenomenal presentation, very important data. There’s a lot to unpack from your study, but a couple of things that jumped out to me that I think we should dive into—particularly among the men with life expectancy less than five years. Proportion receiving definitive therapy for high-risk disease increased by 30%. So who’s driving this? Is this the patients? Is this the physician? What are your thoughts on that specific statistic alone?

Timothy Daskivich: It’s hard to say what’s driving this. There are probably patient-related factors and physician-related factors. But I do think a lot of this is physician-driven in terms of how you present the information to the patient.

We’ve done a lot of studies on communication, how this information is communicated to patients. And often, doctors are nervous to talk about life expectancy with patients. They think they don’t want to hear it. But we’ve actually interviewed patients, and 88% of them say they do want to hear the information.

So I think one part of it is doctors need to give these statistics to patients. And there are particular ways that you can do it that would help make patients more confident in it, feel more comfortable with it, be less anxious about it. So I think that there are ways that physicians can address this.

I also think that it’s partly driven by the new paradigm for treatment for radical prostatectomy and treatment of local disease. I mean, we’re treating a whole lot less low-risk disease, and I personally am treating a lot less favorable intermediate risk disease in favor of either surveillance or ablation therapy.

So all of our treatment—definitive local treatment—is veering towards high-risk. And I think we tend to focus on the cancer more than we’re focusing on the other factors. I think it’s also true that it’s hard to incorporate life expectancy in some ways because there is some variability in these estimates. They’re not perfect, but I would argue that in the obvious situations, like a less than five-year life expectancy for favorable intermediate risk disease, we really should be treating fewer of those patients for sure.

Zachary Klaassen: Yeah, absolutely. I think the other thing that’s interesting is, as you pointed out in one of your slides, the increase is based on radiotherapy for the most part. So we know from 2000 to 2019, certainly radiotherapy has improved—conventional hypofractionation, we’re using spacers in some cases—there’s been a lot of improvements in how we deliver radiotherapy. What are your thoughts in terms of why we’re treating more? Is it because of those factors, or there may be other factors as well?

Timothy Daskivich: I doubt that it’s related to improvements in these therapies, though I think as doctors get more and more confident in their therapies, they’re more likely to give them and recommend them because they’re more confident it’s going to harm patients less. But I would say that the average experience of a patient after any of these therapies is going to be worse than doing no treatment at all if treatment is unnecessary.

My thought is that it’s probably, again, more driven by a lot more focus on high-risk disease. Even the treatments that we give downstream of our local treatments, we’re just getting more aggressive with all forms of therapy.

And I think, I personally see in our tumor boards and treatment of advanced patients, patients who are in their 80s and even sometimes in their early 90s, we’re treating aggressively with multiple hormonal regimens and a variety of other treatments. And I think that translates to our localized setting, too, where we’re getting just more aggressive.

Zachary Klaassen: Yeah, absolutely. You mentioned the discussion in your paper—a really important point I want you to highlight—talking about the Trifecta method for counseling patients and really how to convey competing risks of mortality. So maybe for our listeners, just break down exactly what the Trifecta method is.

Timothy Daskivich: If you will indulge me, I’m going to share my screen again. I put together a slide for this because it’s not intuitive, but I think it’s an important concept. So the Trifecta method was developed after doing interviews with patients and also observing how physicians communicate competing risks in real life. And we categorized how they do it typically and tried to grasp the best practice. And we observed this actually in practice.

So the Trifecta involves three things. It involves quoting the life expectancy as a number of years because that’s how patients want to hear it. The majority of them prefer to hear it in that form because it is easy to understand, and it relates to guideline cutoffs.

Two, giving the cancer prognosis at the time of the patient’s projected life expectancy because it’s a time-dependent risk.

And three, giving the cancer prognosis with and without treatment. And the reason why this is important is that the typical way that doctors present information on cancer prognosis—one of the principal things we describe in a visit—is to give it at 10 years. That’s about a third of doctors who present the data in this way if they’re going to quantify it.

And why do we present it that way? It’s because that’s how the clinical trials all read out. So for an average 65-year-old man, you might say, “Look, the likelihood you’re going to die of cancer without treatment is 18% at 10 years, and if you get treatment, it goes down to 7%.” And that would be a pretty good way of describing it.

But if you actually projected those statistics out to the patient’s life expectancy for a 65-year-old—meaning 18 years—the risk of the cancer harming them without treatment doubles, and the treatment benefit, meaning the delta between with and without treatment, doubles. So it’s relevant to actually give them the data at their life expectancy because it is a time-dependent risk, and it really captures the risk of the cancer over their lifetime.

Now, if you think about this, if you use the Trifecta method, you don’t even necessarily need to mention their life expectancy. You can say, “Look, over your projected life expectancy, here’s the likelihood of dying of cancer with and without treatment.” You’ve done it without even mentioning numbers about life expectancy to them, and you’ve engaged their cancer risk at that time point. And that would incorporate life expectancy in a passive way.

Zachary Klaassen: Great explanation. I think that’s really well done. I always enjoy the conversation—great work. I know you had a great summary slide already, but maybe a couple of quick recap points for our listeners.

Timothy Daskivich: Yeah, absolutely. So I think that our data show that in the VA—which has been a leader in reducing overtreatment of men for low-risk cancer—despite that, the overtreatment of men with limited longevity has increased for men with intermediate and high-risk cancer, despite guidelines recommending otherwise.

Typically, that overtreatment is done with radiation therapy and less with surgery, and that proportion is about three-quarters to one-quarter. And I think the big take-home here is let’s try and incorporate life expectancy into our decision-making, how we communicate to patients, and have upfront conversations where we communicate it using the Trifecta method—meaning project the cancer prognosis at their life expectancy with and without treatment. And in that way, we can really reduce unnecessary treatment and the toxicity related to it.

Zachary Klaassen: Excellent. Tim, appreciate your time. I know our UroToday listeners will enjoy this discussion. Thanks again so much for joining us.

Timothy Daskivich: My pleasure.