Adonis Hijaz: Thank you very much, Alan, for the introduction and for the kind words. I'm joined here with my partner, David Sheyn, and we will be discussing two studies that we've published on the dementia risk with medications that we currently use for overactive bladder. So in, essentially, the background for these studies, we went ahead and did this studies using... First initial study was using the TriNetX database, and the second one was a subsequent study using Cosmos, Epic Cosmos database. And the premise for this study, we wanted to see if we utilize large data to provide some more evidence, like you discussed, to refute or to prove if there is any difference in the management of overactive bladder and the risk of dementia, because some of the retrospective studies of overactive bladder has suggested an increased risk of subsequent dementia. And there are also some in the literature discussion about anticholinergic drugs being some have high risk for dementia and some have low risk for dementia.
So the first study was a study that utilized TriNetX. And TriNetX is a research collaborative network with electronic health records of around 150 million patients across 81 academic healthcare institutions. We included patients with the diagnosis of overactive bladder between 2012 and 2023. Almost 940,000 patients with the diagnosis of overactive bladder. 83,550 patient received medication of overactive bladder. So we split the patients onto those who are on medication and those who are not on medication. And the medications that we included in this initial study were mirabegron, oxybutynin, tolterodine, darifenacin, trospium, fesoterodine, and solifenacin. And we looked at the incidence of dementia as identified by ICD codes. And we did the inverse probability of treatment weighting for age, for sex, for Elixhauser score, anticholinergic burden score. And if you look at the study slide, you can see that except for fesoterodine because of the small N, among patient age of 40 to 80, there was a significant higher risk of dementia relative to no treatment among all medications for overactive bladder, and likely a strong residual confounding effect by indication.
When you look at these categories, you can see the different age groups, and David, you can step in and essentially confirm what my observations are. If you look at, for example, patients at the age of 40 to 59, the hazard ratio compared to control was significant for all medication, as well as the age 60 to 70 and above 80. But when you look at that, you can see that above 80, the risk decreases. And maybe this is, compared to controls, that attributable risk to the medication is less because probably the incidence of aging-related dementia increases. David, do you have any comments on the results here as you look at the different age distributions?
David Sheyn: Yeah, yes. Essentially, the sweet spot is the people that started these medications between 40 to 59 and 60 to 79.
Alan Wein: What was the follow-up for these folks?
David Sheyn: Pardon me?
Alan Wein: The length of follow-up?
David Sheyn: It's between three and six years.
Alan Wein: Gotcha.
David Sheyn: So we followed them as long as they remained in the dataset, or they passed away. So those patients were censored. So basically, none of the medication had a protective effect, they all were associated with dementia with varying degrees of magnitude. Importantly, mirabegron had a fairly substantial odds ratio in both groups. The 18 to 39 risk of dementia, so these are not typical dementia patients, they likely had some sort of other syndrome that predisposed them to dementia. So in these cases, the two drugs that were most associated with it were obviously the ones we would expect. And then in patients over age 80, most likely there's some sort of protective effect. So these are people not at risk for dementia. If they got to 80 before they started a medication, it is unlikely that they are susceptible to whatever is causing dementia in these other groups.
Also, they are not likely to develop overactive bladder until around this age. So average is four to six years. So at best, before somebody gets treatment, they're getting these symptoms in their late 70s. So these are truly age-related changes where there's some sort of pathologic condition that's causing incontinence and dementia. And this is all patients, but when we separate by males and females, the results are identical.
Adonis Hijaz: So this was the first study that we published, and it was an eye-opener, because as we said, we went into the study hoping that we can confirm the results. So it was an eye-opener regarding the risks of dementia with the different medications. So we took on repeating the study using the Epic Cosmos database. And the EPIC Cosmos is EHR data from over 2000 hospitals using the Epic system for the electronic health records, and they opted to be participating in the Cosmos. So with this Epic database, EPIC Cosmos, they're over 300 million lives. And they included patients who had received the diagnosis of overactive bladder within one year of overactive bladder diagnosis between 2012 and 2025. And we split the patients into three groups, the mirabegron and vibegron group, and the group of low-risk anticholinergic. This is the darifenacin, trospium, and fesoterodine based on previous publications, and the high-risk group, which is the oxybutynin, tolterodine, and solifenacin group.
And we identified here the risk of dementia as identified by ICD codes and the prescription for dementia drugs. Again, we did the inverse probability of treatment weighting for years of treatment, age, sex, Elixhauser comorbidities, anticholinergic burden score, social determinant of health, a number of prior OAB encounters. David, do you want to talk about the results next?
David Sheyn: Yep. Here we go. So basically, and we don't have all the really interesting data on here, but essentially, we found no difference. So we compared everything to mirabegron and then high-risk and lower-risk. And the risk stratification for the anticholinergics was based on their structure and their likelihood of crossing the blood-brain barrier. And we follow these patients up to 10 years. And that's the nice thing about Cosmos compared to TriNetX, it has a much longer follow-up and a much more reliable way to capture prescription data.
So we have pretty detailed data on how often and how long patients were prescribed these medications, not necessarily how long they took it. And overall, the dementia risk or incidence was fairly similar across all the groups, 20% for beta-3s and 18 to 19% for the anticholinergics. The thing that we don't have on here is we also did some analyses associated with duration of use and dementia risk, and there was actually no pattern there. It didn't matter if patients took them for two weeks or four years. There was no association between dosing and dementia, which suggested that it's not actually the drugs that are driving the dementia risk, it's something else. Do you want to take it home, Adonis, or do you want me to?
Adonis Hijaz: You can go ahead.
David Sheyn: So basically across both of these studies, I think the punchline is none of these drugs are truly associated with long-term dementia risk relative to one another. In the TriNetX group, they are associated with a higher risk of dementia compared to no medication in patients with OAB. And that may be more of a confounding by indication. These patients maybe had more severe symptoms suggesting some sort of frontal cortical dysfunction, which is something that occurs in dementia that drove them to take these medications. So that punchline, and the main reason that I think these studies were done was I never really bought the theory that these drugs are associated with dementia long-term. Certainly short-term, that's been proven, like short-term cognitive impairment, but there's no actual causal mechanism that's ever been described. And the process that occurs that drives dementia is several decades long.
It doesn't occur in three or four or five years, which is when we're catching these medications. So at worst, some of these medicines are either accelerating or unmasking existing dementia, but I don't think they're causing it. And I think this points to important findings in terms of what the actual cause of dementia is and the relationship between urinary incontinence and dementia. I think it could potentially be a prodrome and just like we can screen smokers for lung cancer and identify other seemingly unrelated conditions to screen for more serious conditions, we can potentially use the onset of urgency incontinence as a way to potentially screen for dementia and intervene earlier.
Alan Wein: That's really interesting because obviously that runs counter to what I call the stick has been for the last few years. Now, was there any difference between mirabegron and vibegron?
David Sheyn: So TriNetX, we were not able to identify that, but in Cosmos, we were, and there was no difference between the two.
Alan Wein: Right. And so basically, the idea or the hypothesis that comes out is that it's not the drugs, it's the overactive bladder that's a signal that something is going on in the brain, at least in a majority of these people, that would later predispose them or predisposes them at the time to some sort of cognitive dysfunction, correct?
David Sheyn: Right, right.
Alan Wein: And in both of these studies, you would publish the actual numbers relative to control, that is the people that had no medications of what the percent of cognitive dysfunction was, and it was greater in all the drugs than in the control, correct?
David Sheyn: Right, right.
Alan Wein: Yeah, so I think people who are interested should really read these articles. Now, has the second one been published, the one that you presented at SUFU this year?
David Sheyn: We're working on that manuscript right now.
Alan Wein: Okay. Yeah, because the first one was in the World Journal of Urology. That was in, I think, 2024. But all the numbers are in the abstract at SUFU if anyone wants to look those up. So basically, the clinical finding would be that although antimuscarinics, they do have more side effects in terms of dry mouth and constipation, et cetera, et cetera, et cetera. As far as cognitive dysfunction is concerned, we should take a strong re-look at that because it looks as though there really may be no substantial difference on which to base the choice of beta-3 agonists versus anticholinergics, at least with respect to cognitive dysfunction.
David Sheyn: Right. I think the choice should be driven by efficacy and by side effects. So if somebody is doing amazing with oxybutynin and they're getting sleep and they're not having side effects, they should stay on that medicine. And that's the other thing I think we weigh. And if somebody has terrible nocturia and we start them with Myrbetriq and their daytime symptoms are great and we don't want to start them on a more effective medicine, not getting any sleep is just as a strong predictor of dementia as any of these other potential drugs.
Alan Wein: Right.
David Sheyn: So I think we have to take it as a whole picture, and I think we have to really try to understand the mechanism. When I talk to patients about urinary urgency incontinence, I tell them, this is not a urologic problem. This is a neurologic problem. All of our third-line therapies are neurologic.
Alan Wein: Right. Yep, exactly. Interesting. Well, I think that's one of the more fascinating things that's come about. Unfortunately, in our field, there are not really very many things that go against conventional thinking that turn out to be right, but this looks like one of them that is going to and in fact is. So I really appreciate you guys coming on and explaining this.
Adonis Hijaz: And I think that this is a great platform for us to present the data and the work that our team has been doing in this. And thank you for shining the light on the study that we have done. And to your point, I think this is going to challenge some of the current understanding in the field, but also open an opportunity for us to research it further and understand more overactive bladder and urgency incontinence, along the lines of what David is talking about. The issues, obviously we go back as a clinical scientist ourselves, is that sometimes answering such big questions, you need randomized clinical trials in order to identify if there is a difference between the two types of management in the incidence of the dementia or the changes with cognition.
However, since such changes could be very remote to happen and the timeframe is too distant in time, then conducting such trials might not be feasible unless you are able to identify a correlate that could be measured early and could be very sensitive and a good predictor of long-term dementia risk. So these are the things that we are thinking of as a group. Is there a way we can identify together with our neurology partners, some correlates that we can look at in order to answer such questions on a short-term in a randomized clinical trial? So these are things we continue to strive to answer.
Alan Wein: Terrific. Yeah, there was a great quote by John Kenneth Galbraith, the economist, said something like, "Accepting the conventional views prevent us from the process of thinking." And I think you guys have thought about it and not accepted the conventional view, and I think everybody should look at this stuff. Listen, thank you so much.
David Sheyn: Thank you.
Adonis Hijaz: Appreciate you, Alan. Thank you.