Skip to content

We’ve talked at length on prior podcasts about the failures of aducanumab, Biogen, and the FDA’s decision to approve it.  

But wait, there’s a shiny new anti-amyloid drug, lecanemab!  (No it’s not just the French version of Aducanumab).  

In an article in the NEJM (a published article this time, wonder of wonders!) lecanemab was shown to slow the rate of cognitive decline by 0.45 points on an 18 point cognitive scale compared to placebo.  Wow!  Wow?  Wait, what?  

On today’s podcast we talk with Jason Karlawish, who we’ve had on previously talking about his book The Problem of Alzheimer’s and with Aaron Kesselhim, to discuss FDA approval of Aducanumab, as well as frequent guest and host Ken Covinsky.  

They debate today’s central question: is it time for geriatricians to get on board with lecanemab?  Along the way we address:

  • Is this degree of slowed cognitive decline meaningful to patients or care partners?
  • What about the burdens, risks, and harms? Every 2 week visits for infusions, regular monitoring for brain swelling and bleeding, case report level risk of death?
  • Did the study do enough to address issues of inclusion and diversity by age, race and ethnicity, and multimorbidity?
  • What does this study say about the amyloid hypothesis?
  • Should the FDA approve, and under what conditions?

Their answers may surprise you.  

As a preview of final thoughts at the end of the episode, Ken and Jason agree that the FDA should approve lecanemab conditional on a post-approval monitoring system and public access to study data, geriatricians should be prepared to have thoughtful conversations with patients about the risks and benefits of lecanemab in view of their values and priorities, and ultimately, that geriatricians should be open to prescribing it. 


The times, they are a changin’.


Eric: Welcome to the GeriPal Podcast. This is Eric Widera.

Alex: This is Alex Smith.

Eric: And Alex, I’m going to love this one. Who do we have with us today?

Alex: This one won’t be controversial at all. We’re delighted to welcome back Jason Karlawish, who is Professor of Medicine and Co-director of the Penn Memory Center at the University of Pennsylvania, Perelman School of Medicine. Welcome back to the GeriPal Podcast, Jason.

Jason: Great to be back, Alex and Eric.

Alex: And we’re delighted to welcome back Ken Covinsky, who is Professor of Medicine and geriatrics at UCSF. Welcome back, Ken.

Ken: It’s an honor and a privilege to be back.

Eric: So we’ve had Jason on before talking about both Dementia, his book, which we’ll have links to in the show notes, and also about Aducanumab. But this podcast is not on Aducanumab. It’s on Lecanemab, which was just published in the New England Journal, anti-amyloid drug for dementia, Alzheimer’s disease.

Before we go into that topic, which we got a lot of ground to cover, Jason, I think you got a song request for Alex.

Jason: Alex, The Times They Are a-Changin’.

Alex: Great choice. Pay close attention to the second verse lyrics. I think they’re appropriate. (singing)

Ken: Bravo.

Eric: Very apropos.

Jason: Brilliant. Just the other title I wanted, which apparently was already co-opted by Sean Morrison was We Won’t Get Fooled Again by The Who. [laughter]

Ken: That is awesome. Isn’t it cool that Bob Dylan won a Nobel Prize for Literature for Lyrics?

Alex: Very cool. Incredible songwriter. Poet, yeah.

Eric: So a couple weeks ago we did a podcast on the need for cultural transformation around dementia care. Actually, we touched upon Lecanemab and anti-amyloid antibodies, but really thinking about other ways we can transform dementia care.

This podcast, ultimately we want to get to the question, should the field of geriatrics get on board with Lecanemab and these anti-amyloid drugs? Because there’s been a lot of pushback from geriatricians, including our national organization, American Geriatric Society, pushing back against, at least, Aducanumab. The question is, should we have that same take on this new drug?

So let’s start off from the beginning. Jason, what the heck is Lecanemab? Am I pronouncing that right? I probably should know how to pronounce that.

Alex: It’s a French version, right, of Aducanumab?

Jason: Yeah, Aducanumab remains the dumpster fire. Lecanemab is… Well, let’s just say its study name says it all. It’s Clarity, was the name of their clinical trial, the Clarity Study.

Lecanemab is an anti-amyloid antibody. It’s delivered by intravenous infusion every two weeks. It targets soluble amyloid species in the amyloid cascade. And it was studied in Clarity AD Phase 3, 18 month long, single dose 50-50 drug versus placebo clinical trial, enrolling people with mild cognitive impairment, mild stage dementia who had confirmed amyloid, elevated amyloid by PET scan. That’s the guts of the study and that’s the drug.

Eric: So it seems to really, again, another drug focused on the amyloid hypothesis. Do we know if it does anything to any other of these biomarkers like tau?

Jason: Yeah, so it is an anti-amyloid drug and one key message with anti-amyloid drugs is they’re not all the same. The amyloid cascade starts with the protein sitting benignly in the cell wall. Then something happens, breaks off, forms fragments, fragments form fibrils, fibrils form oligomers… and so-on, unto the plaques. And depending on the drug, Bapineuzumab, solanezumab, they actually target different points along those protein malformations. It’s just the way the antibodies are designed. So this one goes after, as I say, the soluble fragments, Aducanumab goes along that on the proto fibrils or something. So they all slightly differ there. And this one lowered amyloid as confirmed by PET scan. I mean, fairly notable lowing on the Centiloid measure of amyloid. Below the cutoff considered to move from not elevated to elevated. The secondary data on biomarkers, namely the tau data, the hyperphosphorylated tau that correlates with neuro-generation. That also showed changes of decrease that would be befitting of a reduction in tau level, therefore a reduction in neuro-generation. So it moved amyloid endpoints, as these drugs generally do, but more importantly I kind of think it moved tau measures as well.

Eric: So let’s dive into… So we’re going to start off with the facts, the things that we know. And the great thing is, as opposed to Aducanumab, the drug company actually published their studies first before going to the FDA, before having sift through loads of webpage data from the FDA, we can actually pull up a New England Journal study that gives us some information about the efficacy and safety of this drug. Can we start off with what did they do in this study, this Clarity study? Who was enrolled?

Jason: Well, I’ll continue on, as I say, adults with either mild cognitive impairment, so cognitive impairment that’s detectable but causing inefficiencies in daily life or mild stage dementia, meaning some disability. Mini metal cutoff, I forget precisely, but I think it was low twenties was the cutoff, the base, the lower cutoff. You had to have a definite impairment on a pretty intensive measure of memory called the Logical Memory test. And as I say, had to have evidence of elevated amyloid by a PET scan. Met those criteria, got in. Am I missing anything, Ken or anyone?

Ken: That seems right. So is it fair to say Jason, by and large, people who don’t have Alzheimer’s yet or maybe in some cases ever, and people who… This is a population, by and large, are still living at home, they’re still capable of some independence, maybe even a lot of independence. This is pre-nursing home kind of?

Jason: Oh yeah, yeah, yeah. So this is a group of people that have some inefficiencies or impairments in higher level IADLs. So someone’s got to do the finances, but they’re still doing the cooking, et cetera. No troubles with BADLs stereotypically. Yeah, yeah.

Alex: One other point I’d add is in the Aducanumab study, I believe you had to have the PET scan. Here, you had the PET scan or CSF measurement of Abeta.

Jason: That is correct, yes.

Alex: Which may have implications as far as access to these drugs because PET scans are rare.

Jason: Yes. CMS does not cover PET scans as of yet. They’re going to re-look at that decision in a week or two.

Eric: And it sounds like this drug does a really good job of taking away amyloid. Is that right?

Jason: That’s right. There’s no contesting that. As did Aducanumab, as did many of these drugs. But yes, the reduction of amyloid was well documented and very clear.

Eric: And how well did it do with the… What do we know as far as the data around improving cognitive function and function in general?

Jason: So it didn’t improve. These are diseases that are degenerative, so the trajectory over time is loss of cognitive function and performance on activities of daily living, instrumental and basic. So the measure here was a composite measure called the Clinical Dementia Rating or CDR. It’s a measure that’s used in research only, I don’t think it has ever or will ever be used in clinical practice. And it’s a composite measure that rates someone’s performance on a sort of omnibus concepts of cognition, memory, judgment, decision-making, and then omnibus concepts of function like home and hobbies, personal care.

And that was the primary endpoint and is a very quirky way to look at the CDR, which is instead of the way it was designed, 0.5, 1, 2, 3, normal, very mild, mild, moderate, severe dementia. You can look at the sum of boxes, which is essentially taking the sub-scores that make up those 0.5, 1, 2, 3, and treating them as sub-scores or the boxes, which is a cork of the instrument. And so what they looked at was a difference in the CDR sum of boxes between decline in placebo versus the decline in drug group. That was the primary and single primary endpoint for the study. Which has become the primary endpoint for a lot of these studies.

Ken: So Jason, just to make sure I have my head around this right. So you basically have six pieces of what people consider part of the dementia syndrome. Six impacts, ranging from memory, to orientation, to ability to do activities, to social activities. And they’re each scored on a value of where zero is normal, half is mildly, subtly abnormal, one is definitely abnormal, but mild all the way to three to very abnormal. So you end up with, when we’re sort of like look… When we get to looking at the sum, we’re looking at basically 3 times 6, we’re looking at 18.

Jason: That’s right. The boxes range from zero to 18 and that’s the scale. In this group of patients the range of their CDR sum of boxes was about 0-6 with a mean median of 3. So that reiterates the question you asked Eric, Alex. These were people with mild disease, their sum of boxes was stacked up at the mild stage of severity. Even though the scale does range to 18, this was not a group of people with a median of 9, which would describe a group of people with moderate to severe, as well as mild. This was very stacked in the mild stage.

Ken: So scattering of mild to moderate deficits basically gets you to a three.

Jason: Of a CDR global of three?

Ken: Yeah, a total of three.

Jason: A total of three is severe stage dementia. That would be people with probably that would crosswalk to a mini metal of about 12 or 10 in a well-characterized Alzheimer’s population. In contrast, these people’s global CDR was either 0.5 or 1. So again, this was a mild stage group of people. 0.5 typically describes MCI, although some people with 0.5 are just called having dementia. 1 generally describes someone with mild stage dementia. And this was a group of 0.5 or 1, or a sum of boxes score from about 0-6, again, mean was about 3.5.

Ken: Okay. Yeah, I was thinking more of the mean. So you got to this mean with a scattering of mild. It’s usually a number of mild impairments that, I mean I don’t want to say they’re mild because they matter a lot, but patients are experiencing a number of abnormalities, none of which are severe, but an aggregate might be bothering them a lot.

Jason: That would be correct, yeah. I mean they need caregivers, yes, for monitoring supervision and some assistance. Yeah, that’s why I would call them. Yeah.

Eric: And I can imagine one of the challenges is when everybody’s skewed towards one side, it may be hard to see a difference. So when we think about that, can we even say what a minimally clinically significant change is? Given that we’re starting on one end of the spectrum.

Jason: Well the theology is that a change on the CDR sum of boxes of I think about 0.3 something is considered to be significant. And I kind of listen to that and I say, fine. I think in this study and all these Alzheimer’s studies, you have to have a primary endpoint because you’re going to spend your alpha, your power on that, and your alpha. But there are also secondaries that you want to look at. And ideally you’d like to see all your primary and secondaries line up because it kind of points you in the direction that there’s something going on here. So on the CDR sum boxes, the placebo group didn’t decline very fast in this group as we’d expected. But the drug group declined slower than the placebo group. And if you do, there are various machinations, and I said that. The way they analyzed the data, which is an entirely appropriate intention to treat, showed an overall mean difference of 0.45 on the CDR sum boxes between drug and placebo. That’s the effects size.

Eric: And like you said, on all the secondary outcomes, they also showed similar kind of benefit. Is that right?

Jason: That’s right. So they measured, again, you could debate the particulars of each measure, but they measured a series of standard measures of cognition, instrumental activities of daily living, a kind of composite measure of cognition, instrumental activities of daily living. And on all those measures of cognition and function, the same results were obtained, which was that the people on drug did not decline with the same rate as people on placebo.

Ken: So I’ll just add, Eric, you used the term, you used a rather wonky term, was it minimally, clinically, important difference. And I don’t know what a good lay way of saying that be, okay, we see a number but it doesn’t matter, so what difference matters? This is like in studies of where you’re looking at these kind of difficult to measure things, you get a lot of debates on what that number is.

I’ll just sort of point out my read of… It’s all over the map of what people would say matters on this test. People who do drug trials seem to suggest that the number is lower. People who are less into drug trials seem to say the number is higher. So I’ve seen numbers ranging all the way from a change of 0.3, which this study would’ve met. But I’ve seen a number of other people suggest, well the real number is one. That’s a number, we can’t answer that question. It’s an opinion and it’s a value judgment. But part of where when we debate this, a lot of this really hinges on, what really matters and would a patient or a family see something or feel better because of this change?

Eric: Yeah, I mean it’s funny because dementia literature is ripe with statistically significant improvements but clinically questionable. You look at the Aricepts of the world, cholinesterase inhibitors, Memantine, it keeps on coming up. It’s statistically significant. How important is it clinically? I think we’re stuck in that same rut right now.

Why do you think that is? Why do we keep on falling into this argument? Which, I would say it’s an argument right now. I think you can make a case either way. Thoughts?

Jason: So yeah, I’m not in a rut because if FDA approves the drug, and I think they probably will, unless there’s some data that’s going to come out when re-analysis shows, but FDA will approve the drug. If they approve the drug, I’ll prescribe it at the Penn Memory Center. It’s going to be an interesting discussion because the drug, and we’ll talk about it, does have risks, and I think there’s some subgroups that need to be thought carefully about. But on balance I think crosses the threshold of being a drug that’s efficacious at bay with a drug with risks.

Really for a drug like this, the final proof of its value is not known right now. That is, if you stay on it for say two years or more, what’s its effect on your rate of progression and from a societal perspective, what’s that summative impact on the cumulative time spent caregiving, the cost of caregiving, et cetera. We just don’t know that answer. But that’s what we need to know-

Eric: Well let me ask you about that.

Jason: … that’s what matters in this disease.

Eric: So when I look at the graphs in the New England Journal paper, it looks like most of the change happens between three and nine months. And then the slopes start looking the same between the placebo group and Lecanemab group. Suggesting, again, it’s eyeballing. Potentially suggesting to me is that, you know, you get rid of some amyloid, maybe there’s some moderate clinical benefit. That’s a word that they used. They like the word moderate, I think, in the New England Journal paper.

Jason: Well they struck that from a subs… It was both in and then out.

Eric: Oh was it?

Jason: It was put in, in the final version, but in the previous draft moderate was not there.

Ken: I imagine that was a great source of debate and compromise. I might have opted for minimal to mild, but one can debate death.

Eric: How much should we look at the trend lines? That they’re, after about six months to nine months of treatment I don’t see them separating much more. Suggesting that two or three years later is it going to… Are we really going to see any change? Or have we just changed it by… I mean, this is the maximal effect.

Jason: So what you want to see, and we’ll know this, is and then when the Clarity AD study ended, randomized phase, those who had been randomized blinded to placebo were put on drug, and those on drug remained on drug. And what you want to see is once people who had been on 18 months to placebo on drug, what happens to their curve? Do they jump up to the other group or do they, quote, “Never catch up,”? And that’ll give you some sense of the effect of the drug.

So I think geriatricians need to get on board with this drug and here’s why. The drug is effective. The drug has risks. We will learn from using it the net compendious benefit of this drug to patients. And the kind of conversations that we’re going to have are conversations about, well is it about lowering amyloid quickly early on and then stopping the drug and following amyloid levels? Is it about a persistent lowering of amyloid that you have to keep people on the drug? Is it about lowering amyloid and also seeing tau go down and following tau? All these things are going to be part of the conversation in the next 24 months, two, three years. And this is how I think the thinking about this disease, from the perspective of pharmacologic treatment, is going to change the way we talk about the disease.

And I think our profession needs to be part of that conversation. Needs to be talking that language, understanding the way we talk about that language and contributing our expertise in wisdom to interpreting the data, both to understand its benefits to patients but also how to help patients understand it and families understand it. Because I think we’re uniquely good as a profession, with taking care of patients and helping them understand risks, benefits and balancing those with their other diseases, et cetera. So that’s why I’m pro for this drug entering the clinic and for our profession being prescribers for it.

Eric: Ken, are you going to prescribe?

Ken: Well first of all, I mean I don’t know, I find the term getting on board… I’m not sure I like that term.

Eric: No, it’s a train Ken. Choo-Choo.

Ken: Yeah, I mean, no, it sounds like, I don’t know. I mean, I don’t know, Alex has been telling me for years I should get on board with the Giants and the Warriors. I’ve been in California 20 years, why do I keep rooting for my Chicago teams? I mean the problem is this doctor-

Jason: Because every 70 years the Cubs make it to the World Series. [laughter]

Alex: If you wait long enough.

Ken: Hope, hope, hope. But we see patients one at a time though. So it’s hard to be on board on that level. Here’s my thought-

Jason: Ken, are you going to prescribe it or not?

Ken: Will I recommend it? No, to the average patient. Would I refuse to prescribe it? Unlike Aducanumab where I see no… It would almost be a professional professionally irresponsible to prescribe it. This is one where I do see room for patient preference here.

Eric: Why would you recommend against it, Ken?

Ken: So why would I recommend against it? I think the big question is, in the end, is your patient better off or worse off? So I think one of the mistakes people are making here is, there’s this sense of jubilation here because for years and years we’ve been trying to prove that a drug that somehow involves the amyloid pathway of Alzheimer’s disease, that a pathway that people think but are not sure is the pathway for Alzheimer’s disease, it does something in altering the clinical pathology of Alzheimer’s disease.

And I think for the first time we actually have a drug that actually shows there is something to that. But I think what’s missed is, okay, maybe we finally have crossed that threshold. And this is again hoping, okay, your sports team has put together a good team and in 20 years they’ll win the World Series. Now we’ve reached that point, but I don’t think we have evidence that says this drug is going to make the patient sitting in front of you better. Because I think what we have is what you would say a petite, a teeny piece of fact that may or may not be noticeable to a patient. And significant toxicity. Toxicity that I… And I know we’re going to talk about toxicity later, but toxicity that I think we have reason to worry about, that is understated and that the long-term effects of are unknown. Where we have good reason for other areas of brain science to worry about.

And as even before you get… Remember the whole reason is to make your patient’s life better and to help your patients live a good quality of life. And now we’re saying, oh okay, just come in for this infusion every two weeks-

Jason: So Ken-

Ken: … quality of life to me.

Jason: So infusion every two weeks, that’s a personal choice, as well as, obviously, thinking through the benefits and the risks, particularly for people who are APOE-E4 carriers. But let me jump to a very, very interesting point. You mentioned patient quality of life, actually, among the secondary endpoints in this study was the Zarit rating of caregiver burden and the QOL-AD developed by Rebecca Logsdon. And in multiple studies of cholinesterase inhibitors, it’s interesting, I’ve never seen a cholinesterase inhibitor study that showed an effect on the Zarit scale or the QOL-AD. On this study, patients on drug had a significantly different scores on Zarit burden scale by their caregiver and patient rated quality of life on the QOL-AD. Patients on drug had higher self ratings of their quality of life than patients on placebo. And the caregivers of patients on drug had lower scores on the Zarit burden scale than patients on placebo.

Eric: So two questions on that, because one of the questions from Aducanumab was did people know what drug they were getting based on potentially side effects, including needing multiple MRIs for worries about these brain micro-bleeds? And the second question, well I’m going to ask that question first. What do we know about how good did they blind patients? Do we know anything?

Jason: So in these studies the question of unblinding, obviously, is a big one. Because the side effects are ones that you can feel and there’ll be periods of time where you’re sort of stopped your infusion or go on placebo, et cetera. And the bottom line is, when the analyses take into account the potential effect of unblinding, courtesy of side effects, and they did a variety of sensitivity analyses. I don’t pretend to be able to understand them all, but I’ll take it. And the sense from the reads of the data are, that you can’t explain these data away by simply unblinding, data on the cognition data on the functional measures, et cetera, that there is a drug effect here. It’s not just a confounding by unblinding.

Eric: And the second thought is, and maybe this is the reason why geriatrician should get on board with this drug, is it would be interesting to see what quality of life looks like, not between a group getting placebos twice a week versus this drug twice a week, but a group that’s getting this drug twice a week needing, or sorry, every two weeks, needing go to the infusion center versus a group that is not getting it and not having to travel to the infusion center twice a month.

Jason: That’s a study would be hard to randomize and blind too. We have historical-

Eric: Impossible, right?

Jason: … We have historical data on that. Look, we should probably talk about risks because the drug does have risks. I mean it causes micro-bleeds in the brain and it causes swelling edema as well. And those risks are more likely in people who carry one copy of the APOA4 gene and they’re even more likely people who have two copies of the APOA4 gene. And so you have this difficult situation of a drug that has well-documented, well quantified risks, in particular in people who carry a gene that’s associated with developing the disease. And the force plot data showed a dose decrement of response to the drug, for each copy of the gene you had. So if you were a homozygote, you had less evidence of benefit than if you were a ho heterozygote or a non-carriers. My takeaway from that is, not that the drug quote doesn’t benefit homozygotes or heterozygotes, I think that’s an over read of the forest plots. But it’s pretty suggestive that, and this is going to need some teasing out, but if you are a carrier, you potentially might not benefit as much and you face a greater chance of these risks.

And so I think that does make for a Gordian knotyk of a very ethically challenged risk benefit balance that a patient’s going to face. Again, my view on that is the patient and their family need to make that decision. Obviously, I’ll be there when I help them make it. I think that’s why geriatrician again should be part of this because we’re good at talking to our patients and listening. But it does make for a complicated decision. It does. This is not a risk-free drug. And for people who are APOA4 carriers, it’s a particularly problematic decision.

Eric: There’s some word on the street right now that Eisai may say, recommend against it in homozygous carriers.

Jason: That’s their decision. I personally reading the data would not, I will… If a homozygous patient wants to take it, I would prescribe it. But I think they need to understand that they’re going to need careful monitoring for risk and I may not be able to catch that risk before it becomes symptomatic and causes some problems with gait and falls, et cetera. But again, I would view that as their choice. Yeah.

Eric: And there was also some news about two deaths maybe being linked to getting-

Jason: That’s right. There was this patient in this study who died and then a patient in the open label study who died, both of them in the context of getting ill and it gets a little frustrating when you hear like did Lecanemab cause their death or not? As we all know, in geriatrics the idea of what causes death in patients with multiple problems, it’s quite clear though that exposure to Lecanemab contributed to their death. I mean, it increases the risk of brain bleeds and these people are getting blood thinners, one for a stroke and the other for a heart attack. And so yeah, that probably was not a good combination. Yeah, look, my take on the drug is, this drug has real benefit and real risk. And I think the biggest problem I see is after FDA approval, if it gets widely used by clinicians not skilled in how to use it, we’re going to see problems with it.

So I think actually FDA should put it under a risk evaluation mitigation strategy or REM strategy. And a REM strategy allows FDA to create criteria for who’s a prescriber. It allows FDA to set standards for gathering data for risk in a registry, in a database. So it’s been used for drugs just like this. There’s benefit, but there’s also risk in the early days of MS, it was used for the early MS drugs that cause could cause PML in addition in the brain. Just very similar kind of side effect. In some sense the concern I have is without a REMs, especially in America, wide use by frisky prescribers could cause some real problems here for patients.

Ken: Yeah, there are something, I mean I’m in complete agreement Jason, that for several reasons, I mean think the data registry is super important because if this is approved, we really need to know if the side effect burden is different in a real world practice than in a clinical study. Because these are the sort of side effects for the reasons you mentioned are notorious for being worse in real world application. But also a registry allows us to study some of the long term, both benefits and risks. Here’s one of my concerns, Jason, about risk, is that my understanding of some of these risk factors for brain sciences is from a lot of spheres, whether brain injury is concussion or brain trauma, you know, you think about chronic traumatic encephalopathy in NFL players, whether it is maybe even psychosocial trauma, post-traumatic stress disorder… Or even we know from some of the science, I mean I was taught years ago that when you do MRIs on just regular people and you see these little mini strokes that right now have no clinically meaningful impact that we can measure, they do over time.

So one of my worries here is like we say, okay well it caused a little bit of brain swelling, it caused a teeny bleed in a brain, maybe it was asymptomatic. Don’t worry, you’re pretty little heads over these problems. I-

Jason: I worry about those, I disagree. I think that when I say geriatric needs to get on board with this drug, that means get on board with using biomarkers to understand what the cause of someone’s disease is, in this case Alzheimer’s. And then using MRI technology to better diagnose the disease and look at some of the comorbidities, like micro hemorrhages. In the case of using this drug, if someone develops micro hemorrhages or microadenoma, we’ll have to re-scan them and follow up, potentially stop drug and/or halt it completely, if those micro hemorrhages and micro edema don’t clear. So this is not a drug, you just give it to someone and see them back in six months early on. It requires frequent MRIs and based on those results even more frequent and/or stopping the drug. And I would take that extremely seriously. Yeah.

Alex: Yeah. I want to raise another point about disparities in equity, interested in your thoughts about this. Sharon Brangman wrote a terrific commentary in JAGS where she talked about the need for including persons in clinical trials who are most likely to be affected by the condition. And we know that Alzheimer’s disease impacts African Americans and Latinx patients at rates that are two times and one and a half times respectively higher. And that Aducanumab, they had very low numbers of participants from those communities. I think it was like 19 black people. And in this study they… I’m interested in how well you think they did. They had 20 black people in the Lecanemab arm and 24 in the placebo arm. In terms of Latinx people, they had 12.5%, or 107 in the Lecanemab group and 108 in the placebo group.

Thoughts about… And there are other equity considerations at play here, right? We have access to… You have to have a PET scan or a beta CSF measure to get in and then you require frequent infusions. We have to travel to infusion center, you have to have access to infusion center and then you have to be monitored for these brain bleeds, brain hemorrhages. So there are other issues in terms of access to this medication, if it is approved. Thoughts from the two of you about those issues?

Jason: Ken, did you?

Ken: Yeah, I mean you stated at the problem very well, Alex. You know, I’ll give you a middle take as I at least have to acknowledge that this company that sponsored this trial did way better than the company that sponsored the Aducanumab trial. There was at least a semblance of taking responsibility for inclusivity. That the minority inclusion was not nearly adequate, but at least it was there and taken seriously. It shows that we have a long, long way to go to make this right, but at least there was some effort at minority inclusion. At least there was some effort in it. I actually wish they didn’t have the age cut off at 90. There was no reason for that but oka, they at least went to 90. So they at least included people who were very old, they included people who had multi-morbidity, so by that I mean the diseases of aging beyond dementia. We at least have some data about what happens on people who are on anticoagulants. So that, not enough here, but better. Some progress in study design development.

Eric: That’s the great thing about Aducanumab, it set such a low bar… anything. You got it published in a journal? That’s great. How wonderful.

Jason: Yeah, no, Biogen set a bar so low that anything a other drug company does is automatically virtuous behavior. But I can’t tell you how when you chat with folks in pharma, how much they just roll their eyes what Biogen did. Yeah, I think the challenge that I think we face when we think about the need for diversity in clinical trials is diversity with respect to what, for what reason? If it’s with respect to the reducing the risk of developing disease and who are having disease progression, because these people have already developed the disease in the clarity study, it’s about progression. Then I’d want to make sure that your design is going to attend to what you think are those mechanistic factors that might have persons, say who self-identifies African American, progressing faster. I didn’t see that going on in this study. Instead, I saw diversity with respect to just equity issues of access to clinical trials, right? That all Americans should have access to a trial, assuming they meet the criteria for entry. And it’s odd, as the case with Aducanumab that there are, you know, you could count on one hand the number of persons of color in the study.

And I think there are two very dyssocial issues with respect to diversity in clinical trials. The former actually, it’s very scientifically challenging because if you think about writing a label, so the label would say this drug is for the treatment of Alzheimer’s disease and here’s some other points depending on the person’s self-identified race and ethnicity about how you prescribe the drug. I mean I find that very problematic because behind race and ethnicity are a host of cultural and social constructs. So I’m still struggling with the in drug development thinking about these issues with respect to understanding mechanism or prescribing. But in terms of access to research and its benefits as well as its risks, it’s a shame that our studies tend to simply reflect us only a small slice of America. Absolutely.

Eric: Well let me ask a follow-up question real quick. When I look at the supplement, it seems like, again, caveat is like you look at those forest plots, it looks like this mainly works for white men who are APOE non-carriers.

Jason: Yeah. So this is the issue with forest plots and it’s kind of funny. Well first of all you get power issues.

Eric: It’s not designed for me to make that statement, right?

Jason: But secondly, forest plots, it’s supposed to look at overall am I seeing a trend that favors drug? Or overall am I seeing a trend that favors placebo? Or am I seeing things all over the place? And the forest plots are pretty compelling for a trend that favors drug.

Having said that, I’m the one that fixated on the APOE because of the dose response aspect of that 0, 1, 2, copies of APOA4, you see across all the endpoints, the same dose response, reduction in benefit. That doesn’t lead me to say it doesn’t work for people who have APOA4. It makes me concerned and we need to understand more. What is going on here? The APOA4 carriers may not be as good responders. But I wouldn’t look at those forest plots and say a woman from Europe can’t benefit from this drug, because that’s what the plots would suggest.

Yeah, at some point you’re getting into issues of multiple comparisons. Also, biological plausibility, I mean, can someone limb up some scientific explanation for why Europeans aren’t responding? The answer is, I think it’s like we could have done astral signs and shown that Sagittarius’ don’t respond to the drug. I mean I think we have to be very careful.

Eric: All right, I got two more questions. I know we’re running out of time. Ken, what do you think about this question? Do you think this study is proof that the amyloid hypothesis is correct or do you think that based on these very small changes, this is proof that there’s a lot more going on besides amyloid in Alzheimer’s disease?

Ken: Well you astutely framed the question correctly by not making it an extreme question. Because I think the answer is for the first time we actually have evidence that there may be something to the amyloid hypothesis, that when you change amyloid in a big way, you do something to the clinical parameters of the disease. I think on the other hand it’s also very convincing proof that there’s probably a lot more there. That this study, I mean no one will disagree, as Jason noted. This changes amyloid in a big way. It’s a great… I mean this obliterates amyloid. But if it was all amyloid you would’ve expected, why didn’t the progression of disease stop? So remember what we’re talking about in these two groups. Both groups got worse, one group gets worse less. And that’s important, getting worse less.

Jason: Yeah, I mean Ken’s points are well taken. I think geriatricians need to get on board with this drug because this is the beginning, not the end, of understanding what happens to a human’s brain when you reduce amyloid. And my hunch is five years from now we’re going to be back on this podcast talking about the various subtypes of Alzheimer’s disease, that do and don’t respond to the various drugs based on a variety of factors.

And you know what, that’s been the story in a variety of complex chronic diseases, especially associated with aging. But the effect in this trial, I believe is a real effect. This is not a statistical artifact. I think it crosses thresholds such that it’s effective, it’s got risks. That’s the kind of drug that can be in clinical practice. I’d like to see my colleagues in our field. Because I think geriatricians are the kind of doctors to diagnose and treat people with these diseases. And I hope that our field recognizes that this biomarker of transformation of the disease is something that we can be part of and inform, and improve the way these diseases are diagnosed and treated. Rather than be on the outside saying there’s not enough benefit, or amyloid isn’t the only cause. So I’m not going to pay attention to these data. I just don’t think that that’s clinically responsible and scientifically responsible for us anymore as a field.

Alex: Well, can I follow up with that? And just to be clear, are you suggesting that geriatricians should set up infusion centers and be prepared to send patients for… Screen patients with PET scans or do CSF you-

Jason: Screening? No, I think geriatrician should be prepared… Geriatrician should be in their practice for patients who have MCI or mild stage dementia that looks like the Alzheimer’s phenotype to be comfortable with ordering biomarker tests, particularly amyloid, to confirm the person has amyloid likely therefore has Alzheimer’s and engage in a conversation about this drug. I think they should be comfortable with, I don’t have to have an infusion center, but following up the next steps in terms of MRI imaging, et cetera. I think that these drugs are probably going to be prescribed, in what we’re very good at, multidisciplinary teams, where once the drug is started, our colleagues in their practice can help with monitoring response such as paraprofessionals. But yeah, I think we should be comfortable being prescribers for these drugs in a responsible and reasonable way. Absolutely. Yeah.

Eric: So Ken, I’m going to ask you the same question, but I want you to think about this. Geriatricians are generally one of the highest satisfaction fields, but the lowest paid fields, set up the infusion center, the average salary for geriatricians all of a sudden skyrockets. Should geriatricians get on board?

Ken: Well, first of all, am I released from the debate stance and can I say what I think now?

Eric: Yes, you can say what you really think.

Ken: Okay. That’s an economic issue for our field. I honestly don’t care if we’re the infusion center or we send our patients to the Memory and Aging center. I do agree with Jason that there’s a lot more to learn, and our field should be involved in learning about it.

What I would say is my take on this is, my instinct is that for the average patient with MCI and mild Alzheimer’s disease, this probably will not be worthwhile. But I agree with Jason, is that the evidence is such that I owe my patients a serious discussion and I would ask a lot of questions about their values, what’s important to them, what would matter to them. I would actually ask them how much the protocol would disrupt their life. And if they say they don’t care about that protocol of having to go to a clinic every two weeks, that might influence what I recommend. I’d be open to them that reasonable people do see use in this drug. And if I had a patient after that discussion who wanted that drug, I would facilitate their access to it if that was their decision. After hearing a full discussion,

Eric: Let me ask you the last question. Given that it doesn’t seem like the decision on the FDA, how they decide things does not seem that they take much into account besides the whim of the director. Sorry, I’ll just get off my high horse there. Should the FDA approve this drug? Jason, I’m going to turn to you first. Ken, you can think about it. Jason, I heard with maybe yes, but maybe thinking about-

Jason: No, I think FDA should approve this drug, full approval. I think we should stop this conditional approval nonsense. It needs review for full approval and I think based on the data out there, it should get full approval. I think FDA should approve it though with a risk evaluation and mitigation strategy attached to it or REMs.

Eric: Ken, what do you think?

Ken: Yeah, I mean, I would just sort of say, I mean I have a number of requests. So one of the requests I have is, I know there’s this little box in the study that asked if the authors will make the database available and it’s checked off no. That’s totally unacceptable. So that this database needs to be open for the academic world and the policy world to look at.

I would say probably with a very rigorous risk evaluation mechanism that it needs to be done and we need to assure that it’s approved in a safe manner. That patients are properly informed about the risks and benefits and that anyone getting the drug be entered into a registry where we really understand it. We can’t wait 10 years. If we find the rate of hemorrhage was 3% in a study and is 20% in real life practice, we need to know that soon. We can’t wait to learn that.

Eric: Wow. Ken Covinsky agreeing that the FDA should approve…times are a changin’ [laughter]

Jason: We won’t get fooled again, right Ken? [laughter]

Ken: No, I mean… Yeah, with reluctance, [laughter]

Eric: Like the Oscars, Alex is starting to play in the background.

Alex: (singing)

Jason: Nice.

Eric: Jason, Ken, thanks for joining us on this podcast. That was absolutely fabulous. I learned a lot, as I always do from both of you.

Jason: Well, thanks a lot. I really enjoyed it. Good to be here.

Ken: Thank you. Fun being with you and good seeing you, Jason.

Eric: And to all of our listeners, thank you for joining us on this GeriPal podcast.

Back To Top