 It's my pleasure to introduce Dr. Ellen McVail from the Mayo Clinic, who's going to talk to us about pathology and biopsy. For those of you who've heard the term mass spectrometry here, the Mayo Clinic remains the national treasure and resource for all of us because every time we want to know exactly what the precursor protein is, we send tissue to them and Dr. McVail is always very helpful in sorting that out, so we're all very grateful. Thank you, Matt, for that really nice introduction. Can you all hear me okay? Yeah, okay, well, I'm very, very pleased to be here at this really wonderful and comprehensive meeting, and so I come from Rochester, Minnesota. I am a hematopathologist, so I'm not a cardiologist, I'm even a cardiac pathologist. But since about nine years, I've been part of the mass spec lab and I've been the mass spec medical lab director since 2017. And so since that time, I've seen a lot of cases and I've actually learned quite a bit about the whole amyloid world. So I hope this is helpful and we'll just get started here. So no disclosures. So just as a background, amyloid osis involves basically any organ in the body. And there are now 42 different types that have been described. If you look at the most recent description now, we used to be at 36, but we're up to 42. And they involve basically every organ you can imagine. But fortunately, heart is relatively simple. Because although we have described, we wrote a paper about two years ago. We went and looked at our whole mass spec experience at that point, which was about 16,000 cases. And then we looked at each type separately. So this description here, as you can see, is the heart. And you can see that although we had nine different types at that point, in fact we're up to 10 now, 98% of them were either TTR or AL. So in some ways, heart is really, really easy. Because basically, that's the question. There are other types, the next most common type is ANF, which is a localized disease. And that only involves atrium. So you really don't have to worry about that either. So basically 99% of cases will be TTR or AL. And it's also important to remember that aside from ANF, these are all systemic diseases. So all the patient may not have symptoms in certain organs. Those organs may be involved. So for instance, TTR patients may have prostate involvement, may have gallbladder involvement, may have involvement. And basically any other organ, it just may not be symptomatic. But that's a place where you could possibly go and look to make the diagnosis. And then I think as has already been mentioned, cardiac involvement is the single most important prognostic marker across all amyloid types. So although other organs may be involved and may have some symptoms, the heart is really where the money is for prognosis and outcome. OK, so a couple points I'm going to make today. This may not be maybe controversial. We'll see if this brings up questions. But from my analysis, most patients should not meet a heart biopsy. So we'll start out with using the numbers from the paper that we wrote in 2020. Now, I think TTR patients are actually a lot more common than this. But because they come to diagnosis later, because they're older, because the disease is more progressive and indolent, there tends to be subclinical disease, they don't come to attention, are probably underdiagnosed more than AL is underdiagnosed. But we'll use these numbers just for argument's sake. So about 2-thirds are TTR, 1-third are AL. For the TTR patients, about three-quarters of them will not have an M protein. Again, the numbers vary in different studies. It could be higher, it could be lower. But we'll use 75% as an estimate. And for those patients, as we already heard, the centigraphy scanning should be diagnostic of TTR, and there should be no need for any tissue biopsy whatsoever. About a quarter of the TTR patients will have an M protein. And for them, as we've heard, centigraphy is not indicated. And for the ALs, pretty much all of them should have an M protein. And again, centigraphy is not indicated. So we'll go on. So for those patients who have TTR and have an M protein, those patients will have a TTR present in a surrogate site such as fat pad and bone marrow about a quarter of the time. And similarly for AL, for those patients, again, who have an M protein, so you can't do PYP scan, about 75% of those will have an M, sorry, amyloid present in either the fat pad or the bone marrow. And again, these surrogate sites are overall quite reliable. Not 100%, but overall quite reliable. So that only leaves the 75% of the TTR patients who have an M protein plus the 25% of AL patients who absolutely need a heart biopsy. So if you do the math, this winds up being about 20% of all cardiac amyloid patients. Now the numbers are all estimates, so maybe it's 30%, maybe it's 15%, but it's certainly under 50%. So that leaves most patients, again, should be diagnosed without the need for going to the heart. So as I mentioned, patients, surrogate sites are often acceptable as a surrogate for going to the organ of interest. So again, in AL patients, about 75% of fat aspirates and 70% of bone marrow will be positive for AL amyloid. Similarly for TTR, about 15% of their fat aspirates and 30% of their bone marrow will be positive for TTR amyloid. And then if those sites don't work, there are other sites that have been used with variable success. So lip biopsy is actually a pretty good site for AL. One paper I was looking at showed that about 50% of patients who had a negative fat, 50% of AL patients who had a negative fat and bone marrow did have lip biopsy positive for AL. Again, Tensinovion is good for TTR patients. However, I would not be convinced that a patient who had TTR in their, say, carpal tunnel couldn't possibly have AL in their heart. So I wouldn't use that as a great site for a surrogate. I think it's an indicator, but it doesn't prove it. And then people do use all sorts of other preexisting sites, such as prostate bladder, gall bladder, and GI tract. Whether we should use that or we shouldn't, it's hard to say, but people often will go back to preexisting specimens and see if they can make the diagnosis without having to subject the patient to a heart biopsy. So the next point is about pathology. So if you do need a biopsy, which is gonna be, which is gonna be at least half the time we need a biopsy of some sort, make sure it's reviewed by an experienced pathologist who can confirm the diagnosis. So you'd think this would be easy, but in fact, it's not. So again, the same problem that plagues clinicians, plagues pathologists, is that you have to think of it in the first place. It's not so hard to think of it. If someone tells you that's the question, so for instance, if someone tells you, this is a heart biopsy, do they have amyloid? That's not so hard. What's hard is when you have a patient who has diarrhea as to GI biopsy, because the differential for diarrhea, as you all know, is quite broad. And although amyloids can certainly present with diarrhea, so can a lot of other things. So again, you gotta think of it. And so you need a specialist who has encountered this and knows how to deal with it. The next challenge is the conga red stain. We all talk about that like it's just we do it and that's the end of it. And in fact, it is the gold standard. It's a cheap stain. It's been around forever. It's in some ways easy to do, but it's technically challenging. So everything has to be perfect. The pH has to be perfect. Everything has to be made up quick, freshly. The thickness of the tissue has to be correct. Everything has to be correct for this test to work. Otherwise, you want it with false positive and false negatives. And another problem that we came up aware of recently is that we used to think it was 100% specific for amyloid. In fact, it's not. So the rare diagnoses like fibrillary gluminal nephritis that are conga red positive, it's a fibrillary disease, but it's not amyloid. So again, you need to make sure someone who is an expert is looking at this. And then the next challenge, okay, now you know it's amyloid. What type of amyloid is it? And of course, as this audience knows, the treatment for AL and TTR is so dramatically different that you really want to get the answer right. So, most pre-mass spec, most methods were antibody-based methods. So this included immunohistochemistry, immunofluorescence and what's called immunogol, which is an EM-based technique. And these are all great techniques for a lot of things, but amyloid is not one of them. They're problems with amyloid. That's true. I mean, this is our, the weeds are our workhorses. I see, I mean, it's a chemistry, that's all we do. Basically, that's the workhorse for a backbone of our practice in pathology, but not so much for amyloid. So there's some problems. The first problem is there's a bias toward expected types. So as I mentioned, there are 40 different known types, but nobody out there is doing 40 different antibodies when they come along on an amyloid case. So unless your case is one of the three common types, this is gonna be a problem. So you either will miss the type it is or even worse, misdiagnosis as a type, one of the common types. And then the other problem is the chemistry is non-specific. Again, this isn't a problem for most areas of pathology, but it is for amyloid. So we end up with a very high inclusive rate and a lot of cross-reactivity. So we think that optimal patient care requires a better method. And so it was actually this problem, this very problem that prompted our institution to develop the mass spec test for amyloid. So this was the first test. We launched the world's first clinical test to type amyloid by mass spec based shotgun proteomics back in 2008. And it was developed to solve this precise problem. And it's now considered the gold standard for amyloid typing, now performed worldwide. And so we have, we absolutely did our IHC for all the amyloid proteins back in 2008 because we realized this worked so much better. And why it's a great test, it directly analyzes the proteins of interest. It's unbiased. It unambiguously identifies all amyloid types in a single assay, very, very high sensitivity and specificity. So sensitivity is about 98%. The couple percent we miss is usually because there's not enough amyloid present to do the test. And then specificity is about 100%. It uses a paraffin tissue, so you can go back to specimens that are 30 years old and you can type the amyloid. And it requires very, very little tissue. There are a few cases we can't quite get enough, but they're very, very rare. So even tiny heart biopsies and tiny kidney biopsies, this works fine. So just to give you a little background or maybe my sort of simplistic way of thinking about amyloid. So there are components of amyloid. There are the amyloid signature proteins, which I think of as the straw. So again, that's the stuff that's, they're ubiquitous proteins in the serum that are soluble and they're just there. And so like straw, they can kind of just move freely. And then there's the unique amyloid precursor protein. That's the mud. So again, that's soluble and it's, on its own, is not a problem. But these are not normal proteins. They're abnormal in some way. And when these get combined under the right conditions or the wrong conditions, you wind up with the immovable brick that is amyloid. So here's an example. The amyloid signature proteins that we use are apolipoprotein E, apolipoprotein A4 and serum amyloid protein. So there are actually a couple other proteins that are considered amyloid signature proteins, but those are the three we use in our lab. And when combined with, say, an abnormal lambda light chain, you wind up with AL amyloid lambda type. Here's another example. Same amyloid signature proteins. This time the abnormal protein is an abnormal transthyretin. Put them together and you wind up with TTR amyloid. So here's an example of a case that we ran. I don't want to go through all the specifics of mass spec in the interest of time, but you can see that the image on the left is a blood vessel. This is looking at a conga red under UV light with a fluorescent microscope. So it looks different, it looks kind of bright red. So that red material is the amyloid. You can see a very, very fine sort of lavender line within that red, that's the line of a laser. So what we do is we take the tissue, put it on an uncovered slip slide, and we use a laser. Look at the, we do a conga red on it, and we use a laser to cut out just the amyloid that we can see. And that tissue falls into the cap and that tissue gets analyzed. So it's a very helpful to eliminate, to enrich for the amyloid and eliminate the other material. That's actually a key component to this test and why it works so well. Then we run the mass spec and we wound up with a personalized profile, which you can see here, which are the numbers. We run two samples, you get two columns, each one's a different sample. Just to make sure we're honest. And the numbers are what we call spectral counts, which are a semi-quantitative measure of abundance. So, we can read this. The top three here you can see those are our amyloid signature proteins that have high numbers. This is a way that we proteomically proved to ourselves that this is indeed amyloid. So it is. And the next most abundant protein is TTR. So this is a TTR amyloid case. And yes, there are a few rare gammas and alphas and lambas and capas, but that's background serum proteins. We can't get rid of them 100%. So there are always a few around, but the numbers are very, very low. And we know that's just background. And in addition, this test is kind of even better because beyond just doing our usual search to identify, say the amyloid type, say TTR, we can then use the same, use a different database, which is a curated mutation profile and look at the same information, run the search, and we can identify the mutation. So we do this in every case. So if you send us a case, we do this automatically. And it's not 100%, but it's probably above 95% that we can identify the mutation. And in some ways it's even better because it's disease tissue. So you could have an asymptomatic patient, a carrier who doesn't actually have disease yet, but have the carry the gene. But here we can show here that it is amyloid and it also carries the abnormal protein. So here's an example of a case just to show you the problems that we can face sometimes. So this is a 74 year old man, had a heart biopsy, had amyloid, no problem, gets typed, I don't know if you can read this. This was done somewhere else, actually a very, very good institution. They used amylo fluorescence, which is actually probably better than IHC. And they came with a diagnosis of ALKAPA. And then 31 months later, the very same heart biopsy, so very, very same specimen came to us and they're this signature protein, so we know it's amyloid. And it was TTR. So, beware. So the last point I want to make is although I have already told you that syracous sites are really good, they're not perfect. And so if the clinical picture doesn't fit right, then you're gonna have to go to the endoth organ, which is the heart. So I'll give you three. The first example are rare amyloid types. So again, these are super rare. ApoA4 is probably the most common one that we deal with. That's a problem sometimes because I think even that can give a, maybe Matt, you know, but it can give a false result using PYP scanning, but I think this is anecdotal at this point. So that's a problem, but it's extremely rare. The second problem is discordance. I'll show an example of that. So here's a case, 73-year-old man had a longstanding type 1 diabetes history. He had symptoms that sounded good for systemic amyloid. He did have an emgus, a capyrelight chains and a cap of emgus. He had a fat aspirate, which was positive for amyloid. His bone marrow had 5% monoclonal capoplasma cells. So at that point, the working diagnosis was systemic AL amyloidosis, quite understandably. So he had his fat aspirate typed, and we see the amyloid signature proteins. So they're present, so we know it's amyloid. But what we found was insulin. So this is not AL, this is not TTR, this is insulin. So, an insulin amyloid is actually the fifth most common amyloid type that we see. It's about 1% of cases. This is not a systemic disease. It's a localized disease to where people inject insulin. It causes decreased absorption of insulin. It causes masses usually, but it's a localized process that basically is a red herring. But if you go back to our paper of all our fat pad biopsies, although 60% were AL and 30% were TTR, a full 10% were insulin. So this is not that rare. So it's not unheard of that you will encounter this. So this happens, and basically it's a red herring. So our clinicians knew that, and they said, yes, this patient still could have cardiac amyloid. And so a heart biopsy was done, and it turned out to be TTR. So, in the end, this patient had three things. He had the main thing, which is cardiac amyloid TTR, wild type. He had an incidental localized insulin amyloid, and he had an incidental MGUS. So, yes, there can be just, people can have more than one amyloid type, either in different organs or the very same organ. And we've written about this, other people have written about this. So is it common? No. Does it happen? Yes. So don't be surprised and just keep going if the story doesn't make sense with the surrogate site. And the last example, okay, I forgot about this. So this is a paper we had written about whether you could use surrogate sites like prostate or bladder or tendon as surrogate for cardiac amyloid. Because if you think about it, who gets cardiac amyloid? Well, it's kind of old guys, right? So the old guys often get prostate biopsies and bladder biopsies. So this was kind of an interesting study. And so we had 16 cases that had, or we had a heart biopsy plus one of these sites. And 14 out of 16 were concordant, which is great, but two were not. And so it's important that you know about these. So the last one, which I think is 29, was discordant. This first it had a type of amyloid called seminal gelin, which is localized to seminal vesicles in his prostate. And this has always localized disease. So this one did not pose a problem because you would know that there's no way that's what was in the heart. And the patient had, I think AL, I can't read it from here, AL in the heart. So that's fine. There's no way that could have been the same thing. The other case, case 10 was potentially a problem because he had AL in his bladder and TTR in his heart. So AL can be, in addition to being a systemic disease can be a localized disease. And localized disease, AL in the bladder is a relatively common clinical pathologic entity. So again, this is an example where using the surrogate site didn't work. And the clinical picture didn't fit and the clinicians were astute and went to the heart and diagnosed a separate type there. So this is just to keep in mind that this can happen. Usually it doesn't, but it can. And the last case, this is the one that gave me endless gray hairs. This was a 91-year-old guy who had prominent, I won't go into the whole story because we're out of time here, but he had prominent interstitial amyloid. 98% of these big blobs of interstitial amyloid, which is a typical pattern for TTR. And so he typed it. It was interstitial aggregate. And again, it was TTR. But the patient didn't behave like TTR. He was treated for TTR and he basically fell off a cliff clinically. And so we went back and sure enough, if you really hunt hard enough, you could find rare vessels that had amyloid that looked different. The other interesting thing was this patient had a fat aspirate, which is positive for AL amyloid, which also posed a big problem because we think that fat aspirates as a perfect surrogate. So how could he have a different type in his heart? So we went back to the heart and we found these rare vessels that had amyloid and we dissected those out separately. And they were so rare that we could only come with a single cap. But sure enough, we ran it and it was Lambda. So this patient, I don't know if you probably can't see this very well because the numbers are so small, but there was not a speck of TTR in those vessels. And similarly, there was basically not a speck of AL in the interstitial areas. So they were two entirely different amyloid types, two different anatomic compartments in the same heart. But even though 95 plus percent of it was TTR, he was behaving like AL. So we did get treated for AL and he improved. So again, the amount of involvement of the tissue is not necessarily proportional to what the patient's gonna do clinically. So this is extraordinarily rare. Like this is about the only case ever seemed like this, but if it happened once, it can happen again. So again, I think the surrogate sites are excellent, but if the clinical picture doesn't fit, you have to go back to the heart and again, send it to a very good pathologist because these are hard to diagnose. So these are my take home messages. For one, most patients with cardiac amyloidosis should not need a heart biopsy and about at least half should not need a biopsy at all. For patients who do need a biopsy, surrogate sites like Fat Aspirin, bone marrow are often acceptable. If you need a biopsy, send it to a good pathologist and we recommend typing it by a robust method like mass spec. And finally, surrogate sites, although very good or not perfect. And if the clinical picture doesn't fit, get a heart biopsy. So that's it. Thanks for attention.