​Dr. Gerry Lee (Host): Hello, everyone, and welcome to another episode of Allergy Talk, a roundup of the latest in the field of Allergy and Immunology by the American College of Allergy, Asthma and Immunology. For today's episode, we'll be reviewing three more articles from Allergy Watch, a bi-monthly publication, which provides research summaries to college members from the major journals in Allergy and Immunology. And by listening to this podcast, you can earn CME credit. So, to get more information about that, head over to education.acaai.org/allergytalk. And we also have the ACAAI Community on Doc Matter where we could continue the discussion about today's articles.

Well, hello again. My name is Gerry Lee. I'm an Associate Professor at Emory University and an Assistant Editor of Allergy Watch. And once again, I'm joined by the Editor-In-Chief of Allergy Watch, Dr. Stan Fineman.

Dr. Stan Fineman: Hello, everybody. It's great to be here. I'm also an adjunct faculty with Emory and a past President of the College of Allergy.

Host: And for the third chair, we have the Associate Editor of Allergy Watch, Dr. Shyam Joshi, an Associate Professor and Section Chief at Oregon Health and Science University. Shyam, welcome back to the podcast.

Dr. Shyam Joshi: Thank you so much for having me. It's a beautiful day in Oregon in the Pacific Northwest.

Host: And it's raining here in Atlanta. So, that's how it goes. I think we'll be okay. So Stan, let's get started. I think we always think about bacterial infections with IVIg, but have some data maybe about viral defense. What have we learned?

Dr. Stan Fineman: So, this was an article that was published in the January 2025 issue of the Annals by the group from Johns Hopkins. They did a retrospective cohort review of two of their sites. one was the Johns Hopkins Hospital, and one's called Johns Hopkins Bayview Medical Center. And what they did is they looked at at the 270 patient admissions that were hospitalized for viral infections over a period between 2011 through 2016. So, this is pre-COVID. So, COVID doesn't play a role in this.

So, of the 270 patients, half of them were female and three-quarters of them were transplant recipients. And that may have been, obviously, one of the reasons that they were immunocompromised. And in that group, in 97 patients, who received the IVIg, their doses varied. The varying doses were between 200 and 1800 milligrams per kilogram per dose. The mean dose was 500 milligrams per kilogram. We're going to talk about that in just a minute, how they decided upon the dosage. But the ones who did receive IVIg during the hospitalizations were associated with a reduced length of stay, a lower ICU length of stay of about half a day compared to those who did not.

But on a secondary analysis, which is more interesting, I think, is the fact that the group who received the IVIg within the first 48 hours of being hospitalized, had an even shorter stay in the ICU and shorter hospitalized stay of at least two days. So, the key here is the supplemental IVIg reduced your hospital stay. But if you got it in the first two days of being hospitalized for your respiratory infection, then it was more significant reduction of two days. So, again, this is retrospective, so that's part of a challenge. And the fact that they are not sure about exactly the mechanism, these are viral infections, not bacterial as Gerry said. And the question is somebody said, "Well, could it be passive immunity?"

And there was an interesting editorial in the same issue from Dr. Harville who's at Arkansas, who did talk a little bit about could it have been the numbers of viruses? And do you need to give higher doses of the IVIg? And what he was pointing out was the fact that 0.5 grams or 500 milligrams per kilogram in a single dose may not be enough. And they might have gotten a better effect if they used the one gram per kilogram as a single dose. And part of the reason he says that is because of viral neutralization, because he's calculating the number of viral particles and things like that. This is sort of speculative.

But the bottom line is I think if you have somebody who is immunocompromised, and of course, most of these were because of transplants, then you need to give a supplemental IVIg within the first 48 hours to have some significant effect. And as an aside, even though it wasn't totally studied in the article, I think if you were looking at the editorial, if you were to use a higher dose, like not 500 milligrams, but let's say go to a whole 1 gram per kilogram, you might even see a better effect and a reduction in length of stay.

Host: So Stan, these are transplant recipients, I mean, mostly, but were they hypogam? Do we know anything about their pre-ICU IgG levels?

Dr. Stan Fineman: It didn't talk about that. This was retrospective. They didn't really look at their pre- levels. At least, they didn't report on that. And the fact is, I think, in a hospitalized setting like Johns Hopkins where they do a lot of transplants, it's not that surprising that three-quarters of the patients with immunodeficiency or immunocompromise are going to be transplant recipients.

Host: Well, I shouldn't say that I'm not also surprised they weren't able to get the data on IgG on those patients. Because in my clinic, we have lots of patients getting immunosuppressive treatment and we don't have a baseline IgG beforehand. I think we see that a lot. And I think it's just maybe you wonder how closely we are monitoring the immune function of patients given pretty strong medications. If we know that they can compromise B-cell function, I think we always get those consultations after the fact, but we always wanted to know, well, what about the initial monitoring and so on? And so, you wonder not only the IVIg for those who need it, but the actual being proactive about identifying patients at risk even before we even get to the ICU. I guess, that's the point I was trying to make about this. I'm not saying that that's what's happening, but that's the data that would make me feel more secure, that maybe this could have been even better prevented with a more proactive approach on IgG monitoring. I don't know, Shyam, if you've seen these secondary hypogam patients too.

Dr. Shyam Joshi: I see so many of them, and it's not just checking their IgG level, right? A lot of these medications that these post-transplant patients receive are T-cell-modulating, right? They're affecting T cell function. And we know that T cells are absolutely essential for appropriate B cell functioning. And even if we are checking their immunoglobulin levels and they are slightly on the lower normal end, just because they're on these T-cell-modulating medications, I'm sure that that's affecting their B-cell function and antibody production function. And having this surge right at the beginning kind of makes sense.

Dr. Stan Fineman: Yeah, I don't think that there's a downside. We all know about some of the side effects of giving IVIg. But I think that, in this case, most of these patients were in the ICU for the respiratory illness. So, there was a little comment in the article about the cost effectiveness, and obviously, that's going to improve. The fact that you can reduce your length of stay is going to be even more cost-effective.

Host: I think overall, as immunologists, we'd love to get involved in these cases if they want advice on prevention. Obviously, once they're in the ICU, we're probably not going to be seen in the first 48 hours as a consultant, honestly. But certainly, I think we play a role in the team, if there is recurrent infections. And us being part of that can definitely, perhaps down the road, prevent some of these more severe events, maybe. I would say that at least we're caring about it and we're going to do the due diligence to look for it. So, I think that's another way the immunologist can play a role. So, thanks for presenting that one, Stan. And Shyam, you got one more about the role that allergists can play with asthma by treating the whole patient. So, what other interventions can we do to help the asthma patient?

Dr. Shyam Joshi: Yeah. This is a great article out of CHEST, and this is a quick plug for Allergy Watch as well. CHEST is not a journal that I previously regularly read. But as one of the editors of Allergy Watch, this is one of the journals that was assigned to me. And CHEST has some fantastic articles, really high quality articles, especially in the asthma world. And so, this was by Sharma, et al. It's a group out of the University of Glasgow, and they did a one-year weight management program for difficult-to-treat asthma with obesity.

And so, we've known for, well, decades now that asthma has various phenotypes. We know that there's this comorbid obesity phenotype. And even within that obesity phenotype, there are tons of subcategories, right? There's those that have obesity and ATP, there's those that have obesity and eosinophilia. But this group seems to act a little bit different than our classic pediatric asthma patient. And we know that many studies have shown that even as little as a 10% weight loss in this subgroup does improve outcomes. And we've generally seen more improvement in asthma control tests, patient-reported outcomes, but there's even some data showing that you get an improvement in lung functioning, FEV1, FEC.

So, the question always comes down to is how do we successfully do this? As any clinician knows, having this conversation with patients about weight loss is a difficult conversation to have, and there are various approaches to try to make it work. But it's a hard thing for patients to do. And this has come to the forefront again because of the increase in use of the GLP-1 agonists. And I reviewed another article, or maybe two articles over the past year in Allergy Watch on GLP-1s and how they can play a key role in weight loss, but also in asthma outcomes.

So when we we talk to patients, there's really kind of three main ways we approach weight loss now. We have GLP-1s, we have diet and exercise, and we have bariatric surgery. As we've seen more people use the GLP-1s, we've also seen that some of these patients also have more side effects too. So, it's important for us to continue to have these evaluations of: Are there other ways we can encourage weight loss?

So, this was a follow-up study. the initial study by the same group, Sharma, et al, they looked at this Counterweight Plus Program. It's a CWP program at 16 weeks, and they showed a clear improvement in patients' ACQ scores over that 16-week period. So, this Counterweight Plus Program, it's a pretty intense program. It's not something that we can just tell patients, "Oh, go follow this and you're good to go." But it is a really good method to evaluate true weight loss and a true change in obesity. But it's a three-step program that you go initially to a liquid diet, which sounds terrible, but it's kind of part of the process. And then, you slowly add on more items, and so you're really kind of starting from scratch and building a diet from scratch instead of trying to pull things back, which is just psychologically much harder to do. And so, from their original study, they had 29 patients that completed the full one year part of this.

Interestingly, this is obviously a pretty small study. It's only 29 patients, so it wasn't powered to show statistically significant changes. And they found that the primary outcome, the ACQ-6 didn't show a statistically significant difference between the group that were on the diet versus the group that were on placebo. But they did show the group that were on this CWP program had a 15-kilogram weight loss, versus those in the placebo had no weight loss. But when you dig into the numbers a little bit more, that's where it really gets interesting here. It's that while the ACQ-6 wasn't statistically significant, it was clearly trending that those patients on the program had much better scores. And they looked at some of the secondary outcomes. So, one of the main secondary outcomes was the asthma quality of life questionnaire. And they showed a clear statistical improvement in those on the program versus off the program in all the domains and symptoms and emotional wellness and activity.

And so while the initial study, the overall study is technically a negative study, almost all the secondary outcomes were positive for the weight loss program. And so, why I wanted to highlight this is, yes, we have biologics now, which have changed everything. Yes, we have GLP-1s out there that are helping people lose weight, but we still need to look at the patient as a whole and that there are some really, really great weight loss strategies that are being looked at and that we should continue to investigate because there are patients that could really, really benefit from this and can hopefully get off a lot of the extra pharmacologic therapies that we're giving patients.

Dr. Stan Fineman: So Shyam, I have a question about the study. Did they mention anything about exercise? Because I'm familiar with some of the programs that have instituted exercise in some of these patients who are overweight, and that's shown some improvement in their asthma scores. But could you comment on that?

Dr. Shyam Joshi: Yeah, it wasn't the primary part of the program. The primary part of the program was very diet-oriented. But obviously, they encourage exercise, and it was important. And they had an exercise regimen in both groups. So, it was part of the program, but really they were looking at the diet as the main differentiating factor.

Host: I think that's a good point, Stan. Honestly, we should try to tackle both that kind of go hand in hand. Obviously, weight loss is principally through diet, and weight maintenance is principally through exercise. But Stan, you're probably referring to our previous episodes where Sarah Spriet reviewed that exercise and its relation to asthma outcomes.

So, I mean, yeah, I think we do have to be addressing both sides of the street. The question is the best way to do it. And obviously, Shyam, you told me about some great options. Clearly, we all don't have like a juice bar in our back where we could put someone on a pure liquid diet honestly, but--

Dr. Stan Fineman: I wouldn't want to take one of those anyway.

Host: Yeah. I don't know, there's all these like juicing things or whatever. But the principles stayed the same, right? Like clearly, I want other options than medicine to treat the disease. And certainly, something as simple as diet and exercise not only helps asthma. But I don't know, most health issues that afflict people like cholesterol and heart disease and brain health and so on. I think overall, we could sell it as an intervention, but it's basically a holistic intervention for most people. And I think people do appreciate us trying to encourage them to give them options to treat their disease other than here's another drug.

Dr. Shyam Joshi: There are going to be some patients that are that motivated, that I've seen come off of inhalers because they've lost 30-40% of their body weight and having more of this data and having more programs that are even-- I think one thing that I always find difficult talking to patients is they're often like, "What kind of diet?" And I'm like, "I don't necessarily have the time to go through all of this with you right now, but if there are programs out there that have shown that this diet or this program can work with asthma improvement. I think there'll be some avenues for us to use in the future to refer patients, to be like, "Yeah, this is a program that has been studied, this does work. Let me connect you with them and they can kind of work with you on diet aspect of things, and the same thing for the exercise."

Host: Okay. So, that seals it. So, the green zone of the action plan, in addition to exercise, now we're adding a healthy diet. And that's going to go into that green zone, and we're going to codify that in the action plan. So, for the reminder, Shyam. I am absolutely going to put that routinely in my instructions. Admittedly, I could do a better job than that, and it's absolutely something that I think patients will be very receptive to because it is a non-medicine pharmaceutical option. This is general encouraging healthy living.

So, I got one article to review before we close it out here, and this is discussing different mast cell activation syndrome criteria. This is coming out of JACI, and the title of the article is Clustering of Clinical Symptoms Using Large Language Models Reveal Low Diagnostic Specificity of Proposed Alternatives to Consensus Mast Cell Activation Syndrome Criteria.

So, mast cell activation syndrome, as you know, has been getting a lot of attention amongst physicians and patients. And I think this article tries to look into how there has been a rise in the assignment of this diagnosis in the past 10 years. Now, the original description of mast cell activation syndrome was to describe a syndrome that resembles what systemic mastocytosis patients experience where patients have actually diagnostic proof of an overproduction of mast cells. And the original mast cell activation syndrome criteria back in 2012, Valent, et al, was that you had to have two-organ system involvement of known mast cell activation symptoms of which they define 18 of them. You had to have some sort of biochemical evidence of increase of the tryptase. So again, you obtain a baseline, and then during an episode repeating of the tryptase measurement meets the threshold. As you know, we've reviewed different thresholds. There's the famous 20% increase in baseline plus two. There's the NIH tryptase score. There's the 1.685 multiplier, that's another criteria that's used. Again, there's multiple criteria to use. But again, you have to show baseline versus increase. And then, of course, response to therapy, which I think a lot of people have used as well.

 Now, when we think about the categorization of mast cell activation syndrome, the classic framework is the people who don't meet mastocytosis criteria. So again, they have the KIT mutation, but they don't have the aggregates in the bone marrow, obviously secondary MCAS. Just basically mast cell activation due to a trigger. We're very familiar with that, right? Food allergy anaphylaxis. There's this new one called HαT, so you have mast cell activation syndrome and demonstration of an increased copy number of tryptase. And then, of course, there is the idiopathic version, which I think a lot of people are being diagnosed with.

But the previous Valent criteria is pretty rigid, because we want to be specific to the diagnosis. I mean, sensitivity is very important. We don't want to miss anybody, but we want to be specific because we don't want to have a false positive diagnosis. We don't want to, again, assign a diagnosis that doesn't explain what's going on. But I think a lot of practitioners have felt that the current criteria is too rigid. And so, there was a group back in 2020 who published something called the consensus-2 criteria suggesting some chronic symptoms, less severe symptoms are meted by mast cells. And so if you look at the consensus-2 criteria, they have this table one in their original description of the article, which lists 185 different symptoms that are potentially consistent with mast cell activation syndrome. And this podcast is only between 25 and 30 minutes in length. I'm not able to list all 185 symptoms, but I can give you some examples.

So, some of the things in here, and I'm just looking at the article, would be fatigue, hyperthermia and/or hypothermia, sweats, flushing, rashes and lesions of many sorts, irritated often dry eyes, lid tremor or tic, infectious or serous otitis media or hearing loss, pain or irritation of the mouth, adenopathy, airway inflammation on all levels, presyncope, atherosclerosis, heart failure, allergic angina, dyspepsia, reflux, luminal and solid organ inflammation, interstitial cystitis, migratory bone and joint or muscle pain, tissue growth or developmental abnormalities, sensory neuropathies, dysautonomias, mood disturbances, attention deficit, abnormal electrolytes or liver function tests, nutritional deficiencies, polycythemia, anemia, and the list goes on. So again, that was only a small fraction of the 185. I just wanted to keep this podcast under 30 minutes. But essentially, because of the wide variability of symptoms you could imagine, it's very easy to meet consensus-2 criteria. Essentially, you just had to have symptoms, right? Major criterion. And then, the second, you have one major criterion, which is symptoms, and the secondary criterion includes stuff like response to therapy. So again, if someone has one of those constellation symptoms, they respond to mast cell therapy, they got it.

And again, I think some of the criticisms from some of the allergy community would be, obviously, we know the baseline placebo rate for most therapies is around 20%. And so, we worry about potentially an incorrect assignment of diagnosis. And interestingly, if you look-- and the authors report out data from California from a period from 2016 to 2022-- there was a significant of increase of 12.6 times the baseline of mast cell activation syndrome diagnosis. So again, there was a 12-fold increase in diagnosis rate, partially attributable to looser criteria. So, how do you really prove if criteria is overly generalized?

Well, one way the authors attempted to address this is to use large language models. So, you may be familiar with some of these AI models, ChatGPT, Claude, Gemini, or so on. Basically, as you know, natural language processing takes words and trenches it into data. But then, when you have words as data, you can do tests on them to look at similarities, right? So, what large language models are able to do is they're able to determine if things are similar or things are different, right? And so, one example is a large language model could assign a value to king and queen versus king and car. King and queen are similar. And king and car are different. And so, how is this relevant? Well, you want to know if two different criteria are similar to a diagnosis, unknown diagnosis, or different to diagnoses. And then, you can tell a large language model, okay, if I take symptoms from this criteria, either the 18 from the consensus criteria, the what we're going to call consortium criteria, or the 185 from the alternative criteria, what kind of diagnoses are they consistent with? I.e., as you know, we use large language models to try to make diagnosis and the accuracy has been improving, so that's what they have attempted to do here.

So, the first thing that they do is they try to say, "How does this compare to other control diagnostic criteria?" So, they use stuff like migraine criteria. They have two different lupus criteria. And they say, "Okay, why don't you generate a differential diagnosis if I plug in two or more symptoms from this diagnostic criteria?" And so, from the migraine criteria, you get stuff like migraine, obviously, but then it'll include something differential like brainstem stroke syndrome, malignant neoplasm, multiple sclerosis, suggesting that, again, there's some overlap between different things. If you put in the Kawasaki criteria, the most common diagnosis is Kawasaki, but sometimes you'll get scarlet fever and mono.

But the bottom line is that when you plug in these criteria, there's certain diagnoses that are very prevalent and some that are very less common. They do it like a propensity score. When you compare the consortium standard, the lead criteria for mast activation syndrome versus the alternative mast activation criteria, the number one diagnosis in the differential diagnosis when you apply MCAS consortium criteria is anaphylactic shock, T78.2 unspecified, and that had a very high propensity score. And the other ones are present but lower on the differential by orders of magnitude.

The number one diagnosis identified if you apply two random symptoms from the alternative criteria is Sjogren's syndrome. It says Sjogren's, sarcoidosis, hypothyroidism, chronic fatigue syndrome and lupus. But as opposed to one diagnosis, outnumbering all the others, all of them had the same propensity. And the actual number and diversity of diagnosis you can generate by applying a couple symptoms from the alternative criteria was vast. And at a higher diversity score that greatly outweighed, it's called a Shannon diversity score. The number of diverse diagnoses generated from the alternative MCAS criteria was much higher than the lupus criteria, Kawasaki, migraine and the consortium MCAS criteria, which had same diversity scores, i.e., they had the same number of diagnoses and variety of diagnoses, right?

So, you have this outlier here where the alternative criteria, if you apply the symptoms and ask a large language model, "Give me 10,000 different possibilities of what this is," they're going to find like thousands of different possibilities. When you take the same symptoms from the consortium criteria, you only get a smaller number. And again, the one leading diagnosis ______ that you get is unknown mast cell activation event, which is anaphylaxis.

So, what's the take-home here? We're finding that if you assign a diagnosis from mast cell activation syndrome based on the alternative criteria, it is possible based on large language model analysis of its specificity that you're going to misdiagnose somebody, because so many other conditions overlap with the 185 symptoms on the consensus-2 criteria. And therefore, the patients come, look up their symptoms, they get this answer because it meets the criteria, and then they seek treatments, and they're getting this red herring for what's really going on that's really going to help the patient. And unfortunately, that prolongs their diagnostic odyssey. And unfortunately, it leads to unnecessary testing. And again, we really want patients to get the treatment what they need. But when they have criteria that send them in the wrong direction, unfortunately, they're not getting the treatments that help them. And I think that that is a shame. We really do need to help people recognize what are the clinically accepted criteria versus the ones that, unfortunately, maybe seems like an answer, but unfortunately, maybe just be too broad.

And again, I presented this in journal club recently, if you plug in mast cell activation in, like, TikTok or some sort of social media where people get their information, the top 19 results were not from an allergist. There's only one allergist that comes up on a mast cell activation search, and I think y'all know who it is. It's Dr. Zach Rubin. Bless his heart. Bless your heart, Dr. Zach Rubin, for actually trying to give high quality information for the public for where they live, because it's very important that people get high quality information that's going to help them get the care that they need. So again, I think it was a very interesting way to tackle the problem and to show proof that, if you create criteria that are so broad that, again, you just generate just a litany of different diseases that could actually lead to the same common denominator of something, that does not mechanistically reflect what's going on.

Dr. Stan Fineman: Is this an example of where AI may be more of a challenge for us in terms of taking care of patients who come in with their list of symptoms and saying, "Look, this is what I found"?

Host: I think AI's garbage in, garbage out, right? So if we train the AI that the consensus-2 criteria is the accepted diagnostic criteria, then, yeah, the AI's going to say it. Because guess what? There's like so many different diagnoses that fit this very broad-- I mean, think of how many diseases are in the 185 symptoms I couldn't even get a chance to read to you, right? So again, I think it depends on how we're training these AIs and validating, and they're not hallucinating or giving misinformation. Garbage in, garbage out. We have to give it correct, high-quality, evidence-based information.

Dr. Shyam Joshi: We see this pretty frequently in our clinic as well. These patients have been on a journey for years and have taken every mast cell medication you could think of. And very quickly, we're able to determine, oh, they have paradoxical vocal fold motion plus maybe a flushing disorder or something. And patients are very appreciative of it, honestly. It is taking that few minutes out of the day to be like, "Yes, I've heard that other providers have diagnosed you with this condition, but let's take a step back. And you've been on this journey for a few years now. Let's see if there's something else going on." And it's there, like we see this day in and day out. I know providers are getting frustrated about getting referred for this, honestly, because we see that it's overdiagnosed a lot of the time. But I think this article that you brought up is a fantastic way for us to continue to really dig down to find what is the right way to approach this, because one extreme is maybe too stringent; one extreme, way too liberal. And trying to find that sweet spot in the middle is a challenge.

Dr. Stan Fineman: And I guess patients are going to try to diagnose themselves in a lot of cases. And that's part of the challenge too. And, unfortunately, a lot of providers out there are don't understand the whole mechanism. So, as Shyam said, they're going to get the wrong diagnosis.

Host: Yeah, ultimately, our goal is people get the help that they need. I have listened to the odyssey of many people who have come seeking answers to their symptoms. And again, when I'm able to contribute and help, I absolutely have pointed them in the right direction or, as you're saying, Shyam, found something that is an alternative explanation that does fixed alarming issue.

But I think if any of you have had a chance to see some of the talks by Thanai Pongdee about what the Mayo Clinic's doing, about mailing in urinary samples to look for mast cell activation. Really, we have to recognize that mast cell activation syndrome number one requires a delta, like a significant increase of the mediators during an episode, and I think that is the key piece of information that's going to be essential. Otherwise, without that, I'm just nervous that we're really not getting to the bottom of it. And so, again, I would encourage everyone to sort of look at some of his talks. He gave a talk to my fellows recently. And the Mayo Clinic has been a great resource for us. And again, they're going to have that kit available eventually, where again, as the time of recording, it's not available, but they're eventually going to have that mail-in kit for urinary mediators. So, I definitely will be a hundred percent using it for sure.

Anyways, I hope you've learned something from these articles. And maybe if you've had some advice on the approach on any of these challenging cases, please give us your feedback. That email is allergytalk, one word, @acaai.org. If you'd like what you've heard, please rate us wherever you hear podcast like Apple Podcast or Spotify. And again, don't forget about CME Credit. That's education.acaai.org/allergytalk. And we have other articles we haven't even discussed on this podcast. That website is college.acaai.org/publication/allergywatch. I enjoyed this very much. I'm so glad that you spent the time with us. And again, we'll catch up for the next one. Enjoy the rest of your day y'alls.

Disclaimer: The ACAAI is presenting this podcast for educational purposes only. It is not medical advice or intended to replace the judgment of a licensed physician. The college is not responsible for any claims related to the procedures, professionals, or products or methods discussing the podcast. And it does not approve or endorse any products, professional services, or methods that might be referenced.