I think I'm the only one who's not from Mayo, or has Mayo connections. Mayo ties, yes. Mayo ties. But yeah, I'm privileged to be here, actually. This is a great group. And hopefully, you get something good out of this. So my ask from Dr. Pennington is to make the case for antifungal prophylaxis for invasive fungal disease and lung transplant. This is what we're trying not to have. This is really what we don't want to have. None of us want to deal with this kind of situation. Relatively young, 58 is young, by the way. A relatively young woman who is post-lung transplant had developed graft-versus-host disease, which is a whole different situation. Developed persistent neutropenia. Started with airway disease, as you can see in this left panel here, with ulcerations and lesions in the airway. They had aspergillus, actually, on endobronchial biopsy. Despite dual antifungal therapy, decreasing immunosuppression, giving GCSF, it continued to progress and get worse and develop disseminated disease with brain aspergillosis and died from it. This is what we're trying to prevent. None of us want to see this. So we're trying to see if there is a way to prevent fungal infections and prevent this kind of terrible outcome. So what are we talking about in terms of burden of disease? So this is just a summary data from the TransNet study, which has 15 US centers from 2001, 2006. And this is the incidence of invasive aspergillosis post-transplant type. And you can see, obviously, that the highest column is really a lung transplant. Understandably, lung transplant has the highest incidence and highest risk for invasive pulmonary and disseminated aspergillosis. And the one year survival in this cohort was about 59%. So that's terrible. So if you do have invasive aspergillosis and lung transplant, or any solid organ transplant, mortality is high. The table on the right is a summary of multiple studies. And the reason I have this in there is the time. I want to show you the years, the span of years where this covers from 95, 98, 99. And you can see lung transplant rivals a lot of other organs in terms of mortality and incidence. So it's a big disease. It's a big problem. It has a high incidence, high burden, and high level of mortality. So it would be worth working and thinking about prophylaxis. This is a more recent study, just to complement the older studies. This is from 2020. This is from UPMC actually looking at prophylaxis with isavuconazole versus voriconazole. And it's a retrospective analysis of data. But you can see they were able to identify clear risk factors for breakthrough infections. So pre-transplant aspergillosis colonization, airway dehiscence, using basaliximab for induction, which means higher doses of corticosteroid maintenance immunosuppression, or high amount of red cell transfusion. All of these are clear risk factors for breakthrough infections post-lung transplant. And you can see the mortality is significantly higher in those who developed invasive fungal infection, mostly aspergillosis, 39% versus those who did not develop it. So again, contemporary data shows that the risk is high. We have some risk factors that we can identify. And in addition to mortality, there's also an added morbidity with longer length of stay, almost 100 days versus a month in terms of length of stay. And you can see on the graphs on the right, there's the IFI group versus no IFI and the clear difference in mortality. And then you can see those who received isavuconazole or voriconazole, that's the top lines, had actually a better survival than those who did not receive antifungal prophylaxis. But this is also a retrospective analysis. This is the scheme that we're thinking about. We're trying to lower the incidence of fungal infections without prophylaxis to with prophylaxis, hopefully for the long term, not just for the immediate post-transplant period. And so you could talk about either universal prophylaxis, which means prophylaxis for everybody for a period of time, three months, six months, I don't know. In this scheme, we're talking about six months probably. Or targeted prophylaxis, which means prophylaxis based on risk factors. And we already know a lot of these risk factors. So if you have airway dehiscence, if you're on higher dose corticosteroid, if you have colonization, if you have CMV disease, that's another risk factor, then maybe we should do targeted prophylaxis. And the point is, stop the transformation from colonization to infection. So we don't want infection, we don't want dissemination. If there's colonization, we're gonna try to take care of it in situ. So this is one of the earliest attempt to look at prophylaxis. So this is Dr. Hussain when he was in Pittsburgh before he changed his mind and adopted a different approach to prophylaxis. And in this study, they looked at people who received universal prophylaxis versus who had targeted therapy, meaning they had high risk factors for disease and they received Itoconzole or inhaled Amphotericin versus those who received Voriconzole universally. And you can see the incidence of invasive aspergillosis was 1.5 versus 23%. So that's a big difference. And then when they did a multilevel or multivariate analysis, Voriconzole prophylaxis had a very low odd ratio, 0.08, versus aspergillosis colonization 6.69. So this is essentially one of the earliest attempt to try to see if prophylaxis was helpful. Again, this is retrospective analysis, a small sample, a very small sample. This is a more recent meta-analysis systematic review of data looking at universal prophylaxis and incidence of invasive aspergillosis. This is heart and lung. Mostly lung, but there's also some heart and lung. And you can see there were six studies. The total number of patients that received prophylaxis, 425. Those who received control, in the control with no prophylaxis was 328. And the overwhelming number of the studies, except for one from 2012, showed actually favoring the treatment with reduced incidence of invasive aspergillosis with the odds ratio of 0.378 or 0.19, depending on the adjustment made. But I wanna call your attention to the heterogeneity index here. That's why I have it down here. It's 76%, meaning there's 76% variation in the data that would not be accounted for, which would explain a lot of this. That's why the confidence intervals are wide. That's why there's a question about how valid this data is. This is the same analysis, removing the heart. So this is just lungs. And again, you can see the top four studies, including Hossain's paper, were favoring prophylaxis. The Dofty paper from 2012 was right on the fence. But the overall analysis was favoring prophylaxis. But again, the odds ratio are shown there, and the heterogeneity index is very high. So what's the current practice in the United States? I mean, okay, we don't have much evidence, but what are we all doing? So this is two, the graph is from a study that Dr. Pennington put together in 2019, and you can see the majority of centers surveyed use prophylaxis, either pre-transplant prophylaxis or post-transplant prophylaxis. 97% use post-transplant prophylaxis in the United States. The majority of them actually use it all the way up to six months, three to six months. That's a lot of months of prophylaxis. That's the lowest bottom. Another survey, which is worldwide, about 58%, including United States centers where it's a higher proportion, also use universal prophylaxis. So current practice, there's a lot of prophylaxis. This is as close as we can get to data that are not put together in the retrospective ways or meta-analysis. Again, this is Dr. Pennington's work, very elegantly looked at real-world data from the United States using the OptumLab data warehouse where demographics, medical and pharmacy data are all linked, and there's methods to verify utilization of medications, actual outcomes. So this is the cleanest, closest data to do an actual study comparing. The first thing I want you to notice is the change over time. So from 2005 to 2018, you can see the rising number of centers that are doing prophylaxis versus not. Now it's two to one ratio. And so the total analysis included 662 lung transplants. The vast majority, 58%, did receive prophylaxis with a median number of days of 133, so about five months. And 275 did not receive prophylaxis. And this is the data. So obviously, Dr. Pennington's group looked at one-year mortality, and that's a good outcome to look at. And you can see those who received prophylaxis had a significantly lower risk of one-year mortality versus those who did not. And you could divide it by antifungals in general or mold-active prophylaxis. Either way you look at it, it was favoring prophylaxis. And then the study initially was done using the whole cohort, and then there was a propensity score match analysis to hone it down and clean up, I guess, the variation. And that data actually held up. So hazard ratio 0.54 favoring antifungal prophylaxis. 9.6 versus 18.4, so about twice mortality if you did not receive prophylaxis. What about incidence of IFI? Again, you have antifungal versus no antifungal or mold-active versus no mold-active. And you can see there is a trend towards reduced incidence of IFI but did not reach statistical significance. But when the propensity score was used to match the groups, actually there was a statistically significant difference favoring antifungal prophylaxis. Just to be fair, Dr. Pennington herself also published this analysis. And it's a systematic review of papers looking at the same question. And you can see there were either two studies or three studies, depending on how the questions was asked. And there was no significant difference between prophylaxis or no prophylaxis. I'm highlighting here the heterogeneity index of 94%, 93% in these studies, a little slightly higher than the other systematic review, but also explains the variation in the white confidence interval. So again, we talked about current practices. This is the guidelines. This is what our societies are saying today in terms of prophylaxis. ISHLT 2016, IDSA 2016, AST 2019, all three recommended or endorsed universal prophylaxis. AST, Infectious Disease Community of Practice, led by Dr. Shahid Hussain, favored targeted prophylaxis as a preemptive therapy based on galactomannan and BAL. This is where we are with antifungal agents. So by the way, I'm also an infectious disease doctor. So I like this because it shows me how many agents I have available. We have many more antifungal agents available to us to work with. And the safety tolerability is much higher. We can use agents that barely have any side effects. We have agents that are much more tolerated than amphotericin used to be, IV, or terbenafine, or ketoconazole. So we have higher safety, higher tolerability. We have increased interest in prophylaxis, as I showed you earlier in the study by Dr. Pennington, the ratio of prophylaxis versus no prophylaxis is now 2 to 1. But here's where the evidence is. We have very little to no evidence to support using universal prophylaxis for aspergillosis in lung transplant patients. So the question I was asked to answer, here are my answers and what I'm advocating for. I think having a common disease with invasive fungal infections, they're very common. We know risk factors. So we could use them to target prophylaxis if we don't want to use universal prophylaxis. It carries a high morbidity and high mortality in lung transplant patients. There is effective and safe antifungal agents available. We can use inhaled antifungal agents. There are inhaled azoles being developed. So that actually might be even better. And prophylaxis might actually prevent IFI in lung transplant recipients. It might. I don't know if that's true or not. So this is the stance we have. This is a picture from our picnic this summer. And the vote unanimously is for antifungal prophylaxis. And I have to point out the child in the front with his hand over his face. He's hitting his head saying, are these guys for serious? Are these guys serious enough? Do you want us to use this? And we have no evidence for it. And so goes evidence in medicine. So I think Dr. Pennington has a strong opinion. Thank you. Thank you so much for presenting that side of the argument. I am actually going to take a look at the other side of the argument here and take the position that lung transplant recipients do not require antifungal prophylaxis. I'm Kelly Pennington. I'm one of the lung transplant physicians at Mayo Clinic in Rochester. As you probably know by now, my interest is on fungal disease in lung transplant recipients. As I said, I'm here today to argue against the proposition that lung transplant recipients require pharmacologic antifungal prophylaxis. And I'm going to make my point using three key points. First, I'm going to talk about the efficacy of antifungal prophylaxis in lung transplant recipients. And then we're going to talk a little bit more about the cost of this. And then finally, I'm going to offer you a better way than utilizing universal prophylaxis in lung transplant recipients. So IFIs in solid organ transplant recipients are mainly caused by one of two species. Candida species are aspergillus species. If we look at all of solid organ transplant, candida species are the most common invasive fungal infection. They're responsible for about 5% of invasive fungal infections in all of solid organ transplant recipients. And they most commonly occur within the first month following transplant. We have more recent data from Baker and colleagues at Duke University and Marinelli and colleagues at the University of Toronto looking at the incidence of invasive candidiasis in lung transplant recipients. And we see that the incidence is probably somewhere around 10%. In contrast, invasive aspergillus infections tend to occur in the first six months following transplant with a little more than half of the cases occurring during this period of time. Although it can occur at any time following transplant. The incidence in lung transplant recipients is somewhere between 8% and 15% if we look at the single center retrospective cohort studies trying to identify what the actual case incidence is. When we traditionally talk about antifungal prophylaxis in lung transplant recipients, what we're trying to prevent is invasive aspergillosis. And in our lung transplant population, that usually comes in one of two forms. So that traditional aspergillus pneumonia that you see in other patient populations are myasotomas. But unique to lung transplant recipients, we also see tracheobronchitis. Why are our lung transplant recipients so much more susceptible to invasive aspergillosis than other solid organ transplant recipients? One is that our transplanted lung is in constant contact with the environment. So in immunocompetent hosts or those without structural lung disease, we all inhale fungal spores every day. But we have defense mechanisms to protect against that. These defense mechanisms are unfortunately not intact in our lung transplant recipients. So our barrier defense in lung transplant recipients is altered. We have reduced mucociliary clearance, an ineffective cough, dysmodial cilia, which gets rid of really this first line of defense that we have against fungal pathogens. We then have an airway anastomosis that can be ischemic and necrotic, which sets up a place where fungal spores can lay and then later invade, where they're met with reduced cellular immunity that we're purposely driving down to prevent rejection in our lung transplant recipients. So given the susceptibility of lung transplant recipients to invasive fungal infections and the associated mortality rates that Dr. Hajj presented that were as high as 60%, although I'd argue in the era of newer antifungals are much lower than 60%, it does lead us to ask this question. How do we prevent fungal infections in lung transplant recipients? As you heard from Dr. Hajj, he suggested that prescribing antifungal medications to all lung transplant recipients may be an effective method to prevent these infections. But to paraphrase Jerry McGuire, show me the data. We really have no data to suggest that universal prophylaxis is more efficacious. In fact, since 2013, we've had not one, not two, but three systematic reviews summarizing the data of anti-aspergillus prophylaxis in lung transplant recipients. And all of them have had the same conclusion, that there's insufficient evidence to support or exclude a benefit of antifungal prophylaxis in lung transplant recipients. This is a similar presentation or slide to what Dr. Hajj showed. But you can see that there's high heterogeneity between these studies. And many studies even report opposite results as to whether fungal prophylaxis is beneficial or not. This first study here looked at nebulized embosome and found that giving that to all patients versus no prophylaxis was beneficial. But then taking a very different approach, the second study here looked at voriconazole prescriptions for 12 months following transplant and found that that favored the no prophylaxis group. When we look at different prophylactic strategies, looking at universal versus preemptive, and making preemptive therapy predicated on either BAL, aspergillus antigen, or galactomanin, or fungal colonization, we once again find that there's really disparate results between these single center studies on whether or not fungal prophylaxis is even beneficial to our patients. So in conclusion, lung transplant recipients, we know, are at increased risk for invasive fungal infection. But the incidence being reported as high is about 18%. But we really have insufficient evidence at this time to support or exclude benefit from antifungal prophylaxis. So that makes my second argument against universal antifungal prophylaxis in this population, is that prescribing these medications to lung transplant recipients comes at a cost. And I don't mean just a monetary one. These medications have significant drug interactions. So the most commonly prescribed medications for antifungal prophylaxis in the United States right now are triazoles. These medications are potent inhibitors of the cytochrome P450 pathway, which is responsible for the metabolism of our primary immunosuppressants, our calcineurin inhibitors, so both tacrolimus as well as cyclosporine. Because of these interactions, immunosuppression levels have to be closely monitored at initiation, dose adjustments, and discontinuation of antifungal medications. Failure to do so puts the graft at risk for rejection, puts the recipient at risk for developing calcineurin inhibitor toxicity. In addition to these interactions, these triazole medications are not very well tolerated. The most common side effects that patients report is nausea and GI upset. Itraconazole and liquid posaconazole have difficulty with absorption. And voriconazole has a host of unique side effects, including photosensitivity, periostitis from fluoride toxicity, as well as an increased risk of skin cancer. I'm not even sure that universal fungal prophylaxis is even tolerated very well in our lung transplant recipients and is even possible. When we looked at our data at Mayo Clinic, we had 193 patients that were exposed to antifungals for the purpose of prophylaxis. And when we looked at how well they were able to take these medications to their prophylactic endpoint, we saw that more than half of patients prescribed itraconazole for prophylaxis and more than half of patients prescribed voriconazole for prophylaxis actually discontinued the medication before their treating providers thought that they should be at their prophylactic endpoint. And the biggest reason for this was actually side effects and intolerance from voriconazole, malabsorption from itraconazole, as well as side effects and concerns for breakthrough fungal infection. Now, I will point out, we do have newer antifungal medications that are better tolerated. Istavuconazole has less data for antifungal prophylaxis at this point. But posaconazole, although our N was smaller, was better tolerated in our patient population. But we have to look at the cost, the financial cost of taking these medications for prophylaxis for something that we have very little data for. Just to give you an example, from GoodRx, a one-month prescription of posaconazole 300 daily, which is a pretty common regimen that patients may end up on for prophylaxis, costs $1,300 per month. This isn't just the cost of the drug. This is also lab costs for immunosuppression drug monitoring. We have to factor in costs for the nurse coordinator, as well as the physician and pharmacist, to monitor all of these medication interactions that are brought up by patients. In fact, when we asked transplant centers what their biggest barrier to prescribing antifungal prophylaxis among transplant recipients were, the top three indications they gave were related to costs. So applying for prior authorizations to get these medications covered, the actual cost of the prescriptions itself, and coverage denials. And it would be a mess for us to stand up here and talk about universal antifungal prophylaxis without talking about stewardship. We have to address that prescribing these medications comes at an ecologic cost. While fungal organisms are less likely to develop intrinsic resistance than bacteria and viral infections, we can cause an ecologic shift to more resistant fungal pathogens. And as Dr. Hodge says, if we look at what we're prescribing right now, we can see that there's been a shift since 2005 to more and more lung transplant patients being prescribed these medications. And if you look at the spectrum of activity of the medications that are being prescribed, we're seeing that some of them are actually pretty broad spectrum antifungal medications that are being prescribed. I've illustrated that the cost of antifungal prophylaxis in lung transplant recipients, with little evidence to suggest a benefit of universal prophylaxis, is pretty high. And so now let me show you that there is a better way, and that's individual risk assessment, also known as a targeted prophylactic approach. We don't need cannonballs to kill this mosquito. What we really need is to identify lung transplant recipients that are at the highest risk for developing fungal infection and use antifungal medications that are targeted specifically at these fungal infections that they're at risk for developing. This isn't a novel concept, as you can see from the guidelines. It is a suggestion, either universal or targeted prophylaxis. This has also been pretty well studied in the liver transplant population. As you can see here, there's identifiable risk factors for those liver transplant recipients that are increased risk for developing invasive fungal infections. And then depending on the type of fungal infection they're at increased risk for developing, they're deviated to whatever antifungal would cover that most likely fungal pathogen. So either boriconazole or fluconazole. And if you look at the results of this, there's no difference in mortality between those that get universal prophylaxis or targeted prophylaxis. This here is the incidence of invasive fungal infections between these two groups. And there's really no statistically significant difference between the invasive fungal infections that these two groups get. Now, I will point out that neither of these groups are at 0% invasive fungal infection. So we're still going to see breakthrough fungal infections. While we have some data to suggest what increased risk factors there are for lung transplant recipients to develop invasive fungal infections, this hasn't necessarily been reproduced in all studies or all of these single center cohort studies. But we do have evidence across all solid organ transplant recipients as to what those risk factors may be. And for our lung transplant recipients, this may be a good place for us to start. I hope over the last 10 minutes I've convinced you that lung transplant recipients do not require pharmacologic antifungal prophylaxis. First, because we don't have the data to prove that this is an efficacious way to do things. And it comes at a pretty high cost to our lung transplant recipients. We don't need cannonballs. What we need to do is individually risk assess our patients and treat them appropriately. Thank you. We're going to hold questions until the end. And then we'll have our next speaker come up, which is going to be Dr. Matt Koslow from National Jewish Hospital. And he is going to talk about Bactrim prophylaxis in patients with interstitial lung disease. Thank you, Kelly. It's really great to be here. And thank you, Kelly and Shadi and Dr. Limper for inviting me. I hope I win this, because most of the data or whatever I've read or whatever I've learned is from Dr. Limper, I think, in support of my position. So I'm going to focus on three points. This subject really bothers me. I think about it a lot. And I want to focus in on three things that I think affect my decision. One is compared to patients with HIV. Patients with non-HIV who develop pneumocystis are at a much higher risk for mortality. The second point is that the diagnosis is often delayed or even missed. And the third is that prophylaxis is effective and, I believe, relatively safe. But we'll see what Dr. Limper has to say about that last point. So as far as high mortality, this is data from the United States, from Boston, and on the right from France. Both show a very similar trend. Here you have the HIV incidents and the non-HIV incidents. And what you see in the 90s, and this is when HART was starting to be introduced for HIV patients, is that the incidence of non-HIV is starting to creep up. A decade later from France, you see the incidence overall increasing. And that solid black line creeping up is the contribution from non-HIV patients. So the more we're using immunosuppressive therapy, we're starting to see that the contribution to incidence of pneumocystis is more weighted towards the non-HIV population. And mortality from the Boston data, mortality in non-HIV, nearly 40%. It's actually higher. So patients presenting to the ICU are up to close to 60%. And overall, in the French data, 20%. But again, higher for more sicker, acutely ill patients. Why might that be? Well, one is whether they're being treated the same, non-HIV versus HIV. This is data from 17 university hospitals in France, over 500 patients. And nearly all of the HIV patients were diagnosed within the first day. The non-HIV patients were significantly delayed in their diagnosis and treatment. And you clearly see that that delay in treatment was correlated to the mortality. Almost all of them, these patients developing pneumocystis, are not being treated with prophylaxis. Delayed diagnosis. What about misdiagnosis? This is older data. It's from Sloan Kettering. So albeit all the patient population has malignancy. But the main point that I take away from this study, very good academic center. Mortality 40% to 60% over these 10 years. 20% diagnosed post-mortem. A third of them not even being treated for pneumocystis empirically. So we can make assumptions why that might be. But I think it's reasonable to assume that maybe it wasn't considered when these patients were presenting with signs or symptoms of pneumocystis. So, but this doesn't really answer the question that Kelly asked me to address. So, what about our ILD patients who are immunosuppressed? Well, there's not a lot of data. This was a Cochrane database from 2014, looking at many articles, selecting for high-quality analyses, and none of the studies actually addressed our specific question. But I think that's okay, because the converse issue here is the risk-benefit. What can we still learn from this data? So, what is the risk with prophylaxis? One of the main risks is adverse drug reactions. And most of the studies that report adverse drug reactions do not distinguish between severe and all comers. And the majority of adverse drug reactions are lab abnormalities, whether that be liver enzymes, or pancytopenia, thrombocytopenia, usually completely reversible with drug discontinuation. A few of them do specify severe adverse drug reactions. These top five are looking at trimethylin sulfa versus placebo or no treatment. Over 500 patients. We see that for the treatment group, none were reported for severe. Actually, four in the control group, and the authors make a very important point, which is when we do report an adverse drug reaction, is it actually from the drug? Three other trials were looking at trimethylin sulfa versus quinolones, and there were more events in the treatment group in the trimethylin sulfa, six. Majority of these were thrombocytopenia, and these led to treatment discontinuation. These were all malignancy of bone marrow transplant or hematologic conditions. So, very different from our ILD patients, in the sense that those drug reactions are gonna be much more important to those patients and the treatment they're receiving, and lead to drug discontinuation. So, the authors concluded, number one, that trimethylin sulfa is the most commonly used. It's highly effective. The overall risk combining all of the treatment groups versus placebo or alternative therapy, the risk of a severe adverse drug reaction was 3%, and they suggested extrapolating to other patient populations when you're higher than the risk in the control group, about 6%. So, how do we approach the patients in front of us in the ILD clinic where not all this data specifically relates to our patient population? The first is, how much, the first would be looking at the underlying condition, and this is data from the Hopkins, and it just, I think, best represents that patient sitting in front of us and what we should be considering. Number one is, what's the underlying condition? We see that it really affects the spectrum of autoimmune disease. We see myositis, lupus, vasculitis at the top of the list. There are some reports that these conditions might be at increased risk, but it's really hard to distinguish the condition from the immunosuppression that those patients are receiving whether it's more intensive or not. But I think more important is, about half of them had ILD. So, whether or not they're more susceptible or more susceptible severe outcomes, because here in this cohort, again, we see that high mortality of 43%. The majority are receiving a high dose of prednisone, but some of them are under that rule of thumb that we usually consider, 20 milligrams for eight weeks, and those patients that had a low dose of prednisone were on dual immunosuppression. None of these patients were receiving pneumocystis prophylaxis. So, number one, what's the underlying condition? Number two, what treatment are they receiving? Again, this is data from my esteemed colleague, Dr. Limper, and we know the rule of thumb, 20 milligrams for eight weeks. But from that data, about a quarter of patients were receiving less than 20 milligrams or exposed to less than eight weeks. And I think the important concept there is, when you do have a patient under that threshold, if they're receiving some other immunosuppressive, that risk might not apply to that patient. What about rituximab? It seems like everyone gets rituximab these days. That's almost the first drug we pick, it seems like, sometimes. And this is a really nice story out of Mayo Clinic, where I think Dr. Limper, as well, looked at patients receiving rituximab. Most of this first study were patients with hematologic conditions receiving dual immunosuppression. But three patients who were receiving just rituximab alone developed pneumocystis. This led to a much larger study looking at B-cell lymphoma, where they found a much lower incidence, 1.5%. But other data suggests that when rituximab is a second or third line or more intensive therapy, that incidence is much higher. This was a study just published last year out of South Korea, looking at patients receiving rituximab. And the main thing to take away from here is, they're looking at 3,500 patients. The majority of these were hematologic or other solid cancers, but about 20% had autoimmune disease. And they calculated the number needed to treat to prevent one infection versus the harm. And the harm was defined by 10. Out of 3,500 patients, 10 adverse drug reactions, six were pancytopenia, one severe dermatologic reaction. So they calculated the number needed to treat at 32 versus harm 101. And when they did a subgroup analysis for the autoimmune disease, that 32 was lower at 26. And this benefit-risk ratio persisted after they excluded the patients exposed to hydrosteroids. The last thing is, can some other tool help us address patients that receive other immunosuppression, whether that be azathioprine or tacrolimus? And some have suggested the CD4 count. Most studies don't report data on the CD4 count. This one did have complete data for their cohort. And they found that the majority of patients had a CD4 count less than 300. And they proposed this as a cutoff. However, others caution about using this as a cutoff. And this is a group from Germany where about half of their patients had a CD4 count greater than 300. But that was confounded by concomitant steroid use. So that CD4 count might suggest that, yeah, there's a risk here. But if it's higher than that, I wouldn't use that to make a decision that this patient does not need prophylaxis. So I'm gonna conclude here by three points. Number one, high mortality. Number two, the diagnosis is often delayed or missed. And there is nothing I hate more than being in the hospital and having a patient admitted on three, four, five liters, maybe higher than that, off pneumocystis prophylaxis, and I have to decide, do we do a bronchoscopy or not? I don't wanna be in that position. Prophylaxis is effective, relatively safe. We're talking about mortality versus a very low risk of severe drug reactions. And, but as most things, studies needed to target really who's gonna most benefit versus the risk. And now I'm gonna hand it over to Dr. Limper to bring another different side of the viewpoint. Thank you. Okay. I have to thank people. I have to thank you folks. You're here. You're not out on that beautiful beach on a beautiful Hawaiian morning. I gotta thank Dr. Pennington for putting this together and asking me to use my own data against me. And I need to actually thank Matt Koslow. We trained you well because you have very eloquently shown a lot of our own data over time. But I'm gonna try to take a little bit of a different approach and maybe have a little fun. When you're at a pro-con debate, it's either supposed to be fun and then you're supposed to learn a few things. And really, both of us probably believe pretty much the same thing. And I'm Andy Limper from Mayo Clinic. Really nothing to disclose for this talk, although I did invent the method that we use to diagnose the disease and get a little royalty off of this. So we're gonna discuss, this is about patients that are immunocompromised with interstitial disease, whether or not they need pneumocystis prophylaxis. We're gonna talk about who I think needs it and who really does not need it and discuss when pneumocystis prophylaxis may or may not be necessary. And importantly, I need to try to overturn that prior speaker or at least get you guys to chuckle a little bit, me already got one chuckle out of you and maybe learn a thing or two. So I see some guys out here that are about my age. Now you younger people, this is the original Saturday Night Live cast and Dan Aykroyd and Jane Curtin would go at it at point counterpoint. But at no time am I gonna resort to any cheap tricks like that. I'm not gonna be calling names like they did. I can't use that language. This is a prestigious organization and I have prestigious colleagues here. So we're not gonna do anything like that. And in fact, you know, hey, Matt, I actually gave you Dan and I'm over there with Jane because I have hair jealousy with Jane. And then, sorry, Matt, but I've been watching Game of Thrones the last couple of months. And so I've been crushing through that for a second time. So at any rate, but let's get onto this. Now people that know me really believe that I lost my marbles to allow Kelly Pennington to have me do this. So if you happen to sit at your machine and your desk and you do an EndNote search, which yeah, you could do a Medline search too, but out of EndNote, there's about 129 primary publications on pneumocystis that I've been involved with. And at least 25 of those come out about prophylaxis and they're all in favor of prophylaxis, using more prophylaxis. So I really have lost my marbles to agree to this. But pneumocystis is an incredibly serious disease. I'd like to convince you of that. You know, some people believe, oh, it's not a problem. We have Bactrim, we have all sorts of treatments, but it's an alveolar filling process with lung inflammation, respiratory failure in about a third, and it affects immune compromised patients, usually CD4 defects, but also steroids that have a variety of immunosuppressive effects. Other immune suppressants, we're gonna talk about some of those, chemotherapy. Now in HIV, the mortality is about 10%, which is really low, a lot lower than Matt was showing when the whole pandemic started. But in the non-AIDS population, even today, it's about 30%. And worse yet, if you have respiratory failure, the mortality can be 80%, even in great medical centers. So I mean, I think it's important that we think about this disease. It's a very common disease worldwide. In fact, this estimate from the EMBO Mycology Group that I'm a member of estimates about 400,000 cases a year around the world. So even though we see it, and we're seeing it actually with increasing frequency in non-AIDS related patients in the US and elsewhere, it is a common disease. You have to think about it and take it seriously, particularly in non-HIV patients, but also in HIV patients as they fail their heart, as they stop taking their heart, et cetera. Now who is at risk? And Matthew talked about this. This is the first of a number of papers that we've done at Mayo Clinic. Steve Yale and myself back in 96, old data, it's still good data, 116 consecutive patients without AIDS. Out of that group, only two ILD patients had pneumocystis. One with IPF, and in those days, we use lots of steroids, as you recall, right? That was the treatment of choice back in the moldy, oldie days. And one with hypersensitivity pneumonitis, also big immunosuppression. But you look at it, the most common, if you look at the pie diagram, heme malignancies, the biggest risk factor for this, and then also solid organs and inflammatory disorders, and I agree with Matthew, rheumatologic disorders are really key. So ILD with rheumatology disorders, most or many of those tend to have a higher risk for pneumocystis. I think more about prophylaxis in those. Now, you look at our data, and this has already been shown from that paper, but a 90 plus percent had corticosteroid exposure. Usually, the median was about 30 for 12 weeks, but we've come up with a cutoff of about 20 milligrams over eight weeks, and indeed, there are people that have pneumocystis with lesser doses of corticosteroids, but oftentimes, they're on other immunosuppressive medications. Now, a little more recent study, Maria Calero Bernal and Eva Carmona in our group looked at the next series of cases, 128, really answering the question, did it have to be daily corticosteroids? And of 128 cases, we had a lot of those being, again, hematologic malignancy patients. About 87%, close to 90, were on corticosteroids or having gotten them by themselves or with other immunosuppressive therapy, but interestingly, up to 43% were not on daily steroids. They were getting bursts of prednisone. They were getting prednisone with their chemo. They were getting it intermittently, and we don't always think about that cumulative corticosteroid hit. Keep that in your brain. It's not just 20 mg a day for eight weeks, bingo. I'm trying to get you guys to think in a broader sense that it could be those intermittent bursts. Somebody has an exacerbation, they get steroids, they come off. You need to think about the total hit of steroids and other drugs. From the same paper, you can see that, again, heme malignancies is the biggest group. Inflammatory conditions, still, a lot of them, rheumatologic patients, Dr. Speck's patients with vasculitis, I worry a lot about those patients, a lot about those patients, but also solid tumors. I went back to all of our own data. Again, now the second series out of the Mayo Clinic, only three with ILD. So, ILD is not a big risk category by itself, but ILD with rheumatologic disease and steroids, yes, okay? See, I don't probably disagree that much with Matt. What the heck, he was one of our trainees. So, you know, everybody says, well, what about steroids? We give them a lot as adjunctive therapy, and indeed, we know that when people have an acute pneumocystis pneumonia infection, that the corticosteroids actually can be life-saving. It gets people off of ventilators, keeps them off of ventilators, et cetera. It does have a mortality advantage in severe pneumocystis pneumonia, but that's not the predisposing factor. That's the treatment phase of disease. Now, let's talk a little bit about side effects. I agree, a lot of these side effects are what I would term relatively minor and almost always reverse. Bactrim, skin rash. It can be bad skin rash, but I'll have to say, out of 33 years of doing this, I've only seen one Stevens-Johnson related to Bactrim in a setting of pneumocystis prophylaxis. Increased creatinine, that's really a renal tubular dysfunction or temporary, you stop it, the creatinine goes away. Increased potassium, leukopenia. I think you just need to follow those laboratories to see whether you're getting severe or serious complication and whether they're causing problems. What about other agents? A Tovaquone, I like a Tovaquone, but it tastes bad. You usually give them the liquid. It actually is extremely expensive. It's not always covered by insurance. We have to do a lot of pre-auth, it's just something I absolutely hate, but you know, it could be effective. And it also gives you some of the wildest dreams you ever want to, and I know because I took it as a malarial prophylaxis, and I kind of like the wild dreams, but that's up to each person. Dapsone, G6PD concerns, and then inhaled pentamidine. We're all aware of bronchospasm with application as well as breakthrough pneumocystis up in the upper lobes. Now, just about any of those chemotherapeutic toxins can predispose to pneumocystis. Cyclophosphamide, bleomycin, high dose Aracet certainly can. Cyclosporine, et cetera. We're gonna talk a little bit about rituximab too, and we're gonna talk a little bit about mycophenolate. I wish I could talk about all of them, but for the interest of time, we're gonna try to localize this. So, you see some kind of hazy shadows in the, down in the bases in this patient. This patient was on actually, one of our in this case is Dr. Specks from our paper, a rituximab-associated pneumocystis pneumonia, and indeed the patient was only taking rituximab, had actually been off of corticosteroids and was really doing well until had this complication. And so, Martin Garrido, Ava Carmona Ulrich, Specks, and myself looked at our series, and we looked at all the cases between 1998 and 2011, had 207 non-HIV pneumocystis. 30 cases actually had some associated rituximab use within 12 months. 20 cases, 27 cases also had corticosteroids or cytotoxics. Three cases though, rituximab was the only, the only immunosuppressive agent on board for those patients. And again, bad mortality in non-AIDS, non-HIV-associated pneumocystis was about 38%, and most all of those were not being prophylaxed. I think prophylaxis not provided in 29 out of 30. The mortality in rituximab-associated was no different than non-AIDS without rituximab. So, there's been a lot of work. Everybody wants a magic threshold. And the reality is, I don't think a magic threshold exists. All of these parameters are relative. Everybody thought the steroid thing was really great till we find out it could be intermittent steroids. So, this was a paper by Green and colleagues. They've done this several times, several different places. The first publication they published was in the Mayo Clinic Proceedings. They combined 12 randomized trials of patients that had bone marrow transplant, solid organ, heme disorders, and they found that trimethoprim sulfa had a 91% reduction of pneumocystis. That's great. They looked at the adverse events, and then they calculated adverse events over time that they considered significant, and we could take that with a grain of salt because most of those side effects are reversible. And they came up with these thresholds that in your population of your disorder, we'll say interstitial disease, you need to have at least 2.5 to 3% of those patients over the course of their disease develop PCP, PJP. Actually, it's supposed to be PCP, folks. I was there when we renamed it to Girovacchi, but the PCP went away. PJP, we can call it that. But that threshold is very artificial. How in the heck can the bone marrow transplant patients be compared to our interstitial patients? You can't do it. So we do see that there is mortality benefit when you get above that 2.5 to 3% threshold. So there's a lot of limitations to these studies. They're often reanalyzing clinical trial data. All of you that have worked on clinical trials know how closely we follow patients on clinical trials. We follow patients in real-world settings. We have the nurses calling them all the time. We get to their complications fast. We actually get them treated fast. That makes the mortality of people in clinical trial, the death from complication rate, go down. So it's really, really hard to extrapolate these data to clinical practice. We went ahead and tried to do that at Mayo Clinic. This was driven by our pharmacists, and Jason Barretta is a wonderful guy, and we worked through it. Our ARCHOP treatment, and we found about pneumocystis at 1.5%, that was below the threshold of 2.5% for prophylaxis, and the story's not that simple because we tend to do traditional dosing of ARCHOP, which is usually, I think they do the cycles every three to four weeks, and there are new schedules that do the cycles every two to three weeks, at least in the early induction phase, and lo and behold, when you do more intensive ARCHOP therapy the pneumocystis rate goes up to 6%, and those that have progression and rescue, they go up to 10, 13%, I believe, Matt, that's what you were quoting in your talk. So those rules of thumb are not so simple, so we have to think. Ank-associated vasculitis, there is a paper that has just been accepted, it's in press, looking at 1,400 patients with the EHR data set. They had 37% of them receiving pneumocystis prophylaxis, their cumulative index was about 1.4%. So that's below that magic threshold. Is the threshold for ank-associated vasculitis the same as it is for ARCHOP and people with lymphoma? I don't think so, and we don't really know. So they talked about the side effects of the prophylaxis, the leukopenia, the renal dysfunction, the rash. I would say those are all relatively mild side effects, but if they're severe enough that makes your nephrologist worry, your hematologist worry, then you're gonna probably wish to avoid. So I have actually one real quick, as Kelly said, we had to put one of these in. This is a thinking, there's no right answer. 65-year-old patient, unclassified connective tissue disease, ILD, on two grams of microphenolate a day, do you give pneumocystis prophylaxis? Yes, almost always, no, almost never. Maybe, it depends on what other medication the patient is receiving. Not sure, I'm confused. That's why I showed up this morning. No, I just came to this session for the laughs or the bloodshed. Please use your phones, raise your hand, shoot off a flare gun. Give you a few more seconds. Five, oh, six, come on, we could get to 10. There's 25 of you out there right now. So let's, okay, we'll give it another few seconds. There's no right answer. And I'm gonna show you there's not right answer. Yeah, this is the answer I would answer if I was answering, right? I would say that. And you guys really would agree with me. It depends on the whole situation with the patient. That's all I'm saying with this talk. So what about microphenolate? Well, my good friend and former colleague, he's passed away, Walter Hughes at St. Jude's, actually was the first individual to develop the treatment of trimethoprim sulfamethoxazole. He did it in his rat colonies. He did the first studies of pentamidine, actually brought those in the clinic during the lymphoblastic leukemia, pneumocystis outbreaks that occurred there. And at any rate, he looked at his rat colonies and microphenolate itself did not cause or induce pneumocystis in his rats. And there've been several series, no pneumocystis in 273 transplant patients receiving microphenolate. Smaller study in interstitial disease since that's what I'm supposed to be talking on with no pneumocystis without prophylaxis that was an abstract in ERJ. I've used that for a number of years to have a higher threshold that maybe I don't need it so much in microphenolate. Well, we gotta keep our heads and brains open because then you look at a database study, and I do database studies, and I don't think this has all of the things that I would like to see in a database study, but they looked at autoimmune rheumatic diseases, ARDs, the usual things, you know, RA and lupus and scleroderma and polymyositis, stromatomyositis, vasculitis. They thought that they found the highest risk of pneumocystis in polymyositis, stromatomyositis. Again, the connective tissue diseases, steroids, you know, above 10 milligrams a day were high, but they actually had a significant number of pneumocystis with microphenolate. And then the other things that we think about, cyclophosphamide, biological agents, including rituximab, methotrexate, cyclosporine, et cetera. Limitations. I do large databases. You gotta really think about how they're being done. Again, insurance claims for the autoimmune-related diseases. The autoimmune diseases were adjudicated. They were confirmed. That's not always the case in a lot of studies, but the pneumocystis certainly wasn't. It came from a separate database that they merged, and it had to be just pneumocystis as listed in discharge diagnosis. They could have died of respiratory failure. That could have been their discharge diagnosis. They might have missed it. They might have had it. They did not have to have microbiological confirmation. It could have been clinical pneumocystosis, which there's a whole debate on that. We could talk about that sometime, too. So there's limitations to this as well, but they actually tell you to think about microphenolate. So I'm gonna wrap up here. I love this painting by Francis Picabia. Our heads are round so our thoughts can change directions. And I'm not asking you to stop thinking about prophylaxis. I'm asking you to think about it more, but think about it more intensively. Think about who needs the disease. Maybe they all need it, and maybe they don't. Be selective. Think about that underlying disease. What is their immune state of the underlying disease? What are the other drugs they're using? What is the cumulative steroid hit, not just the daily hit? Be aware of the drug-drug side effects. Look for them. They might just be laboratory. You might stop the drug or change to a different drug. But please pay attention. Keep watching the literature. More details on your specific patient groups I sure hope will follow. But blanket decisions based upon some threshold, I mean, yeah, I look at lymphocyte counts too, but to be honest, there's a lot of failures in the lymphocyte counts, particularly in the heme patients because they have really dysregulated lymphocytes. So you just can't base it on one or two things. We're doctors. We need to use all that clinical material together to make a decision. So not every ILD patient that immunosuppression is mandatorily going to have to be prophylaxed, but many will. So this is the time where I have questions, rebuttals, or retreats. And as you probably heard from what I was talking, I agree with my esteemed colleague, Dr. Koslow, as much as I disagree with him a bit. And so I'll say one of my favorite scenes from the original Star Wars. Then I suggest the change of strategy. C-3PO was saying this to R2-D2. I'm gonna let the Wookiee win. And there we go. So I did get four chuckles out of you guys, and I hope you learned a thing or two, and I'm gonna sit down. Thank you so much. Please evaluate. Thank you.

interstitial lung disease

pneumocystis prophylaxis

mortality

risk factors

corticosteroids

individualize decisions

side effects

microphenolate

careful consideration