Chapters Transcript Generating Guidelines for Managing Lupus Nephritis: A Debatable Issue Course: NYU Langone Seminar in Advanced Rheumatology (2025) So I, um, the, the title for this talk did not come from me, it came from the NYU team, but I think it is apt because, you know, until I got involved in a guideline work with the reproductive health guideline, I used to look at guidelines that were published and think, oh yeah, of course, you know, this is obvious. I mean, how much work. Can this involve? Um, but it really isn't like that, because even the most obvious statements that we all agree should be our standard care have to be researched, discussed, and, and I have found through my experience that, as Peter alluded to, the range of opinions about things that you would think are not controversial. It is really interesting and you know the people involved in the voting panels come from all over the country. They come from different specialties in some cases for this guideline, nephrologists and rheumatologists and so uh there was a lot of debate over our 20 hours of voting panel discussion. So to just to give you an idea. All right. I have no conflicts of interest or relevant disclosures. And part of what I'd like to do today is just give you an idea of how these guidelines are made. Obviously, I'm going to go over the lupus nephritis guideline with you. And, uh, remind you, if you saw our presentation at the ACR about what we think are the most important points. But for the first part of the talk, I do wanna take a few minutes to tell you about clinical, uh, practice guidelines because they are not. Created equal. And I have to say, you know, the ACR is very proud of their process, um, because it really is quite rigorous. So, uh, it's important to understand how they're generated. It's not just people sitting around saying that sounds good. Uh, that's what I do. Um, it's important to identify factors that impact the quality, because just as evidence reports have a certain level of quality or certainty, so do, uh, guideline, uh, recommendations. And then finally, of course, talk about lupus nephritis. So, I really do think there's been quite an increasing need for these guidelines. I feel like every time you get an email from anywhere, it is telling you about publication of a new clinical guideline. And I think the reason for that is that there's really been quite an explosion in data, you know, over the last 15. 20 years. And also, the way we look at data has evolved. I remember as a fellow, not telling you the years, but as a fellow, you know, the best we could do was a good narrative review. And that was really our source of all of our information. And that evolved to a systematic literature review to meta-analysis, and now to incorporating all of that analyzed data in the context of clinical expertise and patient information to come up with these guidelines. So, the goal of the lupus nephritis guideline was to systematically review and synthesize the evidence, uh, developing clear and useful recommendations, and I really highlight that word useful. It is my mantra, because I just feel like complicated guideline statements or, you know, guideline statements that address a minuscule point, those are not so useful to the practicing clinician. So we spent a lot of time thinking about what the most important questions are, looking at the evidence for that, and then discussing it, and giving guidance for diagnosis, treatment, and management of lupus nephritis. So, challenges, um, aside from, you know, the more recent lupus nephritis trials that really guided. Um, some of our recommendations. Other evidence is, is really not, uh, so high level, you know, a lot of case series or cohorts, um, and so that is challenging because that is a lower, you know, degree of certainty evidence and that impacts how you use that evidence in making decisions, balancing conflicts of interest, and I use that term the way it is used by the ACR. Um, it is very important to reduce the appearance of bias. So if the only people who put together a guideline are the doctors who did the trials that looked at the drugs recommended in the guideline, you know, it would be the same guideline, but people might look at it and think, oh. Well, of course they said that, they did the trials. They're not, you know, thinking about other options, but that is why the ACR is very careful to balance all of the groups within the guideline team. And then finally, there was a lot to catch up on because our first and last uh nephritis guideline was published in 2012. So you can, you know, uh imagine how much data we had to look at. So what are clinical practice guidelines? Well, they really started a long time ago with Hippocrates. Um, he summarized medical knowledge and prescribing practices in the Hippocratic corpus back in the 5th century BC. The modern age of guidelines though started within the past 20 to 30 years. Um, the Institute of Medicine first reported recommendations for recommendations, guidelines on guidelines, um, and updated that in, in 2011. And their definition is that these are statements, including recommendations intended to optimize patient. Care informed by a systematic review of evidence, and an assessment of the benefits and harms of alternative care options. And that's very, very important because that takes a panel of people to assess. That isn't something that easily comes out in just the evidence report. And as you can imagine, that will vary depending on the patient's clinical situation and their own. You know, values and preferences. So, these guidelines rely on consensus methodologies, um, and the summarized statements then can provide useful guidance in, uh, both classifying when it's a classification criteria and treating a disease. So, I am by no means expert in this, but I put this slide in just to Again, emphasize that these consensus methodologies, as well as guideline methodologies really do vary depending on what you choose to use. So, the Delphi method, which we've all seen use, there is never an in-person discussion. That is a series of surveys that are. Sent out and, and people respond each time. It's 2 to 4 surveys usually. The nominal group technique is really to prioritize more ideas and issues. So, the purpose there is prioritizing and it's useful when you need a lot of expert opinion. And then the RAND UCLA. Method is closer to what we use for our guideline, which is grade methodology, and that's because it separates out the people who are looking at the evidence from the people who, meaning looking at the papers, extracting the data, interpreting the level of certainty of the evidence, and it separates that from the people who are looking at the report and discussing the clinical implications. So, um, the ACR and many other groups now use grade, um, which is just that kind of methodology, and there are actually three groups. There's a core team who puts together the scope and decides on the important questions to answer. There's a literature review team which I think has the very hardest of jobs because they spend hours going through papers and grading them, um, and then there's the voting panel where they have a difficult job too, because there are 20 of us on a Zoom screen now as opposed to in person. Um, you know, sort of talking back and forth and, and evaluating each other's comments, assessment of the data, concerns about particular patient issues. So, um, there are 4 levels of evidence, high through very low, and recommendations are graded according to strength, which I'll come back to in a minute. So, this paper was published in 2018, and it was looking at ACR rheumatology guidelines up to 2015, and this particular paper just summarized for each of these guidelines the level of evidence. So you can see that The lupus nephritis guideline here did not use grade methodology. They switched to it, I guess, later that year because this OA guideline, also published in 2012, did use grade methodology. But if you look up here, you'll see that for level of evidence, 70% of the evidence that they had available to them was level C. That's case reports, series, standard of care, and expert opinion. It's very different this time around. Um, this is a paper by the same group looking at the quality of the ACR guidelines, and, um, that is done using a standardized agree to tool for guideline assessment. Again, this is a guideline for grading guidelines. The agreed to tool was first published in 2003, and it really, it replaces or was updated in 2010, and it is a way of looking at a guideline and assessing whether it is doing what these experts who put together the agreed to tool, what they feel a guideline should be doing. So, it looks at 6 domains, 23 items, so scope and purpose, stakeholder involvement, rigor of development, clarity of presentation, applicability, and editorial independence. So this is the tool. You can see the domains on the left and on the right are the specific items under each domain. And I wanna, and again, you know, these are things that I think we would all agree are very important to have in a guideline and somewhat obvious. Um, but I want to highlight where um we have changed from the 2012. So, while the guideline group included individuals from all relevant professional groups. In 2012, they did not reach out to patients. In this guideline and in all of our guidelines now, there is a patient panel. We had 14 patients who discussed all of these issues, and two of those 14 were patient representatives on the voting panel. So their votes counted just as everyone else's, and their experience and perspectives really helped guide the discussion. Um, I alluded to the conflict of interest issue, and that is felt to be very important, that if there are people who have so-called conflicts of interest, and mind you that can be quite variable in terms of what falls under that, um, it's important that the readership know that, that, yes, this many people did have some tie in some way to some company that is related to medications that we're talking about. Um, so that's important. And then the only one that we really do not look at is this one. The potential resource implications have been considered, and that is, do you take into account financial considerations. I think. If the ACR did that, this would have been a very different guideline, because obviously there are wide variations in terms of cost of some of the medications that we use, but their policy is to not look at that. Um, so that is one area that we do not get into. Um, this shows you the agreed to scores of ACR guidelines, and I'll just show you 2012 lupus nephritis, a score of 83. As they change to grade, the numbers go up. I assume ours would be higher than 96, I hope now, but we haven't done that process. And then, you know, there are a lot of lupus treatment guidelines out there. This uh paper summarized them, uh, published in 2019, so you can imagine there are that many more now. And looked at all kinds of guidelines related to lupus. So some of them were for general lupus care, some were for lupus nephritis, some were other guidelines that covered lots of different rheumatic diseases, for example, a vaccination guideline that applied also to patients with lupus, and the goal here was to identify unmet needs. So, they looked at in 2019, 23 uh guidelines relevant to lupus. 15 were specific for just SLA 5 for general management as opposed to a specific aspect. Um, only 5 had patient representatives. 1 had a patient panel, and they covered a wide range of topics, and their conclusions were that the current guidelines available at that time for their review in terms of needs for clinicians did not adequately address organ specific manifestations, the issue of non-adherence, duration of therapy, patient input, and then. Other factors that impact outcomes such as ethnicity and socioeconomic status. And for the patients, the ways to improve were really to improve the time and certainty of diagnosis, to find new treatments, to give less steroids, and to focus on issues that the patients find really impact their quality of life, such as pain and fatigue. They also wanted better patient education. So this is our guideline team. Um, as I said, we had a large team, and you can see the core team, voting panel, literature review team. We had a librarian and our patient panel had 14. Um, I talked about the conflict of interest issue. So each team and the team as a whole has to have a majority of Uh, participants who are not conflicted. So for most of this, that really meant 49% versus 51%. And this is where we are in the guideline development process, as Peter told you, we are waiting for the proofs. It's accepted for publication at the ACR journals, um, and so hopefully we'll be published soon. We are up here for the SLE non-renal guideline. Um, and I mentioned that because we scoped for both these guidelines together because we wanted to be efficient in our use of the, um, Literature review, uh, panel, because as it was, their task was overwhelming. So, we didn't want them having to review papers twice that might include some information on nephritis, some on other lupus manifestations. We developed what are called Pico questions, you know, this is a tool used to identify the terms that are used for the literature review. Search. So, for example, in patients with lupus nephritis, does drug X work better than drug Y in terms of reaching a complete renal response? So, that's what the PIO questions are. They guide the search, and they also help guide, ultimately, depending on the results of the search, um, the, the preliminary recommendations. So, um, there are actually two rounds of voting. One is online to get a sense of what the group thinks about the suggested recommendations, and then this, uh, video conference that, you know, was, uh, lengthy. So, we ended up with 105 references for the quantitative synthesis and meta-analysis. And this is the grade process, and I realized looking at this slide, I don't know, it, it can look a little boring, but it really is important to just take away a couple of points. These are the questions. This is where everything gets pulled. The literature review team looks not only at whether the topic is appropriate, but once it is appropriate, they pull out the information and then they grade it. So, you know, if there are multiple. Placebo controlled randomized trials on one particular issue, that'll be a high level of uh evidence. I haven't seen that yet anywhere. Um, the best we've done is moderate. Um, there's a lot of low and very low, and then, you know, these experts can grade up or grade down depending on specific details. In terms of the individual studies that is given to the voting panel. The patient panel information and input and preferences are relayed to the voting panel. Preliminary recommendations are put forward and the voting panel discusses and votes. You have to get to 70% in the direction. So, 70% or more. have to feel, yes, drug X is better than drug Y, looking at all these different outcomes, and the outcomes don't just include something like complete renal response, they include, you know, um, adverse events, you know, toxicity and other issues that have to be weighed against the benefit of a medication, um. So this just sort of summarized all of the different things that that go into that discussion that the voting panel has, and things that can make a recommendation more conditional than strong would be if there's lower quality evidence, if there's more uncertainty. About how the benefits and harms sort of balance out, or if there's a lot of uncertainty or variability in how individual patients would, uh, perceive this. So, a strong recommendation is one where almost everybody really should be following this recommendation. And these are tend to be very obvious things. Um, The majority of our recommendations, 75%, are conditional, and that's because medicine is complicated, uh, lupus nephritis is complicated, and patients are different from person to person, both in terms of how they feel about medicines and in terms of their clinical disease, how severe their disease is, what else is going on. So, these conditional recommendations really demand shared decision making. And that's important to remember when looking at a conditional recommendation. If you look at it and think, Well, I don't think that should be the case for every single patient, we don't. Either. We acknowledge that a lot of patients may not be appropriate for that, but we suggest it as our preferred recommendation, and then it's up to the physician and the patient to decide if it is preferred for that individual person. So, I, I mentioned this. I'll just mention good practice statements. So, when I go over those, you understand what those are. They're basically common sense statements that I think no one would disagree with, but that it seems important to put down in writing. So, um, they're obviously Beneficial. It would be onerous to do a whole review. A lot of the evidence would be indirect, or there may be little evidence, and the example that they use in the grade methodology is this that parachute saving lives is not based on, you know, a controlled study with and without parachutes. So These are our guiding principles. Again, you know, common sense things. We want to preserve kidney function, minimize uh morbidity and mortality from both disease and treatment. We want to include collaborative care with our nephrology colleagues, shared decision making. Um, reducing healthcare disparities as best we can, and we tried to give specific guidance where appropriate for pediatric and geriatric populations. This applies to the full age range for lupus nephritis. So, I told you that, uh, 3/4 were conditional. We had 28 recommendations, 13 ungraded good practice statements. We did not make any recommendations based on race or ethnicity because we felt that, and I think many feel that the evidence in those cases is quite limited and very much confounded by socioeconomic factors. Um, and then we acknowledge these are all the caveats, right? We know that it will depend on the clinical presentation, the patient preferences, the access that a particular patient has to various specialists or therapies. We did not recommend a specific CNI because we felt that we do not have. Comparative effectiveness and safety studies, and we were concerned that accessibility might at times dictate choice. And then, of course, if recommended, if recommended medications are not available or not pursued or desired, then the standard traditional therapies should, of course, be used. So, this is um sort of our, our screening recommendation that all lupus patients without known lupus nephritis should have proteinuria checked every 6 to 12 months and at the time of any kind of clinical flare. For those where you suspect lupus nephritis, these are the common sense statements. You should do a kidney biopsy as soon as you can, and you should treat them with glucocorticoid as soon as you can, and you should not wait to schedule the kidney biopsy or get the results of the kidney biopsy. And that is because this is an inflammatory condition, and you need to put something in place that will quiet that inflammation as you gather more information to decide on. Um, the full panel or regimen of medications that would be most appropriate. Um, kidney biopsy recommended for proteinuria greater than 500 mg per gram, or unexplained impaired kidney function, and in patients with known nephritis, if there is a suspected flare after what seems to be a complete response or a lack of response or worsening after ongoing therapy. We recommend all patients be on hydroxychloroquine and all with any degree of elevated proteinuria on RAS inhibitors, and remind readers to adjust the dosage of medications based on the GFR. We actually, in the manuscript, um, have a table that goes through the commonly used medications and how to adjust them for, uh, changes in GFR. I can't show you that because it hasn't officially been published yet. Um So I think this point about glucocorticoid is the first of 3 important points that come out from this guideline. So, and I think there's 3 important points are these. One, glucocorticoid should be used promptly, and initial therapy should be aggressive with some version of pulse IV, but that the max oral dose should be moderate and not high. And that the taper should be rapid. So, this is what we have here, you know, pulse steroids, oral prednisone 0.5 mg per kilogram per day, but a max of 40 taper to a target dose of less than or equal to 5 mg per day by 6 months. It's a conditional recommendation. It may not be appropriate for everybody. It may not be possible for everybody. We actually had some voting panel members who said, you know, I can't always get my patients pulse therapy. They don't have access to a place where they can get IV therapy. So, it, which, you know, sort of surprises me coming uh from an academic medical center, and I doubt anyone here at NYU has difficulty getting pulse steroids for their acute lupus nephritis patients, but these are guidelines for the country as a whole and for rheumatologists practicing under all kinds of situations and with all kinds of patient populations. So, Uh, this was based on a systematic review and meta-analysis that showed that this regimen maximized renal response while minimizing toxicities. The other very important point that led to this being adopted is that the patient panel participants felt strongly that they wanted less steroid, and if anything, that was probably their most vocal, most frequent concern was the adverse effects of chronic or long term high dose steroids. We know you'll have to individualize the tapering regimen, and I want to acknowledge right here, in case any of you wants to ask a question later, that the data for class 5 are terrible. The only data is very low quality, but does show use of high-dose steroid in conjunction with other immunosuppressives. So, we included class 5 in this with the caveat that This may not be necessary, and this is an area for further research. So, maybe by the time we revise this, we'll change this. We, we have to see. But certainly, I think that, um, that goes under the conditional part of the recommendation, if it's pure class 5. So, this is the treatment overview, and we tried to address. Classes 34, with or without 5, and then pure class 5, and we recommend for all of these so-called triple therapy, and the triple therapy is glucocorticoid with 2 additional immunosuppressant medications, and I'm gonna go through it in detail now. So, for active class 34, whether or not 5 is present, we recommend triple therapy, and there were 3 regimens that count as triple therapy, and these were based on the evidence that we had available to us, the moderate level of evidence, which was the best we had. And so those combinations were mycophenolate analogs with bulimumab, mycophenolate with CNI, and cyclophosphamide with bulimumab. So, looking at those three possibilities for triple therapy, we tried to offer some guidance that would, uh, help for a particular patient in choosing which might be preferable. So, for patients who have a lot of extra renal manifestations, who have a low creatinine clearance, here we've said less than 45, or significant. To control hypertension, we would recommend considering, uh, the bulimumab with mycophenolate combination for those who have significant proteinuria over 3 g, we would recommend CNI regimen. We do recommend mycophenolate over cyclophosphamide simply because of the better safety profile. Um, but cyclophosphamide is certainly a reasonable and, and a good option. If it is pursued, we recommend the euro lupus nephritis, uh, trial dosing. So that is a lower cumulative dose with a better safety profile. Um, there are limited data on the cyclophosphamide arm in the BLISS trial, but did show benefit. We don't have data on cyclophosphamide and CNI of any significant, um, level of certainty. And so this combination is not listed as a triple therapy. It doesn't mean that one cannot use it, and hopefully we'll have more data looking at that in the future. So why triple therapy rather than starting with dual therapy and then escalating therapy? And I think this is an important and really reasonable question. And so I, I think what swayed us was, of course, these randomized controlled trials looking At addition of bulimumab and vocalsporin that showed significantly improved outcomes with initial triple therapy versus standard dual therapies. We don't know how well people would do uh if you waited 6 months and added something. But I thought this, uh, this figure was an important reminder to me, and I think should be to you, that there is risk with not treating aggressively early. So, for a person who does not have lupus nephritis, I mean, I, this goes out to 100, I don't think most of us are going to make it to 100, but if We did, we would be likely to have stage 3 chronic kidney disease. Um, but let's look at 60. Already, you're at stage 2. That's with no health issues. If you have an episode of lupus nephritis that is adequately treated, that brings you down so that then you track along this line, which is similar. But lower. And so, by age 60, you're in between 3 and 4, stage chronic kidney disease. And if you do not control that single episode, or if there are recurrent episodes, you're headed for end-stage kidney disease at age 60. And we felt that we couldn't afford to wait for this nephron loss, um, to begin what has been shown to be the most efficacious therapy. I want to point out that we did do a sensitivity analysis again, coming back to the conflict of interest. We excluded voting panel members who had a conflict of interest regarding these drugs, and we had the same results. Everyone agreed on this therapy, and it was a conditional recommendation still. We thought it was important to do that just to put people's minds at ease, that it wasn't being driven by individuals who Were very experienced with the drug and might inadvertently, you know, um, Vote in favor when others might not. So, this was the Bliss LN trial with Bilimumab. I don't have time to go over the details, but the points here are just what I said. It showed that there are better outcomes, both when you looked at their definition for primary efficacy renal response and. Also, when you looked at standard definitions of complete uh renal response. And importantly, and I think this also was an important factor in decision making, the safety profiles were similar, so you're not sacrificing anything by starting this, except maybe money. But again, we're not talking about that here, so, um. And the Aurora one trials similarly, you know, definitely better outcomes. Adverse event profile was balanced. Um, the extension study, because this was 52 weeks, the extension study showed maintained responses at 36 months. Um, the important point here is that this trial did use the kind of glucocorticoid regimen we're talking about, that is, um, pulse with a lower dose and a more rapid taper, and they still, you know, got these very good results. What about class 5? You know, it is just challenging because it is not as common as class 34, pure class 5, I mean. Um, we voted for triple therapy with mycophenolate and a CNI because of the stabilizing effects of CNI on podocyte cytoskeleton, um, as being preferred over bulimumab regimens, um, and that is for proteinuria greater than 1 g. For less than that, we really had no specific recommendation. So, another important point, I think, is that we recommend that the therapy duration be at least 3 to 5 years. Why is that? Well, of course, we want to get rapid control of disease activity. That's why the steroids, but we want to completely control disease activity until there is sustained and active disease, and a repeat biopsy studies show that immunologic activity. Immune complexes persist for years, that if you withdraw immunosuppression in the setting of known histologic disease activity on repeat biopsy, that predisposes to flare. So we decided 3 to 5 years was the most reasonable interval with the understanding that it would be tapered at some point over that time. It wouldn't be the same initial therapy. And then finally, um, non-response and refractory lupus nephritis. There is essentially very little evidence to guide us in this. And so this was more clinical expertise to, uh, change to an alternate triple therapy. If you're on dual therapy, escalate to triple therapy for. Refractory disease, I think the three important points make sure that, uh, in fact, your patient is being adherent, because that's an important reason for lack of response, and you have to be aware of that. Make sure there's not an alternate diagnosis APL nephropathy, drug-induced toxicity. And make sure that this is not all scarring with advanced chronicity, and assuming that this is still ongoing optimally treated active lupus nephritis, escalate, add CD20 agents, which can also are also, I didn't put it in the slide, but part of the initial non-response uh algorithm as well, uh, combination therapy with more immunosuppressives, no data for that, um, or referral for a clinical trial. And then we also um. Put together a table of good practice sort of suggestions for looking at and monitoring and treating all of the comorbidities and other aspects that go along with lupus nephritis. So, You know, looking at cardiovascular health, bone health, uh, giving vaccines, reproductive and pregnancy considerations. And again, that's a table that will be, that is not, those are not voted upon recommendations, but we thought that was very important information to have as a reference for people. Finally, monitoring, um, every 3 months if you're not at complete renal remission, every 3 to 6 months if you are. And importantly, renal replacement therapies strongly recommend transplant over dialysis. Again, that's sort of obvious, but There is a lot of literature that supports improved outcomes. Um, we conditionally recommend proceeding with transplant without requiring complete clinical or serologic remission, provided there's no other, uh, major organ involvement. If you have a patient who's been quiescent, they have the malar rash they've had every day for the past 20 years, and their C4 is 14. But otherwise they could be transplanted. Please tell the nephrologist it's OK, that they're not flaring. Um, for those who are at risk for end-stage kidney disease, we conditionally recommend a preemptive transplant. Outcomes are better overall in the, uh, transplant literature. And for all of these patients, we recommend regular rheumatology follow-up, because many of them. Get lost to their rheumatologist after they're on dialysis or after they've had a transplant. But they still have lupus, and they still have all the comorbidities that go with having lupus and being treated with steroids for 20 years. And I don't know that the transplant team is going to focus on all of those things. So, I think it, it makes sense, um, to continue this collaborative care. So, these are the 3 most recent lupus nephritis treatment guidelines, and honestly they're very similar, including the emphasis on prolonged therapy, the emphasis on lower dose steroid with a more rapid taper. The only difference really is that Um, Kadigo and Uer offered triple therapy options as equivalent options with so-called standard or dual therapy, and our guideline just went a step further, saying that we preferred triple therapy over dual therapy. Um, so, you know, in conclusion, all clinical practice guidelines are based on the evidence, but also clinical expertise and patient variables and preferences, and it is important to include the patient voice, to be transparent about how all of the. These conclusions and recommendations were reached, because you don't want there ever to be an appearance of bias. Um, and I told you that we emphasize, you know, the triple therapy option as opposed to putting it as equivalent with dual therapy. Um, and the best use of this guideline is to start a discussion with the patient, because again, every patient really is different, both in terms of their clinical status and details, and in terms of how they feel about both their disease and, uh, you know, potential therapies. So, it's a, I think, a really good starting place though. You can at least present this as all of the evidence suggests this is best for most people. Let's talk about you. So I think that's it. Yeah, I am done. Thanks. Published March 20, 2025 Created by
CME Knowledge Hub