JCO Article Insights: IMS-IMWG Consensus on High-Risk Multiple Myeloma
In this JCO Article Insights episode, Michael Hughes summarizes “International Myeloma Society and International Myeloma Working Group Consensus Recommendations on the Definition of High-Risk Multiple Myeloma" by Avet-Loiseau et al. published on June 09, 2025 along with an interview with author Dr Nikhil C. Munshi, MD. TRANSCRIPT Michael Hughes: Welcome to this episode of JCO Article Insights. This is Michael Hughes, JCO’s editorial fellow. Today I am interviewing Dr. Nikhil Munshi on the “International Myeloma Society and International Myeloma Working Group Consensus Recommendations on the Definition of High-Risk Multiple Myeloma” by Avet-Loiseau et al. At the time of this recording, our guest has disclosures that will be linked in the transcript. While some patients with multiple myeloma live for decades after treatment, others exhibit refractory or rapidly relapsing disease irrespective of treatment administered. We term this “high-risk myeloma.” Multiple risk stratification systems have been created, starting with the Durie-Salmon system in 1975 and evolving with the advent of novel therapeutics and novel treatment approaches. In 2015, the Revised International Staging System (R-ISS) was introduced, which incorporated novel clinical and cytogenetic markers and remained, until recently, a mainstay of risk stratification in newly diagnosed disease. Myeloma as a field has, just in the past few years, though, undergone explosive changes. In particular, we have seen groundbreaking advances not only in treatments - the introduction of anti-CD38 agents and the advent of cellular and bispecific therapies - but also in diagnostic technology and our understanding of the genetic lesions in myeloma. This has led to the proliferation of numerous trials employing different definitions of high-risk myeloma, a burgeoning problem for patients and providers alike, and has prompted attempts to consolidate definitions and terminology. Regarding cytogenetic lesions, at least, Kaiser et al’s federated meta-analysis of 24 therapeutic trials, published here in the JCO in February of 2025 and recently podcasted in an interview with associate editor Dr. Suzanne Lentzsch, posited a new cytogenetic classification system to realize a shared platform upon which we might contextualize those trial results. This article we have here by Dr. Avet-Loiseau, Dr. Munshi, and colleagues, published online in early June of this year and hot off the presses, is the definitive joint statement from the International Myeloma Society (IMS) and the International Myeloma Working Group (IMWG). What is high-risk multiple myeloma for the modern era? The IMS and IMWG Genomics Workshop was held in July 2023 and was attended by international myeloma experts, collaborating to reach consensus based on large volumes of data presented and shared. The datasets included cohorts from the Intergroupe Francophone du Myélome (IFM); the HARMONY project, comprised of multiple European academic trials; the FORTE study, findings from which solidified KRd as a viable induction regimen; the Grupo Español de Mieloma Múltiple (GEM) and the PETHEMA Foundation; the German-Speaking Myeloma Multicenter Group (GMMG); the UK-based Myeloma XI, findings from which confirmed the concept of lenalidomide maintenance; Emory 1000, a large, real-world dataset from Emory University in Atlanta; the Multiple Myeloma Research Foundation Clinical Outcomes in Multiple Myeloma to Personal Assessment of Genetic Profile (CoMMpass) dataset; and some newly diagnosed myeloma cohorts from the Mayo Clinic. Data were not pooled for analyses and were assessed individually - that is to say, with clear a priori understanding of whence the data had been gathered and for what original purposes. Consensus on topics was developed based on the preponderance of data across studies and cohorts. In terms of results, substantial revisions were made to the genomic staging of high-risk multiple myeloma, and these can be sorted into three major categories: A) alterations to the tumor suppressor gene TP53; B) translocations involving chromosome 14: t(14;16) (c-MAF overexpression), t(14;20) (MAFB overexpression), and t(4;14) (NSD2 overexpression); and C) chromosome 1 abnormalities: deletions of 1p or additional copies of 1q. In terms of category A, TP53 alterations: Deletion of 17p is present in up to 10% of patients at diagnosis and is enriched in relapsed or refractory disease. This is well-documented as a high-risk feature, but the proportion of the myeloma cells with deletion 17p actually impacts prognosis. GEM and HARMONY data analyses confirmed the use of 20% clonal cell fraction as the optimal threshold value for high-risk disease. That is to say, there must be the deletion of 17p in at least 20% of the myeloma cells on a FISH-analysis of a CD138-enriched bone marrow sample to qualify as high-risk disease. TP53 mutations can also occur. Inactivating mutations appear to have deleterious effects similar to chromosomal losses, and the biallelic loss of TP53, however it occurs, portends particularly poor prognosis. This effect is seen across Myeloma XI, CoMMpass, and IFM cohorts. Biallelic loss is rare, it appears to occur in only about 5% of patients, but next-generation sequencing is nevertheless recommended in all myeloma patients. Category B, chromosome 14 translocations: Translocation t(14;16) occurs in about 2% to 3% of patients with newly diagnosed disease. In the available data, primarily real-world IFM data, t(14;16) almost always occurs with chromosome 1 abnormalities. Translocation t(4;14) occurs in about 10% to 12% of newly diagnosed disease, but only patients with specific NSD2 alterations are, in fact, at risk of worse prognosis, which clinically appears to be about one in every three of those patients. And so together, the CoMMpass and Myeloma XI data suggest that translocation t(4;14) only in combination with deletion 1p or gain or amplification of 1q correlates with worse prognosis. Translocation t(14;20) occurs in only 2% of newly diagnosed disease. Similar to translocation t(4;14), it doesn’t appear to have an effect on prognosis, except if the translocation co-occurs with chromosome 1 lesions, in which case patients do fare worse. Overall, these three translocations - t(14;16), t(4;14), and t(14;20) - should be considered high-risk only if chromosome 1 aberrations are also present. In terms of those chromosome 1 aberrations, category C, first deletions of 1p: Occurring in about 13% to 15% of newly diagnosed disease, deletion 1p eliminates critical cell checkpoints and normal apoptotic signaling. In the IFM and CoMMpass dataset analyses, biallelic deletion of 1p and monoallelic deletion of 1p co-occurring with additional copies of 1q denote high-risk. In terms of the other aberration in chromosome 1 possible in myeloma, gain or amplification of 1q: This occurs in up to 35% to 37% of newly diagnosed disease. It upregulates CKS1B, which is a cyclin-dependent kinase, and ANP32E, a histone acetyltransferase inhibitor. GEM and IFM data suggest that gain or amplification of 1q - there was no clear survival detriment to amplification - is best considered as a high-risk feature only in combination with the other risk factors as above. Now, in terms of any other criteria for high-risk disease, there remains one other item, and that has to do with tumor burden. There has been a consensus shift, really, in both the IMS and IMWG to attempt to develop a definition of high-risk disease which is based on biologic features rather than empirically observed and potentially temporally dynamic features, such as lactate dehydrogenase. Beta-2 microglobulin remains an independent high-risk indicator, but care must be taken when measuring it, as renal dysfunction can artificially inflate peripheral titers. The consensus conclusion was that a beta-2 microglobulin of at least 5.5 without renal failure should be considered high-risk but should not preclude detailed genomic profiling. So, in conclusion, the novel 2025 IMS-IMWG risk stratification system for myeloma is binary. It’s either high-risk disease or standard-risk disease. It’s got four criteria. Number one, deletion 17p and/or a TP53 mutation. Clonal cell fraction cut-off, remember, is 20%. Or number two, an IGH translocation - t(4;14), t(14;16), t(14;20) - with 1q gain and/or deletion of 1p. Or a monoallelic deletion of 1p with 1q additional copies or a biallelic deletion of 1p. Or a beta-2 microglobulin of at least 5.5 only when the creatinine is normal. This is a field-defining work that draws on analyses from across the world to put forward a dominant definition of high-risk disease and introduces a new era of biologically informed risk assessment in myeloma. Now, how does this change our clinical approach? FISH must be performed on CD138-enriched samples and should be performed for all patients. Next-generation sequencing should also be performed on all patients. Trials will hopefully now begin to include this novel definition of high-risk multiple myeloma. It does remain to be seen how data from novel therapeutic trials, if stratified according to this novel definition, will be interpreted. Will we find that therapies being evaluated at present have differential effects on myelomas with different genetic lesions? Other unanswered questions also exist. How do we go about integrating this into academic and then community clinical practice? How do we devise public health interventions for low-resource settings? To discuss this piece further, we welcome the esteemed Dr. Nikhil Munshi to the podcast. Dr. Munshi is a world-renowned leader in multiple myeloma and the corresponding author on this paper. As Professor of Medicine at Harvard Medical School, Director of the Multiple Myeloma Effector Cell Therapy Unit, and Director of Basic and Correlative Science at the Jerome Lipper Multiple Myeloma Center of the Dana-Farber Cancer Institute, he has presided over critical discoveries in the field. Thank you for joining us, Dr. Munshi. Dr. Nikhil Munshi: Oh, it’s my pleasure being here, Michael, to discuss this interesting and important publication. Michael Hughes: I had a few questions for you. So number one, this is a comprehensive, shall we say, monumental and wide-ranging definition for high-risk myeloma. How do you hope this will influence or impact the ways we discuss myeloma with patients in the exam room? And how do we make some of these components recommended, in particular next-generation sequencing, feasible in lower-resource settings? Dr. Nikhil Munshi: So those are two very important questions. Let’s start with the first: How do we utilize this in our day-to-day patient care setting? So, as you know well, we have always tried to identify those patients who do not do so well with the current existing treatment. And for the last 30 years, what constitutes a myeloma of higher risk has continued to change with improvement in our treatment. The current definition basically centers around a quarter of the patients whose PFS is less than 2 to 3 years. And those would require some more involved therapeutic management. So that was a starting point of defining patients and the features. As we developed this consensus amongst ourselves - and it’s titled as “International Myeloma Society, International Myeloma Working Group Consensus Recommendation” - this IMS-IMWG type of recommendation we have done for many years, improvising in various areas of myeloma care. Now, here, we looked at the data that was existing all across the globe, utilizing newer treatment and trying to identify that with these four-drug regimens, with transplant and some of the immunotherapy, which group of patients do not do as well. And this is where this current algorithm comes up. So before I answer your question straight, “How do we use it?” I might like to just suggest, “What are those features that we have identified?” There are four features which constitute high-risk disease in the newer definition. Those with deletion 17p with 20% clonality and/or TP53 mutation. Number two, patients with one of the translocations - t(4;14), t(14;16), or t(14;20) - co-occurring with 1q amplification or deletion 1p32. And that’s a change. Previously, just the translocation was considered high-risk. Now we need a co-occurrence for it to be called high-risk. The third group is patients having biallelic deletion 1p32 or monoallelic deletion 1p32 along with 1q amplification. And finally, patients with high beta-2 microglobulin, more than or equal to 5.5 mg/dL, with normal creatinine less than 1.2 mg/dL. And the question, “How do we use this?” There are multiple areas where we incorporate high-risk features in our treatment algorithm. One of the first areas is where we would consider the induction regimen. If a patient has a high-risk disease, we would definitely consider a four-drug regimen rather than a three-drug regimen, although we are beginning to incorporate four-drug for all groups. That’s one important thing. Number two, those are the patients where we do consider consolidation with transplant or maybe in the new world, considering some of the immunotherapeutic consolidation more early or more aggressively. Number three, these are the patients who get a little bit more maintenance therapy. So normally, lenalidomide might end up being our standard maintenance regimen. In patients who have high-risk disease, we incorporate either addition of daratumumab or the anti-CD38 targeting antibody and/or addition of proteasome inhibitor, either bortezomib or carfilzomib. So you would have multi-drug maintenance therapy in these patients. And in high-risk patients, we follow them with maintenance longer periods of time. One very critically important point to keep in mind is that to get the better outcome in high-risk disease, we must try to get them into MRD negativity because there is clear data that patients who do achieve MRD negativity, despite having high-risk disease, have a much superior outcome. They become near to standard-risk disease. And so, in high-risk patients, I would try to do whatever various options I have to try and get them into MRD-negative status. And when these patients relapse, we do not wait for the classic progression criteria to be met before we intervene. We would propose and suggest that we intervene earlier before the disease really blasts off. And so there are a number of areas in our setting where this high-risk definition will help us intervene appropriately and also with appropriate aggressiveness to achieve better outcome, to make this similar to standard-risk disease. Michael Hughes: Thank you, Dr. Munshi. And thoughts on how to really integrate this not only into academic centers but also lower-resource settings? Dr. Nikhil Munshi: So that’s a very important question, Michael. And when we were developing this consensus, we were very cognizant of that fact. So wherever available, I think we are recommending that over a period of next 2, 3, 5 years, we should begin to switch over to sequencing-based methods because two components of this definition, one is TP53 mutation, which we cannot do without sequencing, and also reliably detecting deletion 1p requires sequencing-based method. So in the low-resource countries - and there are many in this world, and also even in our own country, patients may not be able to afford it - the older method with FISH or similar such technology, which is more affordable, is also acceptable for current time. They may miss a very small number of patients, maybe 2% to 3%, where these finer changes are not picked up, but a majority of this would be captured by them. So the current practice might still be applicable with some limitation in those patient populations, and that’s what we would recommend. What is happening, fortunately, is that actually sequencing-based method is becoming cheaper. And in many centers, it is cheaper to do the sequencing rather than to do the FISH analysis. And so my hope is that even in low-resource centers, sequencing might be more economical in the end. It’s, I think, the access to technology, which is a little bit limited currently, but it’s hopefully becoming available soon. Michael Hughes: Thank you, Dr. Munshi. And staying for a minute and looking at the multiple myeloma subsets which might be missed by this really still very broad-ranging high-risk definition, at least by prior risk stratification systems, right, there is this group of patients who have standard-risk cytogenetics by R-ISS or R2-ISS, but they have primary refractory disease or they relapse early. We call these, as you are well aware, functionally high-risk disease. What proportion of previously FHR, functionally high-risk, myeloma patients do you expect to be captured by this novel definition? Dr. Nikhil Munshi: So I think the newer definition - and we can look at it both ways, but the newer definition should capture most of the functionally high-risk definition. To put it differently, Michael, there are patients who we know are, as you mentioned, functionally high-risk. Those are the patients who might have plasma cell leukemia, those who might have extramedullary disease, those who might not respond to our four-drug induction. If you don’t respond to the four-drug induction, almost by definition, they are high-risk. However, a majority of them have one of the abnormalities that we are describing here. There would be a very small proportion which may not have. And if they do not have, we know one of the important components of this definition here is also that the genome, we know, keeps on evolving. So there may be a very small clone with the high-risk feature which was not obvious in the beginning. Following treatments or following relapse, that clone predominates, and now the patient’s disease becomes high-risk. So the definition would incorporate or would capture these functional high-risk patients, but as you said, in countries where resources are not available, using this functional high-risk would also be helpful and advantageous. Sometimes LDH ends up being a high-risk. In our studies, LDH has not come out to be high-risk anymore because the features we are describing captures most of those patients, but those alternatives, older, can still be considered if other newer techniques are not available. Michael Hughes: Got you. And in terms of these older definitions, yes, that incorporate tumor burden, these empirical observations about how myeloma presents, do you foresee any additional tumor burden indicators being added to future definitions of high-risk disease? Or do you instead see this particular definition as a major waypoint on the journey towards a fully biologically grounded definition of high-risk disease? Dr. Nikhil Munshi: I think your second part is what is going to happen. I think the tumor burden-related definition is being now replaced by the biological or genomic-based definition. And I think at some point, it will be quite fully replaced. One component not here, and it is because one thing, we don’t have enough data; number two, we don’t know how it will pan out, is also the influence of the microenvironment on the risk definition. For example, the immune system, the immune function, etc. But not enough data exists to suggest how it would change the current definition. So in future, would a definition be totally genomic or it could be more integrative? And my personal guess is that it would be more integrative and that some immune features might come into the picture, especially now that we are using immune-based therapy as a very important component of treatment - CAR T-cells, bispecific, and antibody-based treatments. What role the immune system plays in either supporting tumor or what role suppression of the anti-tumor immunity plays? They all will be important how patient outcomes end up being, and which in turn could translate into how patient’s risk stratification might happen. So I think the older tumor burden-related definitions probably will become things of the past. What we have currently proposed and consensus developed is the new path forward, and over time, some microenvironmental influences, if defined and found to be important, may get some more incorporation if it compares favorably with the genomic features. Michael Hughes: Thank you, Dr. Munshi for that enlightening response. To conclude the podcast, I’d like to look to the future and to the immediate future, what are the next steps for high-risk disease definition between now and discussing an integrated genomic-microenvironment-based definition? Will we see attempts to refine? Will we see a multi-level system, things like this? Dr. Nikhil Munshi: Yeah, so I think the current definition will be here to stay for the next 10 years or so. I think this has been developed using a large amount of data, so we do believe that this will remain fine. It has been validated now within the last six months by a few of the other studies. So there won’t be a quick change. But we will try to, all of us will try to innovate. And as you very rightly bring up, the areas of research would include looking at the expression or transcriptomic component. Does that matter? And we do believe a small number of patients will have transcriptomic changes, not looked at the DNA changes, and may play a role. There are newer components, so long non-coding RNA, for example, is going to be an important component to look at, how it impacts the disease outcome, etc. There are also some of the proteomic-related changes which may become important in our studies. And then as we discussed, microenvironment and immunological changes. So these are the future areas of ongoing research where we all should collect data, and then in the next 5 to 10 years, we’ll have another group meeting to see has anything changed or any of the features have become more important. Most of the time, some of the older features are lost because they are not as critically high-risk, and the newer features come in. And so the historical background for just one second, there was a time when chromosome 13 was considered a high-risk disease. We now don’t even mention it because it’s not high-risk. The newer treatments have improved the outcome. t(4;14) used to be a high-risk disease. Now by itself today, in this definition by itself is not; it needs to be with something else. And so I think this is a great sign of progress. As we improve the treatment and outcomes, some of the features will become less important, new features will come up, and we’ll need to keep on evolving with time and with technology and make it better for patients. Michael Hughes: Thank you so much, Dr. Munshi, for your wisdom, for your sagacity, for your historical perspective as well. Thank you for listening to JCO Article Insights. Please come back for more interviews and article summaries. And be sure to leave us a rating and review so others can find our show. For more podcasts and episodes from ASCO, please visit asco.org/podcasts. The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions. Guests on this podcast express their own opinions, experience, and conclusions. Guest statements on the podcast do not express the opinions of ASCO. The mention of any product, service, organization, activity, or therapy should not be construed as an ASCO endorsement.
Brasil