[logo] HealthTree Foundation
search more_vert
close
person Sign In / Create Account

Creating a Risk Profile to Personalize Myeloma Care with Ola Landgren, MD, PhD, University of Miami

Creating a Risk Profile to Personalize Myeloma Care with Ola Landgren, MD, PhD, University of Miami image

Creating a Risk Profile to Personalize Myeloma Care with Ola Landgren, MD, PhD, University of Miami


Jul 15, 2024 / 11:00AM MDT
HealthTree Podcast for Multiple Myeloma
link icon facebook logo X logo linkedin logo reddit logo email icon

Also listen to us on:

spotify apple podcast youtube

Episode Summary

Dr. Ola Landgren shares a clinical trial about smoldering myeloma. Join the REVIVE trial study for high-risk smoldering myeloma study here. 

He also share about his research partnership with HealthTree on a personalized medicine study that creates a computerized model of risk prediction and treatment recommendations for newly diagnosed patients. 

This is a study that is open to all myeloma patients regardless of where they are being treated. The only involvement by patients is to sign a consent form and connect their medical records which requires less than 5 minutes of participation. 

Full Transcript

Join Dr. Landgren's Personalized Medicine Study

Jenny: Welcome to today's show of the Health Tree Podcast for Multiple Myeloma. I'm your host, Jenny Ahlstrom, and I'd love to welcome you to today's program. 

Today we have with us Dr. Ola Landgren, who is at the University of Miami leading that program and has done an astounding job creating this program really from scratch. Dr. Landgren, welcome to the program.

Dr. Landgren: Thank you very much, Jenny. It's really a great honor being on your program. I don't know how many times we have done this before. I feel it was yesterday we started, but I think we have done it for many years. It's really a great pleasure working together with you. 

Jenny: Yes, you were one of our very first shows, so I'm thrilled. And that was when you were back at the NIH. Let me give you an introduction for you before we get started, and then we will jump in because we have a very full program to talk about today. 

Dr. Ola Landgren is Professor of Medicine at the University of Miami and head of the program at University of Miami. Prior to his appointment, he was head of the program at Memorial Sloan Kettering Cancer Center, Professor at Weill Cornell Medical College and the National Cancer Institute and Karolinska University Hospital in Sweden, where he's from. 

He is an author of over 403 publications and received the American Society of Clinical Investigation Award and many NCI awards including the NCI Research Highlights Award, the Bench to Bedside Award, the Intramural Award, and the Director's Intramural Innovation Research Award. 

At the University of Miami, as I mentioned before, he is truly building an astounding team, an amazing program, with a specific focus on the genetics of multiple myeloma, in addition to his precursor condition work on MGUS and smoldering myeloma, and so much more. And we are going to talk about both of those topics today.

So Dr. Landgren, as we think about research, patients think they know that research is happening behind the scenes on their behalf. They are thrilled to participate in research if they're aware of it and they understand it and they are able to contribute. 

And really the first path that was available to patients was clinical research or traditional clinical trials to get a drug to market and tested in different scenarios. 

Can you help us understand clinical research in general and how that in particular advances cures for myeloma patients? 

Dr. Landgren: I think there are many different types of research that are available for patients that are interested in participating. can, as a patient, be part of a prospective clinical trial. So clinical trials, historically, have been developed for later lines, but we have also trials for newly diagnosed patients. 

We also have for patients in this disease for precursor conditions with smoldering myeloma. So what the trial does that you cannot get outside of trial is giving access to a treatment that's not yet FDA approved. Many times a drug can be in trials for years before it becomes available. 

For the newly diagnosed setting has been the case for over 10 years if you look historically how the endpoints were developed. 

That is about to change and we had a discussion with the FDA through the ODAC committee earlier this spring where minimal residual disease (MRD) has been now voted 12 -0 in favor of MRD as an early endpoint for accelerated approval. So historically it's taken many years. 

So going on a trial, in the old era, so to speak, has given patients many years earlier access to newer drugs. With MRD being an endpoint for accelerated approval, the expectation is that drug development will go much, much faster, which will allow patients to get access to newer therapies much faster. 

In addition to clinical trials, there are also real world studies that are ongoing. So as a patient, one does not have to go on a particular study, but many times you have to sign consent to participate in these retrospective studies. Or institutions can also use some of the data based on all the restrictions for scientific review and ethics review board and so forth to just look in bulk how patients are doing. 

You can look to see for patient treated with a certain indication for how long the durations and the progression for his survival last in the real world. So that's outside a clinical trial.

Some of these studies will require a patient to sign their consent, but some of them could also use some more basic variables that could still be very useful. 

And I think over time, what we have learned is that the clinical trials historically have really been the gold standard, but the real world evidence has increasingly gotten attention, I think for the right reason, because the clinical trials capture all the new data on the new drugs, but the real world shows really how those drugs once they're FDA approved, how they actually deliver. 

And many of the clinical trial patients are selected based on inclusion criteria. So some patients may not make it to the clinical trial. So when you look at the data from the clinical trial, it looks maybe better than the real world. And that's very important to know as a treating physician, how good is actually a particular drug or a combination of drugs in the real world because that's where you're going to use it. 

So I think they are very complimentary to each other and more more focus over time has been on these real world studies. I'm not saying one is better than the other. I would say both of them are very, very important. 

Jenny: Well, let's talk and use some examples. So I know you've done a lot of work in the precursor condition space and you have for decades. And this could include monoclonal gammopathy of undetermined significance (MGUS) or smoldering multiple myeloma (SMM).

I know in these precursor conditions, clinical trials are really so key. With smoldering myeloma, it's not a guarantee that somebody's going to progress (to active myeloma), but then in high-risk smoldering myeloma, for example, you might be within a two-year period of progressing. So can we shut things down beforehand? Is there a curative path for these precursor conditions?

There's so many important questions when it comes to these precursor conditions. So can you give us an example of a study, maybe an MGUS or smoldering myeloma where you're trying to achieve these findings? 

Dr. Landgren: I spent a lot of time thinking about those questions. I probably worked on it for actually over 20 years. I was the lead investigator over 15 years ago for the study that was able to prove that all myeloma patients have a proceeding phase of monoclonal gammopathy. 

So that we published in 2009 based on over 77,000 people from the PLCO cancer screening trial. And I did that study in collaboration with the Mayo Clinic with Bob Kyle and the team there. And we showed that all patients with myeloma previously had a precursor phase. And that was a study that I think was very important because now we have a path towards the disease.

We know that if you go through the precursor phase that some patient will progress and some patient will not. Before that study was done, the thinking in the field was that many myelomas just showed up in the middle of nowhere, like they just happened. But that study showed that that's not the case. They are all slow progressors for the most part, and some of them are faster progressors on a timeline. So the next step then would be to study mechanisms of progression.

And also there have been a lot of initiatives and I worked on that for years, trying to develop treatment strategies. I think I've by now done four or five or more intervention studies. I did the first three drug combinations therapy, which also 10, 15 years ago, I did a Kyprolis, Revlimid, dex (KRd) study for smoldering myeloma and we showed amazing responses, deep responses. We actually updated it about a year ago, showing that eight years later, 70% of patients remain MRD negative, which is quite amazing. 

We treated them for eight months, two years of maintenance with single drug lenlidomide, and then we stopped the therapy. So that's a pretty amazing result. 

But I've been wrestling myself with the question, should we really develop more more trials in that indication, or should we try to identify multiple myeloma earlier? 

I develop a lot of the genomic tests and we have a trial called the TRANSFORM study, which is a biomarker study without intervention. So the idea is to pick up genomic changes in people that are programmed to progress. And that really triggered me to think about what should we do if we find someone having higher risk of progression, or if someone just comes to us here in Miami and says, I have high risk smoldering myeloma. Well, the FDA doesn't have any approved drugs for that space.

And as much as I think the KRd results are amazing, it is quite involved therapy with very many infusions and visits and things like that. 

So we decided to open a new study that's been open for a few months. It's called the REVIVE study, where we give bispecific antibody once a week for one month. And we give it in combination with a CD38 targeted monoclonal antibody, also the same once a week for a month.

But after the first cycle is done, that combination is only given once a month. So it's a very low intensity treatment for the individual patient. Low side effects, not very involved, it's subcutaneous injections. And the study allows for up to two years of treatment, once a month. And after that, the goal of the study is to see if we can bring over 70% of patients into MRD negativity. 

And after stopping therapy, we are seeking to see if we can also keep that MRD negativity beyond five years in over 70 % of patients. That's the goal of the trial. That study is open here in Miami and patient can either live here in the Miami area or even travel here because it's only once a month dosing. So that's an example of a recent study we have. 

Jenny: And bispecific antibodies haven't been tested a lot in high risk smoldering myeloma or any smoldering myeloma, right? 

Dr. Landgren: That’s correct. They are FDA approved. As you know, about one and a half years ago, the first bispecific antibody was approved for multiple myeloma for patients who failed four or more prior lines. So it's a pretty late line therapy still. 

And about a year ago, there were two additional bispecific antibodies again, developed and approved for that later line.

They are in trials for investigation for patients with one to three prior lines. So that will still be relapsed disease. 

And there are ongoing investigations to even try it in the newly diagnosed setting. That's going to probably take longer. So going all the way to smoldering myeloma is obviously a bold move. 

And our thinking was given how effective those drugs are in the relapse refractory setting, and given what we have investigated and published on the disease by already being less aggressive, less developed, less advanced in the small room space, our thinking was, can we even eradicate the disease? 

So I don't know that at this time, but that's the hypothesis we want to test to see if you can really potentially get rid of the disease.

Jenny: Well, that'd be amazing. So no one actually progresses to active myeloma.

That's the dream, right? It'd be really nice to achieve that. It's incredible. 

Dr. Landgren: I can obviously not promise that until we have the results from the study, that's the hypothesis that we have and that's how we have developed the study. 

Jenny: Wonderful.

Dr. Landgren: Once it's a trial, it's obviously voluntary to participate. There is no randomization. So all patients are treated with the same drug. So every patient who goes on the trial will know what he or she gets.

Jenny: We will definitely include a link to that trial when we have our full transcript after the show's complete so patients can learn more about that study and send you questions if they're interested in joining. So thank you for sharing that. And that's the difference too between what you call an interventional clinical trial, right? And where you're actually intervening with drugs or treatment and an observational clinical trial where you are not necessarily providing any medication, but you want to test and see outcomes. 

So let's talk about that next, because you have a very important project that you're working on and you're working on building what you call the IRMMa project. 

So maybe you want to explain the idea behind this and what you're attempting to learn and what you're building based off of this information. 

Dr. Landgren: So the URMA study is really centered around the idea of trying to think outside the box. And it goes for the newly diagnosed patients. 

And I think the whole literature is built on averages. And that's something that really has bothered me a lot. So no one is average, everyone is unique. 

And you can look in one study and you see that on average, this or that therapy lasts for say five or six years, or on average, this therapy lasts about eight years. It sounds like the latter of those two would be the better, but no one is average. 

So within both the first and the second example, there would be patients that do better and that will be patients that do less well than the average. And you don't know who you are in that kind of setting - if you are the average, if you're above or below the average. So if that was a more individualized risk prediction, of course you would pick that. Anyone would do that. 

So how could you get that? So what we were thinking is instead of just taking all the patients together and look for one therapy and comparing them with every patient together with another therapy and look for averages, let's look within each of these compartments. And not only looking at age, gender and race and things like that, looking in full detail of the disease biology. 

So we were able to gather large numbers of patients with whole genome and whole exome data. And we did that in collaboration with our long-term friends and collaborators I worked with for very many years around the world in Europe and here in the US and elsewhere. 

So people were willing to share protocol data based on IRB approval and scientific review approval. And it took some time to put all the paperwork together, but once all that has been hammered out, we could then share data as part of this collaboration. 

We don't know the individual patients, but we knew enough variables to use it for the purpose of building a large database. 

So about 2,000 individual records were included. And the beauty of having a collaboration with so many centers is that very many centers have one strategy, while other centers have another strategy.

Some other centers had a third strategy and so forth. There were so many ways of treating multiple myeloma. And then on top of it, the disease is very different between patients. So you're basically looking at different diseases treated in different ways. 

And that is really what we took advantage of. So building this large database where we factor in all these things allowed us to build it almost like a Chat GPT database, AI database. So you could now add a new case to the database and say, given everything that's in the database, here we have this new case that has this profile, what would be the optimal therapy to give this case given what's in the database? 

And now the computer could look at what's known about this case, compare it to all the cases in the database, and find the ones that are either exact or similar, and see how they were treated differently and see how that translated into longer progression free survival and say based on what's in the database, if you were to do this approach, this is what you would expect in terms of outcome. 

But also the database would say, if you pick this treatment option or this treatment option or this treatment option, this would be the difference in the clinical outcome. 

So you basically compare an individual to all those individuals that the same, similar to that individual case. So that's what an individual risk model is. You could also ask questions that are more centered around a particular therapy. You could say, let's look in the database, show me cases that were treated with a certain therapy. 

I'm just making an example now with the addition of a transplant where there are also cases that look genomically the same way as the first group I mentioned that were treated with the same combination therapy without a transplant. And the outcome in terms of clinical outcome is basically the same. Well, the computer will say, well, if you have this genetic makeup, these are the cases. So that answers the question, is there a group where the addition of transplant doesn't translate into longer progression free survival? And the answer is yes and the computer will tell you what that looks like. 

And you could make the same type of questions, say adding a CD38 targeted antibody, adding another drug, et cetera, other cases where that doesn't really make any difference. And those are very important questions. That's how I see the field needs to go forward. 

Jenny: Well, I think it's extraordinary because Benjamin Diamond at your center presented some slides at our HealthTree Roundtable when we came to Miami.

And it was astounding to see how you would map patients into these different columns. And each patient had all these different genetic markers, some that we don't even think of. Like, you know, we're used to talking about like a 4;14 translocation or an 11;14 translocation or a deletion 17P, but you had probably, I don't know, 30 or more genetic markers. 

And then you were showing below the treatment that they received and if they did well or didn't do well. So you can now take all these different myeloma patients, which are so unique and individual and put them into more cohesive buckets. But for that, you need a lot of data, right? Once you're subsetting patients with those specific genetic markers.

Dr. Landgren: You're absolutely right. You need larger databases because you're looking at more distinct groups and you compare distinct small subsets with maybe an individual patient or looking at distinct small groups with another distinct small group. So with a small database, you run into the problem of statistical power. 

And what we mean with that, and we do scientific studies is there will be a certain influence by randomness. And we don't like randomness. We like to know for sure. So we apply statistical models. And if you have a large enough database, you can pretty much rule out randomness. So that means that the finding you have is a true finding. And that's kind of the basis for all statistics. So having a large database removes the influence of randomness. We have over 2,000 cases.

I'm constantly talking to groups around the world and I ask most of the drug companies if they would be willing to share their data sets. People are excited about it and I think that's also part of the future that there are so many questions and as you learn more and more about these details, you will split everything into subgroups. So no one can any longer solve the important questions on his or her own. You have to work with others.

So collaboration is critical. I always say internally that the future is really based on collaboration. It's based on transparency. I'm telling you here today what I'm thinking. I'm not keeping anything secret. I just have no secrets. I always am very transparent and high speed forward. So collaboration, transparency and high speed forward. Those are the key ingredients.

You could then say, well, how could you do that? You give away all your ideas. Someone can steal your ideas. Well, I mentioned that the third ingredient was high speed forward. And we are really intending to run fast. So if you want to compete with us, you better run very, very fast. And if you win sometimes, or we win sometimes, whoever delivers first, that's great. And it's great for patients. 

So I'd rather say, why don't we work together than trying to compete with each other and for the most part, people want that, but there are always some groups they want to compete and that's great. And that's just keeps the speed higher and higher. So that's sort of how I think about the future. 

Jenny: That's great. Well, you know, over the past 10 years, I would say we've heard a lot about precision medicine and the idea that you have these certain types of genetics and then we're going to give you these targeted therapies become a very challenging or difficult thing and it hasn't worked that well in my opinion because you have patients with maybe five different genomic features presented at diagnosis and then that evolves over time and then which clone is the biggest clone and how do you go after the clone that's actually driving the disease forward versus the clone that's not. 

So to me, this model makes a ton of sense because you can look at it and say, what are all the tools that we have to work with in the myeloma space? And then what do you have, but what worked for other people that look like you? And to me, it just seems very practical and very incredible. So, we wanted to support you in this effort. We see it. 

Do you want to talk also about not over or under treating patients and what you're learning from this? 

Dr. Landgren: So I think a consequence of knowing more precise information is that you can better tailor the treatments. 

And I mentioned, if you ask the question, is there a group of patients where you give a combination therapy plus something else that could be an antibody or a proteasome inhibitor or transplant or various types of additional therapies? Is there a group of patients where you make that addition versus you don't have that addition that the outcome is the same? The computer will be able to answer that question to you. 

So that's an example where you could say you could avoid over -treatment because you show that the outcome is not different if you add.

Similarly, you could ask this model to make sure that you don't give inferior therapy. So if you just give standard therapy because you think that's the standard, we know unfortunately that there are patients that don't have the same great outcome as the majority of patients have. Being in the field for many years, I have seen how overall survival has just kept on being better and better and better. 

But I've also seen that there is a very small group of patients right out of the gate have problems with the treatment and they don't do as well. It's a small group, but there is a group of patients that have really not benefited from all the new therapies that have been developed. 

So again, it's a reflection of this average measure of success. So the average survival is much better. The average progression for your survival is better. But for patients that are in that category, very little has really improved.

And I think a big task for the future will be to solve the mystery of the high risk disease biology. I think all the FISH and cytogenetic tools (genetic tests) are very outdated. I would not make treatment decisions based on FISH and cytogenetics because FISH and cytogenetics are basically like a black and white TV. You're looking for say 20 different markers. Yes, no. And if you do whole genome sequencing, pretty much every patient has over 5,000 mutations. 

In addition, you have skewing of the mutational signatures, you have these complex events with chromatripsis and so forth. So you basically comparing an old black and white TV with the latest high definition TV in terms of resolution. And I have seen it firsthand treating so many patients for so many years when you give the best therapies, many patients that have what historically has been called high risk, they have excellent outcomes. 

Many patients do. I've had patients coming to me, they have been even off therapy for years and they have asked the same question every time I have seen them. How could this happen? I have high risk disease. 

And my answer has been is because you don't have high risk disease. Well, the first doctor that saw me said I had high risk disease and my answer then would be, well, you know, the markers for high and standard risk disease are not that great. And when they told you that they thought that because that was the average. 

Knowing what we know today is that you have been off therapy, say for five years and you still have no detectable disease. So clearly you don't have high risk disease and that's a good thing. 

I know also firsthand, having treated thousands of patients that there are patients that don't have high risk disease markers that fare poorly and they have these other features that I'm talking about. 

So a big task for the future is to better understand the high-risk disease. We talked about statistical power before, and one of the challenges here for the high-risk dilemma is that the therapies are better and better, so many patients can do really well. So individual centers, thank God, have fewer patients with bad outcome. 

But from a scientific and from a problem solving point of view that makes it also very difficult because now we have so few cases to really investigate. So if you don't collaborate with others, you will never be able to solve it. 

So working together, building big databases and trying to understand better the biology of the high risk disease is going to be critical. And I think these individual risk models such as IRMMa are critical steps forward in this direction.

Jenny: Well, I think what's key also, you mentioned a lot of things that I want to just comment on. You you have about 80% of patients being treated in the general oncology community and they don't see this kind of pattern recognition that you've seen over the course of 30 plus years in the space. So when we did our 50 city tour and we first introduced HealthTree Cure Hub, we saw, I don't know, I would say at least a third to a half of patients, that appeared to be getting under treated.

And so these practitioners are treating 20 plus different kinds of cancer. They're using maybe NCCN guidelines or things like that to treat. They don't necessarily know how to treat myeloma in this nuanced of a way. And so when you talk about high-risk disease biology patients or even standard risk disease biology patients, back then we were even seeing patients just on doublets, not on triplets. Now we're talking about quad therapies.

The whole community needs to get up to speed more quickly with a model like this. To me would be really astounding. To be able to share that model and say, look, all these samples have gone into it and we have thousands of samples, we've developed this model and allowing other people to use that model to make key decisions for their patients would just be completely game changing in my opinion. 

Also, it brings up the fact that The Health Tree approach, because you're going to centers and inviting their collaboration, we're going directly to patients and saying, you have control over your data, you have control over your samples, it's your data sitting in the hospital, you can do anything you want with it, and we'd like to invite you to participate in a study like this. 

Maybe we can jump into our collaboration and how you see Health Tree helping advance this project. 

Can you explain the project, our collaboration, how it works, and then why you think it's important? 

Dr. Landgren: I’m really very excited about this collaboration that we have formed and we have worked on for a long time. have talked about it, but now really is sort of off the ground and we are working full speed for it. I think this will be the 2 .0 of these IRMMa project in terms of deliverables. 

So you have built a fantastic network of all the patients and as you just said patients have a lot of information sitting at the institutions. The institutions have the data too but when you put together many institutions that's when you have the power along the lines of what I was trying to outline before. Every institution has its way of managing the disease and the disease biology is different from patient to patient so when you start putting all the pieces together that's when you start seeing all these patterns, you can actually generate more individualized outcomes and you can build these types of models. 

So the collaboration that we are working on to outline that for listeners here is based on our 2.0 initiative where we are using stored biopsies, patients that were newly diagnosed were worked up to confirm the diagnosis.

The institutions have what's called paraffin blocks where these biopsies are sitting in the pathology department. That was how the initial workup was done. The bone marrow biopsy was done. The core biopsy was embedded in paraffin and the laboratory, the pathology lab at that institution would cut thin slices, put them on a piece of glass. They would color them and look in the microscope and they would confirm the diagnosis. That's part of the standard workup.

So what we have done in my lab is to develop strategies to use some of the unstained parts of the core biopsy. So from these biopsies, you have hundreds and hundreds of slides that are potential to cut off. You can take the block and cut off a few slides and you could analyze that with new approaches that are not yet standard. And that's what we have developed.

So we are asking patients to give consent to HealthTree Cure Hub. HealthTree can ask on behalf of the patient to send just five unstained slides to our laboratory. That's part of the research. And we are then using our technology to analyze these slides. There are potentially hundreds and hundreds of slides sitting in these core blocks. So we are asking for five. And with those five, we are then characterizing thousands of cases. And we have already started generating preliminary data where we show that with the approach we are taking, that you can basically recreate a lot of that information that more advanced profiling is doing. So even FISH, if there was no FISH done, we can with the technology we have developed, take these slides and we can with a very high likelihood predict what type of FISH changes there are in these slides. 

Now we're going towards more advanced genomic changes. So basically the idea is to say, if you have just a few slides that you can use the technology that we have developed to basically predict disease heterogeneity, the unique patient disease biology. And again, coming back to what I talked about before, if you know how the treatment was given, you can use that to predict clinical outcomes. 

And you can also see how different therapeutic choices potentially could impact long-term. For people that are worked up, that have these biopsies sitting, I'm not sure exactly how this information for that individual could be useful, but it for sure could be useful for patients in the future. 

What I'm hoping to generate from it is also signatures if you want to change therapy in the future. So it's really an initiative in the direction of precision medicine. So it's a big project. We already have already several hundred consents for the HealthTree collaboration. We are including thousands of cases. My hope is that we could potentially just use the slides to get the information that would cost thousands of dollars to do with these expensive assays that most cases have not been worked up with. 

Jenny: This is a combination of the original bone marrow slides that you're testing genetically to see what kind of genomics these patients have at diagnosis. And then we're also inviting patients to connect their EHR or medical records data to that project. And then you can see over time what people had for therapy, how they responded to therapy, how long their remissions were, what's their time to next treatment, and all those types of factors because then you have the beginning and how long people are living with that particular treatment paradigm. 

And then you can go back and say, okay, now the computer is building this model out to say, well, for patients that have these types of genetics, you're likely to have this kind of outcome or response. So it's that combined piece. So HealthTree is now reaching out to all myeloma patients. This is a way where we can get people in every state participating in this study because you don't really have to be in Miami to participate in this, right? 

Dr. Landgren: You don't have to be even close to Miami. You could be in Seattle or Los Angeles or in the middle of the country or New York. You could be anywhere. It could be in a big city or it could be a small town. It doesn't matter if you have a diagnosis of myeloma and you had a biopsy at the time of the initial workup and you want to participate. can consent to the study that HealthTree has the consent for and we have a collaboration between HealthTree and the University of Miami. 

So that consent allows HealthTree to share the slides and the data. We won't know the individual patient's name and details, but we will have enough variables to predict clinical outcome the way you outline. So we again, are trying to look at what we statistically call the interaction between the disease biology that's different in every patient - difference in treatment that varies from center to center to see if another patient that shows up with a particular signature that fits with a subgroup of the patients in the database, which treatment based on what's known in this database would take the patient to the furthest time point out, which would have the best prognostic delivery, which is the optimal therapy for this patient. And also to ask the questions that you asked before.

How could you avoid over and under treatment? So those are very similar questions that we are trying to address. And you really have to use computerized models nowadays because the treatment landscape is changing so fast. 

Jenny: Bispecifics are coming on that you talked about earlier. CAR -T therapy is here. Vaccines and NK cells, I mean, all this new stuff is coming and you're not going to be able to iterate fast enough to be able to run individualized clinical trials for each different treatment combination for each type of patient. It's just too expensive. And once the drug is approved, typically the pharma company might not be that incentivized because it's approved and people can use it. And yes, other indications are necessary for those drugs. But this could give you a really fast real world modeling.

Dr. Landgren: Absolutely agree on that. think that the future is going to rely much in the future, we're going to rely much more on these virtual control groups, if you want, where you could compare data from a study with data from a computerized model where you can much better tailor the control group to your treatment group. Right now, the gold standard and has been for a long time to randomize patients between therapy A and the control arm. You have the experimental therapy and the control arm. 

I think in the future it's very likely that you could have single arm studies where you give the same therapy to every patient and you just have enough information about the disease biology and all the details for the patients and you could match those in the computer to individuals that match the individuals. So the field is not yet there.

And you know, I worked for a long time with the FDA on the MRD project that we delivered in April of 2024. Every time I talked to the FDA, I said to them that MRD is important, but the next thing will be to think about these virtual control groups and they are smiling, they're saying, we're not yet there. I'm gonna keep on pushing for that because I think that is the future for drug development. So think how you could really pioneer the field if you had virtual control groups. 

You don't need to randomize patients to substandard therapy and you probably get the results much faster because it's going to take half the time to enroll the patients. You don't need to enroll twice as many patients, first the experimental and then the control arm. You can just do the experimental arm and you can just get the results very quickly. 

Jenny: We're completely on board with that as an organization. That is our goal is to be able to provide that kind of data because we have 14,000 plus patients who have joined this platform and are helping to share their data. But it's so important to do projects like this with investigators because you have that myeloma expertise and the research expertise and all of that. So just to help people who are listening understand how we're involved, you kind of designed the entire platform. We are inviting patients to participate. 

We make the request once they consent to their facility to get those unstained slides. We send those slides to you. You do the genetic testing on those slides. 

And then a patient is also consenting to share their medical records so that you can see the long -term outcomes of that particular therapy. So that's how we're facilitating. This is what I wanted when I was diagnosed, I would say, because the fight of the time, you know, I was 43.

I was younger, I had a high risk genetic feature show up on my next generation sequencing, my gene expression profiling test, my GEP test, but not on my FISH test. And the fight at the time was single versus tandem transplants. Now to me as a patient, I looked at it thinking, why is there a fight? Why isn't there just data around this to show me how long I would live on both protocols and then let me make the decision? 

An 80 year old might not want tandem transplants, but I was 43 and that was my best shot at the time. There was no daratumumab, there was no nothing. And so, you know, you had like RVD maybe, or some of those basic myeloma combinations that are not curing. So like that was my best shot. So I went for it, but not with the data in hand. And that's what I really wanted.

If we as patients can help support you as a researcher as building this out, I think it's magnificent and this is one of the most important projects that I think we're helping with. And I really want to encourage patients to participate in this. 

Join the Study

So where can you go to sign up might be a question. If you go to healthtree.org/myeloma/curehub you can create a CureHub account and in Health Tree CureHub, there is a section on the lefthand side that says Accelerate Research. So we have a list of all of our open partnerships with investigators like Dr. Langren, and you can go click on the one that's talking about personalized care. And your name is on there, so it'll be clear which one is, that patients can sign up for.

That's how it works, just practically speaking. And what else?

Dr. Landgren: I would like to add that the project does not require any further biopsies. It doesn't require any further blood tests. It doesn't require any further testing or any intervention. The only thing that we need access to as part of this collaboration is basically access to the data that's restricted, that's IRB approved, the Ethics Review Board and our Scientific Review Committee has reviewed and approved that. 

So the limited data points and access to those five unstained slides that are sitting somewhere in a pathology department where that individual that consents to participate was initially worked up. So there is no additional testing required. It's only available data that we need. 

Jenny: That's a great point. And just to further explain what IRB (Institutional Review Board) approval is. So, Health Tree, the whole platform for Health Tree Cure Hub was IRB approved very long time ago. Every time we do a new project with an investigator, we will go and get project level IRB approval. And then University of Miami also got their own IRB approval. So, we're now double approved for this research project. 

But we're so thrilled to be facilitating. We're very excited and we did have some patients who registered for this event to have some questions. So if you don't mind if I ask some of those questions. So Darlene is asking, will you be testing risk for multiple myeloma and amylodosis or just multiple myeloma? 

Dr. Landgren: So the project we are working on together is built on multiple myeloma. You could technically use the platform to study actually probably any disease. It could be other plasma cell disorders such as AL amyloidosis or it could be other diseases as well. But the way we have designed it is to focus on multiple myeloma. 

That comes back to what we talked about before that we need to have the statistical power. So if you start doing a little bit of everything, you're going to have a few cases here and there, but it's not going to be enough to rule out randomness.

It's only multiple myeloma for now. 

Jenny: Thomas was asking, will the model help us understand when to stop maintenance therapy?

The study will not be able to say this is what you shall or shall not do because it's a research project. The individual doctor and individual patient will have to make those decisions, but it will be a powerful research tool that will suggest maybe there are certain disease biology subtypes where if you do one thing versus the other, it's say you stop maintenance or you continue maintenance that if a high percentage of patients remain free from disease many years out, that would be a strong indicator. And those could be proposals where we currently don't have any data. 

Dr. Landgren: You talk about that, Jenny, that having data is really critical. And I 100% agree with that. And you can only build data by having large numbers and long follow -ups. So hopefully the model will be able to generate that, then the doctor and the patient will have to use that for decision making. 

We cannot give 100% direction. We are not going to replace all the doctors, that's for sure. 

Jenny: No, I mean, everything is so personalized. I think giving that guidance though, to say these are the five different options you could consider. Let's look at how patients like you are responding to those options. I think the extension of that would to be include side effect data, which some of the information that we're gathering and say, okay, here's the efficacy of this particular therapy. Here's the side effect burden. Now you go and talk to your doctor about what's right for you because this is your situation. 

This is how far you live from a treatment center. This is how old you are. These are your life goals. Maybe it's quality of life. Maybe it's length of life. Everybody is so unique and different that we need the experts and we need our doctors to help us do this.

So I agree with you, the doctors are never going away, but I can see how this would be so helpful to further that kind of discussion with your doctor based on your personal priorities and the doctor's knowledge of this disease. 

So it's all needed. I think it's humanly impossible right now to be able to put all this data together in your mind. You can use clinical practice and you can use you your pattern recognition based on treating thousands of myeloma patients that you're seeing. And that's why experts, you live longer if you see a myeloma expert than if you're with a general oncologist. But building some of these tools so everybody is benefiting is really totally amazing. Okay, Vicki had another question. She said, can an early sample be used if the first biopsy sample is not available?

Dr. Landgren: So again, it's all about modeling and available data points. So the way we have built the model is that we say if there is a biopsy done around the time of the diagnosis and we have the follow-up variables that we have as part of the research study that we can then use that to model. So if there are people that are having outlay biopsies say there was a biopsy done two or three years later.

That would be something that would be very different from the rest of the database. And that would not really work. So let's say half the database was like that and half was a diagnosis. It would be hard to make sense of the data. You could of course ask the question, what happened if that was a biopsy done around the time of diagnosis and there were some issues with it. And then there was a new biopsy done a month later or some weeks later. So I think on a pragmatic note, it probably very similar, so it probably will work. But if there are multiple samples available, we obviously would like to have the one closest to the diagnosis. The way the database is set up, that we would have information on the approximate time of the diagnosis, and we would have information on the approximate time of the collection of the sample. So that would allow us to do statistical kind of adjustments to take this into account.

When you do these types of analysis, and I've done many of them, you can do what's called sensitivity analysis. So you can say the main analysis we do, the main model will only use the cleanest level of data. And then you run your model and then you see what the results are. And then you could add in those cases where there may be a drift off with some weeks or even some months, let's say. And then you rerun your analysis and then you look - did that change the results or not? 

And if it doesn't really change the results, maybe that will relax your criteria. But if you see that it will change the delivery of what you're trying to accomplish, that tells us that that would be biased. We would of course not include such data in the model. This is a very typical type of approach. If you look at papers that are published in all the big journals, they many times say the main analysis showed blah, blah, blah. And in sub analysis, when we included this or that, the results were virtually the same. And they many times say that are not shown, but they just basically say that we did these types of exercises and we have confidence in what we have done. So it's a very common type of procedure. 

So on a practical note, I would say if there is someone who wants to participate in the study, to want to sign the consent and is aware, the diagnostic sample was not perfect and there was a biopsy done say within some window after, we would certainly appreciate if we could get access to that as well. And we will take this into account for these types of models I mentioned with sub analysis or sensitivity analysis. 

Jenny: That makes sense. So it's better to say yes and consent and see on your end if it fits versus not. Okay. We also had a question from Paula and she was asking about the latest in genetic testing. So you talked earlier about how the FISH test is kind of like a black and white TV and you're using these new models. Do you see those models advancing in all clinics just in research right now or what do you think the future path will be for those types of new models?

Dr. Landgren: Well, I think that there are a couple of different variables going. So the FISH and cytogenetics have been around for a long time, over, I think over 40 years or so. And they are set up and people just do the same. They keep on doing the same, the same, the same. They don't question why we do the same. The same is true for the treatments. Many of the treatment steps are the same, the same, the same. So as you know, I tend to ask those questions. Why do we do this way? Why are we not using newer technology if I know that there are kind of emerging data supporting it. So we have pushed for that. We have started doing whole genome sequencing for research where you have the whole genome be characterized. That's of course much more informative. 

The downside with that is that there is not yet a path forward to have it covered financially, it will have to come out of research grants. So that means that I have to first have the hypothesis, the idea, and have to write a grant. And then you'll go through all these grant cycles where unfortunately, many grants are being declined because either they don't understand the grant or they don't think it's important enough. we spend a lot of time writing grants and they unfortunately are repeatedly rejected until we eventually can get it approved.

From a hypothesis to get something approved could sometimes take years. That's the brutal reality. So the other path is if you can partner with companies that are diagnostic companies and there are companies that are using more advanced technologies. Currently there are no whole genome available, at least not for myeloma, but you have targeted panels where you could capture say 100 or some genes and you could also capture say the whole genome for gains and losses, which is what the FISH could result in if you do additional chromosome three and five and seven, the hyperdiploid, or you could see a chromosome 17P deletion, let's say. So those are the panels you can do, but then you have also in myeloma, half the cases have translocations with chromosome 14. So unless the companies that do these assays have set up pipelines so they actually analyze the data so they can answer the question whether there’s a translocation or not, you will or will not get it. And for now, most of those companies are geared towards solid tumors. So the solid tumors don't have chromosome 14 translocation. That's sort of quite unique to myeloma. It's not completely unique, but for myeloma, that's a big chunk, it's 50 % of all the cases. So that's not the case in solid tumors. 

So you can get all the mutations, you can get gains and losses, but you won't know if there is a translocation. And that has been considered to be quite important. And then you have all these other information I mentioned before with the chromatripsis, you have the mutational signatures and things like that. That's usually not covered. I think what's happening right now in the field is that there are more and more of these diagnostic companies that have set up whole exome sequencing.

And that could technically cover all the things I'm talking about. And I think we will soon see whole genome sequencing companies delivering diagnostic workup. I think eventually it will replace everything. And ideally, if these companies could partner with academic institutions such as ours and others, where you could be tested and it could generate the standard of care information, but also the data if the patient consents could go in for discovery into databases where you could do data mining and you could find new markers that we currently don't understand if they are important or not. So that's how I see the field going forward. 

So basically the answer to your question is that all the new technologies, yes, are they implemented? No. 

What are the reasons? It's partly kind of a tradition that people like to do the same thing over and over again, but it's also partly due to the fact that the mechanism to get the data set up are still complicated. And you need to have the analysis on the backend of these companies. And then you have to have these collaborations, if you want, between these companies and academic institutions. So all this is happening and it's happening probably across the majority of cancers as we speak. A lot of groups are trying but it's going to take time to sort it out and then you have a lot of other practical things on top of it. 

So it is happening, but it's not happening fast enough. So I'm kind of a little bit impatient. I'm trying to push as fast as I can. 

Jenny: Well, it makes sense. I think your project design is fantastic. Okay, last question. Carolyn and Lisa have the same question. Can you identify a risk modeling using this for smoldering melanoma patients?

Dr. Landgren: …If we can use an individualized model to predict smoldering. So that's actually something we have worked on for quite some time. And we are working with many other groups. And I had this idea many years back ago, and I work with some of the bigger centers. And we have data that we have generated from collaboration with many of the other large institutions. So we hope within six to 12 months that we will have a model put together from a large collaboration where we hopefully with more advanced technology with whole genome and whole exome sequencing can better predict progression patterns. We have some preliminary data, but we need a larger study and we are working on that. I would think that this study would also help inform that on the, as you look at it retrospectively too, when you know patients are, you see them transitioning from smoldering to active myeloma.

And this is to create that model for the newly diagnosed patients. So it will be helpful in that treatment journey, regardless. I think that all these things are probably linked together. I would think that what you can gain and learn from the newly diagnosed, I believe that that probably will have impact also in the future for, say, for the next line of therapy.

But we may need to have some additional time points to make the case even stronger. So we're not going to just assume that the baseline will be good enough. We're going to continue to build with multiple time points. But I think there is a lot of data in the baseline sample that will be useful from both the newly diagnosed but also for later lines. 

Jenny: Well, we have talked about a lot and both your interventional study for smoldering myeloma and this project for active myeloma patients. And I would just like to invite patients who are listening to become these Cure Contributors. What you're building could be completely practice changing and helpful for all physicians, regardless of the facility, based on thousands of patients' worth of data. So if we as patients want faster cures, we can become these cure contributors by participating in projects like this to help support you in your work. And that's our invitation to all of our patient listeners. 

So thank you, Dr. Langren for the amazing work that you're doing. We're just so impressed with it. We want to help as a foundation and we know patients want to help you too. 

Dr. Landgren: Thank you very much for having me and thank you for fantastic collaboration. It's really a great honor working together with you and your team. You're just amazing. 

Jenny: Thank you for listening to the Health Tree podcast for multiple myeloma. Join us next time to learn more about what's happening in myeloma research and what it means for you. 

Have Any Questions?

Thank you for your interest in the event. If you have any questions, we would love to help!

Feel free to give us a call or send us a message below.

support

Get In Touch With Us

phone

1-800-709-1113

email

Support@healthtree.org

More Podcast Episodes

newsletter icon

Get the latest thought leadership on your Adult T-cell Leukemia / Lymphoma delivered straight to your inbox

Subscribe to the weekly newsletter for news, stories, clinical trial updates, and helpful resources and events with cancer experts.