Considering Venous Thromboembolism Risk Factors and Prophylaxis in Multiple Myeloma With Rajshekhar Chakraborty, MD, and Rahul Banerjee, MD

In this interview in honor of Multiple Myeloma Awareness Month, Dr. Rahul Banerjee, a Clinical Fellow in Medicine at the University of California, San Francisco, and member of the Oncology Data Advisor Editorial Board, speaks with Dr. Rajshekhar Chakraborty of Columbia University Irving Medical Center about risk factors for the development of venous thromboembolism (VTE) in patients with multiple myeloma and the considerations for thromboprophylaxis during their treatment.  

Dr. Banerjee: Welcome to Oncology Data Advisor, a digital resource for the multidisciplinary cancer team. My name is Dr. Rahul Banerjee. I’m one of the editorial board members. In March, in honor of Multiple Myeloma Awareness Month, I’ll be speaking with various faculty across the country who are experts at important topics in myeloma that we sometimes don’t talk about as much.

Today, it’s my honor to speak with Dr. Rajshekhar Chakraborty, an Assistant Professor at the Columbia University Irving Medical Center. He is an expert in multiple myeloma and other plasma cell dyscrasias, including amyloid light-chain (AL) amyloidosis. He’s also published extensively about the topic that we’ll be discussing today: clots, venous thromboembolic events, and arterial thromboembolic events. Dr. Chakraborty, Raj, it’s pleasure to meet with you. Thanks for taking the time to do this.

Dr. Chakraborty: Thank you, Rahul. I’m not an expert by any means, but thanks for the invitation. It’s a pleasure to talk about thromboembolism in myeloma today.

Dr. Banerjee: Absolutely. I think the prelude to this is that it is important for two reasons. One is that blood clots, either venous or arterial, can be life threatening. Second is that a lot of our patients are reflexively put on prophylaxis—aspirin or sometimes even a direct novel anticoagulant, which we’ll speak about later in this interview—and that’s an extra pill for them. Polypharmacy is a big problem for our patients with myeloma. Many of them are taking dozens of pills a week, excluding even the dexamethasone they take. Making sure that we are not overprescribing or underprescribing prophylaxis is important on both fronts.

I’ll start with venous thromboembolic events. In my mind, I often say, “Lenalidomide, aspirin; no lenalidomide, no aspirin.” Is it more complicated than that? How do you gauge who’s at risk of VTE in myeloma and who isn’t?

Dr. Chakraborty: That’s a great question. Definitely, lenalidomide or immunomodulatory drugs (IMiDs) are a risk factor for venous thromboembolism. Going back a little bit, myeloma by itself is a risk factor for venous thromboembolism. There was a large study done in Sweden that pretty much predated the IMiD era, in which they found that patients with multiple myeloma had about a 7.5-fold risk of developing VTE in the first year compared with matched controls. So we know that myeloma by itself can be a risk factor for VTE. Then fast forward to the 2000s and early 2010s when we knew that IMiDs are a risk factor for venous thromboembolism. It’s a class effect seen in all three of the IMiDs we have.

That’s why the International Myeloma Working Group (IMWG) formulated this consensus guideline for thromboprophylaxis with immunomodulatory drugs in myeloma around 2008, which was later adopted by the National Comprehensive Cancer Network (NCCN), as well. There were two clinical trials done in the UK. One was pre-IMWG era and -thromboprophylaxis era, and one was post-thromboprophylaxis era. One was the Myeloma IX trial, and one was the Myeloma XI trial. Although we don’t have a randomized controlled trial, this was a natural experiment in the setting of a clinical trial as to the rate of VTE before and after these thromboprophylaxis guidelines were adopted in clinical practice.

They saw that although the rate of VTE decreased somewhat, the decrease wasn’t that significant. If you look at the numbers for the pre–IMWG thromboprophylaxis of the Myeloma IX trial, the rate of VTE with regimens like CTD (cyclophosphamide/thalidomide/dexamethasone), which was commonly used in the UK for induction, was roughly in the ballpark of 16%. The post–IMWG thromboprophylaxis era was roughly in the ballpark of 10% to 13% depending on what population you looked at. It decreased somewhat, but 10% to 13% is still a significant number. It begs the question: can we develop better risk stratification tools to predict who is at the highest risk of developing thromboembolic events like VTE in multiple myeloma?

We now have two risk stratification tools which were rigorously developed and externally validated and are already published. One of them, the PRISM score, we presented at the American Society of Hematology (ASH) 2020 meeting, and a manuscript with the external validation is currently in review. I’ll go over them briefly to discuss the risk factors. Let’s talk first about the two scores which are already published. One is the SAVED score, and the other is the IMPEDE VTE score. Both of them were published in the past few years, and they are now in the NCCN guidelines, as well.

Basically, when I look at a newly diagnosed patient with multiple myeloma and I’m trying to figure out if they are at a higher risk of developing VTE, one of the important things that I look at is what are the risk factors that are common to all of these three scores—SAVED, IMPEDE VTE, and PRISM? Out of the ones which are common, number one is prior VTE. Any patient with prior history of VTE has a very high risk. Second is prior surgery within 90 days. There is an overlap with patients who have orthopedic surgeries to do fractures like pelvic, femur, or hip fractures. Then IMiD is a risk factor, just yes versus no. Anybody who received an IMiD and has one of these additional risk factors is definitely at a high risk, no matter which risk score you use.

Then the other thing is the treatment-related risk factor, which has been consistently shown as high-dose dexamethasone that is more than 160 mg of dexamethasone per cycle. That’s how high dose was defined in the derivation cohorts of these scores. Then there’s also the use of multiagent chemotherapy or doxorubicin. Both high-dose dexamethasone and chemotherapy use are fairly rare nowadays in newly diagnosed myeloma treatment, as you know. But if there is a patient for whom I’m using high-dose dexamethasone for some reason, like nephropathy or acute kidney injury, in combination with IMiDs, that I would definitely consider a very high risk of venous thromboembolism.

Another interesting negative risk factor is Asian and Pacific Island race. In both the SAVED and IMPEDE VTE scores, because they used a large population database, they had a large number of Asian patients, which we did not have in the derivation cohort of PRISM; our population was about 80% White, 19% African American, and 1% all other races. Both the SAVED and IMPEDE VTE scores found that patients who were Asian or Pacific Island race had a significantly lower risk of VTE, so it’s a negative risk factor compared with non–Asian or Pacific Island race.

These are the risk factors that I look at if somebody’s on an IMiD, on top of the additional risk factors of prior surgery, prior VTE, or high-dose dexamethasone versus any chemotherapy. Those I usually consider as a very high risk of VTE, and those who don’t have any of these shared risk factors, I consider them at a lower risk. In the PRISM score, we also found that abnormal metaphase cytogenetics was a risk factor for VTE in the deviation cohort. We will see if that holds up with the external validation cohort. If it does, then that would be an additional disease-specific risk factor for VTE. But we did not find high-risk fluorescence situ hybridization (FISH) cytogenetics or high International Staging System (ISS) stage to be a risk factor for VTE in the deviation cohort of the PRISM score when we presented.

Dr. Banerjee: Absolutely. I will say another negative factor would be choice of proteasome inhibitor (PI). I think a lot of us, when we see a quadruplet versus a triplet—or more particularly, carfilzomib versus bortezomib versus bortezomib—do consider a higher risk or consider stronger prophylaxis, which we’ll speak about in a second. In some institutions in New York, for example, in the MANHATTAN trials, Memorial Sloan Kettering used rivaroxaban for everybody. Can you speak on the evidence base for specific PIs and risk of VTE?

Dr. Chakraborty: That’s definitely a great point that among PIs, not all PIs are the same as far as the risk of VTE is concerned. We have definitely seen that the risk of VTE is higher with carfilzomib compared with bortezomib or ixazomib. In many of the trials, when carfilzomib was combined with IMiDs, the risk of VTE was high. Most of the data on this has come from Sloan Kettering. As you know, they have been using carfilzomib-based induction for quite some time. Dr. Neha Korde and colleagues have recently presented a paper in the British Journal of Haematology where they looked at VTE rates when KRD (carfilzomib/lenalidomide/dexamethasone)–based regimens were used with aspirin prophylaxis versus when KRD-based regimens were used with rivaroxaban thromboprophylaxis. Although it’s not a randomized trial, I think it’s a very, very good natural experiment in that they switched from aspirin to rivaroxaban because they noticed a high risk of VTE with aspirin alone.

The difference was actually dramatic. There was a 16% risk of VTE when KRD was used with aspirin, and it dropped to around the ballpark of 4% when KRD was used with rivaroxaban. This is actually very comforting data showing that when we are using carfilzomib in combination with lenalidomide or any IMiD, for that matter, we should probably be using something more than aspirin to prevent VTE. Another point I would make is that neither IMPEDE VTE nor even PRISM had a significant number of carfilzomib-treated patients in the derivation cohort. Carfilzomib is an additional risk factor that is currently not captured by many of these scores, and we have to keep that in mind when we are evaluating patients, newly diagnosed patients in particular.

Dr. Banerjee: Absolutely. I’ll switch gears slightly. We’ve been speaking about induction therapy for triplets and quadruplets for the last five to 10 minutes here. On the other side, now that they’ve gotten transplant or finished their induction therapy and are on lenalidomide maintenance, can you talk through two things? One, your approach to thromboprophylaxis in the risk-conscious lenalidomide maintenance, and two—only because this came up in a spirited debate on Twitter earlier this month—when you stop the lenalidomide, how quickly do you stop the prophylaxis thereafter?

Dr. Chakraborty: It’s always interesting because we know that based on the natural history studies of VTE in myeloma, most of the VTE events happen in the first six months. This has been shown in the Myeloma XI trial, where they have looked specifically at VTE, and the majority of the events were in the first six months. In the Cleveland Clinic cohort, during my fellowship, we looked at VTE incidence in approximately 1,000 patients who were diagnosed at the Cleveland Clinic between 2008 and 2018. We looked only at the first year, and we had found that most of the VTE events happened in the first six months. Now, having said that, we used lenalidomide in maintenance. The question is, is there a high risk of VTE with lenalidomide in the maintenance setting?

I think the cleanest data for this comes again from the Myeloma XI trial, because they randomized patients to lenalidomide versus observation. They looked at the risk of VTE in their protocol regarding thromboprophylaxis, because you have to interpret that in the context of what thromboprophylaxis you would be using at that time. In their protocol for the Myeloma XI trial, they had specified thromboprophylaxis as per the IMWG guidelines in the first three months. Beyond three months, to my understanding, it was per investigators’ discretion, because of the reason that most of the VTE events happened in the first three months. The manuscript by Dr. Bradbury in Blood specified that in the first three months, it was strongly recommended; then after that, I’m not sure whether they were using aspirin with the lenalidomide or not. Maybe some investigators were using it, and some were not. They found the risk of VTE with lenalidomide maintenance to be around 4%, whereas it was 0.6% with placebo. The risk was fairly low compared with induction, which had a rate of 10% to 12%. It was still 4%, so it was not zero, but I would say it’s a fairly low risk of VTE during the maintenance.

During maintenance, I never use anything more than aspirin for thromboprophylaxis, because you have to weigh the risk of bleeding when you are using things like rivaroxaban or apixaban. I usually use aspirin, and I typically stop as soon as I stop the lenalidomide. That’s what I typically do. It’s going to be hard to have any more randomized data, because most thromboprophylaxis trials will typically look at induction. I don’t think anybody will look at maintenance, but I think these numbers from the Myeloma XI trial help a lot in putting the risk benefit in perspective.

Dr. Banerjee: I totally agree. This is very informative. Thank you, Raj. Just to clarify, for the 4% risk with lenalidomide maintenance, is that per year, or is that over the duration of this period as they’re following Myeloma XI?

Dr. Chakraborty: To my knowledge, that was the total incidence over the entire trial period.

Dr. Banerjee: This is very informative. Thank you. The last thing I’ll do is quickly pivot to one of your other papers on a topic that we really don’t talk about that much in hematology at all: arterial thrombotic events. I feel like the cardiologist and the neurologist normally take those under their wing. You’ve also published about the risk of arterial events in these patients. Can you speak more about that?

Dr. Chakraborty: Sure. In the same study that we did at the Cleveland Clinic looking at 1,000 patients and the risk of thromboembolism, we actually looked at both venous and arterial at the same time. In that, we found that the risk of arterial thromboembolism (ATE) in the first year was roughly about 2.7%, so it was much lower than what has been published previously in the era of cytotoxic chemotherapy. For example, there was one published in which patients were treated with IMiDs and doxorubicin-based chemotherapy, and it was in the range of 5% to 6% at that time. It was much lower.

In the Myeloma XI trial, they looked at arterial thromboembolism rates, and they were roughly 1.5% to 1.7%, which is understandable given that in clinical trials, they’re probably going to have somewhat healthier patients than in the real world. That’s why the incidence was slightly lower compared with the real world. We only looked at one year from diagnosis, so that’s a caveat in Cleveland Clinic cohort. We did not look beyond one year. It was roughly in the ballpark of 2.7%, and most of the events were either stroke or acute myocardial infarction.

Another thing I would like to point is that we always focus more on the first year, but this is an elderly population overall—the median age of myeloma is 69 years—so there is a risk of arterial thromboembolism even beyond the first year. In another recent paper from the UK, a very large paper published in Lancet by Helen Strongman et al, they looked at the risk of cardiovascular events—which included not only arterial thromboembolism but also things like arrhythmia or congestive heart failure (CHF)—in patients with who were one-year survivors of several cancers. It included around 1,900 patients with multiple myeloma who were one-year survivors. They matched extensively with not only age and sex, but they also matched for all body risk factors. The analysis was very robust, and they found that patients who were one-year survivors of myeloma had a hazard ratio for developing coronary artery disease (CAD) of roughly 1.8; for stroke, it was roughly 1.7 compared with matched controls.

Even in one-year survivors, when patients are on maintenance or post-transplant and the dust is settled, it’s important to go back to the basics and maybe do a basic cardiovascular risk assessment to check for modifiable risk factors like smoking, hypertension, and hyperlipidemia. Maybe have them see cardio-oncologists for risk factor modification. It’s important because their risk of ATE persists over time and is more than what the age- and sex-matched controls have otherwise without myeloma. I think that’s an important thing to remember.

Dr. Banerjee: That’s very, very helpful. In your study that you had done at the Cleveland Clinic, could you tell that there were certain risk factors, be it smoking or be it use of carfilzomib or prior cardiac history, that put people at higher risk of arterial events? Or is it difficult to say with that analysis?

Dr. Chakraborty: It’s difficult to say, because although we had 1,000 patients, which seems like a lot, we had only very few events of arterial thromboembolism. We could only do a univariable analysis; we could not do multivariable analysis. Having said that, we found that the strongest risk factors were prior VTE, prior arterial thromboembolism, or prior CAD, which makes sense. Then there were all the usual suspects: smoking, chronic kidney disease (CKD), diabetes, and hyperlipidemia.

The only one which was a little bit unexpected was in patients who had an infection within 90 days of diagnosis. They had a higher risk of having VTE. Maybe there’s a strong correlation in those with diabetes or comorbidities or something inflammatory; maybe they’re at high risk of infection. We’re not sure, but hopefully, with larger cohorts, maybe someone could do a multivariable analysis with a larger number of events. That would be really helpful to find out those risk factors.

Dr. Banerjee: Agreed, and it’s a good plug that we should talk about smoking with our patients. As a non-thoracic oncologist, I think our transplant coordinators go, “Stop smoking before your transplant.” We try to get them to stop, but we just do it logistically for the transplant, not so much for the long-term risk. As you mentioned, there’s a 1.8-fold ratio of arterial events, and that’s important to remember in our patients.

I think those are all the questions I had. This was very informative. I certainly learned a lot. Any closing remarks or anything else you’d want to add?

Dr. Chakraborty: Maybe just one thing regarding VTE: I would say that one of the things we have noticed in myeloma, as in other solid cancers, is a strong correlation with patients who have ATE having worse survival compared with patients who do not have ATE. The Myeloma XI trial, which was a very large dataset, and our Cleveland Clinic cohort had a similar finding that VTE in the first 12 months did not actually have an adverse impact on overall survival. It’s possible that this VTE may be reflecting an IMiD side effect and not adverse biology. I’m not sure what the reason is, but both the Myeloma XI trial and the Cleveland Clinic cohort showed that in patients who have VTE in the first 12 months, it was not associated with worse overall survival; ATE was, but VTE wasn’t. I think that’s an important thing to remember because it’s counterintuitive to what you would think in other solid tumors.

Dr. Banerjee: That’s very, very true. Thank you again for your time Dr. Chakraborty. This was very, very eye opening for all of us. To the audience, thank you all for listening. I’m Dr. Banerjee, and this was Oncology Data Advisor. Thank you again.

Dr. Chakraborty: Thank you.

About Dr. Chakraborty and Dr. Banerjee

Rajshekhar Chakraborty, MD, is an Assistant Professor of Medicine in the Division of Hematology/Oncology at the Columbia University Irving Medical Center in New York. He specializes in the treatment of patients with multiple myeloma, amyloidosis, and Waldenstrom macroglobulinemia. His research interests are in clinical trials and outcomes research in plasma cell disorders

Rahul Banerjee, MD, is an advanced clinical fellow in Bone Marrow Transplant/CAR T-cell therapy at the University of California, San Francisco (UCSF). Most recently, he has joined the multiple myeloma faculty at the University of Washington Fred Hutchinson Cancer Center in Seattle. His clinical interests are in multiple myeloma and AL amyloidosis, and his research interests are in CAR T therapy, digital health, and care delivery.

For More Information

Kristinsson SY, Pfeiffer RM, Björkholm M, et al (2010). Arterial and venous thrombosis in monoclonal gammopathy of undetermined significance and multiple myeloma: a population-based study. Blood 115 (24):4991–4998. DOI:10.1182/blood-2009-11-252072

Palumbo A, Rajkumar SV, Dimopoulos MA, et al (2007). Prevention of thalidomide- and lenalidomide-associated thrombosis in myeloma. Leukemia, 22(2):414-423. DOI:10.1038/sj.leu/2405062

National Comprehensive Cancer Network (2022). Clinical Practice Guidelines in Oncology: cancer-associated venous thromboembolic disease. Version 1.2022. Available at:

Bradbury CA, Craig Z, Cook G, et al (2020). Thrombosis in patients with myeloma treated in the Myeloma IX and Myeloma XI phase 3 randomized controlled trials. Blood, 136(9):1091-1104. DOI:10.1182/blood.2020005125

Chakraborty R, Rybicki L, Valent J, et al (2020). Abnormal metaphase cytogenetics adds to currently known risk-factors for venous thromboembolism in multiple myeloma: derivation of the PRISM score. Blood (ASH Annual Meeting Abstracts), 136(suppl_1). Abstract 905. DOI:10.1182/blood-2020-137525

Piedra K, Peterson T, Tan C, et al (2022). Comparison of venous thromboembolism incidence in newly diagnosed multiple myeloma patients receiving bortezomib, lenalidomide, dexamethasone (RVD) or carfilzomib, lenalidomide, dexamethasone (KRD) with aspirin or rivaroxaban thromboprophylaxis. Br J Haematol, 196(1):105-109. DOI:10.1111/bjh.17772

Chakraborty R, Riaz IB, Malik SU, et al (2020). Venous thromboembolism risk with contemporary lenalidomide-based regimens despite thromboprophylaxis in multiple myeloma: a systematic review and meta-analysis. Cancer, 126(8):1640-1650. DOI:10.1002/cncr.326

Transcript edited for clarity. Any views expressed above are the speaker’s own and do not necessarily reflect those of Oncology Data Advisor.

Related Articles


Your email address will not be published. Required fields are marked *