Accurate Risk Assessment in MDS

Transcript:Vinod Pullarkat, MD: How important is it to accurately assess risk when you see the patient? And what implications does it have for prognosis?

Rami S. Komrokji, MD: I think this is a principle that we use in oncology in general. I always teach the Fellows the first step, you want to make sure the diagnosis is made right. And in MDS, we all know that could be the most challenging. The art is probably in making the diagnosis. Once you step back and you are comfortable with the diagnosis, I think your next step is risk assessment for the disease. This is important information that you want to convey to the patients and their families, and this is the tool that we use to basically tailor our therapy accordingly. And in MDS, most of the time we are talking a go/no-go transplant decision down the road. So, there are several tools basically, including originally the WHO subtypes. But what got adopted internationally is the IPSS, the International Prognosis Scoring System. All those models look at the idea of several variables to incorporate a scoring system to classify patients into risk groups where you can predict survival and AML transformation.

The IPSS depended on blast percentage, cytogenetics on the bone marrow, and cytopenias. This has been the gold standard since 1997, but we all have used the system and we know that it has shortcomings, that it could underestimate the risk of the disease and, in less common situations, overestimate the risk of the disease. Basically, there had been revisions for that to address some of those shortcomings. There is an IPSS revised that still depends on those same variables, but has a little bit more details. So, if a patient had a platelet count of 20, they are regarded different from a patient that had platelet of 90. In the IPSS, they were looked at as the same. But, we know clinically, from what we see in practice, that those patients don’t do the same. The cytogenetics are much more detailed, so we’ve learned more about the heterogeneity and the cytogenetic abnormalities among those patients.

The blast percentage also is adjusted for, which could be tricky because now we have more than 2, less than 5, a category that most of the time the pathology records would not mention. But, nonetheless, it’s been validated now by several groups that the IPSS revised is refining. There are several other models also that have been proposed: the MD Anderson model, there is one model for lower-risk, higher-risk. The idea is the same basically or just refining those variables and how to look at them. The MD Anderson model incorporates age and performance status as well, but if you look at any of those models, what we used to think in low-risk MDS patients that we bring in clinic and say, “I’ll see you back in 6 months,” there is around 20% to 25% of those who don’t behave at that lower risk. Those are patients maybe at least we should pay more attention to the follow-up for them. And there are patients who we thought were higher risk, in 10% to 15% of those cases, their disease is better and may not need the transplant directly.

The complexity gets even more. Like once you finish with all the clinical variables, we have all these data now on the next-generation sequencing, incorporation of value of somatic mutations. There had been that seminal paper by Rafael Bejar looking at 5-gene mutations. If they are present, they upstage the patient one grade. I think we are starting to have also some things to address. Those are all disease-related risk factors, and I think there are some incorporating patient-related factors. Maybe Ellen wants to comment a little bit on comorbidities and frailty?

Ellen K. Ritchie, MD: We have to think hard about what the comorbidities are for these particular patients because how they’re going to respond to therapy and how they’ll withstand therapy really depends on what their comorbidities are. Especially since this is an older population, not only do we need to keep in mind what their comorbidities and their other medications are—and how many medications they’re taking, and how well organized they are to do that—but whether or not they have a caregiver or someone to help take care of them, which I think is a really important part of the prognosis of an older person that you’re treating with any of the agents that we use to treat myelodysplastic syndrome. And if we’re thinking about potentially a transplant or something else down the line for these patients, again, whether or not they have an appropriate caregiver and whether or not their other comorbid illnesses are well managed is a very important part of their success down the road.

Rami S. Komrokji, MD: I learned, which I never thought of before, that there are differences between comorbidities and frailty. So, you could have patients that have 10 comorbidities but are still fit and able to do things, and you could have patients that have no comorbidities and in the exam room, they cannot get from the chair to the exam table. Those all count. There are some models now incorporating those that had been looked at and validated.

Ellen K. Ritchie, MD: Well, the geriatric assessment in AML, what they really had found is that certain physical performance measures, as well as cognitive ability, really play a large role in how successful older patients are in withstanding treatment. This hasn’t been applied specifically to myelodysplastic syndrome, but it’s probable that that person who can’t move, doesn’t have the grip strength to actually really grab on to something, or whose cognitive ability is compromised, probably won’t do as well as other patients.

Transcript Edited for Clarity