Harry P. Erba, MD: Thank you for joining this OncLive^® Peer Exchange^®, which explores current clinical practice strategies for managing chronic myeloid leukemia. The use of tyrosine kinase inhibitors targeting the BCR-ABL protein is the hallmark of therapy for patients with chronic myeloid leukemia and has transformed this disease from a life-threatening to a chronic one for most patients. We continue to refine our approaches to monitoring CML and treatment resistance to further improve outcomes for our patients. In this panel discussion, my colleagues and I will discuss the latest advances in the field, as well as how to apply the new clinical data to clinical practice.

I am Dr. Harry Erba, and I am a professor of medicine and director of both the Leukemia Program and the Phase I Program in Hematologic Malignancy at Duke University in Durham, North Carolina. Joining me on this distinguished panel are Dr. Jorge Cortes, chief of the CML and AML Sections and deputy chair for the Department of Leukemia at MD Anderson Cancer Center in Houston, Texas; and Dr. Michael Mauro, leader of the Myeloproliferative Neoplasms Program at Memorial Sloan Kettering Cancer Center and a professor at Weill Cornell Medicine in New York, New York. Thank you so much for joining us. Let’s begin.

Jorge, why don’t you first start by telling us a little bit about how common CML is.

Jorge E. Cortes, MD: CML is not a common malignancy. It represents a small percentage of all the leukemias. About 1% to 2% of all the leukemias are CML. In the United States, in the more recent statistics, it has remained pretty constant; we get about 5000 new patients per year. I think what has changed significantly is the prevalence of CML because nowadays, as you established during your introduction, the patients have been able to survive longer and it has become a chronic disease. Patients who are diagnosed, fortunately, live longer. We think that they’re getting to live their normal life expectancy. Therefore, the prevalence of patients who have been diagnosed with CML at some point has increased dramatically over recent years, so we all have more and more patients whom we follow in the clinic with CML.

Harry P. Erba, MD: So let’s talk a little bit about how these patients will present. About 50% of patients may just present to their internist and have routine blood counts showing a leukocytosis; rarely there actually might be an isolated thrombocytosis. I’ve seen a few patients like that. But it’s generally picked up on a blood smear or in the blood count of a patient who is asymptomatic and being evaluated for something else. And then the other 50% of patients will have constitutional symptoms like night sweats and weight loss, fatigue, or symptoms related to splenomegaly, early satiety, and abdominal fullness. This will lead to the evaluation.

And the question that often comes up is, What is the appropriate initial evaluation for a patient with chronic myeloid leukemia? And I’m looking forward to hearing your opinions on this, but I’ve seen the tendency for some patients to be referred to me after having just peripheral blood analysis. And that might be by PCR or by FISH analysis. I actually still perform a bone marrow biopsy in the vast majority of my patients at diagnosis for a couple of reasons. One is that I think it’s important to do a cytogenetic analysis on patients.

Now, I understand that we’re following patients by PCR now. But if at some point you do want to know whether the patient has the Philadelphia chromosome and if it’s gone into remission, you need to know that baseline. There are about 5%, maybe up to 10%, of patients who may not have a Philadelphia chromosome in the 9;22 translocation seen on a metaphase karyotype. The only time you’re going to be able to see that, or be sure to see it, is at diagnosis.

On top of that, we know that one of the definitions of accelerated phase is whether there are other cytogenetic changes there. Now I understand there are data that show that patients who have accelerated phase just on the basis of another cytogenetic change may do just as well with ABL TKIs as patients in chronic phase, as long as they don’t have any other hematologic parameters. But remember, those patients were treated with higher doses of ABL TKIs, and so that might be very important to know. So, I do a bone marrow to get a complete cytogenetic analysis, which you may not be able to get on peripheral blood unless there’s a significant number of circulating blasts that divide. The other thing that you get is the marrow morphology, which might give some hint toward accelerated phase and the rare patient with blast crisis based on fibrosis and clustering of blast and things like that.

I also perform a FISH analysis, which I’ll be interested to see whether you do that. And then finally, I do PCR. I do RT PCR for BCR-ABL. And the reason I do that is not really to get a quantitation of the amount of disease at baseline; it should be on the International Scale somewhere around 100%. If it isn’t, we could talk about what that means. But it’s really to make sure that the patient has an identifiable B382 or B2A2 fusion, which is picked up by the commercially available primers. Because if they don’t have that, and then you don’t know that at baseline and then go to follow them later, and you get a very low value or a negative value, you might be led astray and think that the patient’s in a molecular remission, but it was just noninformative. So, that’s my typical approach to diagnosis. Any comments, Mike?

Michael Mauro, MD: So I think the devil is in the details. When you have a patient, you probably want to risk-stratify them as well. I think people often overlook that. I think it’s important for prognosis, for treatment choice. I think a lot of the data on treatment choice in the frontline are that we have good descriptors around risk stratification. Now you can do that from the blood. So, that being said, I agree with you. I think a bone marrow is definitely something you should do at diagnosis.

And when I said the devil is in the details, I think we’ve learned a lot more about what exactly clonal evolution means. There are probably certain clonal markers—Jorge, I know you have data on this. Certain things are probably worse than others. Historically, we kind of wrestle with them. Maybe some aren’t, but some clearly are. And then even the type of infusion matters. Does someone have an E13 or an E14A2 fusion? That probably has some prognostic significance, too.

So, you want to do all your tests at diagnosis without a doubt. Ironically, you can get risk stratification with just blood and a physical exam and high-quality blood testing—meaning good molecular studies confirming the pH. But I definitely support cytogenetics, bone marrow evaluation, ruling out coronal evolution, ruling out accelerated phase or a colt blastic phase, and to remind people that risk stratification matters. It’s often forgotten.

Jorge E. Cortes, MD: I agree. A couple of things that I would add. You can do risk stratification but not staging. In some instances, the blood and the bone marrow do not correlate with the percentage of who is in blast, and you need that to see the patient’s accelerated phase or chronic phase. So, you can make the diagnosis, you can do risk classification, but you cannot properly, unequivocally stage every patient without the bone marrow. So, at diagnosis, I think it is necessary.

What Mike mentioned about the unique chromosomal abnormalities, yes, I think we have come to understand that with the chromosomal abnormalities at time of diagnosis, some of these patients can do very well with TKIs. But there are chromosomal abnormalities—specifically, chromosome 17, isochromosome 17; chromosome 3 and chromosome 7 abnormalities. Those 2 do very poorly when they’re present at a time of even when just present at the time of diagnosis. So, it is important to know those things. For that patient, you may want a different approach. The others, trisomy 8 or the straight Philadelphia chromosome, you’ll do TKIs just as the others, but it is important to detect those patients.

Michael Mauro, MD: I know we’re going to get into this later, too, but I think that looking at the patient’s baseline level, the BCR-ABL transcript is also important because confusion can then follow from there with regard to how well a patient’s doing at 3 months, and the early molecular response definitions really require baseline assessment.

Harry P. Erba, MD: Yes, along those lines about the level, I occasionally get consultations about patients where they appear to have the disease, not based on cytogenetics or FISH analysis but by a PCR. But when you actually look at the level, the level of BCR-ABL transcript is very low, maybe 0.1% or below. And maybe that’s a false-positive. Maybe it’s a variant fusion that is not being efficiently amplified. And I think that patient deserves further evaluation before just assuming that they do or don’t have CML.

Transcript Edited for Clarity