Harry P. Erba, MD: Let’s get to a problem that we don’t want to have, and that’s loss of response or failure to respond. Jorge, do you want to walk us through what are some of the causes of that? What should a clinician be thinking about?

Jorge E. Cortes, MD: We’ve been talking about the monitoring and this is when it becomes critical. You want to identify this, and you want to identify it early. The criteria, I think, are standard. I think the audience can look at these things, and we know what it means. And there’s a difference between not achieving the endpoints we want, which is what we call a primary resistance, and losing the endpoints you had. So, somebody who loses complete cytogenetic response, that’s a secondary resistance.

So, both are important, but they may represent the different mechanism. The mechanism that we understand the best is mutations. We know that patients develop ABL kinase domain mutations, and that is a reason for developing resistance. That was kind of the genesis of developing these second-generation TKIs that we know can overcome some of these mutations, and they have different profiles of which mutations they address and which ones they don’t. We even have at least 1 drug now approved that works against the T315I, and that’s ponatinib. So, we can use that.

Now, the problem is that not everybody has a mutation, and probably about 50% of the patients have no detectable mutation by the standard sequencing, which is what most of us get from the clinical labs. Some studies have suggested that if you use next-generation sequencing and other new methodologies, you will identify mutations in a subset of those negative patients at very low levels. The question is, is a mutation that’s present in 1% of the malignant clones sufficient to trigger the resistance or not? And that, to me, is an open question. But there are still some patients who don’t have mutations.

We’re starting to understand that some of these patients have a disease that’s more genetically heterogeneous than we had earlier thought. Some physicians, particularly our colleagues from Australia, Sue Branford and Tim Hughes, have shown that those patients who progress very early, if you go back to the original samples, have some mutations in some of the genes that are associated with other leukemias, TET2 mutations and ROS1 mutations, and things like that.

So, it looks as if this disease that we thought was very molecularly homogeneous—it’s not always the case. So, we’re starting to understand. There are other things that have been suggested, like the transport of the drug into the cell, these receptors that transport the Src. There was a story that came originally when dasatinib was being developed, but there has never been a full assessment of how often that happens. Some of these mechanisms we don’t understand. Right now, the one that has clinical value significance is the mutations.

Harry P. Erba, MD: Specifically, Mike, how do you decide when to move on from one ABL TKI to another based on loss of response or failure to respond?

Michael Mauro, MD: There are some pretty black-and-white decisions. If someone doesn’t have hematologic response, that’s a real problem. That’s pretty rare. I think I’m convinced by the fact that someone who doesn’t have early molecular response is in trouble, although I think the timing of when you switch is up for debate. And that’s when we get into some of the things we already talked about. Do you have a real sense for the change over time? How much has the BCR really reduced? Can we use a 10% milestone? If you’ve started a patient on imatinib, for example, why should you be hesitant about using a second-generation TKI and switching early if the next patient who came into the clinic you started on a second-generation TKI at the beginning? So, I have a fairly low threshold. People waiver. “Oh, 3 months, isn’t that too early to switch?” I’m like, “No, not particularly.” So, you kind of dial up with the patient. You want to make sure they understand the risks and the benefits moving forward. But there are early molecular response failures, again, especially if it’s after with imatinib.

A second-generation initial treatment is a little bit trickier, because are you going to abandon that now and go to another second-generation drug or to ponatinib? So, a little bit more thought is required. And then I think it gets even more subtle when you get to how fast does a patient get into major molecular remission or deep molecular remission. I tend to be algorithmic, so I think we set a goal early to say we want to have early molecular remission, we want to have prompt MMR, we’d like to get to deep MR. With the growing link to data on safety and tolerability, although we obviously we have some things we have to watch out for, I think switching patients based on guidelines and response milestone failures is important, with the subtleties being that I would lean heavy with EMR. And I would probably be a little bit more watchful waiting with patients in deeper molecular response, especially around MMR or clearly around deep MR. That’s really an experimental question to optimize.

Harry P. Erba, MD: So, let me ask you. If a patient is in an MMR, what would be a trigger for you to reevaluate that patient and potentially switch therapy? How much of a rise do you need to see?

Michael Mauro, MD: Loss of MMR, I think, is important. I think there are data showing that sometimes mutation testing will give you important information. As Jorge just mentioned, that’s sort of the one thing we can look at. I’m not sure if it always describes what therapy you should take, but it certainly will guide your therapy choice, because if you see one mutation or another, especially in a dominant clone, that’s going to drive your decision. But if a patient is in MMR and maybe it plateaus, things like that, those are different scenarios. But I think loss of MMR is an important threshold to pay attention to.

Jorge E. Cortes, MD: A quick comment about that. I’m not always sure that I’d change somebody who lost MMR. I get the point, but I’ll just emphasize a confirmed loss of MMR. Not just one time.

Harry P. Erba, MD: I agree.

Michael Mauro, MD: Given all the nuances of PCR.

Transcript Edited for Clarity