Chronic Myeloid Leukemia: Selecting Patients for TFR
Panelists: Harry P. Erba, MD, Duke University; Jorge E. Cortes, MD, MD Anderson Cancer Center; Michael Mauro, MD, Memorial Sloan Kettering Cancer Center
Harry P. Erba, MD: We’re going to talk about the data. I’ll let you move right into that, Mike. But first, since we brought up treatment-free remission, they might achieve that beyond a TKI for at least 3 years and MR4.5 for the last year. Can you predict which patients are likely to remain in TFR? Are they prognostic factors? And then maybe tell us about the studies.
Michael Mauro, MD: I think we’ve been trying, that would be our Holy Grail. Can we, based on a patient, read the tea leaves? And I think it’s evolved through the different studies of treatment-free remission. In the beginning, we were dealing with 1 population: imatinib with or without interferon exposure. So, did prior interferon matter? It seemed to in some trials, but I don’t know if it was confirmed. We then asked a panoply of questions. What about the age, the sex, the so-called risk assessment? I think we had mixed feelings on those. I think there was some sense that high-risk patients may not do as well, particularly from the STIM trial. I think in the end, we’ve learned that most of the things early in CML tend to be erased, and we mainly have focused more recently on how long has someone been in a deep remission and how long have they been on treatment. But moreover, how long have they been in a deep remission?
That being said, the other questions that we needed answered in trials were if a patient enters this endeavor, what is the chance of mutations, resistance, and loss of chronic phase CML, which fortunately has been exceedingly low? Re-treatment success I think through all the trials has been exceedingly high with regaining MMR, and MR4, and MR4.5 at a very high rate, north of 90%, 95%. And those are the kind of questions people in the clinic will have. If I do this, what are my chances of losing my response, developing mutations, or not getting into the same quality remission I had before?
The largest data set we have from Europe, which I think we want to cover, are from the EURO-SKI trial. I think we’ve now learned that you probably need to be in a deep remission for a critical amount of time, particularly if you’re an imatinib-treated patient, that really matters, 3+ years. You probably need to be on a TKI for 5+ years. That’s how long it takes you to get into deep molecular remission and then sustain it for 3 years, so it’s pretty logical.
What’s interesting is that in my view, I think the bar has been moved back with aggressive depth of remission. You just heard, you guys just both mentioned MR4.5, MR4 for NCCN guidelines, I don’t know if we know the right answer there. I think we’ve been more generous and more liberal in who we can consider TFR. Honestly, I think the devil is in the details. An MR4 patient who’s sitting right at MR4 is kind of sitting close to it, it may be different than someone who is below MR4 but not consistently at MR4.5. But that has maybe fallen away. And then I think with newer trials, we’re also seeing that higher resistance may be falling away, which is most intriguing but probably not the most brittle form of resistance, patients with no response. Currently, we’re not going to do this endeavor in patients who have transformed, but I think we have data from, for example, optimizing with nilotinib. Some patients from dasatinib trials—where they had switched for intolerance, in some cases resistance—those patients have similar success. So, we’ve gone through all these iterations of the data, and we’ve come up with some conclusions. I think it continues to get more generous, which is encouraging for patients, but we have to be careful.
Jorge E. Cortes, MD: I’d like to make a couple observations. One is the duration of deep molecular response. I think that we’re starting to differentiate between the minimum requirement and the optimal requirement, just like we have for the response, the optimal versus the warning. The 2-year was the initial criteria set by these studies and it works, so that’s good. But we learned in our series that we’ve published earlier, and we updated recently at ASH, the patients who have a 5-year duration of MR4.5 or longer, they have a very low risk of relapse after treatment discontinuation, about 10% to 15%. So, I think that underscores that duration as an important factor, the longer, the better.
The other question is the level of response. And as mentioned, the earlier studies went to even 5 logs, and then 4.5, and other studies with 4. It’s difficult to assess because there are subtleties between one study and the other and how to tell. But when I look at the data of the bigger studies, for example, the STIM and then the EURO-SKI, the EURO-SKI is 4 logs. And I see 2 things that are a little different than the STIM. One is that the curve has continued to drop little by little, but it has continued to drop, whereas the STIM has appeared to reach that plateau where there has been a relapse after 2 years, I believe, or something like that. Whereas the other one, little by little, has continued to drop. And the other thing is that it has dropped a little bit below the 50%, whereas the STIM is sitting at around 60%. So, I’m extrapolating, but I think that I feel more comfortable with 4.5.
And particularly, do you do this in clinical trials or you do it in standard practice? I think that if you’re going to do it in standard practice, I prefer to stick into the more established conventional criteria. In clinical trials, I think it is a very worthy question, MR4, 1 year, 2 years. Now, I don’t have a doubt that we’re going to have patients who, with a lower level of response with a more superficial response and less durable, are going to be able to maintain the response, and we don’t understand why yet, by the way. But some are going to do it. But the question is, are we going to get to the same levels, globally, of responses? And that needs more follow-up, more studies.
Michael Mauro, MD: One other thing to mention that we learned in earlier studies was I don’t know if we have all of the pieces of the puzzle yet, and I think there may be something different about the patients who do or don’t relapse, irrespective of the CML, irrespective of the time and their depth of remission—being their immune system and their immune repertoire. So, I think that potentially needs further exploration and may give us the answer.
Harry P. Erba, MD: OK, guys. I do this every day, and you 2 have just given me a big headache because what I’ve heard is this is a goal of our therapy and we have all of these clinical trials. But the length, the duration that people have been on a TKI, the duration they’ve been in a deep response, how we define the deep response, and when you reinitiate therapy, these vary between all of these studies. And so, yes, we can do it. So, for the practicing clinician who wants to do this, we need to have some clear recommendations. The NCCN has some and quite honestly, one of them that I agree with is to get a second opinion from doctors who’ve done this. Your experience, Jorge, is remarkable. I mean, 200 patients stopping a TKI is similar to some studies that have added to the nilotinib label. So, getting that opinion from someone who does this all the time before just embarking on it, I think, might be very important.