Transcript:Jorge Cortes, MD: Now, Kevin, it’s been brought up a couple of times, but let me just ask you a little bit more detail. We look at these deeper responses, the deepest, the MR 4.5s, for example, in terms of treatment discontinuation. We’re starting to talk about this issue of treatment discontinuation. So, the question is, when should we consider treatment discontinuation? Under what conditions? Is it something that you just do on patients or is it in clinical trial? What can you tell us about this?
Kevin Kelly, MD, PhD: Well, yeah, it’s a fascinating concept that maybe we can take these patients off treatment. I think it’s going to become increasingly relevant as these patients get older and they develop more comorbidities. And we are starting to see more of a cardiovascular signal with some of these agents, particularly the second-generation TKIs that might lead to more risks as patients age.
So, I think at the moment, it’s certainly only in a context of a clinical trial where I would personally consider taking a patient off treatment. I know some people are doing it out of the context of a clinical trial. The only time that I would really consider patient discontinuation is, as we mentioned earlier, if somebody wants to become pregnant. In that situation, you would like to see a complete molecular response that’s sustained preferably over a two-year period before you would start discontinuation. And I think a lot of the trials follow those guidelines as well.
Once you take the patient off treatment, they need to be monitored very carefully because they can relapse, and they can relapse with blast crisis, and they can relapse even late after discontinuation with lymphoid blast crisis. So, you have to watch those patients even more carefully than the patients who are on treatment. Usually, the recommendations will be to check the BCR-ABL at least every month for the first few years until those patients are stable. But they cannot be lost to follow-up because they can relapse at blast crisis, and that can be devastating for the patient.
Jorge Cortes, MD: Yeah, absolutely. You made some very good points. One is that, clearly, clinical trials right now, I think, should still be the standard because we still have many questions that we haven’t answered. How deep the response should be, how sustained should it be before we attempt discontinuation, when should we restart treatment when it starts going back up? All these things. That is very important.
The other thing is the very close monitoring that that requires once you stop therapy, which is even more frequent than when we’ll be doing the treatment. The way I tell my patients is you’re getting rid of your drugs; you’re not getting rid of me, because I’m going to be doing your PCR a lot more often than I’m doing it now. In terms of the restart of treatment, just briefly I want to ask David: we started by restarting treatment when their transcript levels went from undetectable to detectable, and now there’s some studies where we let them go until they lose MMR before we restart. The studies are being done so that we can understand what is the right time. But what’s your gut feeling on what would be more appropriate: just when they start going up, when they get all the way to major molecular response? What do you think?
David Snyder, MD, FACP: Looking at the other way around: when I’m starting treatment and looking for the milestones, I want to get patients to below an MR level and I’m comfortable with that. I don’t need to get them to MR 4, MR 4.5. And I recognize there will be some fluctuation. But as long as they’re below MMR, I feel comfortable. And so on the flip side, if they’ve been undetectable, have discontinued treatment and now they’ve become detectable, at that range, I’m not going to be so worried if it’s stable fluctuating at that kind of low level between MMR and MR 4.5.
But I think what I’ve seen so far with some of the studies that have been done and the studies that are ongoing: patients who break through from being undetectable tend to do it fairly early within the first six to 12 months after stopping, and, I believe, tend to do it at a fairly brisk rate of rise. In other words, they’re going to reach MMR and above fairly quickly. I may be overstating it, but I think most of them, if they’re losing the response, tend to lose it all the way up there.
Javier Pinilla-Ibarz, MD, PhD: I completely agree. In most of the data that has been shown, almost 90% of the patients will, if they’re going to lose the response, it’s going to happen in the first six months. However, I want to emphasize what Jorge would say and in the past we used to say: TKI discontinuation, don’t try at home. So, I think we need to still emphasize that this is something that has to be done in clinical trials where there is very, very good monitoring. I think after so many years, really emphasizing adherence and asking patients to take the drug, we cannot really let this happen really outside these settings because it may be, as Kevin said, quite devastating.
Harry Erba, MD, PhD: I’d like to challenge the panel.
Javier Pinilla-Ibarz, MD, PhD: You always do.
Harry Erba, MD, PhD: Because it begs the question. I understand there are still very important research questions, but we do have the STIM trial and we know that patients who have been on imatinib for a very long time and in a complete molecular remission and in good labs for a couple of years, 60% of them will relapse at a molecular level. But, as you said, very quickly: all of them regain their responses. And then there are these other patients, about 40%, who continue in a completed molecular remission or blip up, as we say, and then go back down.
So, for me, we already have 100 patients or more in that one trial and we have five years of follow-up now. What are we waiting for? So, a patient comes to your office and says, ‘I’m sorry, the quality of my life has just not been the same on an ABL TKI,’ and you’ve done everything you can to get rid of maybe the fatigue or the aches and pains, and it doesn’t work. Is it reasonable to stop therapy? And I agree, it’s not ready for prime time. If you’re going to do this, we need to monitor them very closely.
I’d like to present one scenario that I’ve done recently. See if you agree or not. So a woman who’s been on imatinib for 10 years, in a complete molecular remission for the last five years, and she has a very unusual toxicity that her skin turned blue. And I’m assuming it’s from the drug; she’s not on anything else. We’ve checked silver levels or whatever you check, and nothing. I gave her options. I said, ‘We can go down on the dose of imatinib, but if you are a little less blue would you be happy and what dose would you need? And at what dose would you not be controlling the disease anymore? We could switch you to another ABL TKI. If it’s c-KIT that’s causing this, maybe it will cause the same thing, maybe it won’t in you. Or we could stop and see what happens, and when it goes back up, then we could switch to something else or use a lower dose. At least then I can monitor for your response. If I just switch you now, you’re in a complete molecular remission so I don’t even know if you need that drug.’ So, I offered those options to the patient and the patient chose the treatment-free remission. It’s been over a year. Her hyperpigmentation has abated almost back to normal and she’s in a complete molecular remission with monthly checking of PCR.
Javier Pinilla-Ibarz, MD, PhD: So, I have to re-say that, and I’m going to go after this wonderful case. That’s fine, and I don’t have any problem for you or any of us to do that. But I think the problem is, the monitoring in the United States is quite poor, right? Our patients are not getting the right monitoring. But my concern is that we really do these outside at a controlled setting. People can really have a bad event that we really, really regret. So, there’s no doubt that these things can happen that we discuss in a very specific setting.
Harry Erba, MD, PhD: Well, keep in mind, our patients might start doing it without us knowing, and so it’s better for us to know when monitoring closely. But I agree with you this is not something that we should be recommending.
Jorge Cortes, MD: Before we move on to the next topic, I do want to emphasize a couple of things, and it’s pretty much been said by Javier, but I think it is important to emphasize a couple of these points. No. 1 is that the STIM trial had very specific criteria for consideration of treatment discontinuation. They use a PCR with 5-log sensitivity sustained for at least two years. They restarted treatment when the PCR became positive. In that study, that’s what they did, and under those circumstances, we know what happens. The other thing that I think is that we need to be aware that the follow-up of five years, it’s long, but probably not long enough. There’s, for example, after transplant, occasional [relapses], but some relapses that happen much later.
Javier Pinilla-Ibarz, MD, PhD: 10%.
Jorge Cortes, MD: So, all these things need to be taken into consideration that it’s under those circumstances. The other important thing is that we need to remember: when we talk about the outcome of patients that are treated in the real world, it is not as good. Perhaps part of the reason is because we don’t do the rigorous follow-up that we do in clinical trials, and the same thing may apply in this setting. If we discontinue outside of a clinical trial, but we do it with the same rigor as the people in France did on the STIM trial, we are likely to get the same results. But if we do it different, if we don’t monitor, we can transform a very good response into a disaster because we didn’t check.
Harry Erba, MD, PhD: And it could be a disaster, and the reason why it should not be done just routinely is the patients I’m most concerned about are the 40% who are on unmaintained remissions. And what I’m concerned might happen is that they won’t just relapse like these others did quickly, but they’ll actually progress into accelerated phase and blast crisis. In fact, your colleague, Dr. Talpaz, when he was at MD Anderson doing interferon work, published on a couple of cases just like that. That’s my greatest concern. The problem I have, though, is there won’t be a clinical trial that will ever satisfy us that it’s safe to do from that perspective because it would have to go on for many, many years.
Transcript Edited for Clarity