Select Topic:
Browse by Series:

The Use of CTL019 in Lymphoma

Panelists: Krishna V. Komanduri, MD, Sylvester Comprehensive Cancer Center; David Maloney, MD, PhD, Fred Hutchinson Cancer Research Center; Sattva S. Neelapu, MD, University of Texas MD Anderson Cancer Center; Michael Pulsipher, MD, Children
Published: Tuesday, Feb 20, 2018



Transcript: 

Krishna V. Komanduri, MD: Steve, I want you to talk about the JULIET study. Obviously, this is your baby.

Stephen J. Schuster, MD: This is, yes. JULIET trial is a trial of the tisagenlecleucel construct, a 4-1BB CD19-directed CAR. It was based on a study that we did at the University of Pennsylvania that we began in 2013. And basically this JULIET trial was a global trial, an international trial that involved 27 centers in 10 countries on 4 continents. What really excites me is they got essentially the same results that I got at the University of Pennsylvania. So, it’s pretty unusual to go to a multicenter study and get the same results that you got in a single-center study. It’s almost always much better at a single institution, the results. So, that was really encouraging. By pheresing patients and shipping cryopreserved peripheral blood nuclear cells to the production facilities, we were able to manufacture CARs, distribute them globally, and have patients treated all over the world.

The success rate looks great. What I observed in my study was very durable responses, and what’s really important is to look at the 3- and 6-month responses. Yes, we had 50% complete responses on JULIET, 53% if you look at 1 month. On my own trial, it’s probably higher, but I don’t even look at that because what matters is where you are at 3 months and 6 months. They’re the patients who have a durable response that goes on for years. We have follow-up in JULIET. I guess the follow-up now will be a little beyond 9 months, but the percentages of responses are the same. In my own trial—the stuff that is going to be published this weekend—the median follow-up was 28 months, but we have patients out beyond 3 years. And I can tell you that the remissions are durable. Relapses are not the problem. The problem is getting people into remission, which happens, like I said, about 30% to 40% of the time. We can successfully do that with a single infusion of CAR cells without any additional checkpoint inhibitor or manipulation. The future that I see is trying to get that initial response higher than 40%, and that’s the direction that I’d like us to move in. And I’m sure we’ll see late relapses, but for a few years now, these have been quite durable.

Krishna V. Komanduri, MD: Congratulations. I think this is really amazing.

David Maloney, MD, PhD: I’d like to just add one point. In the field, none of these are randomized phase III studies. Everything is being approved on a phase II study that shows really encouraging response rates and some of these are obviously durable. In many cases, they’ve been compared to historical standards, and I think there’s a danger in that. Patient selection does occur, and that can occur for multiple reasons. Patients cannot go on the study because they can’t get a slot, the study can be gated and you don’t have an enrollment, and patients can die waiting. And so, you cannot factor that into your control population to really be able to do this. So, this is a somewhat selected population. Now, that said, I have seen absolutely miraculous responses in people who you would not predict would be alive in a month or so. So, this can work in just amazing cases. But this is not a randomized phase III study. So, we really don’t know how much of that selection bias we’re putting into our patients to keep alive to get collected, to wait the 17 to 28 days to get their cells and not progress or not die from intervening chemotherapy. There’s a big falloff, but I know in the JULIET trial, there was quite a big falloff because of manufacturing slots and timing. And this is true about all the studies, regardless of whether they actually impose this wait or not because you wouldn’t know about it. So, it’s false. To compare between trials, they’re all very similar and I think you have to recognize this. This is kind of like transplant. These are people who got the transplant compared to people who didn’t and that’s always artificial in terms of the population. That’s a key point.

Stephen J. Schuster, MD: That’s the most important take-home message. I think, for example, the JULIET trial was a registrational trial, so it required a statistical test. So, our statistical test was the CR rate at 1 month, 4 weeks, which I just told you…

Michael Pulsipher, MD: On infused patients.

Stephen J. Schuster, MD: Infused patients, correct.

Michael Pulsipher, MD: That’s the very thing that David is alluding to.

Stephen J. Schuster, MD: Yes, and it’s compared to historical control of response rate to salvage chemotherapy in a similar population of less than or equal to 20%. It’s a low bar, and it’s too early of a readout anyway for any kind of meaningful response. So, these are the kind of, I think, artificial things that we use when we’re trying to get something registered.

David Maloney, MD, PhD: But going forward, it has got to be an intent-to-treat population eventually to put this into the context of the…

Michael Pulsipher, MD: If you were going to compare 2 constructs.

David Maloney, MD, PhD: Right, yes.

Michael Pulsipher, MD: Head-to-head. It needs to be intent-to-treat from first encounter.

Stephen J. Schuster, MD: Exactly, but even with the existing data though, there needs to be more transparency. So, we need to know the date of screening, the date a patient was enrolled, and the date the production happened. If you’re looking at a bunch of charts and you see a lot of sick patients as you’re screening them and you know you have one spot, and you pick that chart and enroll that patient, there’s a selection bias. And I think that’s exactly why you can’t compare the studies. In the JULIET trial, what we did is being very enthusiastic: we wanted to treat as many patients as we could all at once. So, we ran out and we pheresed everybody with a refractory diffuse large B-cell lymphoma that was capable of being pheresed, and we had a lot of T-cell products to make a limited production capacity. It led, at the beginning of the trial, to a backup in patients actually getting infused. If you compare that with a trial that didn’t enroll a patient until there was a spot, you get very different results. So, I think this kind of transparency needs to be available to us when we talk about these studies, and nobody should be comparing apples and oranges.

Krishna V. Komanduri, MD: I think to summarize, I think we have 2 remarkable, pivotal trials, including the JULIET study.

Stephen J. Schuster, MD: There are 3 remarkable trials.

Krishna V. Komanduri, MD: Three remarkable trials, but I think that all of them actually are pretty amazing. And what was started as single or a few institutions that had highly specialized CGMP labs and clinical staff weren’t able to be generalized with remarkable results, but I think with all those caveats going forward…

David Maloney, MD, PhD: I think it’s also early in the game, right? They’re second-generation CARs, but really this is the first construct getting FDA approval. There are so many things down the road that people can come up with, armored CARs and all kinds of things where you can make the cells resistant.

Stephen J. Schuster, MD: This is probably not even the end of the beginning yet.

David Maloney, MD, PhD: Yes, so we’re really, really in the game.

Transcript Edited for Clarity 

Slider Left
Slider Right


Transcript: 

Krishna V. Komanduri, MD: Steve, I want you to talk about the JULIET study. Obviously, this is your baby.

Stephen J. Schuster, MD: This is, yes. JULIET trial is a trial of the tisagenlecleucel construct, a 4-1BB CD19-directed CAR. It was based on a study that we did at the University of Pennsylvania that we began in 2013. And basically this JULIET trial was a global trial, an international trial that involved 27 centers in 10 countries on 4 continents. What really excites me is they got essentially the same results that I got at the University of Pennsylvania. So, it’s pretty unusual to go to a multicenter study and get the same results that you got in a single-center study. It’s almost always much better at a single institution, the results. So, that was really encouraging. By pheresing patients and shipping cryopreserved peripheral blood nuclear cells to the production facilities, we were able to manufacture CARs, distribute them globally, and have patients treated all over the world.

The success rate looks great. What I observed in my study was very durable responses, and what’s really important is to look at the 3- and 6-month responses. Yes, we had 50% complete responses on JULIET, 53% if you look at 1 month. On my own trial, it’s probably higher, but I don’t even look at that because what matters is where you are at 3 months and 6 months. They’re the patients who have a durable response that goes on for years. We have follow-up in JULIET. I guess the follow-up now will be a little beyond 9 months, but the percentages of responses are the same. In my own trial—the stuff that is going to be published this weekend—the median follow-up was 28 months, but we have patients out beyond 3 years. And I can tell you that the remissions are durable. Relapses are not the problem. The problem is getting people into remission, which happens, like I said, about 30% to 40% of the time. We can successfully do that with a single infusion of CAR cells without any additional checkpoint inhibitor or manipulation. The future that I see is trying to get that initial response higher than 40%, and that’s the direction that I’d like us to move in. And I’m sure we’ll see late relapses, but for a few years now, these have been quite durable.

Krishna V. Komanduri, MD: Congratulations. I think this is really amazing.

David Maloney, MD, PhD: I’d like to just add one point. In the field, none of these are randomized phase III studies. Everything is being approved on a phase II study that shows really encouraging response rates and some of these are obviously durable. In many cases, they’ve been compared to historical standards, and I think there’s a danger in that. Patient selection does occur, and that can occur for multiple reasons. Patients cannot go on the study because they can’t get a slot, the study can be gated and you don’t have an enrollment, and patients can die waiting. And so, you cannot factor that into your control population to really be able to do this. So, this is a somewhat selected population. Now, that said, I have seen absolutely miraculous responses in people who you would not predict would be alive in a month or so. So, this can work in just amazing cases. But this is not a randomized phase III study. So, we really don’t know how much of that selection bias we’re putting into our patients to keep alive to get collected, to wait the 17 to 28 days to get their cells and not progress or not die from intervening chemotherapy. There’s a big falloff, but I know in the JULIET trial, there was quite a big falloff because of manufacturing slots and timing. And this is true about all the studies, regardless of whether they actually impose this wait or not because you wouldn’t know about it. So, it’s false. To compare between trials, they’re all very similar and I think you have to recognize this. This is kind of like transplant. These are people who got the transplant compared to people who didn’t and that’s always artificial in terms of the population. That’s a key point.

Stephen J. Schuster, MD: That’s the most important take-home message. I think, for example, the JULIET trial was a registrational trial, so it required a statistical test. So, our statistical test was the CR rate at 1 month, 4 weeks, which I just told you…

Michael Pulsipher, MD: On infused patients.

Stephen J. Schuster, MD: Infused patients, correct.

Michael Pulsipher, MD: That’s the very thing that David is alluding to.

Stephen J. Schuster, MD: Yes, and it’s compared to historical control of response rate to salvage chemotherapy in a similar population of less than or equal to 20%. It’s a low bar, and it’s too early of a readout anyway for any kind of meaningful response. So, these are the kind of, I think, artificial things that we use when we’re trying to get something registered.

David Maloney, MD, PhD: But going forward, it has got to be an intent-to-treat population eventually to put this into the context of the…

Michael Pulsipher, MD: If you were going to compare 2 constructs.

David Maloney, MD, PhD: Right, yes.

Michael Pulsipher, MD: Head-to-head. It needs to be intent-to-treat from first encounter.

Stephen J. Schuster, MD: Exactly, but even with the existing data though, there needs to be more transparency. So, we need to know the date of screening, the date a patient was enrolled, and the date the production happened. If you’re looking at a bunch of charts and you see a lot of sick patients as you’re screening them and you know you have one spot, and you pick that chart and enroll that patient, there’s a selection bias. And I think that’s exactly why you can’t compare the studies. In the JULIET trial, what we did is being very enthusiastic: we wanted to treat as many patients as we could all at once. So, we ran out and we pheresed everybody with a refractory diffuse large B-cell lymphoma that was capable of being pheresed, and we had a lot of T-cell products to make a limited production capacity. It led, at the beginning of the trial, to a backup in patients actually getting infused. If you compare that with a trial that didn’t enroll a patient until there was a spot, you get very different results. So, I think this kind of transparency needs to be available to us when we talk about these studies, and nobody should be comparing apples and oranges.

Krishna V. Komanduri, MD: I think to summarize, I think we have 2 remarkable, pivotal trials, including the JULIET study.

Stephen J. Schuster, MD: There are 3 remarkable trials.

Krishna V. Komanduri, MD: Three remarkable trials, but I think that all of them actually are pretty amazing. And what was started as single or a few institutions that had highly specialized CGMP labs and clinical staff weren’t able to be generalized with remarkable results, but I think with all those caveats going forward…

David Maloney, MD, PhD: I think it’s also early in the game, right? They’re second-generation CARs, but really this is the first construct getting FDA approval. There are so many things down the road that people can come up with, armored CARs and all kinds of things where you can make the cells resistant.

Stephen J. Schuster, MD: This is probably not even the end of the beginning yet.

David Maloney, MD, PhD: Yes, so we’re really, really in the game.

Transcript Edited for Clarity 
View Conference Coverage
Online CME Activities
TitleExpiration DateCME Credits
Community Practice Connections™: 21st Annual International Congress on Hematologic Malignancies®: Focus on Leukemias, Lymphomas, and MyelomaApr 27, 20182.0
Oncology Briefings™: Updates in Therapeutic Options and Approaches for Tenosynovial Giant Cell TumorsApr 28, 20181.0
Publication Bottom Border
Border Publication
x