The Uromigos Episode 105: A Phase 2 Study of Novel Immune Combinations in Bladder Cancer

Arjun Balar discusses this novel immune modifier in bladder cancer. Despite a negative trial, T cell stimulation in addition to checkpoint inhibition may be worthy of future study. View the ASCO abstract.

Episode Transcript

Tom:
Arjun Balar Welcome everybody I’m joined by Arjuna Bala, a great friend of ours. In fact, Arjuna, I think you were our first guest on our first podcast, and as Brian said earlier, it really has enhanced your career, which we have received no gifts, no great gratitude for which is surprising.

Nevertheless, we’ve invited you back because of the excellent work that you will present in Glasgow this year. Arjuna, do you want to introduce yourself quickly and just talk about the work you did?

Arjun Balar:
I am a Medical Oncologist and I direct the GU program at NYU, and I do bladder cancer research.

Brian:
So, Arjuna, talk about your abstract about this Induced One study. It’s a drug and a mechanism that’s probably not familiar to most, so maybe you could start with that and then just give us a summary of the data.

Arjun Balar:
ICO stands for the Inducible T-cell Co-stimulatory Receptor. It’s a member of the CD28 IG receptor super family. It was discovered quite a while ago, but I think probably close to a decade ago. It’s a pharmacodynamics biomarker targeting anti-CTLA4, so initial work actually done by Pam Sharma a while ago, found that it’s unregulated on the surface of T-cell when you target CTLA4 and it’s a T-cell agonist.

The signs behind it, the theory behind it is that if we target some of these agonistic receptors on T-cells, we can actually further drive anti-tumor T-cell responses against tumors and companies have been working on this pathway for a while.

Jones Therapeutics was one, they had a drug, I don’t that they made a lot of progress in getting the drug to demonstrate clinical efficacy, but GSK developed a drug called feladilimab, and we at NYU lead their phase one dose escalation, dose expansion study, this broad program called Induce One.

What I presented at ASCO, were cohort expansion study for the urothelial cancer patient population, which included two cohorts, one group, which was the monotherapy cohort. This is the patient population with advanced disease, heavily pretreated, PD1 experienced with monotherapy, with Feladilimab. Then the other cohort, which was the 2B, which was the patients who were pretreated, but PD1 naive, and this was Feladilimab in combination with Pembrolizumab.

Tom:
So, Arjuna let’s start just by going back a time, it’s essentially an agonist resulting in T-cell activation. Is there any idea of how that process of T-cell activation occurs? Is it a balance between T-reg cells and T-effector type cells or is it specifically targeting one population?

Arjun Balar:
This is specifically targeting one population, which is that you’re really targeting the effector T-cell population. Ideally exclusively just that. So, when we think about the whole cascade of priming and activation and antigen presenting cells, NAT cells prime them against specific tumor specific antigens and then those activated T-cells are now driven against tumors, then they will give an offshoot of a memory T-cells.

Now what we hope is that those activated T-cells upon interaction with the tumor cell, they can be further driven to target those tumor cells with engagement with an agonistic antibody that’s going to target ICOS. This does not have any, theoretically should not have any effect on any of the suppressive immune populations. This is really supposed to target the effector population.

Tom:
And is there a phase one data? Have you got the dose of the drug, right?

Arjun Balar:
Yes, they [inaudible 00:04:20] levels, 0.3, 1mg/kg, so they tested variety of dose levels, and they settled out at both, and we tested both at 0.3 and 1mg/kg, and 0.3mg/kg was the dose level that we settled on. There are consequences to overdosing, because you can develop T-cell exhaustion by overdosing the agonistic molecule, so you do have to get it just right.

Tom:
And are there biomarkers to show that you’ve got it just, right?

Arjun Balar:
I’m not sharing some of that data and GSK has not, we’re probably going to put it in the manuscript, I don’t have that data ready on hand to speak to it, at least not certainly during this podcast.

Brian:
Let’s go to the monotherapy clinical activity and the adverse event profile. What about that monotherapy cohort to start with what did you [inaudible 00:05:10]?

Arjun Balar:
Essentially in the monotherapy cohort, we had 14 patients that were treated. These were all heavily pretreated platinum experienced PD1 antibodies in a variety of others, in some cases and the activity I have to be honest, is modest, but what we did see is clear activity, at least in one patient here with a partial response, a patient with visceral disease involving fairly sizable lung metastasis in which I describe in our presentation.

An 81-year-old gentleman with a fairly large lingular metastasis that responded quite well, durable response up to 36 weeks out into treatment and then have ultimately progressed. [crosstalk 00:05:55] So there was one partial response.

Brian:
Tangential question about,

Arjun Balar:
Sure.

Brian:
Monotherapy activity in a drug, maybe you don’t expect a lot of (monotherapy) activity, but you want to test it anyway for a variety of reasons. Just from a drug development standpoint, how much stock do you put in activity or lack thereof or is the plan all along combinations with PD1 or CTLA4 for the mechanistic reasons you mentioned?

Arjun Balar:
Yeah, [inaudible 00:06:19] this is the challenge in drug development in general with IO based therapies is that the eventual development strategy tends to be combination with PD1 and CTLA4. However, I think most of us want to see at least some evidence of monotherapy activity before we get too excited, because at the end of the day, you want to see that you’re engaging the immune system and that the immune system is leading to an anti-tumor immune response.

Show this activity is proof of concept. Now, the fact that it’s one response out of 14 tells us that maybe this activity is not going to be earth shattering but to see a visceral response in a patient like this is noteworthy.

Brian:
If you didn’t see it, would biomarker activity be enough? If you show effects on T-cells or whatever the appropriate biomarkers, just general question, not even for this drug. I’m just wondering, because like you say, it is a challenge in developing sort of add-on immune therapy.

Arjun Balar:
It’s not enough for me, and unfortunately, I think that was kind of the nature of immunotherapy for years until the first generation of checkpoint antibodies kind of really hit the scene.

Tom:
Yeah. So, I agree with that. I think the principle of bringing these two drugs together, neither of which have very much activity, and you suddenly have a synergistic explosion of activity seems we haven’t shown that in bladder cancer, in fact, in any tumor type with these immune therapy drugs that I’ve seen. There’s something called Fermi’s Paradox, which is actually about life on the universe with all these billions…

Brian:
Oh boy. Here we go.

Tom:
With all these billions of planets and…

Brian:
This why Arjuna hasn’t come back in on our podcast.

Tom:
Well, this is important, right? Because with all these billions of universes and planets, how come we don’t see life or how come we haven’t been visited by extraterrestrial life? Probability suggest it should be out there. It’s the same sort of principle, if it was out there, how come it’s so quiet? If there were these synergistic combinations out there, we would know by now because so many people are testing them.

I think the point that Arjuna is making is absolutely right, is we have to have monotherapy activity, otherwise because if we did have these synergistic dynamic duets put together, God knows we’ve been testing them for long enough periods of time, and it’s been awfully quiet.

Arjun Balar:
The story with IDO is very telling, right? We made a lot out of epacadostat and look what happened in the randomized study. Either you put your money where your mouth is and run the randomized studies or we stop putting too much weight on these single armed trials of combinations and say, “Look, oh, this is doing much better than we would expect with monotherapy PD1”. I think most of us look at that data with a lot of skepticism nowadays.

Brian:
So, talk about the combo data, talk about the combo cohort and the activity and specifically how you interpret it in light of what you might expect from Pembroke alone.

Arjun Balar:
Right, and that’s exactly it. So, I actually put much more weight on the monotherapy data. So, to be clear, so the monotherapy was 1mg/kg Feladilimab, the combo data was 0.3mg/kg, and Pembroke was dosed at 200 milligrams every three weeks and the response rates in the combination cohort data, clearly show that it’s about what you would expect, a response rate was about 22%, no different from we would expect in the second line treatment or factory setting with Pembroke alone, and it’s durable and that’s what you would expect.

I don’t make too much out of that. I really think the basis for this and the reason I think it was selected for oral abstract is really, is the monotherapy data to show single agent activity for a novel checkpoint, this is an agonistic molecule and activity in a visceral patient I think that’s what’s really the attractiveness.

Tom:
Adverse event profile?

Arjun Balar:
Overall, well tolerated in our presentation we are careful to say that immune related adverse events formally from a data collection standpoint were not evaluated, but when we look back the overall rates of treatment, [inaudible 00:10:17] low grade, low frequency, very few grade three and fours, and potential immunology were low. Not much added on top of what we would expect with Pembroke alone, and certainly monotherapy was really well tolerated.

Brian:
I have a two-part question. One is sort of, generically, sort of next steps given everything we just said about combo and monotherapy, and given what you said about potential relevance with CTLA4 targeting, are there plans to combine with a CTLA4 inhibitor?

Arjun Balar:
Not that I’m aware of, the combination cohort, there is some information that is publicly available, which is that the first part of the GSK ICOS program to launch was the INDUCE three and four programs, which is a head and neck randomized trial of Pembroke with and without the Feladilimab study in head neck. There was a press release that they closed that study based on an interim analysis, the IDMC evaluated the data based on response which was the interim analysis.

They decided not to move on with that trial. Now, the implications of that for the broader program is not yet clear, GSK will continually evolve and that’s kind of the company line at the moment which is look at the program as a whole as they collect more biomarker data from that study as well as more follow up from the current phase one, hopefully by next year, it’ll be more clear in terms of what their plans are.

Brian:
Are there any biomarker data in terms of what you can say what’s promising or is that revealing at all in terms of where to take it next steps?

Arjun Balar:
So, in the induced one study and what I did present is some data with regard to immunohistochemistry in looking at basically the PDL and positive cohort and the ICOS positive cohorts. So, they are looking at a proprietary antibody, looking at measuring both PDL and status and ICOs expression status, and tumors that are collected at baseline.

Essentially, and again, post talk review, small data sets that’s retrospective, these are always problematic and we all know that PD [inaudible 00:12:35] is problematic to begin with, but essentially what it did find is that if you looked at these so-called double positive patients, those who are both PDL and positive and ICOS positive at baseline, it did associate with improved survival as compared to those who were double negative.

That’s not surprising I think when you see higher expression levels of the target protein of interest that’s generally associated with a higher likelihood response and maybe that translates to better survival, but we know that these immune based biomarkers are highly inconsistent and poorly replicable.

Tom:
Arjuna, if I put it to you that I can look at this data and say, “I’m not thrilled by it” and you might come back to me and say, “Tom, actually, I think most of your data I’m not thrilled by either.”

Brian:
And I would. I agree with you, but it’s yeah, [crosstalk 00:13:26] go ahead.

Tom:
I’d accept that. I guess one [inaudible 00:13:31] to ask you is what is the sum of our effort with immune combination therapy in urothelial cancer? Now we’ve had a lot of different goes of it. Are we wasting our time? Are we not, or are we doing it wrong?

Arjun Balar:
I think it’s interesting. I actually had another interview recently regarding a broader topic of immunotherapy with The Washington Post, of all people, but,

Tom:
They’re not quite as well versed as us, we keep saying no, but it’s fine.

Arjun Balar:
But the broader question was asked about kind of the immunotherapy and are we hitting in kind of an inflection point of checkpoints and I think the answer is yes, based on what is currently available, are we in hitting an inflection point of both enthusiasm and the current drugs that we have and perhaps we are, that goes beyond just urothelial cancer, right?

And maybe there’s a next generation of perhaps cellular based therapies that we are waiting for on the horizon over the next five to 10 years and we know there are a lot of biotech’s that are invested in cars and whether they’re CARs and CAR-NT or CAR-NKs and other adoptive cell therapies, maybe that’s the next generation that we should be looking forward to.

Tom:
Arjuna what do you think about CK4 activity in urothelial cancer? The DANUBE trial looked okay but was negative for various reasons to do with statistical design and some of the IPI niveau data also looks okay and pattern refractory disease that Matt and Pam have done. What’s your take? Do you think…because that’s when I said, are we doing it wrong? Do you think there is a role CK4, and we just haven’t gotten there yet?

Arjun Balar:
Absolutely I do. That’s the one study that I think is going to be the difference if we were to fast forward, let’s say three years, three or four years from now, I think it’s EV Pembroke or most patients, but for a pure IO type of role, it IPI niveau, or if we kind of redo Dree a higher dose CTLA four plus PD one absolutely has a role in frontline bladder cancer and I’m, I’m cautiously optimistic about 901 for that reason.

Tom:
Interesting.

Arjun Balar:
For that very reason, I think it’s going to be very similar to what we do in kidney right now, which is, it’s sib or let me rephrase, VEGFR TKI plus PD one for some patients it’s CTLA four plus PD, one for other patients, it’ll be very similar, I think we’re going to something very similar in bladder.

Brian:
Do you want to take the last two minutes and just talk about the chemo RT?

Arjun Balar:
Yeah.

Brian:
The abstract, yeah.

Arjun Balar:
That’s the I guess my morning presentation on Monday will be about that. So, this is kind of a this a bit of a cliche, but it’s a labor of love that we started back in 2014, 2015, when we had this concept that, Merck was very excited about supporting their MIS P and saying, come with us with concepts that we wanted to do was bladder preservation, hyper fractionated, radiation schedules were all the rage as immune synergizing.

So, what we put together was pembrolizumab in combination with a four-week bladder-only radiation schedule with twice weekly gemcitabine, the idea then was gemcitabine low dose twice weekly, great for patients, maybe immune synergistic mouse models suggested depleting MDCs, a lot of good hand waving made that trial really attractive.

And, and we designed the study to collect, tissue and blood in just the right way, and five, six years later we treated 54 patients and most of the patients were patients who elected bladder preservation, so we were not focusing just on those who were non poor surgical candidates.

The data basically shows that, we updated it by the way. So, what’s in the abstract and, and what’s in our final presentation is a little bit different I apologize. But what we are seeing is in the early analysis.

So, this is about 15, close about 15 months follow up, is that our one-year bladder tag disease free survival rate is about 88% in the efficacy cohort, and if we look at the entire cohort of 54 patients it’s about 89% the Treatment’s very well tolerated patients did develop immune related toxicities, but well managed with steroids.

This is the basis that I took to merge for the basis for keynote 992, which is the ongoing randomized phase three, which includes hypo fractionated radiation, bladder only RT.

Tom:
And what’s that randomization just if you just expand on that study?

Arjun Balar:
So also SW 1806, I think are two very pivotal studies but the broad concept is bladder preservation therapy, concurrent chemo radiation with, or without immunotherapy in keynote 992, it’s either conventional radiation schedule of six weeks RT with concurrent chemotherapy or four weeks radiation schedule with concurrent chemo chemotherapy, your options include weekly cisplatinum five, a few mitomycin or twice weekly gemcitabine and with immunotherapy in SW 1806 you have your choice of otelixizumab and in 992, you have every six weeks pembrolizumab

Tom:
And your data as it currently stands, supports the combination. Let’s hypothetically say that the data comes out and it shows that combination is better, would that then challenge cystectomy altogether, or do you think this will always be an alternative and less attractive alternative to cystectomy?

Arjun Balar:
So that’s the problem was that we’ll never have a randomized trial of surgery versus chemo radiation, it ends up being an issue of how we present our options to our patients and how our patients are actually presented options.