Immune Checkpoint B7-H3 - APCCC 2022 - Emmanuel Antonarakis

October 16, 2022

Emmanuel Antonarakis joins Alicia Morgans to discuss the immune checkpoint B7-H3 and its implications for prostate cancer treatment. Dr. Antonarakis provides an overview of clinical trials that tested B7-H3 and the encouraging findings of those trials. He also details some of the toxicities noted during the trials and plans for future studies given the measurable disease responses and PSA responses experienced by patients in these trials.


Emmanuel Antonarakis, MD, Clark Endowed Professor of Medicine, Division of Hematology, Oncology and Transplantation, University of Minnesota

Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, Massachusetts

Read the Full Video Transcript

Alicia Morgans: Hi, my name is Alicia Morgans and I'm a GU Medical oncologist at Dana-Farber Cancer Institute. I'm so excited to have here with me today, Dr. Emmanuel Antonarakis, who's a Professor of Medicine at the University of Minnesota, where he's also a GU medical oncologist. We're here at APCCC 2022, so excited to talk with you about a presentation you had at the translational symposium prior to the main meeting where you really talked about B7-H3, and its implications in therapeutics for prostate cancer. Can you share a little bit about that, please?

Emmanuel Antonarakis: Good morning, Alicia. Thanks for having me. And of course, it's nice to be here in Lugano and the Italian-speaking part of Switzerland.

Alicia Morgans: Yes.

Emmanuel Antonarakis: And yes, I did talk about B7-H3, which I don't think we've mentioned to your audience before here, so I think it'll be a new topic. All of us by now are very familiar with the immune checkpoints, B7-H1, which is also called PD-L1.

Alicia Morgans: Yes.

Emmanuel Antonarakis: And PD-L2, also called B7-H2. Very few of us are familiar with the third immune checkpoint as sometimes called PD-L3, but also known by convention now as B7-H3. So turns out that B7-H3 is much more highly expressed on prostate cancer than PD-L1 and PD-L2. For the longest time we've been focusing on immune checkpoints, but maybe the wrong ones in prostate cancer.

So, I'll give you some examples. So PD-L1 is expressed anywhere between 20 to 40% of prostate cancers. B7-H3 is expressed in more than 80%, closer to 90%, and the prevalence may go up in metastatic castration-resistant prostate cancer compared to localized.

So, before I moved to the University of Minnesota, I used to be in Maryland and there's a company called MacroGenics in Maryland, and they have two compounds, both of which target B7-H3, and I had started to work with them and I continue to do so now from University of Minnesota. The first is an antibody against the B7-H3, you can think of it as an inhibitory antibody. So we did a window of opportunity neoadjuvant clinical trials. So, these were patients who were going to have surgery for intermediate to high-risk localized prostate cancer. And we took 32 such patients.

And prior to their prostatectomy, we treated them with enoblituzumab. And enoblituzumab is the B7-H3 directed antibody. It's a naked antibody that has no payload. The goal of that study was to see what effects the enoblituzumab Anti-B7-H3 has on the prostate cancer tissues at the time of prostatectomy.

The way that the drug is given, it's an intravenous infusion and it's given weekly. So these patients had six weekly infusions of the enoblituzumab, and two weeks after the sixth infusion, they underwent radical prostatectomy. We found a number of interesting things. One was several of these patients, about a third, had reductions in their PSA level just induced by the antibody, the enoblituzumab, prior to ever getting their prostatectomy. So that was exciting. We also saw a number of patients having one point and two point downgrades on their Gleason score. So for example, patients having a Gleason score of 9 on their prostate biopsy downgrading to a 7 at prostatectomy. And this happened about 45% of the time in the 32 patients. A little bit more than we would've expected by chance.

And then focusing on the immune analyses, we looked at, of course, CD8 T cells, which is a marker of an excited immune system. And then we compared the prostatectomies from this trial to another group of matched prostatectomies that were done in a contemporary fashion matched by Gleason score. And we did find that indeed this drug enoblituzumab did increase CD8 T cells, compared to untreated prostatectomy samples, as well as a number of other immune activators, including granzyme B, for example, which is an effect of immune function. So we are excited about that. This is unpublished information actually presented here for the first time, and we will be detailing the full data at ASCO 2022.

The second compound that we worked on is a B7-H3-directed antibody drug conjugate. So an ADC using the payload called duocarmycin, a chemotherapy agent, which is delivered into B7-H3 expressing cells, and this is also a MacroGenics compound. And we conducted a phase two study in approximately 40 metastatic CRPC patients, so the other end of the spectrum. These were all patients who had progressed on enzalutamide, abiraterone, and a taxane. So, this was a third line population.

And in those 40 patients, we saw a 54% PSA response rate, which is quite exceptional in these patients. And also in the 16 of the 40 patients with measurable disease, we saw objective tumor responses either at partial or complete responses in four of the 16, so 25% objective response rate, small sample size. But looked exciting.

Now, unlike the antibody which had very few toxicities, the ADC had more toxicities. And the three main ones were palmar-plantar erythrodysesthesia, also called hand-foot syndrome. The second was pleural effusions, which is not a side effect that we're used to seeing and managing in prostate cancer. This was probably an off-target effect of the payload.

But B7-H3 seems to be expressed in the pleura and maybe the lungs. And the third was thrombocytopenia, including some grade three in grade four thrombocytopenia. So definitely more toxicity with the antibody drug conjugate compared to the naked antibody. But we are now exploring both of those in other studies in prostate cancer.

Alicia Morgans: That's phenomenal. I'm really excited because, as I'm sure the audience knows, really targeting PD-L1 has not been as effective as we hoped in the majority of patients, at least. Can be highly effective in some patients. But this is really, really exciting.

Alicia Morgans: And I would just want to clarify, in your sort of new adjuvant study, it sounds like all patients made it to surgery and that they made it there in an acceptable time because six to eight weeks, I mean, that's pretty standard for localized disease. Can you just clarify that?

Emmanuel Antonarakis: Yeah, that's really important. So just imagine you're a patient, you've just been told you've got Gleason 9 prostate cancer and you've met with your surgeon, and he says, "We can take out the prostate gland." One of the considerations is, a lot of those patients are not willing to wait a long time. The other consideration is that these patients are otherwise healthy. They have a life expectancy that's measured in decades, not years or months. And so any neoadjuvant approach has to be done relatively quickly. I would say within three months. You wouldn't want to delay the surgery by more than about three months, maybe up to six at the most. And also the safety profile has to be excellent. You don't want to use a drug that has a 10 or 20% of grade three, grade four toxicities in that population. It won't be acceptable. So these patients, in that particular trial, had their surgery delayed by eight weeks. It was not really a delay.

Now we were not giving concurrent hormone therapy, so we were just giving the B7-H3 antibody by itself. There are some studies that delay surgery by more than three months, but they do incorporate some form of ADT or anti-androgens as well.

Very importantly, as you mentioned or are implicated, of the 32 patients, we only saw one immune-related toxicity. Now it was myocarditis. So this was a patient who had no cardiac history, presented with a slight shortness of breath, and was found in the emergency room to have a troponin leak with no other identifiable cause. We did decide to treat the patient with steroids, assuming immune myocarditis. And the troponin leak did improve, took about three months for the troponins to become undetectable.

So that was the only immune-related toxicity we saw. Could be scary to see myocarditis in a new adjuvant trial. Thankfully that patient did well. But yes, these are all important issues when you're consenting patients for presurgical studies.

Alicia Morgans: Absolutely. And even in the more advanced setting, it sounds like the drug had toxicities, some of which were not as familiar with, but are certainly manageable because it doesn't sound like there were grade five events and there were some grade three, maybe four events, but again, they were manageable.

It sounds like there were some pretty phenomenal measurable disease responses and PSA responses in a relatively heavily pretreated population. Where do you go from here with that agent and of course, with the new adjuvant approach as well?

Emmanuel Antonarakis: Yeah, so the antibody drug conjugate, by the way, has a name it’s called MGC 018. This is an ADC. In prostate cancer, as of yet, we don't have any FDA-approved or EMA-approved antibody drug conjugates. These, as a class, tend to have many of the chemotherapy side effects because there is a payload, which is a chemotherapy agent, which is delivered by the antibody. And the hand-foot syndrome is something that we're very used to managing with our TKIs in renal cell carcinoma, but is not something that many of our prostate cancer patients suffer from. Maybe in the era of cabozantinib where we were using that in prostate cancer, maybe that's coming back. We saw some hand-foot syndrome. But I have to say that there was that 10% pleural effusion rate, and some of these patients had to undergo thoracentesis to drain the effusion.

So that is quite bothersome to a prostate cancer patient. And then some of the myelosuppression is something we have to look out for, especially thrombocytopenias. The company behind MGC 018, MacroGenics, is excited enough to be developing a phase three program with this drug. Now that program is not cemented. There are two potential thoughts here. One is to do a randomized phase three study in a population where you compare against a chemotherapy agent, such as docetaxel or cabazitaxel, that would make sense to me.

The alternative is to look at patients who have progressed on one androgen receptor therapy like enzalutamide and randomized to the alternative one, abiraterone. We have seen a lot of that. Some people have worried about the ethics or the efficacy of back-to-back energy receptor-targeted therapies, but that would clearly be an easier bar, a lower bar in a phase three study. I think the bigger question is, do we need to select patients based on B7-H3 expression? So there are immunohistochemical stains for B7-H3, They can be measured and even quantified. None of them are sort of FDA-approved companion diagnostics.

And we're not sure what the optimal cut points are. So the studies that we conducted, we measured B7-H3 by immunohistochemistry, but we didn't require that for eligibility. So that's something that the field I think has to figure out. Are we going to go in an unselected population and examine B7-H3 post hoc? Or are we going to specifically select patients who express B7-H3, which is about 80 to 90% of metastatic prostate cancers?

Alicia Morgans: Well, exciting questions and exciting developments, I think, in the near future. So you'll have to keep us posted, and we'll certainly talk with you about those trials and progress as they come about in whatever form they arrive.

But thank you so much for really enlightening us on something that we, as a field, have not really dug into before, as you mentioned. I sincerely appreciate your time and your expertise.

Emmanuel Antonarakis: Thanks very much for the invitation, Alicia.