The Use of Imaging for Patient Selection for PSMA Directed Therapies – Jeremie Calais

November 4, 2021

Alicia Morgans speaks with Jeremie Calais about the evolving landscape of theranostic and imaging modalities in prostate cancer. The discussion primarily focuses on the role of PSMA PET imaging in patient selection for PSMA-directed therapies like lutetium PSMA. Dr. Calais elaborates on the criteria used in the VISION trial and the complexities surrounding the necessity of targeted imaging for targeted therapy. He also discusses the predictive value of PSMA PET scans in determining patient response to treatment. The conversation delves into the potential differences between various PSMA tracers and the importance of understanding disease heterogeneity through PET imaging. Dr. Calais emphasizes that while there are no clear answers yet, PET imaging remains a crucial tool for understanding the biology of the disease and guiding treatment decisions.


Jeremie Calais, MD, MSc, Director, Clinical Research Program, Ahmanson Translational Theranostics Division, Department of Molecular and Medical Pharmacology, UCLA

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 you here with me today, Dr. Jeremie Calais, who is the Director of the Theranostics Program at UCLA. Thank you so much for being here today, Jeremie.

Jeremie Calais:
Hi, it's a very great pleasure for me to be here and share my experience. Thanks for inviting me.

Alicia Morgans:
Of course, it is always a pleasure to talk to you and really to understand your knowledge on this rapidly evolving theranostic and imaging landscape that we are facing in the era of PSMA treatments and imaging modalities. So today I'd like to really focus on the imaging aspect and talk to you about how we use imaging, potentially to select patients for our PSMA-directed therapies like lutetium PSMA. So can you talk to me a little bit about the criteria that were used in the VISION trial to select patients and what your thoughts are as a nuclear medicine physician in terms of how helpful they were in identifying this population? And maybe if you have a comment on it, whether they might really be necessary from your perspective, of course, personal opinion, in this patient population that is so heavily pretreated.

Jeremie Calais:
Yes. That's a hard topic. It is a very, a question that is a little bit everywhere. Do we need targeted imaging to do targeted therapy? There is a lot to say about it. I will try to put some elements together, and then people are free to decide whatever they want and think whatever they want. So in the VISION trial, first of all, let's say you have a patient, you want to know if the target is expressed to be able to treat this patient with the targeted therapy. So here we are talking about PSMA metastatic castrate-resistant prostate cancer patients.

And so in the VISION trial, PSMA PET was used as a biomarker, mostly not to include patients, I would say, but more to exclude patients who would have very poor target expression or a low-level target expression. If I'm correct, it was an 87% inclusion rate or screen failure of 13% of patients.

The main criteria were to have at least one lesion with visible PSMA expression above the liver reference and no lesions that are measurable by CT. So let's say pretty big already with the level of expression below the liver, what we call PSMA negative lesions. And these patients, they are pretty aware, especially when you take at the patient level, this patient has a lot of lesions. And it's easy to find at least one lesion with uptake above the liver. And so that's an easy inclusion criterion. So people say, okay, so if you manage to put overall survival improvement in that patient population in 87% of the patients, while you do not even need the PET scan screening, maybe this 13% that's okay to treat. And then you just removed one step on the patient's journey.

I understand the point. And now I'm going to describe a little bit of experience with the PSMA targeted therapy and what you can see based on the PETs and how we can try to move further from that. I know that a PSMA PET scan is predictive of a very good response profile. If you see very intense lesions, very homogeneous across all the lesions in the lymph nodes, this patient will respond very well, usually. Of course, there is no 100% rule in anything, but you have a more great likelihood of the chance that the patient will respond well. If the patients start to have more disease and a different shade of gray on the PET, meaning a different level of target expression, a little bit of heterogeneity, the patient can respond well, but a little bit less, or you have, let's say fewer chances for a less long time or in fewer number of patients.

And then at the end, if you see really like a low level of expression of disease in the liver or other prognostic factor, you know that your patient is less likely to respond well. But of course, in each of these categories, you do not have 100%. And then you give a treatment. Do you need to exclude patients to receive the treatment? The ones that you know are less likely to respond, but it's never 100%. So how can you say, I will not treat this patient because you are less likely to respond? It depends. If the patient has no other options, then of course you still may want to try because you need to give something to the patients that want to be treated. You need to give some hope, maybe to improve their quality of life, to improve bone pain. All these parameters are also important. But in my experience, usually, when it doesn't work, you see it pretty well. And it doesn't work very well.

I think for the approval of the FDA, I don't know what the FDA will say, of course, I will never know. But if we take the example of [inaudible 00:05:34], it's the level of the indication is somatostatin receptor-expressing tumors. They don't say how, and if you need to assess the expression, but that's the indication of the drug. So my assumption is that we will have something similar with PSMA, that we know that the treatment is effective in PSMA-expressing tumors, but then whether a physician needs it or is required to assess the expression of the PSMA, that I don't know how it will play. My assumption is that on the insurance level, this can play a big role. Will insurance, will they cover the treatments if there was no assessment before? That, I don't know. Will it be required?

It can save a lot of money to insurances if you require first to scan, that costs, I don't know, $5,000. If the patient is really not suitable for therapy, you can save up to, I don't know, a few hundreds of thousands of dollars, depending on the cost of the cycles, because you will remove the patient from the treatment list. So insurance may have some interest to have the PSMA PET requirement.

And then you have other tracers, people are talking about FDG and things like that. So in the end, if I have to summarize the three patterns, you have a very homogeneous intense uptake, an intense target expression level. This is a very good response profile. Then when it gets [inaudible 00:07:09] lesions in the bone, it varies, you have a much less good response rate. And then when it goes to liver lesions or low uptake, you have a poor response profile.

When you add FDG on top of that, you even see more, let's say heterogeneity of the disease, you see more PSMA negative lesions, the ones that you were not able to see on CT and bone scan. So you can select, even more, good responders. So the ones you would select would respond even better because you select them more. But of course, you would exclude more patients. Then it's always about this trade-off how you are selective. If you want to have perfect response rates, I know how to select these patients, but at the end, the pattern of these scans, I see it in maybe 15% of the patients only, but then that's not what you want. You want to treat as many patients as you can.

And of course, the less selective you are, the worse your response rate will be, and your response rates are. So it's a medical art. This can have consequences on the insurance coverage, at the end, I'm not sure really what will happen, but all these arguments need to be taken into account to get an opinion on how we use a biomarker for target expression.

Alicia Morgans:
I think that's a great point. And we certainly, as you mentioned, do not know what the FDA is going to say, but I think it's interesting, the label, because we can assess PSMA expression from tissue expression, even using a pathology specimen and thinking about IHC or some other way to assess within the tumor itself, perhaps could be an alternate strategy, but who knows. We will have to see what the FDA says and then think creatively about how we can help most patients actually have access, as will benefit, which is, as you said, always quite an art.

When you are thinking about using PSMA tracers to identify these patients, if that is the biomarker that we need to use, or that we choose to use to identify those patients who will benefit, there are multiple PSMA tracers available for PET imaging. And you are very, very well aware of that. Of course, you helped contribute to the gallium PSMA approval, which is wonderful, but not available at this point in time at all centers. Whereas PYL, PSMA will be available as sites ramp up.

From your perspective, are there differences between these and potentially other PSMA-directed PET tracers that may make a difference when we are assessing patients and using this potentially as a biomarker for patient identification?

Jeremie Calais:
To answer your question, when you say, are there differences? I would say, yes, there is some difference. And you say, would that make a difference. I'd say, no. That won't make any difference. In the end, you're talking about still the same target, PSMA. You're not looking at something different with these PSMA PET tracers. You still deliver a tracer that emits radiation on the site of the target, and then you localize where it comes from. And you can estimate the quantity of it. That is the same principle. And even if there is a tiny difference because it's not exactly the same PET emitter, it's not exactly the same chemical formula, in the end, it shows the same thing.

So long story short, I don't think there is a major clinically significant difference between whether you use DCFPyL or gallium 68 PSMA-11. These are the two main PSMA PET tracers that will be or are available in the US. There are many others, thousands of others in the world, but I think in the US, at least for now, these are the two main ones. And between these two, I don't think there is a major difference. The availability prevails. What matters is to have an assessment of the PSMA target expression, and that the patient gets a PSMA PET imaging scan, whether you get it with gallium 68 PSMA-11 or the DCFPyL, I think it doesn't matter that much.

Alicia Morgans: And that's very, very helpful. So thank you. So another hot topic question, as you know, the TheraP trial was designed using both a PSMA targeted PET tracer, as well as an FDG PET-targeting tracer. So from your perspective, is there a disease state in prostate cancer where it might be really important to use both of these approaches in combination, as was done in the TheraP trial to really understand the disease heterogeneity? Or are there states where it may be less important? Or is this something that we are still trying to understand as the field is moving forward?

Jeremie Calais: Yeah, I think we are clearly still trying to understand. First of all, I want to say that PET imaging phenotyping for me, was one of the key reasons why I choose nuclear medicine and theranostics. It is so impressive to see visually this biological mechanism in the body of a patient. You can see the heterogeneity then when you add multiple tracers, you can see mismatch and you really understand the disease at the whole-body level, which makes really the beauty of PET imaging. Now, do we know the best action based on these findings? I think the answer is no. And that's the problem why people say, why do you need that to do the treatment? It's just maybe, one day we will know what is the best treatment management based on each PET pattern. But for now, we do not know it.

And maybe this day will come. We will need definitely the scans. But because now we don't know exactly the exact correct recipe. People, they say the nonbeliever people can make the arguments. Do we not know how to use it? You don't need it, basically. Me, that's not my point. The more you can get information on the patient, the more likely you are to better treat your patients. It's just, we have to find out what is the best.

So again, as I said, just before, when you combine PSMA and FDG, so you have the target expression and then you have the glycolytic activity. Of course, the FDG indicates high glycolytic activity and aggressiveness. So mostly the lesions that will drive the prognostic of the patient because those are the ones that are very aggressive. And by combining the two, you increase the number of heterogeneous sites you can discover.

And again, you increase the number of patients who are less likely to respond, you can see better. Imagine what would have happened if, with any other treatment chemotherapy or second-line androgen deprivation therapy, you had such PET imaging biomarkers. You could predict before where your treatment will go and if there is a lot of heterogeneity or not, how this would have impacted the indication of these direct steps. It's just that we do not have these PET imaging agents available for these other tracers, but I think that's the beauty of theranostics. And I think we should really do the baseline imaging assessment and understand the biology of the disease in the best of the world. I would really like to see, you have individualized management based on the patient's disease, that would be assessed with multiple tracers, indicating multiple phenotypes at different levels.

I heard a good idea, like when you have FDG positive lesions, the ones that are most aggressive, maybe you can hit them with external beam radiation therapy because these are the ones that are less likely to respond to PSMA targeted therapy only. So you can combine molecular systemic PSMA targeted therapy with focal external beam radiation therapy, to the most aggressive [inaudible 00:15:08] sites. That may be an attractive approach. Maybe you can combine it with, I don't know, even chemotherapy or other immunotherapy to overcome the resistance of non-target expressions at these sites. That may be another approach that can work. I think the more that we advance, we will find out. For now, there are no clear answers, but I do not think that it's just because that there is no clear answer right now, that we should not continue to use PET imaging for phenotyping.

Alicia Morgans: I think that's a great point. There is still so much to learn and maybe a process by which we use biopsies to understand the molecular heterogeneity as it relates to what we are seeing from these PET tracers would be a great step, and then understanding how patients respond to different therapies as they have these different expression patterns, because I'm sure that a patient would more likely rather undergo PET imaging than undergo multiple biopsies to understand the genetic heterogeneity of their disease.

So very, very interesting. And as you say, I agree, this is the beauty of PET imaging and really understanding our patients on a larger and more clear level is our goal. So thank you so much for sharing your expertise with us today.

Jeremie Calais: Very welcome. And I'm happy to come back anytime. It was a pleasure.