Imaging Prostate Cancer
Posted: Nov 01, 2018
POSTED: November 15, 2018
Thomas Hope, MD, of UCSF and the San Francisco Veterans Affairs Medical Center, is keenly interested in novel imaging agents and therapies for prostate cancer and neuroendocrine tumors.
Prostatepedia spoke to him about novel imaging for prostate cancer.
Dr. Hope: I do a lot of radiopharmaceutical imaging. We inject radioactivity into people, and then we image it with a positron emission tomography (PET) scanner to locate where the radioactivity has gone. We label these small molecules (proteins) with the radioactivity and use those proteins to target different places in the body. In this case, we try to figure out where prostate cancer is.
There’s been a whole host of developments over the past 20 years of increasingly improved detection strategies for prostate cancer. The old-school fluorodeoxyglucose (FDG) PET imaging technique has been around for 30 years in the United States, and they have used sodium fluoride worldwide for just as long. For bone imaging, sodium fluoride can tell you where obvious metastatic bone disease is. FDG is actually the stable for the majority of PET/ CT imaging we do in the world, but it’s used primarily for other cancer types that are hypermetabolic or use a lot of glucose. When prostate cancer is in the earlier stages, it typically does not use a lot of glucose.
There’s been this hole in prostate cancer treatment for patients with biochemical recurrence. These are patients who have undergone definitive therapy and have a rising PSA. Neither of those two imaging modalities really help them. But a couple of new agents have been developed.
Choline-based agents, such as fluorocholine and C-11 choline, have been used in the United States and Europe. The Mayo Clinic brought choline C-11 to market in the US. Those radiotracers are certainly better than FDG PET or sodium fluoride PET in localizing particularly soft-tissue metastases, but they fail at lower PSA values.
When your PSA gets below one, you really don’t see much disease, and the studies are also quite difficult to interpret. The next imaging agent, is Axumin (fluciclovine). Fluciclovine is another amino acid tracer, just like C-11 choline, that’s used in biochemical recurrence. It was FDA-approved two years ago and has been used fairly frequently with biochemical recurrence. It’s probably, in my mind, equivalent to choline imaging. Fluciclovine itself is not really used outside of the United States because, if you have the availability of other radiotracers, you wouldn’t use fluciclovine. Yet in the United States, fluciclovine has become the mainstay because it is reimbursed by Medicare and readily available. Prostate specific membrane antigen (PSMA) compounds have been developed by a number of groups and companies over about ten years. The big change came with the gallium PSMA-11 compound, which is a small molecule that was first developed at the University of Heidelberg in Germany. Gallium PSMA 11 is a unique compound in that it wasn’t patented. No company controlled it, and so any site that wanted to use it could just sign up and get the precursor delivered to them. Very quickly, a large number of sites around the world started using PSMA-11 to image patients with prostate cancer. It is not approved in many counties, although technically, it is approved in Switzerland and Israel. Outside those countries, it is used on a compassionate-use basis. In the United States, it is being used under Investigational New Drug (IND) authorization from the FDA. The fact that it was so quickly adopted and widely used led to a huge number of articles in the literature.
In addition to PSMA-11, there is a whole host of other PSMA compounds. Gallium PSMA-R2 is being developed by AAA, DCFPyL is being developed by Progenics, PSMA-1007 is being developed by ABX Chem. There is a whole family of PSMA compounds coming to market on the back of the experience of PSMA-11. There are questions as to which is better, and although there is not a lot of head-to-head literature published, it’s fairly clear that PSMA 11 is better than, for example, the choline radiotracers and fluciclovine. The question is: how do these other PSMA tracers rate against one another? In my mind, they’re much better overall as a class, but I’m not sure there is a huge difference between them in terms of detection activity. We’ll find more about that as things progress.
Dr. Hope: You have to go country by country, so it gets complicated. In the United States, for example,
PSMA-11 is not owned by a company, and there’s no company paying for clinical trials. Centers like ours are running trials through Investigational New Drug (IND) authorization, which means it’s being studied in clinical trial aiming to get FDA approval. In the United States, everything is done under a clinical trial. There are a couple of methods to pay for the studies. There are a few insurance companies that will pay for these imaging studies under a trial setting. But I would say that the majority do not, and patients end up having to pay out of pocket.
The FDA allows you to use a cost recovery mechanism if you are acquiring data to eventually support an NDA application, and that’s how the majority of these studies are paid for. There are other institutions that use research funds in order to have a small number of studies performed. The two major institutions in the United States are UCSF and UCLA, and each uses cost recovery mechanisms and billing patients’ insurance companies directly in order to perform the study.
Dr. Hope: Yes. I would say they range between $3,000 to $5,000 apiece, so they’re quite expensive. There is clearly an ethical dilemma in having patients pay for an imaging study that’s not FDA-approved. What do you do with that? I think it is a reasonable approach as long as the institutions are actually using that data in the way that they state they are, which means that it’s up to us to use the data to get the agent approved. If the data isn’t used productively to get the drug approved so that insurance companies will pay for it, then I have an issue with the ethical aspect of it. As long as I’m doing the work, then it may be reasonable, although different people might disagree.
Dr. Hope: There’s no right answer. For example, two weeks ago, we went to the FDA for our pre- NDA meeting and presented all of our data, which we are doing in collaboration with UCLA. The FDA was very positive and said that we had enough clinical data to support an NDA application, which is pretty exciting. Hopefully, we can get the drug approved within the next 6 to 12 months.
Dr. Hope: These newer techniques are changing current patient care. But is that actually improving the outcome? For example, if you have a low PSA, and your PSA is 0.2 after radical prostatectomy, the standard treatment is to radiate the prostate bed and maybe the pelvic nodes. Now you get a PSMA PET, and it shows a node somewhere over here. So, now the radiation oncologist zaps that PSMA-positive node. Everyone thinks we did a good job, and maybe we did. But we just don’t know.
What we don’t understand fully is whether or not PSMA PET is the tip of the iceberg. If you have a PSMA positive node, are there many nodes we do not see, or does it mean that those nodes are the extent of the patient’s disease, which we can potentially cure if we hit it? Right now, the care is changed in maybe over 50 percent of patients who get a PSMA PET, but whether that change in care or treatment planning has improved outcomes, no one has a handle on that yet. There are some clinical trials starting that use varying radiotracers. The question in the community is: how does PSMA PET impact this care, and does that change improve the outcome of the patient who we’re imaging?
Dr. Hope: Yes, but it is not that straightforward. You cannot take a cohort of patients who got PSMA PETs, check what happens to them, and conclude that things got better. You have to do it in a trial setting with a cohort of patients who do not get PSMA PETs and a cohort who did, and see if there’s a difference between the two. Otherwise, there are a lot of biases if you have a one arm study. You cannot tell if the patients have improved outcomes for other reasons or even how you compare the data. You really do need a randomized trial in order to demonstrate this improvement in outcomes.
That will come, but those trials will take a very long time to perform. These drugs will all be approved well before the length of time that these trials take to perform. This becomes a big issue. If you have an imaging agent, and we all believe it’s better than the previously existing ones, how do you randomize patients to not get it once it is FDA-approved? We are going to face difficulty showing that PSMA PET improves patient outcomes because we are going to be bottlenecked based on the availability of agents in the near future.
Dr. Hope: It has happened in imaging over the years. Take sodium fluoride PET, which was never approved. It was grandfathered into FDA approval. No one ever did any clinical trials showing impact and outcome, and that is why Medicare has chosen not to reimburse sodium fluoride PET CT. This has happened over and over again.
It is mainly because imaging trials are unique. Drug trials must have outcome benefits as the endpoint in order to obtain approval. Imaging trials only need to show that we saw something we thought we would see. For example: “I think there is prostate cancer, I looked at a cohort of patients, I biopsied them, and the biopsies came back as prostate cancer. Therefore, this imaging study is good.” But that doesn’t work in a therapy world. Therapy data is a lot stronger.
Dr. Hope: That is a hard position to be in right now. Think about it in a different setting. Let’s say you were at your institution, and you were thinking about participating in a clinical trial for an investigational therapeutic agent. Most men would not travel too far outside their institution for that therapy. With PSMA PET, patients are traveling all across the country for this agent and paying out of pocket for it. It’s an unusual circumstance. Two years ago, it would not have occurred to people to do this.
I think in the United States, you have to think about the cost and the marginal benefit. It really depends on your PSA. It depends on discussions with your oncologist or urologist in terms of where you are and what type of therapies you are thinking of. Outside of imaging studies, there are therapeutic aspects of PSMA targeted radionuclide therapy. That becomes a much bigger issue. Outside of the United States, the vast majority of sites that offer it do so outside of trial settings.
There are potential huge ethical issues with doing that. Sites are treating patients with therapies that have significant toxicities, and that data is not being collected prospectively, is not being reported, and the trial is not being done in a way that will lead to data that will help us determine what to do with patients moving forward. Centers should run clinical trials and publish results so that we learn, but there’s a large number of centers around the world offering some of these agents out there to treat patients with limited to no follow-up.
It’s really important that, if we’re going to treat patients with a non-approved drug, the trial or the setting where it’s administered does so in a way that leads to actionable, usable information for the community at large, and not just the individual institution or patient.