Read the Full Video Transcript

Andrei Iagaru: Thank you very much, Jeremie. Thank you, Jeremie, Tom, and Johannes for the very kind invitation to speak here, also to the Prostate Cancer Foundation for having me talk to you about something that has been around for almost half a century. So it's not bone scans then that were around for as long. But hopefully we'll come to a conclusion and agree that actually FDG-PET helps here.

And I'm going to give you the disclaimer that this is about FDG in the VISION population. So I was as surprised as you were this morning that FDA announced today—we were expecting early April. So I'm going to talk about VISION population.

What I'm going to talk about applies probably to this jurisdiction in the US and perhaps to academic centers more than in private practice. So, how do we find the right balance? What is the right amount of imaging?

There are people who say there's no need for imaging, that the targeted expressed will give the treatment, and we'll see the response. And then there are others who say we need to image every patient with every possible agent that's available out there because that's how we interrogate the biology.

I do believe that we need to tailor our approach based on the patient phenotype and what we know about the disease, and find the right medium. So you've seen this in people who say no imaging is needed. In the VISION trial, 85% of patients had at least one lesion, so they qualified. So some argued no need to do PSMA imaging.

Johannes and others, ourselves included, wrote that this is ludicrous, because one of the principles of cancer treatment is to know the extent of where cancer is present. And then imaging has been described as complex, expensive, may limit access to care even more.

I will try to convince you otherwise. And in addition to the PSMA, which is something that you are now very familiar with, we're going to talk about the drug that has been around since the late '70s: FDG, fluorodeoxyglucose, that goes wherever you have higher glucose consumption—in this case, more aggressive prostate cancer.

This has been around for a while. And our colleague here across town at USC, Hossein Jadvar, built his career on doing FDG in advanced prostate cancer. And he has shown before others did that there is a lot of value in using FDG-PET in advanced prostate cancer. And he wrote this nice editorial post-VISION about the use of FDG.

I'll also give you the disclaimer that I am fully aware that the label does not require you to do FDG-PET imaging prior to deciding to treat someone with lutetium PSMA. But is it the right approach? So let's talk about a few things—access, cost, and reimbursement.

This is data from the IAEA. You can go find your jurisdiction, and you'll see how many PET scanners and SPECT scanners are available where you live. And you see the US both for PET and for SPECT as green, no pun intended, as Australia. So if they can do multiple scans prior to deciding if a patient is eligible for lutetium PSMA, what's the argument that we cannot?

Now, cost—these are not Stanford charges. I want to make that very clear. There's no way I have access to what Stanford charges every insurance company. But these are estimates of CMS reimbursement. And if you look at the cost of adding FDG-PET CT, it's less than 1% of the cost of six cycles of lutetium PSMA. Less than 1% to determine if someone should benefit from the drug or not. So why not do it?

As far as reimbursement, let's dispel that urban myth that it's not reimbursed. CMS has decided, rightfully so, that the initial diagnosis for prostate cancer FDG is not needed but that biochemical recurrence initially with coverage determination—and then since 2013 with actual determination—CMS covers for FDG-PET biochemical recurrence. And I have to say this is data that was studied here by UCLA and the Institute for Molecular Imaging, Academy for Molecular Imaging, with Professor [INAUDIBLE] and Barry Siegel, who did the NOPR, the National Oncologic PET Registry—tens of thousands of patients across the board, uniformly. About a third or a minimum of a third of patients benefited from FDG. That is the basis why CMS pays for FDG, including in prostate cancer.

This is our workflow. Maybe we image a little bit too much at Stanford. But the truth is we don't know what to image with what, so we will learn from looking at the data. And we deliver on what we said. These are patients who go through all these cycles, and we try to get the patients in and out within 90 minutes. And we have images that are perhaps not as pretty as at four hours or 24 hours, but they give us information that has predictive value.

So when should we use FDG-PET? Definitely for eligibility check before selecting someone for lutetium therapy. I would say if this delays starting treatment, OK to skip it, but otherwise it can give you useful information. And it's not only what Michael has shown in his trial. It's what we see in day-to-day practice.

So in this case, there are lymph nodes that are biopsy-proven metastatic prostate cancer in the head and neck. There are other lymph nodes that are also prostate cancer in the neck that are FDG-avid and not PSMA-avid. Is it worth treating these patients? We ended up treating, and then we decided to treat the patient with radiation for the other lymph nodes. If we only had the PSMA here, these patients qualify by VISION, right? There are PSMA-avid lesions. But now that you have the FDG, how many people in the room would feel that lutetium PSMA is the appropriate treatment? Not too many.

And if you didn't scan, again, this patient would have undergone treatment that's expensive and has toxicities. And this patient even more so. There's no PSMA-avid disease, and there's a lot of FDG-avid disease. And these are just examples from day-to-day practice. So perhaps ignorance is blissful, but we're not in the business of being ignorant. We're in the business of being educated and informed, and making the best decisions for treatments.

Another scenario—FDG can help us troubleshoot during the courses of radioligand therapy or targeted radionuclide therapy, however we want to call it. This is an example where there was PSMA-avid disease. After cycle one, treatment went there, but there are some funny new lesions on the SPECT-CT. Then after cycle two, we decided to do an FDG. At that time, there's more FDG-avid disease than PSMA-avid disease. Is radioligand therapy the right course here by itself? I think these patients would benefit from other treatments, perhaps enrollment in a trial.

In another example, great response after cycle two and three, but the PSMA was stubbornly high. And there was still disease on PSMA-PET, and that disease did not match the appearance on FDG-PET. So this patient, again, probably should not continue with all full cycles, definitely not by itself. This patient would probably benefit from being enrolled in a trial.

And I know that Tom will talk to us about SPECT. But I want to share with you some very recent results from my colleagues, Farshad and Tomoaki, who used this novel method of generative adversarial networks to create SPECT images that look just like PET. So on the left, we have a PSMA-PET—high image quality. One hour post-infusion of lutetium PSMA-SPECT shows that the treatment went or intended to go. And I say that image on the right—this newly developed algorithm—makes a SPECT image look just like PET. And it's a SPECT image that has been acquired in two minutes per bed. So this is done in under 10 minutes, and it looks just like PET. And, again, no excuse to not image, because imaging is allowing us to provide accurate information and change management when changes are due.

And we're coming close to the end. So in conclusion, CMS reimburses for FDG-PET in advanced-stage prostate cancer. I would suggest that you use it, because you learn much more than if you don't use it. Don't delay treatment if it takes five to six weeks to schedule your FDG-PET scan, but otherwise I would encourage you to do it. It can help troubleshoot. And the more we know about the biology of cancer, the better we are.

Many thanks to my 90-plus colleagues in nuclear medicine at Stanford. And many thanks to you for listening to my talk.