In this interview, Mark Frydenberg, MD, discusses the evolution of PET scanning, its clinical indications, its advantages and drawbacks, and key questions that remain about its use.
Imaging with positron emission tomography (PET) in prostate cancer represents an important clinical advance, especially for identifying metastases. Its use has expanded significantly, especially outside of the United States. In this interview, Mark Frydenberg, MD, discusses the evolution of PET scanning, its clinical indications, its advantages and drawbacks, and key questions that remain about its use. Dr. Frydenberg is professor of surgery, Monash University, and academic director of urology, Cabrini Institute, Cabrini Health, Melbourne, Victoria, Australia. Dr. Frydenberg was interviewed by Urology Times Editorial Consultant J. Brantley Thrasher, MD, executive secretary of the American Board of Urology, Charlottesville, VA.
To start, tell me what you consider positron emission tomography (PET) scanning today and how it has evolved.
PET scanning today has moved on from where it was previously with vastly improved tracers. We’re really looking for an imaging tool that better identifies metastatic spread of any malignancy, in this case, prostate cancer. Urologists have been well aware of PET scans in the past and had used PET choline scans, but they never became popular because they weren’t particularly sensitive and specific. Choline is also quite difficult and expensive to produce; it requires a cyclotron on site.
Since then, other PET tracers have been studied, including sodium fluoride, which is very good at detecting bone metastases. More recently, people have looked at fluciclovine F18 (Axumin), and in Australia where I practice, the most common one is a Gallium 68 PET PSMA scan, which is prostate-specific membrane antigen.
The dissemination of PET scanning now in Australia is well ahead of the United States. How is it being used in Australia?
Its use has really exploded in Australia over the last 5 years or so. The reason is that most of the nuclear medicine departments actually have their own gallium generators. As such, it is relatively quick, easy, and inexpensive to manufacture and deliver, so the dissemination has been very rapid in both public and private sectors.
The main use for PET initially was in biochemical recurrence following definitive treatment for prostate cancer, predominantly radical prostatectomy, but also primary radiotherapy. It’s also been used to monitor metastatic disease, especially in the setting of castrate-resistant disease, and more recently it’s been used in primary staging, in particular in patients with high-risk prostate cancer, to look for oligometastatic spread.
There’s worry always in the U.S. about cost. What would be a similar cost for a PET scan in U.S. dollars, and how are you able to justify the cost in certain patient groups?
The cost in Australia would be approximately $500 U.S., so it has become quite affordable. It is not something that the government pays for the patient right now. Government approvals for payment of new technologies often take 2 to 3 years so it may be some time before it is covered for the general public by a government rebate. However, it is readily available in virtually all of the university public hospitals in Australia. Many of the private providers also have access to PET PSMA, with most patients being reasonably comfortable paying for it privately if it’s not covered through the university hospital.
It obviously is important to use it wisely because there is a direct patient cost associated with it, especially in the U.S. As such, it’s important to try to choose the patients who are most likely to have a positive scan. In the primary staging setting, there’s no doubt we really need to look at the high-risk cases only. For the intermediate- and low-risk cases, less than 5% of patients will have a positive PET scan, and it will probably not be a cost-effective strategy for them. However, most of the studies looking at high-risk prostate cancer have shown 30% to 50% of patients do show oligometastatic sites on a PET PSMA scan, so it’s definitely worthwhile to consider it in that population.
The other group where it should be considered is obviously those with biochemical recurrence, and we’ve learned there is probably little point in doing a PET PSMA scan when PSA is below 0.2 ng/mL. The likelihood of a positive result is very low in that group, but it starts increasing from 0.2 ng/mL onwards. Most of the data show that about 30% of patients with PSA between 0.2 and 0.5 ng/mL will have a positive PET PSMA scan. That will increase to about 60% when PSA is between 0.5 and 1.0 ng/mL and will increase to about 80% when it’s above 1.0 ng/mL.
Next:"With PET PSMA, there are numerous other organs that will demonstrate uptake in addition to the prostate"Does it display the prostate only, or do other organs light up?
With PET PSMA, there are numerous other organs that will demonstrate uptake in addition to the prostate. Probably the most troublesome are the kidneys and the associated urinary excretion because it will actually show up in the ureter and bladder. One of the tricky aspects of PET PSMA is that a peristaltic wave of urine across the pelvic brim can be mistaken as a pelvic lymph node quite easily, so you need to be careful of that. It shows up in the bladder, which can obscure the detection of local recurrences, and is also taken up by the bowel and salivary glands.
Interestingly, it’s been shown that PET PSMA can be a useful tool for kidney cancers as well, but a lot of further work needs to be done in that area before it can be used clinically for that indication.
Do you put in a Foley catheter to decrease the possibility of hydronephrosis or something similar that would throw off the sensitivity of the scan?
Usually we don’t put in a catheter, but we do virtually always give Lasix to get some urinary clearance, then do the scan soon after that.
Our group did a study years ago showing that the CT scan in high-risk disease doesn’t offer much past the bone scan, and this has been confirmed by others. Could you see PET replacing CT?
Very much so. In fact, in Australia that largely has already happened with high-risk disease. We’re waiting with interest the results of a trial being done in Australia, in which I was a co-investigator, called the ProPSMA study, which is looking at conventional imaging versus PET PSMA for staging high-risk prostate cancer. It’s examining the utility of PET PSMA for identifying metastases but also for determining the likelihood of altering treatment. The main question that still arises is whether it makes any difference to long-term patient survival, but there’s no doubt the more information and the more accurate information you have, the better choices you can make for the patient.
One of the big criticisms with PET scanning is that we don’t have histologic confirmation when something lights up. In the ProPSMA study, will we have histologic confirmation?
Patients on trial who have gone on to have a radical prostatectomy and extended node dissection will have histologic confirmation of the positive nodes and how they correlated with the histopathology at the surgery. Hopefully we will get some more information because we’ve made the decision that all the patients in the trial will receive extended node dissections since they’re considered high risk. Even if they’re PET negative, we hope to also identify those patients who are missed because we know that the sensitivity is not perfect with PET PSMA scans.
However, if you look historically at the data when there has been correlation with histopathology, the specificity is excellent-often 95%-plus. Even with the data we have to date, we can be fairly comfortable that if something is positive on a PET PSMA scan, it will be histologically malignant.
For me, the bigger issue is the negative PET PSMA scan in a high-risk patient because we know that 30% to 40% of patients will not have a positive PET PSMA signal even though there might be disease present, which largely has to do with tumor volume. If the patient has lymph node metastasis that’s less than 2 to 3 mm in size, for example, that isn’t going to get picked up on the scan. But if you wait for it to get over 5 mm or so, there’s a much greater chance it will show up. So the problem is that we will end up missing small malignant nodes with a negative PET PSMA scan.
When do you think we will see the data from the ProPSMA trial?
We’re currently in the follow-up phase, so all the recruiting is finished. We’re basically looking at the 6- and 12-month reviews since patients have gone on to treatment, and we would expect that data around the time of both the European and American urology meetings in 2020. It’s very likely that data will be presented as an abstract at both the EAU and AUA annual meetings next year.
Let’s talk about PET’s use in oligometastatic disease or biochemical recurrence. First, when do you generally use it? Is there a PSA threshold that practicing urologists might take home if they have access to the scan? Second, what are you doing when you find oligometastatic disease or biochemical recurrence-surgery, radiation therapy, or continuing hormonal therapy?
It’s a very challenging area. The whole problem with the PET PSMA scan is it improves our ability to stage the disease or to restage the disease, but we have no idea whether the interventions we’re studying now have actually changed long-term survival. That remains a challenge for us moving forward.
A number of studies have shown that the PET PSMA results change management. If we look at biochemical-recurrent patients, about 30% to 80% have their management changed as a result of the PSMA scan. Typically what will happen is that instead of there being a presumed local recurrence, we find a pelvic node, an extrapelvic node, a bony metastasis, or sometimes a combination. As such, instead of radiation to the prostate bed, which would normally be standard of care for PSA rise post prostatectomy, we are then faced with the decision whether this should be done, combined with stereotactic ablative radiotherapy to involved nodes, or consider a salvage lymph node dissection, androgen deprivation therapy, or combination of therapies.
I don’t think there’s a belief that we’re going to cure the disease with ablative radiotherapy or salvage node dissections alone because usually the PET PSMA scan underestimates the degree of metastatic spread. However, we have data that we hope to publish soon that will suggest that it at least delays androgen deprivation therapy in about 50% of patients by at least a couple of years. That may at the very least provide the patient some quality of life benefits.
For a young, healthy patient with 20 years life expectancy, do you see a role for surgical removal in a patient with a positive PET scan? Let’s say it’s right at the bifurcation of the aorta and you’ve got one nodal spot that lights up on PET.
It’s clearly another strategy, as is stereotactic ablative radiotherapy, and it’s certainly been done in the U.S. but especially in Europe. Again, the results haven’t been that spectacularly successful because PET PSMA underestimates the amount of metastatic disease that’s present, and the small additional lymph nodes that are present just don’t demonstrate uptake. So you can do an extended node dissection and even a retroperitoneal node dissection, but there may well still be micrometastases in nodes further upstream that you haven’t visualized and hence haven’t been able to tackle.
I think this is an area we have to look at. The role of either radiation or aggressive surgery is very ripe now for clinical trials. But I think the take-home message is, what you see on the PET is probably just the tip of the iceberg, and there’s probably more microscopic disease beyond what is seen. Also, I certainly would not recommend a node pluck to take out the lymph node. You must do a proper node clearance in that area if you’re going to make any impact at all.
Next:"There are two aspects of the future: therapeutic and further imaging"Let’s talk about the future. Do you see new radionucleotides or new tracers coming out for PET scanning or imaging in general?
There are two aspects of the future: therapeutic and further imaging. With regard to further imaging, I think there will definitely be a swing toward using a fluoridated PSMA tracer rather than gallium. Fluoridated PSMA is a little more difficult to manufacture because it requires a cyclotron, but it has a very long half-life, which means it can sit on the shelf for quite a long time and can be made in large batches. Thus, a lot of nuclear medicine departments can store it on site. The advantage for me is that there is no urinary excretion, and the ureteric peristalsis mimicking a pelvic node won’t occur. It should be much better at picking up the local recurrences at the bladder neck after radical prostatectomy.
Interestingly, we’ve found in Australia that a multiparametric MRI of the pelvis probably actually identifies local recurrences often a lot better than a PET PSMA scan, purely because of the problem with urinary excretion. You can still pick them up on the PET scan, but an MRI is often better.
The other future direction relates to theranostics or treatment of advanced disease. We’ve been running some trials in Australia using lutetium as the agent, which is bound onto the PSMA and has effectively been used to treat castrate-resistant prostate cancer when all therapeutic options have been exhausted. There have been some very impressive results; we’ve had at least 50% PSA drops in well over 50% of patients and about 10% to 20% have extraordinary drops where the disease virtually melts away with a couple of treatments with lutetium. It’s generally relatively well tolerated. This is another very ripe area of research.
What you’re talking about is what some say is “calling in the fire department after the house has burned down.” Do you see a future where we take a nuclear tracer and attach a therapeutic molecule of some type at an earlier stage-perhaps biochemical recurrence with nothing showing up beside a small area?
I would agree with you. I think there’s a real chance of bringing these in much earlier in the treatment paradigm compared to where we are. In the current trials, we are using lutetium to look at toxicities and make sure it’s safe, and it appears to be very well tolerated. Once we confirm that it’s safe, it would make sense to be using it earlier.
Also, in patients with widespread metastatic disease, we often have to do a PET 18F-fluorodeoxyglucose (FDG) scan with glucose as well as the PET PSMA scan. In fact, we’re finding that some patients will have completely separate sites of disease and the PSMA and the FDG sometimes don’t match, and if this is the case the effectiveness of lutetium theranostics is reduced.
That’s interesting because we’ve always thought that it’s not glucose avid.
I think as it dedifferentiates, it actually becomes glucose avid. So when you’re using lutetium PSMA as a therapeutic target, it will track to the PSMA-positive sites, but if there is non-concordant FDG-avid disease those areas will not be well treated. Those patients actually don’t do well with lutetium PSMA because you’re not actively treating all of their disease, but you only see that in very end-stage disease.
Is there anything you’d like to add regarding imaging in prostate cancer?
I think it’s an exciting time. Both the advent of MRI and these new PET tracers like PSMA have altered the way we stage and restage prostate cancer. It does create a number of questions that we have to answer, including whether or not it ultimately leads to improved survival. Hopefully further trials will help answer the questions that have arisen.