18F-fluciclovine PET/CT in Patients with Suspected Recurrent Prostate Cancer - Bital Savir-Baruch

Phillip Koo welcomes Bital Savir-Baruch who is presenting a review of prostate cancer imaging with 18F-FACBC fluciclovine (Axumin®) in men with suspected prostate cancer recurrence.  Dr. Baruch reviews the benefits of this technology including earlier detection of recurrent disease as compared to conventional imaging, inability of the tracer to be metabolized and the slow urinary excretion. While detection rates improve as PSA increases, there is no absolute PSA threshold for obtaining a scan.  Based on data from the LOCATE study,18F-FACBC fluciclovine scans may result in a change in the treatment strategy for 60% of patients and is therefore recommended prior to the initiation of salvage therapy. 

Biographies:

Bital Savir-Baruch, MD, Assistant Professor, Radiology Loyola University Medical Center Maywood, Illinois

Phillip J. Koo, MD Division Chief of Diagnostic Imaging at the Banner MD Anderson Cancer Center in Arizona.


Read the Full Video Transcript

Philip Koo: Hi, this is Phillip Koo, the Editor for the Center of Excellence on Imaging at UroToday.com. Today, we're very fortunate to have with us the second lecture in the new SNMMI lecture series on prostate cancer imaging and therapies. We're very fortunate to have with us today Dr. Savir-Baruch, who is the Assistant Professor of Radiology at Loyola University Medical Center and an expert on fluciclovine imaging. Thank you very much for joining us.

Dr. Savir-Baruch: Thank you very much, Phil, for the introduction. As you mentioned, I will speak today about prostate cancer imaging with fluciclovine, also known as FACBC and Axumin® PET/CT scan. Before we start, it's very important to emphasize the FDA approval for this modality, it is for men with suspected prostate cancer recurrence based on the elevated blood PSA level following prior treatment. In other words, we're using this modality for recurrent prostate cancer and not for initial staging.

One of the main benefits of fluciclovine is that it actually may be positive when other conventional imaging is negative. There is no absolute PSA threshold to order that scan, but one should remember that the positivity rate is increasing with increasing PSA level. Now that being said, we do recommend to consider that scan before obtaining salvage therapy. That's for accurate treatment planning and I'll show you some slides that actually emphasizing those recommendations.

Without going much into chemistry, FACBC is an amino acid analog, a synthetic amino acid analog. The most importance of that slide is that it will go into the cell but will not be metabolized, which is a great benefit of a tracer.

In terms of the whole body distribution, as you can see, there will be intense uptake within the liver and pancreas. We will see some heterogenicity in the bone marrow. Then, you can see we have increased tracer uptake with time within the skeletal muscles. We won't have much of uptake within the brain or the lungs.

Now, another benefit of fluciclovine is its low and slow urine excretion over time. You can see that we will eventually see some filling within the bladder, but usually, and I'll show some pitfalls later, usually, it will not be as intense as that of FDG, which allows better visualization of the pelvic organs such as prostate.

Now, amino acids are in demand for both anabolism and catabolism, but in prostate cancer specifically, there will be upregulation of the amino acid transporters. Now, it is important to understand that the mechanism of the amino acid transporters is based on 1:1 exchange. For every amino acid to go in, one has to come out. This is why we see a very nice peak on early time images, but then there is a downsloping time-activity curve. Despite that, the uptake within prostate cancer cells will be higher than those of benign and the background, but we do recommend early imaging to actually catch that peak of concentration within the cell.

Now, going into the clinical data was done before the FDA approval and those data are based on the publication in Urology, with additional final analysis that we'll publish sometime in the future. We do see that the positivity rate of the overall is 83%. That means that out of the 128 patients in this trial, 83% of those scans were positive. It's a common trait of looking at performance of scan. Another thing that we observed is that the positivity rate is increasing with increasing PSA level, and we can see when the PSA is less than one, the positivity rate is around 40%.

Now, going into the diagnostic performance is where we're taking those scans and we're comparing it with the truth. We do see that for prostate localization, the sensitivity is very high with 90%, yet the specificity is only 35. That means that we do have some quite high numbers of false positives. Now, we do another breakdown of this population. We do know today that for patients that underwent a prostatectomy, the study may be more specific compared to those that had radiation therapy as initial therapy with a little bit higher false positive. Now, going into the extraprostatic localization, we do see that the sensitivity is 57%. Again, we need to remember we're looking at a population that out of conventional modalities are usually negative, so this quite high sensitivity rate. Now, the specificity here is 95% and that means that we may not see all disease, but once we see, it is actually very quite positive.

Going into now the cases that we obtained as part of clinical practice, and again this is less selective population compared to the prospective study was done prior to the approval. We actually got almost the same positivity rate in 152 patients that we scanned here in our institution. We have 81% positivity rate. When we looked at the parameters that actually influencing the positivity rates, we found again that PSA has a linear relationship with positivity rate. I would like to take you into that side of the graph where PSA is less than one. In this population, finding the disease is very hard. We have slightly higher numbers of positivity rates compared to the prospective study. Also, we have almost 40% of patients that actually presented with extraprostatic disease, and this was statistically significant.

Now, going farther into another of the analysis of the doubling time, PSA doubling time because we know that PSA kinetics actually is a risk factor. We couldn't find any statistical significance. In fact, you can see it's pretty fluctuating. But, when we actually took this parameter and the doubling time, and we put it together into different groups, first of all, we couldn't find any linear correlation, but you can see that when PSA is above one, no matter what the doubling time is, we have quite high positivity rate. Now, I want to take you to the left side of the graph. We actually found 40% patients where the doubling time was less than three and the PSA was less than one. Even though we couldn't prove any statistical significance, we truly believe is that we have actually less population to reach significance, so we will do this analysis in the future to see if there is any correlation.

Going into the Gleason score, and this is Gleason score at the time of diagnosis, we couldn't find any a relationship with the Gleason score or Gleason risk groups. What we could actually find, there is a nominal association between increasing Gleason score and the odds of positive extraprostatic disease, which is that kind, this slide, this area of the graph.

How do we take all of that into clinical practice? I think this is the main question, and I do want to take your attention to this publication of the New England that was published just after the FDA approval. Now, this population is an initial population. It's not a recurrent population, but I'll read the conclusion. It says that "at a median of 10 years, prostate-cancer-specific mortality was low irrespective of the treatment assigned, with no significant differences among the treatment." When you see all the approach of the molecular imaging, if it's fluciclovine, choline, PSMA, one should ask the $1 million question, do we really know what we didn't know today. One of the biggest questions that we should ask, we choose wisely, or we can only know what we know because we don't know what we don't know or maybe we think we know because we only can make an assumption of a disease pattern. This is how I think we should actually apply the benefits of molecular imaging.

This was actually proven by the LOCATE study for the recurrent disease population with biochemical recurrence. The LOCATE study was the multi-center trial that looked at the population that was supposed to have some kind of salvage treatment. They had a treatment plan before the fluciclovine PET/CT scan, and a treatment plan after. According to that graph, you can see that 60% of that population actually had changed in management. Now going into the breakdown of that, we have a major change and we have other change. The major change was 78% of them. What was the major change? It's a complete change of the treatment regimen. For example, someone's supposed to go to hormonal therapy and now goes into surveillance. Someone's supposed to go into surveillance, now goes to radiation therapy or systemic therapy. That's a quite high number. Now, the other changes are small changes within the treatment plan. For example, one's supposed to have a whole pelvic radiation treatment plan and now going into a specific zone or vice versa.

Let me show you some examples. This is a patient that presented initially with the low-risk prostate cancer adenocarcinoma with Gleason 6. He underwent external beam radiation therapy that completed in 2008. His lowest PSA level was 1.7 and now presented in 26 in 2016. He had multiple prior attempts to actually localize the disease within the prostate bed with biopsy. Just before he had the scan, he had a positive biopsy for local recurrence in the prostate bed, negative CT and negative bone scans. Obviously, he was a candidate for salvage radiation therapy.

He had the scan and let me take you to the left side. There is a left deep pelvic lymph node that is very specific for recurrent disease, as I've showed you before. This would be covered within the radiation zone. However, he does have also in the middle slide, he has also a retroperitoneal lymph node that was highly suspicious for malignancy. On the right side, there was a very, very subtle mixed less dense sclerotic lesion. I'll show you another example afterward that had very much abnormal optics. Obviously, if this patient will go into radiation therapy, he will have biochemical recurrence again. This treatment was changed and this patient was placed on hormonal therapy. At the beginning of the time of that scan, I got a lot of questions of what do you do with that, what's the next step. Now, we are looking at patients that we didn't know they have oligometastases sites. What are you going to do?

We're looking at patients where their PSA is still low. We're not looking at a thousand, so a lot of them will go into hormonal therapy. Now, there are some studies looking at salvage lymph node resection. It's not yet recommended treatment, so there are sites that looking at that as diagnostic resection to get the truth of the exam and also to follow up PSA. Also, modification of radiation, and not just within the pelvis, to approach it in a different way to do extra pelvic radiation therapy. But again, you have to look at consistency. With all those treatments, the main question is what about quality of life, so this is the question that in the future with molecular images we will have to answer. For example, for this purpose, this patient was on hormonal therapy.

This is another case where a patient had initial prostatectomy and now presented with a PSA of 0.4 with a doubling time of three months. Obviously, he was a candidate for external beam radiation therapy just because the assumption is that you have a local recurrence. This patient was presented to us, and you can see there is a very specific, abnormal aortocaval lymph node. Obviously, this patient would fail radiation therapy. His physician decided to take him into excisional biopsy from the diagnostic purposes and just to follow up PSA. That's another example of guided salvage lymph node resection.

What about radiation therapy? This is an example of a patient that had salvage radiation therapy after radical prostatectomy back in 2009. His initial Gleason was 7. On the scan, that he presented PSA of 0.4. We did actually find a five-millimeter presacral lymph node that was very specific for recurrent disease, as you can see. The radiation therapy was modified into the presacral region compared to a whole pelvis, again. This patient already had radiation therapy before, so that's not without any risk. His post-treatment PSA was undetectable, so that's where radiation change was the main player.

One of the questions that we keep getting from referring physicians: Can that replace bone scans? Let me let you judge. Going into the overall experience, there was no good perspective study for those. You can see that there is intense uptake within lytic lesions. You can also have intense uptake on those isolated bone marrow lesions without any changes on CT. We will have some moderate mixed lesion, mixed up, sorry, moderate uptake and mixed sclerotic and lytic lesions. This is like the example I showed before in our initial case. Unfortunately on a very dense sclerotic lesions like these, we may have false negatives, so in some cases for those who do not have any changes on CT with intense uptake, and those that have changes on CT without any uptake, maybe those cases will need to have further evaluation with MRI or other bone scans, so that will be a case scenario.

Going into a false positive, we do have the example of a very positive looking lesion, but it's too good to be true. It looks like a drop within the right iliac crest, but it does follow criteria for positivity, and therefore we needed to do an MRI and we could prove that this was just a bone marrow regeneration without any evidence of cancer involvement. With experience, we learn that we do see them on the iliac crest and also on the acetabulum, so you do need to make sense of the case before calling isolated lesion metastatic disease.

Another pitfall is the bladder issue. As I showed you before, the fluciclovine has a low and slow excretion, yet we do get some outliers with very intense uptake. On the left side where the green arrow is, you can see that this is intense uptake that actually masking the prostate bed, so you cannot evaluate it. Then on the upper images, you can see that that can actually mimic local recurrence within the prostate bed for someone that had a prostatectomy. Also in the lower end, you have ureteral optic that can mimic positive lymph nodes. Those cases may need to have greater evaluation. What's the solution? We did. We looked at that here as part of our quality analysis, and we could actually see that if you ask patients not to void prior to the exam, and now we're thinking it's around an hour before the injection time, you can actually have much more patients in the group of the insignificant and mild excretion compared to the intense one. Therefore, you can actually control that situation to some extent.

What the referring physician needs to know, the overall take-home message is that the FDA approved indication is for recurrent disease. We don't have enough for initial disease, but I hope that will come in the future. Fluciclovine PET scan whether it's PET CT or PET MRI may be positive even when conventional imaging is negative, so it's worth giving it the benefit of doubt. Again, there is no absolute PSA threshold to order the scan, but you need to keep in mind, as I showed before, the positivity more likely to be when the PSA is above one, but in real life scenario, majority of the patient wants to catch it much before, predominantly for the prostatectomy population. You need to remember that probably the kinetic test has some role, but we have quite a bit high positive cases when the PSA is less than one. The last thing is that you should actually consider to the order the scan before salvage therapy, that'd been proven by the LOCATE study, that will allow accurate treatment planning. Thank you very much for having me, today.

Philip Koo: Thank you so much for that wonderful presentation. Your data on the doubling time and no associations with the positive scan was very interesting. Your thoughts on the use of doubling time, do you think that needs to be investigated more or do you have an explanation why you think that doubling time doesn't really affect the positivity rate?

Dr. Savir-Baruch: Yeah. First of all, if you look at the slide, we actually took some breakdown where those doubling time showed to have influence of the disease prognosis, so it wasn't just random thinking. Why is that? It's good. I don't know how to answer that because other studies with other tracers actually got the same conclusion, that kinetics is not a predictor for a positive scan. However, I truly believe that we don't have enough population, so we actually planning to go back, to look at that and to try to increase that number of groups, patient to maybe reach significance. I think it's more of a population rather than reality, because the studies that actually showed that these breakdowns actually has some prognostic effect, the number of patients is much higher.

Philip Koo: From your own experience, will you scan patients with a PSA lesson one and do you feel like that we should be scanning patients with a PSA less than one and even less than 0.5 because I think we all had those cases where we see these patients with very low PSA values and positive scan?

Dr. Savir-Baruch: In fact, we did publish a small abstract because of the limitation of time. Our positivity rate for less than 0.5 is more than 50% as well. The answer is absolutely yes. I wouldn't bother. I don't think it's the PSA. I think it's the treatment. If the physician decides that he going to go into surveillance because of different reasons, maybe it's not a correct time to do the scan, but even then maybe the scan will find something that will take the physician away of putting that patient on surveillance. I think it's more of the planning, the treatment planning.

If someone wants to treat their patient with salvage treatment, I think getting that scan is going to be a huge help to actually stage more accurately because we don't really know if those tiny lymph nodes that we see on CT images are cancer or not. This is where the size of those lymph nodes that we see are ranging between four, five millimeters to three centimeters, but the three centimeters will be caught by CT. Those are the substantive lymph nodes that we actually want to target to see if to change the treatment. Yes, the PSA is a predictor. We will have more positives, but I think it's more of the treatment plan rather than the PSA itself, so just to play it safe for the patient.

Philip Koo: Last question is: Can you give all the listeners advice with regard to how aggressive should we be with biopsies? You talked about the specificity and the specificity being pretty high, especially for extraprostatic disease. Should we still pursue biopsy of these lesions if they are accessible?

Dr. Savir-Baruch: This is a great question because that's reflecting I think the entire field of imaging. How much can you believe for a scan after it goes into approval, and now it's been read not by, some of them have more experience than other readers. I think that if it's isolated finding ... For example, someone's supposed to go into radiation therapy and all of a sudden, like the case I showed, you see aortocaval lymph node. You know you can't give radiation with the aortocaval lymph node, and those lymph nodes are reachable, so sometimes you will have to actually prove the scan right or wrong before you change management. But, that case of the aortocaval, it's abnormal by CT and it's abnormal by fluciclovine, so if this patient is ... the risk and benefit from biopsies too much, I think you should believe the scan, but if it's a five-millimeter lymph node in the retroperitoneal that takes update, that's the $1 million question. You can't really biopsy it. You can try, but what do you do?

I don't think there is a good question for this, but if I would need to choose one answer is that yes, you need to go and try to prove the scan if there will be, if that will influence a major treatment plan. For example, palliative or salvage, I don't know. It's a major question and we will have to work really hard to answer that. In the prospective analysis, I can tell you that when the exam was not FDA approved and we did it, part of the clinical trial, we had to prove our point, so we took patients' excisional biopsy and to CT guided biopsy and we have there I think 97% of histological confirmation, if I'm not wrong, don't catch me on that word, but it was very high. I think it's important to continue that, but after you have the diagnostic performance, I think it's a lot of pieces of a puzzle this should put together, and it's a case-based scenario.

Philip Koo: Thank you very much, and thank you again for your time. We hope you'll join us again on UroToday sometime in the near future.

Dr. Savir-Baruch: Thank you so much. Thank you.