Urologist Ashley Ross, MD, PhD leads a discussion of the benefits and limitations of PSMA-PET in patients with prostate cancer, as compared to older imaging modalities.
David Albala, MD: Ashley, give us a little insight on the limitations and clinical challenges you’ve had with these with the modalities we just talked about. We talked about PET [positron emission tomography] imaging. The largest experience with PET imaging has been in the recurrent setting. We’re starting to see it in early diagnosis, as Gary alluded to. What challenges have arisen when you try to get the studies done?
Ashley Ross, MD, PhD: I’ll start with the challenges once you get the study. For every patient I’m going to diagnose with prostate cancer, I try to get an MRI first. Similarly, because it’s the most accurate scan, any patient with unfavorable intermediate-risk cancer and above, if it’s newly diagnosed, I’m trying to get a PET-PSMA [prostate-specific membrane antigen] as my first imaging modality. That’s because when you look at studies, some of this experience is outside the United States. Compared with conventional imaging—we’ll get into terminology later—the accuracy of the PET-PSMA is higher both in sensitivity and specificity, particularly if you look at an all-comers study and not a pre-selected population.
Even though PET-PSMA is more accurate than the MDP [methylene diphosphonate] bone scans and the CT scans we’ve had, there are still limitations. The positive predictive value is pretty high. If you have a patient who has high-risk, presumptively clinically localized prostate cancer, and they have a lymph node show up in the pelvis on the PET-PSMA, the chance is pretty high that that lymph node is positive and that other lymph nodes in that area are positive. The negative predictive value isn’t always perfect because it still has a limitation. You usually have to have a lesion that’s 1 to 2 mm to see it on PET-PSMA.
Some cancers won’t express as much PSMA on the surface as others. For example, if I have a high-risk patient with a Gleason score of 4 and a PSA [prostate-specific antigen] of 7 ng/mL, and on T2a exam I feel a little nodule and the PET-PSMA is positive in the prostate only, I’m not going to omit the lymph node dissection. The negative predictive value in some of those presurgical trials was good, but it was still over 80%, so I’m not going to omit my lymph node dissection.
Similarly, in the biochemical recurrence setting, you’ll get a PET-PSMA. [Let’s say a man has an] adverse pathology: T3a disease and a positive margin. You’ve waited till the PSA was 2.3 ng/dL, and then you get a PET-PSMA. The PET-PSMA is clean, meaning there’s nothing in the prostate bed and nothing in the lymph nodes. I’m not telling that person they should omit salvage radiation therapy. There’s a limit of detection of these scans. Actually, the people who benefit the most from salvage radiation therapy are those who have a negative PET-PSMA in that setting or ones with prostate that’s only positive.
Gerald Andriole, MD: Can I ask Ashley or Gary about a recent patient that I had? I did biochemical recurrence after radiation therapy about 8 years ago. His rectal exam and MRI were very worrisome for a local recurrence. PSA was about 4 or 5 ng/dL. His PSMA-PET scan was completely negative in the prostate. I did a biopsy. Every core had Gleason 4+5 or 5+4. What do you think of that? What’s the effective radiation? Is this 1 of those terribly dedifferentiated cancers that didn’t make PSMA?
Ashley Ross, MD, PhD: I’ll put in 1 or 2 cents and let Gary take the rest. One thing I’ve seen in the postradiation setting—I don’t know what the course looked like—is that often you’ll see the density of cancer cells intercalated in the tissue is less, even if it’s a big mass less. You have these cells in this sea of stroma where there was dead stuff, and they’re not as densely clustered together. I’ve seen some of these false negatives in that setting myself. I believe that you’re not getting enough cellular density to get a bright signal. It would be interesting to do a fluciclovine or some other prostate-specific scan and see if that lights up where the PSMA doesn’t. Gary, what are your thoughts?
Gary Ulaner, MD, PhD, FACNM: Neuroendocrine-differentiated tumors are often not PSMA positive, so check the path if that was a neuroendocrine differentiated tumor. Ashley already gave my 2 cents: this is the category for which you may want to use a fluciclovine scan. For a patient you suspect or know has disease, if it turns out the patient doesn’t express PSMA, I’ll go to 1 of the other agents in our historical armamentarium.
David Albala, MD: In some respects, this is the ideal patient for this fluciclovine study because not all these tumors are PSMA sensitive. Some people think that’s a historical study, but in this case Gerry, that would have been a great study to order. You may find things because of the sensitivity in the prostate.
Gerald Andriole, MD: Now I know why you guys are on the panel. That’s exactly what we did.
David Albala, MD: That’s right.
Gerald Andriole, MD: The fluciclovine showed activity within the prostate and also showed bone metastases. It’s very sad.
David Albala, MD: You can’t rule out these older studies and imaging modalities because they clearly have a place. The pendulum has swung, but I don’t think it’s swung completely toward this change in imaging modality. That was a great case, Gerry. That’s how we’re going to use this in the future. That case illustrates it quite nicely.
Transcript edited for clarity.