Outcomes of Patients From PSMA Targeted Radionuclide Trials - Panagiotis Vlachostergios

August 8, 2020

Panagiotis Vlachostergios, MD, Ph.D., describes a study that he conducted reviewing the outcomes of patients who participated in PSMA targeted radionuclide trials over the past 18 years. Dr. Vlachostergios generated a visual score of PSMA expression to determine an association with response to PSMA-radio-nucleotide targeted therapies. Higher PSMA expressers showed an association with PSA response. Alicia Morgans, MD, MPH, asks how this PSMA visual score method could be implemented by nuclear medicine physicians in the clinic in a discussion with Dr. Vlachostergios.


Panagiotis Vlachostergios, MD, Ph.D., Medical Oncologist, Department of Hematology/Oncology, New York-Presbyterian Brooklyn Methodist Hospital, Weill Cornell Medicine, Brooklyn, New York, USA.

Alicia Morgans, MD, MPH Associate Professor of Medicine in the Division of Hematology/Oncology at the Northwestern University Feinberg School of Medicine in Chicago, Illinois.

Read the Full Video Transcript

Alicia Morgans: Hi. I'm excited to have here with me today, Dr. Panagiotis Vlachostergios, who is an Assistant Professor of Clinical Medicine at Weill Cornell. We are going to talk today about PSMA and some of the implications of that. Very exciting in the world of prostate cancer. Thank you so much for being here today.

Panagiotis Vlachostergios: Thank you for having me.

Alicia Morgans: Great.

Panagiotis Vlachostergios: So PSMA targeted radionuclide therapies are increasingly being tested. They're not yet approved, but there's a lot of research and clinical trials going on both at the early setting to look at PSMA imaging, but also at later settings in the metastatic castration-resistant prostate cancer patients. In terms of theranostics and really their role in therapeutics for these patients. At Cornell, with my mentor Scott Tagawa, we looked into all the patients that had previously participated in PSMA targeted radionuclide trials over the past 18 years. And that included patients who received either alpha emitters like actinium, which was a minority or beta emitters like lutetium or atrium or lutetium-J591 was like the majority of patients. And all of these patients were selected for their PSMA expression in terms of non-invasive radiographics.

So their semi-quantitative PSMA expression on imaging, we tried to look at their visual score. We generated a visual score of PSMA expression on either planner camera or gallium-PSMA PET scans and looked at whether there's any association between the degree of expression or the visual score and the response to therapies. We mostly used a PSA response of 30% or 50%. We did find that there was a significant association between the higher PSMA expressors, meaning those that had a visual score between two and four. And that was based on their expression compared to the liver in most cases, versus the patients that had a visual score of zero or one.

Interestingly, we did find about 13 patients that despite the fact that they had zero PSMA expression, zero visual score, those patients still had a PSA response. So our conclusion was we can certainly say that patients will have a lower visual score, will be less likely to respond, but it is likely that some minority of patients will still respond with a zero PSMA expression. So they should not be excluded from future trials.

Alicia Morgans: Well, and like many biomarkers, it sounds like it's not exactly perfect, but it certainly gives you an idea when you're trying to predict for a patient whether he may respond. Especially as we move these theranostics earlier in the treatment paradigm, I think that might be an important piece of what we do.

Panagiotis Vlachostergios: Absolutely. I mean within the same effort of trying to identify biomarkers now shifting to molecular biomarkers. We did look into a minority of the subset of these patients, about 53 patients to see whether there's any association between these DNA-damaging therapy, PSMA radionuclide therapy and molecular alterations in their homologous recombination repair genes. So we did identify that patients with a defective BRCA2 were more likely to respond to PSMA target brand-equipped therapies. The opposite was the case when we looked at patients that had androgen receptor amplification or MIC amplification. Those patients did worse.

Alicia Morgans: So important for us to think about as we're trying to personalize medicine. Good news for those patients, at least in this particular trial that had a BRCA2 alteration that they would have potentially a better response to a PSMA targeted therapy. Whereas those patients who have AR amplification or MIC alterations, they're having not necessarily as robust of a response, which is important I think as we think about escape mechanisms from PSA targeting and certainly approaches in the future, things that we need to kind of address in the future as we're moving this therapy forward. Any other messages that we should take from this exciting study that you've done?

Panagiotis Vlachostergios: I think it's promising data. We definitely need to confirm in prospective studies. The fact that PSMA zero visual score patients had some response. Look, a minority of them suggest that there's still maybe some micrometastatic disease that we're not able to detect. And probably the future is combining PSMA with other either FDA approved or experimental therapies to try to increase the yield of responses. But definitely in a biomarker-driven way and not blindly.

Alicia Morgans: And if somebody wanted to implement one of these visual scores in their own clinic, is it something that is software-based or is it clinician-based? How do you calculate that score?

Panagiotis Vlachostergios: It's more clinician based by two different assessments done. And basically it's a scale from zero to four either on a planner camera or on gallium-PET, comparing the expression of PSMA, the radiological question of PSMA with the avidity of the liver.

Alicia Morgans: So a nuclear medicine physician in the future might be able to implement this score as they're trying to help you prognosticate.

Panagiotis Vlachostergios: Absolutely.

Alicia Morgans: Well, very good. Thank you so much for sharing this study with us. Thank you for doing the work in the first place, and thank you for taking the time to talk with me.

Panagiotis Vlachostergios: Thank you very much.