Advancing Prostate Cancer Surgery with PSMA Imaging and Fluorescent Guidance "Presentation" - Peter Carroll
February 7, 2024
At the 2024 UCSF-UCLA PSMA Conference, Peter Carroll highlights the shift in prostate cancer treatment from surgery for low-risk patients to active surveillance and targeted treatment for higher-risk groups, emphasizing the importance of novel imaging like PSMA PET in improving outcomes. He advocates for personalized treatment strategies based on detailed risk assessments and the potential benefits of surgery for certain high-risk patients, supported by UCSF's extensive data and collaborative research efforts.
![UCLAUCSF video logo stroke](/images/UCLAUCSF_video_logo_stroke.png)
Biographies:
Peter R. Carroll, MD, MPH, Ken and Donna Derr - Chevron Distinguished Professor, Taube Family Distinguished Professor - UCSF, Helen Diller Family Comprehensive Cancer Center, San Francisco, CA
![UCLAUCSF video logo stroke](/images/UCLAUCSF_video_logo_stroke.png)
Biographies:
Peter R. Carroll, MD, MPH, Ken and Donna Derr - Chevron Distinguished Professor, Taube Family Distinguished Professor - UCSF, Helen Diller Family Comprehensive Cancer Center, San Francisco, CA
Read the Full Video Transcript
Peter Carroll: Is it moving forward? I think the knowledge that lower-risk patients are candidates for active surveillance. So, 20 years ago, the ideal patient for surgery was a patient with low-grade cancer, organ-confined. We've learned now that those patients don't need to be treated initially. Higher-risk patients may benefit most from surgery. I'll show you the data and rationale for that. Novel imaging and molecular biomarkers have transformed the way we treat prostate cancer.
This is a patient sent to me who had had three or four negative biopsies, had a PSMA PET, identified the disease, and he underwent surgery. This is our data on active surveillance here at UCSF. Even though we're a very busy surgical center, one of the busiest, and we have a high number of patients undergoing radiation therapy, about one-third of the patients that I see undergo active surveillance. We probably have about 3000 men on active surveillance. I'm showing you here Kaplan-Meier upgrade-free and treatment-free survival for men with low-risk disease and favorable intermediate-risk disease. So, about half the men can avoid treatment at 7 to 10 years on surveillance. A big controversy recently about whether surveillance is durable; in about half the men, it may be very durable. So, these patients do quite well, and when they do progress, they tend to get treated with the same thing they would've been treated with initially here. The overall metastasis-free survival rate in our cohort at 10 to 15 years is 1%. So, it's a very safe form of treatment.
This just shows the CAPRA score. CAPRA is a well-validated system of risk assessment validated around the world, and increasingly over these years, we've gravitated from really giving surveillance to low-risk patients only now increasingly to intermediate-risk patients, usually favorable intermediate-risk patients, whom we think are good candidates for surveillance.
This is data we've looked at. We run a very large registry at UCSF called Capture. This is about 15 to 16,000 patients treated around the country, 40 sites, followed for many years, started probably in the early '90s by me and others. What I'm showing you here is just the CAPRA score on the bottom, 0, which means a low-risk patient, 10, a high-risk patient. I'm showing you prostate-specific mortality rates free from death in men treated with either surgery, in blue; radiation therapy, hormonal therapy, brachytherapy, active surveillance, and watchful waiting. We did this at 10 years, and we just updated it at 25 years recently published. We were agnostic to the outcome here. We controlled for everything: insurance status, demographics, T stage, PSA, use of hormonal therapy. What we found was that men with very high-risk disease seem to have a survival benefit with surgery. Again, that doesn't mean we won't operate on every man with high-risk disease, but at least it's changing our concept about doing surgery. 80% of the patients in the US diagnosed are on the left-hand side, so high-risk patients constitute about 20-25% of patients currently diagnosed in the US.
This is my surgical series over the last 20 years. You can see here in blue and at the bottom, these are low-risk patients; in green, intermediate-risk patients; and in lighter blue, high-risk patients. You go back 20 years, fast-forward, and there's been a big change in the way we manage prostate cancer with surgery at UCSF. We've stopped operating on low-risk patients. There's still a small number of low-risk patients, about 6%. They tend to be patients who are carrying mutations, abnormal genomics based on volume. So, it's a very rare patient who gets in here to surgery with low-risk disease. Increasingly, we're operating on men with higher-risk disease. The green there is mostly unfavorable intermediate-risk patients. So, we've gone from operating on low-risk patients to operating on high-risk patients because there may be a benefit for these patients. That doesn't mean we don't offer radiation. I think I'm the single biggest referrer to radiation oncology at UCSF.
But the strengths and limitations of surgery are that surgery, rather than other treatment options, may have value, although lymph node dissection described as extended is recommended by guidelines as the therapeutic benefit is not well established. At UCSF, when we looked at the impact of lymph node dissection, the number of nodes on outcomes, we could not clearly show a therapeutic benefit, and it is associated with increased morbidity. In men with high-risk cancer undergoing surgery, 50-40% will either have positive margins or miss metastatic disease in lymph nodes, both of which impact outcomes.
Now, do we abandon surgery? I don't think we abandon it, but I think we try and make it better, and I think better imaging may be critical to do this.
You'll see this slide over and over again. I won't go through PSMA any more here, but PSMA is an important biomarker, as you've already seen. Our thought is that intraoperative imaging, or I should say intraoperative tumor localization, may complete better local and regional cancer excision. That's the hypothesis. This is work done here showing that about 30-40% of the positive nodes sit outside the standard landing fields for either surgery or radiation therapy. Again, we've learned this from PSMA PET that we are missing nodes because we're relying on old templates that don't encompass all the nodal disease.
I borrowed this slide from Rob Reiter, The Future of Radical Prostatectomy? "Color-Coded" Surgery. We do a lot of robotic surgery. On the robot, we have an advanced Firefly camera, which we can turn on to allow for fluorescent imaging. This is a phase I study recently published looking at an intraoperative fluorophore, a PSMA fluorophore. This was a study done in tight collaboration with Intuitive Surgical, and the primary objective, of course, this was a phase I study, was to determine the safety of the drug. There were secondary endpoints to look at the sensitivity, specificity, margin rates, and all these patients had high-risk tumors. These are what we call high CAPRA scores that take into account all the variables which predict outcome.
These are some images here, intraoperative images. There's a prostate I'm holding up right there, and we're lighting that tumor up. It's very interesting to see this during surgery. Another image here. This is a patient, the prostate is out, that's the urethra down below here. We turned the camera on and identified this 2 to 3 millimeter thing that I would've missed using white light. I was very surprised to see this, that was biopsy-positive and excised. So, we're identifying residual disease, which we would've missed, I think, with conventional white-light imaging.
This is a patient with a lymph node here along a left external iliac artery, there. We light it up, and you'll see me, I'm removing that lymph node shortly here. It's rather remarkable to see this in action. Residual disease was seen. We had, in, I think, 24 patients, residual local disease or regional disease was identified in 33% and 9% of the patients, respectively. Now, this was a small phase I study, but it was very encouraging.
This is a patient I treated recently. We just started a phase II multi-site trial we're leading here, having included Mayo Clinic and Hopkins along with us. We're far along with our cohort here. This is a patient with a node sitting very low. For the surgeons in the room, this was sitting between the iliac artery and the bifurcation of the iliac vein, a very difficult place to get to. This was excised. If you look, our pathology analysis confirms target localization. So, you're seeing these whole mounts, and you're seeing the fact that these light up rather nicely in vitro.
What we found was the AEs were really blue-colored urine for 24 hours. There are really no side effects. We do use different doses here, and we found the 25 dose was best. A small study, but we were asked by the reviewers to show the negative and positive predictive values were 97% and 45% respectively for lymph nodes and 100% and 80% for residual/loco-regional disease. What we found was it was safe, well-tolerated, and had the potential to enable intraoperative imaging that could not be identified using white light. Again, we're in the middle of a phase II trial and hopefully, we'll see, this is looking at positive margin rates as well as adverse events, and we're well on our way to doing that. The nice thing about this is it could be readily deployed because on the robots you have this Firefly camera. So, in fact, if this is helpful, it could be deployed rather rapidly around the world.
This is a patient of mine. I just want to point out that this is a 62-year-old man who underwent surgery by me, had T3a disease, negative margins, 25 lymph nodes removed, all clear. He had a residual PSA of .58, repeat, 0.6, very, very favorable kinetics, and he was actually carrying a CHEK2 mutation. On him, again, this is a patient who would classically go to radiation therapy, but we did a PSMA PET, and he had this small lesion. This is not a ganglion. This turned out to be positive. A small ganglion in the pre-sacral space. This would've been very difficult to... That's interesting. Yeah, yeah, it is. A radiation therapist got a hold of that there.
But let me just say, that patient, as a surgeon going after that lymph node, you could not have seen it. That's an area we have a normal-sized lymph node, pre-sacral space, very difficult to identify. Actually, I couldn't get youth on a compassionate basis for the intraoperative imaging agent, so he flew to Germany, had radioguided surgery, open radioguided surgery, positive lymph node. He's at 3 years now, maintains an undetectable hypersensitive PSA. So now, in doing salvage surgery, I don't think I would want to do that if I didn't have any intraoperative localization. You'll have other people talk about this later.
I'm talking about fluorescent imaging, but I want to point out, there's a lot of other things out there. They're looking at radioguided surgery. I'm not sure that we can tell which one is better here. I do think some technology is going to be very important to do surgery better, and certainly in the salvage setting if we're going to do surgery, I think you need some guidance along with it.
In conclusion, I think studies range in size and use of either diagnosis or recurrence. All these imaging agents, such techniques better identify disease compared to conventional techniques. Resolution may vary. Benefits, I think, can be seen both with excision of the primary tumor and regional disease. I used to think it would be regional disease, but I think actually excision of the primary may be aided by localization. Oncologic benefit is not known with certainty. Even though I think these are very impressive results, we have to wait to complete the trials that show an oncologic benefit. Not so much for this group, but I always say that prostate cancer is a spectrum of disease and should be treated as such. I think PSMA refines our patient population very well. Treatment decisions should be made on cancer characteristics, I think PSMA helps us with that, and outcomes should be recorded very, very carefully.
I work with a great team here at UCSF, both surgeons and I want to point out the team from Intuitive. I think many are in this room here. It's been a very fruitful collaboration, and we're looking forward to the completion of this phase II trial, and I'm hoping it'll lead to a phase III trial. Thank you very much.
Peter Carroll: Is it moving forward? I think the knowledge that lower-risk patients are candidates for active surveillance. So, 20 years ago, the ideal patient for surgery was a patient with low-grade cancer, organ-confined. We've learned now that those patients don't need to be treated initially. Higher-risk patients may benefit most from surgery. I'll show you the data and rationale for that. Novel imaging and molecular biomarkers have transformed the way we treat prostate cancer.
This is a patient sent to me who had had three or four negative biopsies, had a PSMA PET, identified the disease, and he underwent surgery. This is our data on active surveillance here at UCSF. Even though we're a very busy surgical center, one of the busiest, and we have a high number of patients undergoing radiation therapy, about one-third of the patients that I see undergo active surveillance. We probably have about 3000 men on active surveillance. I'm showing you here Kaplan-Meier upgrade-free and treatment-free survival for men with low-risk disease and favorable intermediate-risk disease. So, about half the men can avoid treatment at 7 to 10 years on surveillance. A big controversy recently about whether surveillance is durable; in about half the men, it may be very durable. So, these patients do quite well, and when they do progress, they tend to get treated with the same thing they would've been treated with initially here. The overall metastasis-free survival rate in our cohort at 10 to 15 years is 1%. So, it's a very safe form of treatment.
This just shows the CAPRA score. CAPRA is a well-validated system of risk assessment validated around the world, and increasingly over these years, we've gravitated from really giving surveillance to low-risk patients only now increasingly to intermediate-risk patients, usually favorable intermediate-risk patients, whom we think are good candidates for surveillance.
This is data we've looked at. We run a very large registry at UCSF called Capture. This is about 15 to 16,000 patients treated around the country, 40 sites, followed for many years, started probably in the early '90s by me and others. What I'm showing you here is just the CAPRA score on the bottom, 0, which means a low-risk patient, 10, a high-risk patient. I'm showing you prostate-specific mortality rates free from death in men treated with either surgery, in blue; radiation therapy, hormonal therapy, brachytherapy, active surveillance, and watchful waiting. We did this at 10 years, and we just updated it at 25 years recently published. We were agnostic to the outcome here. We controlled for everything: insurance status, demographics, T stage, PSA, use of hormonal therapy. What we found was that men with very high-risk disease seem to have a survival benefit with surgery. Again, that doesn't mean we won't operate on every man with high-risk disease, but at least it's changing our concept about doing surgery. 80% of the patients in the US diagnosed are on the left-hand side, so high-risk patients constitute about 20-25% of patients currently diagnosed in the US.
This is my surgical series over the last 20 years. You can see here in blue and at the bottom, these are low-risk patients; in green, intermediate-risk patients; and in lighter blue, high-risk patients. You go back 20 years, fast-forward, and there's been a big change in the way we manage prostate cancer with surgery at UCSF. We've stopped operating on low-risk patients. There's still a small number of low-risk patients, about 6%. They tend to be patients who are carrying mutations, abnormal genomics based on volume. So, it's a very rare patient who gets in here to surgery with low-risk disease. Increasingly, we're operating on men with higher-risk disease. The green there is mostly unfavorable intermediate-risk patients. So, we've gone from operating on low-risk patients to operating on high-risk patients because there may be a benefit for these patients. That doesn't mean we don't offer radiation. I think I'm the single biggest referrer to radiation oncology at UCSF.
But the strengths and limitations of surgery are that surgery, rather than other treatment options, may have value, although lymph node dissection described as extended is recommended by guidelines as the therapeutic benefit is not well established. At UCSF, when we looked at the impact of lymph node dissection, the number of nodes on outcomes, we could not clearly show a therapeutic benefit, and it is associated with increased morbidity. In men with high-risk cancer undergoing surgery, 50-40% will either have positive margins or miss metastatic disease in lymph nodes, both of which impact outcomes.
Now, do we abandon surgery? I don't think we abandon it, but I think we try and make it better, and I think better imaging may be critical to do this.
You'll see this slide over and over again. I won't go through PSMA any more here, but PSMA is an important biomarker, as you've already seen. Our thought is that intraoperative imaging, or I should say intraoperative tumor localization, may complete better local and regional cancer excision. That's the hypothesis. This is work done here showing that about 30-40% of the positive nodes sit outside the standard landing fields for either surgery or radiation therapy. Again, we've learned this from PSMA PET that we are missing nodes because we're relying on old templates that don't encompass all the nodal disease.
I borrowed this slide from Rob Reiter, The Future of Radical Prostatectomy? "Color-Coded" Surgery. We do a lot of robotic surgery. On the robot, we have an advanced Firefly camera, which we can turn on to allow for fluorescent imaging. This is a phase I study recently published looking at an intraoperative fluorophore, a PSMA fluorophore. This was a study done in tight collaboration with Intuitive Surgical, and the primary objective, of course, this was a phase I study, was to determine the safety of the drug. There were secondary endpoints to look at the sensitivity, specificity, margin rates, and all these patients had high-risk tumors. These are what we call high CAPRA scores that take into account all the variables which predict outcome.
These are some images here, intraoperative images. There's a prostate I'm holding up right there, and we're lighting that tumor up. It's very interesting to see this during surgery. Another image here. This is a patient, the prostate is out, that's the urethra down below here. We turned the camera on and identified this 2 to 3 millimeter thing that I would've missed using white light. I was very surprised to see this, that was biopsy-positive and excised. So, we're identifying residual disease, which we would've missed, I think, with conventional white-light imaging.
This is a patient with a lymph node here along a left external iliac artery, there. We light it up, and you'll see me, I'm removing that lymph node shortly here. It's rather remarkable to see this in action. Residual disease was seen. We had, in, I think, 24 patients, residual local disease or regional disease was identified in 33% and 9% of the patients, respectively. Now, this was a small phase I study, but it was very encouraging.
This is a patient I treated recently. We just started a phase II multi-site trial we're leading here, having included Mayo Clinic and Hopkins along with us. We're far along with our cohort here. This is a patient with a node sitting very low. For the surgeons in the room, this was sitting between the iliac artery and the bifurcation of the iliac vein, a very difficult place to get to. This was excised. If you look, our pathology analysis confirms target localization. So, you're seeing these whole mounts, and you're seeing the fact that these light up rather nicely in vitro.
What we found was the AEs were really blue-colored urine for 24 hours. There are really no side effects. We do use different doses here, and we found the 25 dose was best. A small study, but we were asked by the reviewers to show the negative and positive predictive values were 97% and 45% respectively for lymph nodes and 100% and 80% for residual/loco-regional disease. What we found was it was safe, well-tolerated, and had the potential to enable intraoperative imaging that could not be identified using white light. Again, we're in the middle of a phase II trial and hopefully, we'll see, this is looking at positive margin rates as well as adverse events, and we're well on our way to doing that. The nice thing about this is it could be readily deployed because on the robots you have this Firefly camera. So, in fact, if this is helpful, it could be deployed rather rapidly around the world.
This is a patient of mine. I just want to point out that this is a 62-year-old man who underwent surgery by me, had T3a disease, negative margins, 25 lymph nodes removed, all clear. He had a residual PSA of .58, repeat, 0.6, very, very favorable kinetics, and he was actually carrying a CHEK2 mutation. On him, again, this is a patient who would classically go to radiation therapy, but we did a PSMA PET, and he had this small lesion. This is not a ganglion. This turned out to be positive. A small ganglion in the pre-sacral space. This would've been very difficult to... That's interesting. Yeah, yeah, it is. A radiation therapist got a hold of that there.
But let me just say, that patient, as a surgeon going after that lymph node, you could not have seen it. That's an area we have a normal-sized lymph node, pre-sacral space, very difficult to identify. Actually, I couldn't get youth on a compassionate basis for the intraoperative imaging agent, so he flew to Germany, had radioguided surgery, open radioguided surgery, positive lymph node. He's at 3 years now, maintains an undetectable hypersensitive PSA. So now, in doing salvage surgery, I don't think I would want to do that if I didn't have any intraoperative localization. You'll have other people talk about this later.
I'm talking about fluorescent imaging, but I want to point out, there's a lot of other things out there. They're looking at radioguided surgery. I'm not sure that we can tell which one is better here. I do think some technology is going to be very important to do surgery better, and certainly in the salvage setting if we're going to do surgery, I think you need some guidance along with it.
In conclusion, I think studies range in size and use of either diagnosis or recurrence. All these imaging agents, such techniques better identify disease compared to conventional techniques. Resolution may vary. Benefits, I think, can be seen both with excision of the primary tumor and regional disease. I used to think it would be regional disease, but I think actually excision of the primary may be aided by localization. Oncologic benefit is not known with certainty. Even though I think these are very impressive results, we have to wait to complete the trials that show an oncologic benefit. Not so much for this group, but I always say that prostate cancer is a spectrum of disease and should be treated as such. I think PSMA refines our patient population very well. Treatment decisions should be made on cancer characteristics, I think PSMA helps us with that, and outcomes should be recorded very, very carefully.
I work with a great team here at UCSF, both surgeons and I want to point out the team from Intuitive. I think many are in this room here. It's been a very fruitful collaboration, and we're looking forward to the completion of this phase II trial, and I'm hoping it'll lead to a phase III trial. Thank you very much.