We and others did studies such as this that showed the ability of PSMA PET to detect even very small amounts of disease at low PSA values. And we're just amazed and for sure thought this was going to be a game-changer for these patients. But several years on, we're beginning to see some troubling things that I think we need to come to grips with. First of all, remember that prostate cancer is a very long duration process. So the majority of men with biochemical recurrence actually do very well at a time course of seven to 10 years. 85 to 90% of those patients will do well either on observation or low-intensity ADT. Some patients progress inexorably to metastatic disease as defined by bone scan and CT. That's the data that we have. And there's evidence-based trigger for treatment, intense treatment at that point. But the problem is PSMA is coming in very early in this course. And many of these people will do very, very well, but many clinicians are alarmed by the fact that there's either a PSMA finding or that there's PSMA progression.
Let's just look at some numbers from, this is just one of many studies that looks at the long-term outcome of patients after radical prostatectomy. 21,000, almost 22,000 patients with radical prostatectomy, 11% BCR, 10-year overall survival, 70%. But remember, we need disease-specific survival. Among the BCR patients, 1,140 died, among those BCR deaths, 224 were from prostate cancer. So a minority from prostate cancer and the overall prostate cancer-specific overall survival, 98% at five years, 94% at 10 years. So these slides just say the same thing that with conventional imaging, we are detecting the disease later, but we have good data from multicenter trials that show that treatment at that point is efficacious. What we have with PET imaging is a diagnostic shift to the left, and we have negative data for treating patients at that stage, at least multicenter patient data, which is based again on CT and bone scan. Now, we don't have the data that treating patients in this intermediate group, PSMA-positive, conventional imaging-negative, is a benefit to those patients. And yet, we know that there are significant side effects to those treatments. So before PSMA PET, BCR was monitored with bone scan or CT based on the PSA kinetics.
Observation was considered appropriate. Treatment when done was often based on PSA doubling time and therapies were dose reduced often with intermittent treatment or holidays given. After PSMA PET, these findings are taken much more seriously by clinicians. BCR is detected, treatment is given immediately. PSADT is put off to the side sometimes. Therapy is more intense, and we don't know if there's an actual benefit to that. We know there are actual harms, and I don't need to belabor that. The lowly PSA doubling time, a very inexpensive test is extremely predictive in this setting, and we have many decades of data supporting the use of PSA doubling time for predicting outcomes. So one of the other aspects of this is BCR patient leads to a PSMA scan that's positive. Panic OMG, initiate earlier treatment and then follow PSMA looks great. PSMA scan gets better. PSAs go down. Patients do very, very well. So that when the next patient rolls around with BCR, you say, "I know exactly what I'm going to do. I'm going to treat."
But that is a non-virtuous cycle because many of these patients really don't need this treatment in the first place. So is it really a crashing dam that the dam is about to burst or is it more like a glacier that's moving quite slowly? And everything we know about the natural history of prostate cancer suggests that it's the latter. So, okay, I'm complaining. I'm whining. What can we do about this? So there are a number of things. I mean, the problem with this space of patients is that the disease doesn't progress very fast. So the trials are expensive and long, and the benefits are way in the future. So what about a prospective PSMA natural history study to see what happens to patients who don't get treated but have PSMA-positive scans? So we've opened such a trial. The inclusion criteria is a history of primary treatment for prostate cancer, PSA at least 0.5 with normal testosterone. And we've begun to study these patients. It accrues actually quite well because many patients are told they need treatment, but they understand that the benefits of treatment are uncertain. So they come to our study to hedge their bets. Here is a patient who shows some lymph nodes, a year later, some more lymph nodes, a year later, some more lymph nodes.
And he's like, "Are they insane at NCI for not treating this?" Well, the doubling time is very long in this patient and there are no symptoms. There's no necessarily good data to support treatment at this moment. Another case, lesions on the periphery of the liver. Oh my God, is this peritoneal metastasis? React, react, react. But look at this patient over time. It essentially doesn't change over a period of two or three years. So we don't have a ton of follow-up data at this point, 21 months, but we're gathering it. And here's what we have so far. So when the PSMA is only positive in the prostate gland, most of those patients are selecting monitoring without therapy, and we've had one patient progress to metastatic disease, so rate of 4.5%. When there's PSMA-positive findings beyond the prostate, as you can see, most of them are, again, not being treated. There's a 3.6% rate of metastatic disease.
Now, I would say, truly, I would rather these numbers be zero, but it does indicate that the vast majority of these patients are not progressing to metastatic disease at a rapid rate, and there's some happy medium between doing nothing and doing everything. So in conclusion, PSMA PET in BCR is leading to earlier, more intensive treatments without necessarily evidence of benefit. The risk of prostate cancer related death is very low in this population. The risk of adverse consequences from ADT/ARPI are very real. Conventional imaging and PSA doubling time are the only evidence-based metrics to guide therapy in this population. So we need more data in this space about the benefits and the risks of earlier treatments. And in that sense, I think we need to discourage the overuse of PSMA and the overuse of reaction to positive PSMAs.
And lastly, I think there are important lessons for the future of molecular imaging probes in general, because we have the dogma that the more we see, the better, but we have to view these findings in the context of the larger context of medicine where people are going to be reacting to what we're showing them. And I'm thinking about FAPI probes and all kinds of other probes that are being developed. So I began with an ancient quote. I'll end with a modern quote. Sometimes the bravest scan is no scan at all by ChatGPT. Thank you.