Almudena Zapatero: ... finally, our nurse speaker about treatment directed for biochemical relapse is going to speak about PSA-directed is Neha Vapiwala from Philadelphia University, traditional oncologist.

Neha Vapiwala: My thank you. I've been assigned to team PSA, so I'm going to do my best to make that argument although I suspect we'll find common ground in the end.

Really this conversation it's about timing. When do we actually decide that there's a problem we want to solve? Do we decide with locally-advanced disease on initial staging and perhaps even a tri-modality approach of surgery with likely postoperative therapy? Do we decide based on high-risk features discovered on surgical pathology as you heard from Dr. Yamoah? Excellent conversation about the considerations there. Do we decide only when you have the above, but in conjunction with genomic classifier score that's concerning? And you'll hear more about that from Dr. Fang. Is it when there's evidence of abnormality either on imaging or on physical exam, or perhaps both? The argument that was made by Dr. James. Or perhaps the first sign of symptoms? And for that, I might ask, does provider and patient anxiety count as a symptom? I think it certainly does for many of us in our clinics and actually there's some unique work done in Germany on a data set of long-term prostate cancer survivors after prostatectomy and significant association of increased anxiety reported by patients in particular based on their higher PSA levels during follow-up.

I would argue, rather than being reactive, perhaps being proactive with early salvage radiation therapy. And now the question becomes how early is early enough? Is it the first sign of PSA rise with our ultra-sensitive assays and five, six significant figures? Is it with or without the high-risk genomic classifier score? Should we stick with the AUA definition? And then what about those patients whose PSAs never reach that magical zero? This is where we're going to focus our conversation today.

The Tendulkar update is multi-institutional data updating the original Stephenson nomogram. And for this, we can learn some lessons here. In particular, you can see that even in the patients with the lowest PSAs prior to receiving salvage radiation therapy, the incidence of distant metastasis is not zero. In fact, although these are low numbers, and in the number of patients with PSAs prior to Salvage RT less than 0.05 was actually quite small.

Nonetheless, you do see that there is, as I say, continued incidents of distant mets. And when you actually try to argue, "Well, where can I balance the risk benefit ratio?" You don't want to overtreat, you don't want to pull the trigger too soon, but on the other hand, where might we make a difference? And you can see a clear inflection point here, particularly in patients with Gleason 7 or 8 at time of pathology, where you might decide that's the sweet spot, that's the point at which I want to intervene. If you look at patients who present with a PSA over 2 by the time they receive salvage radiation or who present with pathologic Gleason score 9-10 as you heard in the last session shared by an audience member, that's where we might argue either it's too late for salvage radiation therapy, or you should be offering it adjuvantly from day one.

And there's of course, a multitude of publications on this. This is work by Fossati and colleagues that also argued that the earliest sign of PSA rise and the presence of these pathologic findings should argue for intervention. And this is just to drive the point home again from the Tendulkar paper as you can see the impact, excuse me, of pre RT PSA and how that relates in terms of contribution to risk.

We want to take the PSA seriously, and maybe it's not just the absolute value, we have to think about, as you heard earlier, bad kinetics and doubling time, short intervals in surgery and our clinical instincts that tell us when something doesn't look right.

We also have guidelines to turn to, and you can see here that ASTRO/AUA guidelines which go so far as to define post-op PSA recurrence and recommend monitoring and perhaps restaging, whereas the European guidelines do say that in fact, at a PSA level of less than 0.5 should be your target. They go so far as to actually define that cutoff for us.

And then, as you heard earlier, the adjuvant versus early salvage radiation therapy trials also had to define thresholds at which patients were to be offered Salvage RT. And we can turn to these as an idea in your clinic of where you might decide you want to set your cut point. And you can see here in red, the median PSAs of the patients in the studies at the time that they actually receive salvage radiation therapy. Again, somewhere around this area is probably where most of us feel the need to at least act.

How do we act? And in many cases, restaging is part of the workup. And this is where we all agree that better staging and better patient selection through imaging should yield better outcomes. It just makes sense because presumably you are tailoring your treatment more appropriately, and this inverse relationship that I just showed you of pre Salvage Radiation PSA and outcome might really just simply be a reflection of unappreciated, undertreated macroscopic disease. We know that conventional imaging, sensitivity, and specificity, particularly for nodal mets leaves a lot to be desired. We know that even with the new molecular imaging and PSMA PET/CT, that detection rates of macroscopic disease still remain lower than we would all like. And multitude of publications have again shown the PSA limitations. And you also heard earlier about relative-to-pelvic nodal sampling, a sensitivity of only 40% in most publications.

The danger, then, of waiting and seeing and hoping that you'll find something on staging is potentially seen in this relatively small series, but can be, I think, reproduced in many other cases. As your PSA increases, your chances of seeing disease on PET goes up. We know this, we've established this, but it's pretty remarkable for patients below and above a PSA of 1.

But what happens when you wait? What happens when you wait until the disease can be seen? And think about the fact that as you just heard from Dr. James, I think he actually made this point for me, that increased risk of spread beyond the salvage radiation therapy field should be taken seriously. Not only does it add more treatment that might be needed, but the toxicity that comes with it. That's exactly the argument why you should want to be proactive, not reactive.

What's the danger in believing only what we see or believing only what we don't see? Work by Johnny and colleagues. And you heard about the EMPIRE 1 study earlier, a very important, albeit small, phase 2-3 randomized clinical trial in which patients were randomized to treatment decisions for Salvage RT based on conventional imaging versus PET. The median PSA of patients in both arms was 0.34, and the three-year event free survival favoring PET, but on analysis, PSA still remained on multivariable analysis, a significant factor in event-free survival.

And furthermore, the decision-making, and this is where we really have to exercise caution, the decision-making pre-PET for prostate bed and pelvis was reduced to either prostate bed only in 10 patients or to know XRT, meaning disease was seen outside of the pelvis in four patients. And when we reduce or withhold definitive intent local therapy, we better feel confident that we're doing the right thing for those patients.

I will point out that the absence of PET-defined local involvement, we know is not proof of disease absence. We also know that the clinical significance of PET-detected disease that lacks a conventional imaging correlate either separate conventional imaging or the CT component of the PET has unknown clinical significance. And those four patients who did not receive XRT are not included in the survival analyses of these studies. We will have to see how they fare, but every individual patient matters.

Here's an example of a 74-year-old gentleman I saw who presented after prostatectomy T2, Gleason 7, positive margins, he'd been observed for five years, steadily rising PSA, negative conventional imaging, periodically, no treatment during this entire time, he had a PET for the first time at a PSA of 5.3 postop, which showed small nodes, no suspicious tracer uptake, thus no treatment. This is putting all your eggs in the imaging basket.

He presented to me eight months later, his PSA increased to 6.8. This is the first time I met him. I ordered a PyL PET and the nodes, interestingly, anatomically unchanged from the prior PET, and yet, now market tracer uptake. Is this tracer differences? Yes, we have studies that will be evaluating this. Is this just a reflection of that or is it that he was more curable eight months ago and we missed that window because we were just relying on the imaging and not what the PSA was telling us? I don't know. We won't know, not from that example.

How do we resolve this debate? PSA is a phenomenal biomarker. You cannot get around that. In the post prostatectomy setting, it is, but so is PET and it can be an incredibly important biomarker and we should think about how to use both. One way to evaluate this further is to think about patients who have biochemical recurrence, negative PET, receiving salvage radiation therapy alone, and look at the PSA response as a rough surrogate measure of potential subclinical, extra prosthetic disease.

And in this small series of 164 patients who received PSMA PET, of the 60 patients who had a negative PET, and mind you, they were controlled for Gleason score PSA, T-stage and margins, 45% who received radiation therapy alone had tremendous treatment response, 85%. The 55% who did not receive any treatment, nearly two thirds had continued PSA rise. We know that this can be useful. We also know that treatment response in PET positive patients after salvage radiation therapy alone, it makes a difference where it was positive. If it's fossa only, 81% response. As you heard, as the disease gets outside of the pelvis, we are doing the patient's a disservice by having waited.

We can absolutely incorporate PSA, but not overly rely on the imaging. It can be independently predictive of treatment response to SRT alone, but a negative result does predict a high response to SRT and we have to think about, in the end, what is the right answer to when for each patient?

Imaging detection threshold of which we know there are limitations does not always equal clinical relevance in all cases, you have to have your own threshold, you and the patient have to have that threshold. And use of salvage radiation therapy either with or without systemic therapy to reduce distant mets from suspected localized disease should be and can be based on clinical judgment, on clinical pathologic data, genomic classifier score if available, the PSA and what that biomarker is telling you, a relatively cheap biomarker I might add, with or without imaging evidence to support what you already believe or suspect or know. And at the end of the day, it's all about balancing that safety profile with the risk that you're taking. But I would argue that awaiting imaging evidence at the risk of distant metastatic development from untreated, undetected disease is too big a risk for most patients.

Many thanks.