Celestia Higano: Thank you, and thank you to the organizers for inviting us back to this fantastic meeting. It's amazing how much you can learn here that you didn't know about before, and I think that's the beauty of this gathering, and hearing it from our multidisciplinary colleagues.

So, I'm going to supposedly speak about the cons, particularly maybe some of the more important considerations to take into account when you wish to prescribe these medications for patients with non-metastatic CRPC, and what does this all mean in this era of molecular imaging. Here's my disclosures.

So, the standard definition of non-metastatic CRPC that we have used since we really realized this was an entity, which some of us realize when that was in the late '90s. So the standard definition was serial rising PSAs despite a castrate level of testosterone and no evidence of metastatic disease by conventional imaging. And by that we meant Technetium bone scan and CT scan of the chest, abdomen, and pelvis. As Dr. Hussein just mentioned, in these Phase III trials that gave us the SPA drugs, the inclusion criteria for those trials specifically included only those patients with high-risk, non-metastatic CRPC, and that was defined as PSA doubling time is less than or equal to 10 months and a PSA of two or greater.

And you know there is a reason for that. It had to do with statistical design and some of the things that Dr. Scher mentioned with respect to reaching the endpoint, and a lot of this was based on historical data, which I'm going to briefly show you again. But as Maha just said it is important to understand in the context of the approval of these three drugs that the FDA label does not restrict the use of these to this high-risk population.

Okay. And the reason I'm emphasizing that is I don't think that people in practice, I mean this kind of excludes this room, but that people in practice really understand this and they, I think perhaps erroneously think that this data means we should treat all patients with non-metastatic CRPC with these drugs. And I'm not saying we should or we shouldn't, but I think that we have to think twice about treating that population.

Actually the whole issue of PSA doubling time came up even before the denosumab trial that Howard mentioned. It came up really in the failed trial of looking at Zometa® in a non-metastatic CRPC setting to prevent or delay bone metastases. And so, Matthew and some of the rest of us looked at the placebo arm of that trial really in order to better understand the natural history of patients with non-metastatic CRPC. Because at that point in time we really didn't know what to expect, and that was part of the problem. The trial was underpowered and there were other issues.

But so, the graph on the left was the very first indication of what to expect of this population of patients based on PSA doubling time. And then Howard showed you the subsequent probably better-powered study where it really showed that inflection point at the PSA doubling time of around 10 months. And then in a non-trial setting, this is the graph on the far right is from a 440 patient VA study that looked at patients in the VA system with documented non-metastatic CRPC, and looked at the outcome of their PSA doubling times with respect to all-cause death. And it's very easy to see that those with a very short doubling time of less than three months did significantly worse than the other groups and so on and so forth. So, not only in clinical trials, but also even in a VA practice this thinking applies.

So, we've just talked about the importance of metastasis-free survival. And I don't think anybody disputes the findings of these three trials (SPARTAN, PROSPER, ARAMIS). I mean this is the first time this magnitude of difference has been seen in trials in this population. I mean we've tried before, there was a little hint with denosumab, but this is the very first time. And I don't think we can underestimate that this is a very substantial difference and it's a meaningful difference. So, but even without the OS data, which we don't have yet, I think the difference in delaying time to metastasis is definitely meaningful to patients. But you know, there is a bit of a proviso there, and that is as long as they do not experience substantial toxicity and they can afford the price, which in the US is sometimes as high as $10,000 per month. Now there are assistance programs and so forth, but these are some of the issues associated with treating these patients.

The other thing, so with respect to the toxicities, we know that there is toxicities and these may be more important when it comes to this group of patients who don't have metastatic disease. At least they don't have symptomatic metastatic disease, and they feel fine otherwise. So, if a drug starts making them feel worse, that becomes a lot more important issue than patients with metastatic disease. So, especially we know about the fatigue. But especially wanted to highlight falls and fractures, because especially in an elderly population, falls can kill you and they do kill you and we know that. So, I really think especially when we're trying to decide whether to treat, which drug to use, et cetera, we really have to take this into account.

Now, it is impossible to, it's not proper really to compare across trials the differences in toxicities but, thank you. But basically there is no head to head comparison regarding the toxicities and tolerance, the impact on quality of life and other functional measures has not been compared. And then how about other underlying co-morbidities which are very important and these are just listed here, but we don't know between drugs what the differences might be there.

There's, so if you just ask yourself how many people are there with this high risk, non-metastatic CRPC? And it really varies. It depends on what series you look at. It might be as low as 40%, it might be as high as 60%, but that means there's the opposite of low-risk patients. So, treating all patients with non-metastatic CRPC could expose a significant portion of patients to longer durations of therapy and more problems with toxicity.

So again, I'm referencing the Fendler paper and I thank that group for allowing us to look at this data, but I think this is starting to bring up some questions. Clearly PET scanning will identify more patients with non-metastatic disease who have early M1 disease. Some of the questions should M1 non-metastatic CRPC patients be treated with therapy for metastatic CRPC, or should they be treated with the second generation ARTs? We don't have data and I'm almost done. And can PSMA-PET imaging with or without PSA doubling time be used to identify patients who could delay systemic therapy and perhaps be treated with SBRT or other salvage approaches.

So, what are the cons of using second line ARTs for non-metastatic CRPC? There's definitely toxicity at individual patient level that we don't yet understand. We don't know who will have it and how reversible it is. The longterm effect on the natural history of metastatic CRPC is unknown if patients are so treated in earlier disease. So, will the aggressive disease phenotype develop earlier in metastatic CRPC? Treating men with non-metastatic disease and longer PSA doubling times I believe has a significant risk of doing more harm than good, and a safe choice of agent and monitoring these agents requires urologists to practice internal medicine.

And finally, there can be a significant financial toxicity as I mentioned earlier. So, going forward I realize that these three trials are strongly positive and that these agents should be used in appropriate men, but I do not believe that all non-metastatic CRPC patients are suitable candidates for therapy. We need a better understanding of the post-market setting about the toxicities of these agents relative to each other. And are there genetic or pharmacogenetic or other biomarkers that can predict these patient's specific, sometimes rather significant toxicities?

And finally, I think we need to consider the role of conventional imaging in concert with molecular imaging to better characterize the different subtypes of heterogeneous group of patients with non-metastatic CRPC in clinical trials, and to determine the role of salvage approaches guided by PSMA-PET in non metastatic CRPC.

Thank you.