When Low-Volume on Conventional Imaging Goes into High-Volume on Next-Generation Imaging in mHSPC - Treat like High-Volume APCCC 2022 Presentation - Ian Davis

September 5, 2022

At the 2022 Advanced Prostate Cancer Consensus Conference (APCCC 2022), Ian Davis presents when low-volume disease on conventional imaging becomes high-volume on next-generation imaging in mHSPC, we should treat these patients like high-volume patients.  This is a part of a point-counter-point discussion "When Low-Volume on Conventional Imaging Goes Into High-Volume on Next-Generation Imaging in mHSPC - Treat like Low-Volume"  presented by Karim Fizazi.  


Professor Ian Davis, MB, BS, Ph.D., FRAC, P FAChPM, Professor of Medicine, Monash University and Eastern Health, and Head of the Eastern Health Clinical School (EHCS)

Read the Full Video Transcript

Matthew Smith: Our next speaker is Ian Davis. We'll be seeing his presentation in pre-recorded form, and then Ian will join us live by Zoom for the Q and A in discussion. My bell will be ineffective.

Ian Davis: Hello everybody. My name is Ian Davis. I'm a medical oncologist at Monash University and Eastern Health in Melbourne. I'm also chair of ANZUP Cancer Trials Group. And I'm devastated that I can't be with you in person in Lagano, but thank you for the opportunity to contribute to what's effectively a rapid crossfire debate and it's always a joy to be speaking in the same session as Greg, even if I'm taking the opposite position to him.

So I've been asked to talk about this situation of treating patients who have apparent low volume metastatic hormone sensitive prostate cancer on conventional imaging, but a high volume pattern of disease on modern imaging, such as PSMA PET. And here are my disclosures. I'm member of several advisory boards, but I don't receive any personal remuneration for that.

So if we're going to treat these patients as high volume disease, what does that actually mean? Well, we know that we should be offering these patients Triplet therapy with androgen deprivation therapy plus effective androgen receptor inhibition, plus docetaxel and ARASENS and PEACE-1 have shown this, but with the caveats that in ARASENS only 86% of patients, not all, had synchronous in one disease. And by definition, all patients were deemed to require docetaxel. In Peace-1, all patients had 100% synchronous in one disease. So have no data in the metachronous setting, and the overall survival benefits so far has only been shown for high volume disease. So probably Triplet therapy, but at the very least Doublet therapy.

We also know that we should not be offering patients androgen deprivation therapy alone. And we certainly know that we shouldn't be doing this for patients with low volume disease on conventional imaging. We know that for Abiraterone, for Enzalutamide, for Apalutamide for Darolutamide. We've got plenty of studies here showing us that intensification of androgen deprivation therapy for low volume disease is appropriate.

There's a question open for these patients as to whether they should be receiving radiation to the primary. Of course, we do that for patients with low volume disease. Should we be doing it for these ones if we're treating them like high volume disease? I would argue perhaps that this is the subset of patients in this low volume group who are least likely to get a survival benefit from radiation to the primary. If the rationale for radiation is to try to prevent future metastasis. We've already shown in these patients that they probably already have metastasis, but that's open to contention, open to discussions.

So the next couple of slides have shown that the APCCC meeting previously, but I've modified them somewhat. So are we just talking about stage migration? Is this the Will Rogers phenomenon? So if you look at the diagrams there on the left, we've got conventional imaging on the far left and what we get with more modern imaging, like PSMA PET/CT on the right. And the patients with the blue dots were previously called low volume disease, but with the modern imaging, these are the ones that we're considering treating as high volume disease.

So they were low volume, but they were truly high volume. They had a lower burden of disease than the other high volume patients. They were previously worsening the prognosis for the low volume group and we've removed them and now they improve the prognosis for the high volume group. So we've diluted that. So that means that the outcomes for both subsets will do better by this very, very classification.

So is this simply stage migration? Well, yes, of course it is. This is the very definition of stage migration, but I would argue actually that what we're talking about here is personalized risk adapted therapy that we may in fact be identifying patients with different biology. These are patients who've got disease that is evident on PSMA PET compared to the ones who do not. So it's already behaving differently. And of course, we're using the best tools available to us, which we should always do.

 Is it lead-time bias? Well, let's pretend we've got a therapy that does not alter survival depending on when you give it. We've got a cancer, if you look at the bottom, that's detected in this patient and we'll say for the sake of argument that the five year survival is 0%. All patients are dead by three years. Now, we bring in a new modality like PSMA PET/CT that finds patients with high volume disease much earlier. And now, the five year survival is 100%. So this is how lead-time bias works, but of course the outcomes are no different for these patients. We define that.

Does this matter? Well, maybe it is lead-time bias, but I'm not sure it does matter. The patients on the bottom have already had the necessary treatment for high volume disease. They were just found later. So they've had the benefit from that. The patients on the top have also had a similar benefit. They just haven't lived longer than they would have otherwise. In other words, these patients need to have high volume type treatment anyway, at some point to extend their survival. Maybe they're just getting it a bit earlier than normal. So maybe it doesn't matter that we're finding it earlier.

The final point I'd like to make is that this is actually reminiscent of Pascal's Wager. And if you're not familiar with that, these Pascal on the right there, and he said that, "A rational person should live as though God exists and seek to believe in God." And the reason for that is that if God does not exist, you haven't lost very much. Perhaps, you haven't gone out and partied quite as much as you might have, but you've been a decent human through your life and there's really not much harm in this. But if God does exist, then you of course get infinite gains in heaven and you avoid going to hell.

So how does that apply to the topic at hand? Well, if it doesn't really matter whether we treat these patients as high volume disease, then maybe we're overtreating a few of them with Triplet therapy, but we're unlikely to be doing them significant harm, and really these patients would probably need to have the treatment eventually. They might just be getting it a little bit sooner than normal.

On the other hand, if treatment as high volume disease is important, then we maximize the probability of benefit because we're treated earlier when they're at the lowest volume with the least chance of inherent resistance. We maximize the probability that will actually give the treatment because the patients are younger and fitter with lower volume disease, more likely to tolerate it well, and it avoids the harm of not treating when we should have been treated much earlier in the setting.

So to conclude, modern imaging is here to stay, and we're going to be seeing these patients. Yes, this is stage migration, but I would argue that doesn't actually matter if we're defining a different disease state. Although, I would concede the caveat that perhaps the same rules might not apply if we're talking about metachronous versus synchronous disease.

This imaging modality in this approach now means we can better select for treatment. We can choose those who are most likely to benefit and target them for more effective therapy. And also perhaps choose those that are least likely to benefit, the ones who do not have apparent high volume disease on modern imaging. Maybe they are the ones who don't need to have intensification of therapy. Maybe we can even deintensify treatment. Lead-time bias doesn't really matter if the treatment is going to be needed anyway. And if you look at Pascal's Wager, you would have to conclude that a rational person would treat these patients as though they had high volume disease. So thank you very much for your attention.