Read the Full Video Transcript

Ian Davis: So I've been given the task of convincing you what the ideal sequence is of treatment once you have had treatment for metastatic hormone-sensitive prostate cancer with a doublet. And Karim will follow me, talking about triplet therapy. Here are my disclosures. None of these are relevant to today's talks.

But I will start off with a spoiler alert. I've been asked to tell you about the ideal sequence. There is no ideal sequence. But I think there are some less than ideal sequences, and I'll hope to talk to you a little bit about that.

Now, obviously being a little bit silly here, we have to go to the source of truth. So I went to NCCN. Surely, there's guidance in NCCN about what we do. And here it—oh. So it appears if you look at everything except the top left panel, then you've got a fair bit of choice and not much guidance here about what to do.

But hang on, there's a footnote. Perhaps there's guidance here about what we can do for sequencing. And oh, OK. So we can continue through all the treatment options listed. So I hope that's helpful for you. This is where NCCN has left us at the moment.

So I think this is a bit of a data-free zone. So I'm going to try and talk about some of the key principles to consider as you're thinking about selecting that first treatment and subsequent treatments for patients in this situation, remembering you can only ever give one treatment for metastatic hormone-sensitive prostate cancer, unless you change for toxicity, because after that, it is CRPC, and all the rules are then changed. So every subsequent treatment is for that.

And we know, as I'll show you later, that there's an overall diminishing benefit as you progress through various lines of therapy. So the goal is to make sure that we get the very best value that we can for this individual patient over the entire course of their life, not just for the next treatment, but for the treatments to follow as well, recognizing that there may be windows of opportunity—a treatment that you may be able to give now, you might not be able to give later.

So a clinical trial is talking about what's the probability of benefit here. We are talking now about what is the probability of benefit, given that you have already had some prior treatment. And what is maximum benefit? Are we looking for improved survival? Are we looking at delayed progression? Are we looking at quality of life? All of these things are important. So clearly, we need to use the most effective treatments and use them in the most effective sequence.

Now, this is a slide I put up at an APCCC meeting many years ago. I won't dwell on it, but it's just important to remember that there are situations where the order of treatment is important. If you have two treatments that are completely independent, it probably doesn't matter. You can use them in whichever order you like; you get the same benefit.

But if you have interdependent mechanisms of action, then it becomes very important what the order is, because if you use it in the wrong order, you don't have the opportunity of getting benefit from that other treatment. And I've listed some examples there on the bottom left of treatments that we know have got some interdependence. So I'll come back to that shortly.

So here are some of the options for CRPC. I won't go through this in detail. We heard fantastic presentations about this this morning. But the five at the top are clinical trials that were done when there really were no treatments for mHSPC beyond ADT at that time.

If you look at the timing of those studies, this was before CHAARTED and STAMPEDE came out with the docetaxel data. The ones at the bottom there—and we'll hear more about these later in the meeting, including this session—are in CRPC situations, where there have been various levels of treatment in the hormone-sensitive situation. And I've listed some of those there.

Now, the point of this slide is not for you to remember these figures or be tested on them. I've highlighted some of the points here. The point I want to make out of this slide, which I'll make available to anyone who wants it, is that the efficacy of these agents depends very much on the clinical state, and how the cancer has been selected through exposure to previous treatments.

So if you look at the docetaxel data, the abiraterone data, the enzalutamide data, the absolute magnitude and proportional benefit that you get in the hormone-sensitive state is much greater than in the CRPC state. So we have altered fundamentally the biology of the cancers so that these exact same agents, working through perhaps similar mechanisms of action, have different outcomes. And I've listed, just for the sake of completeness, the non-metastatic CRPC data on the right there.

So we have got various challenges in interpreting the literature. The HSPC protocols do not control the sequencing of post-trial therapies. We know from where these trials have been reported, and even in my case as co-chair of ANZUP, that we saw quite a bit of regional variability in the uptake of docetaxel and in access to subsequent therapies. We know that's the case.

And of course, we've got very diverse patient populations, as I'll show you in a moment, meaning that we cannot assume that each individual trial is representative of the broader population. We cannot directly compare the HSPC trials. And of course, we have to take into consideration patient-specific factors when making treatment selections.

So I've ordered these studies here in HSPC according to the time that their recruitment finished, from top to bottom. And you can see there that some of these studies were done in the time when there really were no options for prior treatment.

But big differences in the patient populations, in the proportion of patients with synchronous M1 disease, in the mix of high or low tumor burden—however that was defined—and in access to later use of any life-prolonging therapy, although as a general rule, the control groups had greater access to subsequent life-prolonging therapies, at least in the first post-trial therapy, than those in the active group, because they needed more of it.

So again, when we're talking about the challenges in interpreting the literature, can we learn anything from the older CRPC sequencing data, where these did not include patients who had intensified treatment for HSPC? And perhaps this is still relevant, given that we know that even today, depending on the series, more than half of patients still only receive ADT alone for HSPC, although we heard some encouraging news about that this morning.

The older reports of sequencing in CRPC suggest that docetaxel is less effective if used after an ARPI, although cabazitaxel does retain activity. It makes no sense at all to move from enzalutamide to abiraterone, although there's perhaps some justification for moving from abi to enza.

And in the so-called real-world situation, we know that prior treatment affects the probability of response to later treatment. The patient factors and accessibility are also important. And you can see in the bottom line there that we get diminishing yield as we move down the sequence of treatments.

So this is a study looking at US Medicare data. And it may be difficult for you to read. It is basically looking at the choices of therapy in the first, second, and third-line setting. And this is from US Medicare data. Large numbers of patients here.

And the messages I take out of this is that the ARPIs are very much the preferred first-line treatment, with docetaxel being used only in 16%. Even in second line, there's obviously a switch between ARPI and docetaxel—only 15%. And you only start to see docetaxel being used really in the third-line situation.

There are differences. Chemotherapy tends to be used more in those younger and higher risk. And I think we'll learn a lot from Iron Man in that situation and, perhaps, some differences in practice between Europe and the US, where Europeans are perhaps more likely to swap mechanisms of action.

So what are some of the specific situations? And warning that these are opinions, and you take them accordingly. It's based on evidence-based medicine. If you have a situation of no DDR or TMB high, perhaps the PARPi monotherapies or combinations are less likely to be helpful. And there's little evidence yet for immunotherapy there, although I'm still confident that that will change.

If you have a setting of early or rapid progression, perhaps it makes less sense to use treatments targeting the androgen receptor. Perhaps you should consider other mechanisms of action, such as taxanes, radioligand therapy, other targeted approaches.

The radioligand therapies are logical and now evidence-based choices for later lines of therapy and moving earlier in the treatment sequence, although we need to do that cautiously. And some of the combination approaches may help address tumor heterogeneity. And of course, there's still always time and appropriateness to put patients on clinical trials.

So what are some of the less than ideal approaches? I think it's not ideal if you do anything that removes a later useful option. And that might not be an obvious decision. Failure to use docetaxel early means that you might not get the opportunity to use it later as the patient deteriorates. Similar for cabazitaxel. It makes no sense to use a less effective treatment before a more effective one in therapy, which did not show an overall survival advantage for lutetium PSMA over cabazitaxel, though it showed superiority in other areas.

We should not, I think, be looking at ARPI switch before cabazitaxel. The trial showed that very clearly. We should not be using new approaches just because we can, just because they are there. And really, if you're thinking about it, if we are moving to a situation where doublet therapy for mHSPC is perhaps the default in many situations, docetaxel might then be the first-line treatment for CRPC. It frees up that effective treatment now for a later line of treatment. The corollary of that is listed there.

And finally, a challenge to all of us: I think that we need to call out and resist the tendency for clinical trials to be designed using ineffective or toxic placebo endpoints. And I'm looking squarely at ARPI change in that situation. And with that, I thank you for your attention.