Read the Full Video Transcript

Sam Chang: Hello. My name is Sam Chang. I'm a urologist at Vanderbilt University Medical Center, and we have the honor of having Professor Robert Huddart, who is the lead investigator of a RAIDER II trial, which focuses actually on adaptive radiotherapy for patients. And we're quite fortunate to have Professor Huddart, who is from the Royal Marsden in the UK, and we are going to ask him to actually give us an update based upon a recent publication from late 2024, looking at the phase II results. So Professor Huddart, thank you for spending some time with us, and we look forward to your presentation.

Robert Huddart: Thank you. Thank you very much for that introduction. My name is Robert Huddart. I am a clinician running clinical trials based at the Royal Marsden Hospital and the Institute of Cancer Research. This trial was run by the Institute for Cancer Research, Clinical Trials Unit, with which you have worked very closely, along with investigators from Australia and New Zealand through the TROG group.

And what we were addressing in this trial was how best to use radiotherapy as a way of preserving the bladder in patients with muscle-invasive bladder cancer. And over recent years, there have been quite a lot of technical developments that allow us to image the patient each day to see the actual bladder in situ. And we've realized that the bladder moves quite a lot, and it changes in shape. And so techniques have been developed of what we call adaptive radiotherapy, where you make a compensation for changes in the shape and position of the bladder on a daily basis.

And one of the common ways of doing that is to make a series of plans and then choosing the best one each day—the so-called plan of the day. And what we hypothesized in this trial was that by doing so, we can make the treatment more accurate. And that would also allow us to be able to see where the tumor bed was and to boost that area with a higher dose of radiation. And by making the treatment more accurate, by limiting the volume of the high-dose volume, that would allow us to dose escalate.

And we did a pilot study, which showed at the Royal Marsden—that we were successfully able to dose escalate patients. So we designed this study to test this hypothesis, and we particularly wanted to ensure that we could dose escalate without increased toxicity. And so we designed a study where we have three arms. We have a standard arm giving radiotherapy the way it’s normally given using a whole bladder, and then we had an arm where we were doing a tumor-focused boost, treating the bladder to about two-thirds of the dose and then the main dose given to the tumor bed—what we call the SART Arm or Standard Dose Adaptive Arm. And then with an arm, which we did the same as the SART Arm, but we increased the dose.

So we increased the dose by about 15% from 74 Gray to 70 Gray—32 fractions—or from 55 Gray to 20, 60 Gray, 20 fractions. The arms were randomized on a 1-1-2. So half the patients had dose escalation. And we had two dose cohorts, as I mentioned—a 20-fraction cohort and a 32-fraction cohort, which reflects what was common practice at the time across the UK, with about half the centers using 32 fractions and half using 20 fractions.

Subsequent to the trial, there has been some data suggesting that the 20-fraction arm might be the optimum way of delivering treatment. So we’re glad we included that within the trial. We randomized patients from across many centers across the UK and also from Australia and New Zealand. And this is a very large randomized phase II design, but it had 345 patients.

And what we wanted to do is exclude a significant rate of late radiation toxicity in the data. And for that, we calculated—we needed 57 evaluable patients in each fractionation cohort. So a total of 114 patients having high-dose radiotherapy. And they have to have follow-up for at least six months following the end of treatment. And actually, that requirement meant we increased the sample size to ensure we had that number of patients.

We also collected data on disease control outcomes, including local, regional, and invasive disease, bladder-intact disease-free survival, overall survival, and other toxicities. So what did we find? Before we go there, this is just to illustrate the technique that we used. So in the start of the DART Arm with two volumes, we had a whole bladder volume, and we had a volume which covered the tumor that was derived by the clinician with decent quality assurance to make sure everyone was doing the same thing.

And after dividing the actual volume, we then made a series of three plans. The margins are shown here, so we had a small, a medium, and a large. And you see the impact of increasing that volume size. And then before each treatment, trained clinicians and radiographers chose the most appropriate plan for that particular day. We had a training program to ensure that was done correctly, and we had an ongoing quality assurance program and actually made adjustments halfway through the trial when we found that things weren’t going exactly how we would ideally want them to.

So when we looked at actually what people chose, this is the adaptive plan selection across the trial, so in the two adaptive arms. So overall, 58% of treatments were given either with a small or large. The medium is closest to what you might consider to be a standard radiotherapy volume. And interestingly for me, only less than 2% of patients actually only used a single plan all the way through. And in fact, 70% of patients used all plans through the course of treatment. So actually, this showed there to be a need for this to optimize treatment. I should also say that in the standard arm, we did allow image guidance, and over 80% of patients had daily cone beam CT scans. So this is not a trial of image guidance and adaptation versus—it’s image-guided radiotherapy in the whole bladder versus adaptive arms in two other arms.

So what about our endpoints? Our primary endpoint was this one, which is the emergence of grade 3 toxicity. And in the DART Arm, we had one patient with radiotherapy-related grade 3 toxicity. It’s 1.7%. And that was well within the upper confidence limits that we set, which we tried to exclude 20%. So well within that, and a very low rate, and no evidence actually that the DART Arm had increased radiotherapy—grade 3 toxicity—compared to the other arms in the trial. And if we include all grade 3 toxicity, that then was 3.5% in the 32-fraction arm and 8.6% in the 20-fraction arm, and in both arms, well below the 20% grade 3 toxicity, and with the confidence limits below this level.

So we can exclude that level of toxicity as being prevalent. And these toxicity results compare very favorably with previously published work, including a previous trial we ran at the Royal Marsden in the UK called BC 2001, where these levels are significantly lower than that report. And then in terms of disease control, the trial was not powered to show a significant difference between the arms.

So what we have—this is bladder-intact event-free survival. So this is not relapsing and not having a cystectomy. And we can see that the two-year event rate was 66% in the whole bladder and SART Arm. We combined the standard-dose arms, and then the high-dose arm, it was about 5% higher at 71.7%, as a hazard ratio of 0.81, but not statistically significant. You can see the type of recurrence that we see.

Importantly, the number of pelvic-only recurrences was low across the trial in both arms, and that’s important because we’re making the treatment more focused. And this was a bladder-only targeted treatment. We didn’t increase the pelvic lymph nodes, and this is an area of great debate. There were patients who were—other patients who had pelvic nodal recurrences, but they were at the same time as other sorts of relapse, particularly metastatic relapse. And there was no real evidence that the targeted treatments had worse invasive recurrence rates than the whole bladder arm.

At the same time, there wasn’t a clear reduction in invasive recurrence rates, but we did have this small difference in these event-free rates. And importantly, across the trial at the current time, at the time of analysis, only 13 patients—that’s 4% of patients—had a cystectomy. Eleven of those were for disease recurrence. Two were because of people electing to do that earlier on.

Quality of life was pretty well maintained across groups. What we saw was some evidence of possibly small deterioration in quality of life at the end of treatment, as we’ve seen in previous studies. But over the subsequent two years, quality of life was pretty flat, and no evidence that this radiotherapy causes deterioration in symptoms. This particular one is a symptom severity scale related to urinary symptoms, but we have a number of other quality of life measurements, which show the same data, really.

So in conclusion, we concluded that adapt—98% of patients seem to benefit from adaptation. That this sort of radiotherapy has a very low rate of G3 toxicity. It was safe and feasible to deliver that higher dose. We couldn’t show that this was better, but the data was encouraging. A health-related quality of life is maintained, and the local control for this radiotherapy seemed to actually be better than we had. And the toxicity rates were better than we’d had in previous historical studies. And this is the reference for the full paper. Thank you very much.

Sam Chang: Professor Huddart, that was a great overview of the phase II trial. A couple of questions then to ask you. You talk about the vast majority of all the patients in the trial, and all three arms basically had image-guided therapy each day. How much extra work from either manpower or time regarding doing the adaptive plans? Is it—I’m trying to figure out the next steps in terms of actually integrating into a clinical practice.

Robert Huddart: Yeah.

Sam Chang: Is there much more effort in doing this?

Robert Huddart: So, the effort for doing it is there’s a bit of extra effort up front, because you’re manufacturing the three plans. When we started doing that when everything was much more manual, that was quite a bit more time. But now, you’ve got computers that push a button and can produce a plan in virtually no time. That is actually a lot less work than it used to be.

But you do have to do that planning and then the quality assurance of those plans, so a bit more effort in terms of the planning up front. When we did our pilot data—we don’t have very good data from the trial about how long it took, but it took on average—I think it was 12 or 13 minutes compared to a normal treatment slot of about 10 minutes. So you might need to do two treatment slots rather than one. It depends on your practice.

And some places in the UK, we’re in a very busy public health service, and there’s quite a lot of pressure on slots, so it’s a little bit longer, but certainly, you can do that within a—if you’ve got a 20-minute slot, you can do that. The actual scan—the cone beam CT scan you do takes one or two minutes extra, then you’ve got to match, and then it’s a question of experience and the computer system you have in terms of doing that matching.

And the selection usually is reasonably quick. So, we normally can do—you can do that within five minutes or so of the patient, and you’re quite right to point out this. It’s not only important for the actual department, but when you’re treating a patient with bladder cancer, the bladder is filling up with urine, so you can’t hang around for half an hour. If you hang around for half an hour, it causes problems.

We have a newer technique using MRI-guided radiotherapy, where we’re doing real-time adaptation. And that’s one of the things we’re fighting against all the time. It’s the actual—the bladder is not static, even if you take static pictures at the beginning. It’s not necessarily the same at the end. Having said that, we did quite a lot of pilot work—doing scans at the end of the treatment. And when we’ve selected using the protocol we used, the coverage was very good at the end. So it was like 98% coverage with that. So even with that allowance, I think using this technique is very reliable in terms of covering the bladder.

Sam Chang: And as you do this, do your surgeons—obviously, not for this trial, but how strongly do you urge your surgeons to do some type of marking of the sites of location of tumor? Be it with oil, be it with whatever other type of marker. You all routinely doing that in the US. There are definitely some centers that are really doing that more and more. We have not yet routinely. It just makes sense. It would only make this, I think, even more accurate. Tell me your thoughts regarding that.

Robert Huddart: When we were setting up the study, we were doing it quite a lot. And we did quite a lot of work, and we were—within the trial, we were trying to encourage people to do it. But I have to say the take-up was very poor, and the number of people who had it. So actually, these results are based essentially on people who haven’t had that done. The thing that works in our favor, compared to more historical series, is that increasingly people are having CT urograms, pre-TURBT.

So generally speaking, we have a pre-resection volume in which to use and identify patients. So in the distant past, before those days, you had a TURBT, and then someone had a scan. You were looking at, well, where was the tumor if it’s been completely resected? Those people didn’t have a complete resection, you can see the tumor, but those who haven’t—so the modern practice of doing a CT urogram worked in your favor for doing that.

And so actually, we’ve done very little of that in the recent past. We like MR when we could do it, because I think that gives you better location. But actually, in terms of identifying it, the CT urogram is probably the most useful tool. And then doing a visual match onto your planning CT scan. It’s quite difficult to do an actual match because the bladder isn’t always exactly the same as—

Sam Chang: Yeah, absolutely. Yeah, it’s not a static target for sure. And in terms of—the last question then is, OK, what next? Honestly, I had hoped for even better local control rates. Obviously it was not powered for that. This was more for, obviously, a safety adverse event type of thing. But you would hope with the boost that you’d get—and there was a slight difference, but there was still significant invasive recurrence of disease. What’s the next step in terms of surgery as we continue to try to maintain quality, decrease side effects, but perhaps even increase effectiveness?

Robert Huddart: Yeah. No. It’s a great question. We had to make a decision partway through the trial about do we take this into a full phase III, and it was—what was difficult was to make the case until you got the results. And then we got the results too late to take it into the phase III, and it’s difficult to see whether anyone is going to in the current climate want to do another 800,000-pound patient study in this setting. So this might be the best data we have on that. So it is a little bit disappointing for me that we haven’t got something more conclusive.

You probably—we can show you can take the dose. So maybe that sort of 60, 70 Gray, depending on your fractionation, might be an optimal dose if you’re doing a boost. And then if you’re going to go beyond that, it may well be the combination treatment. We saw quite a big effect for adding chemotherapy to radiotherapy alone in our previous trials, and it might be within the era of immunotherapy and other drugs that we’re now looking at, actually, more combination treatments.

And it may be that we’ll have a bigger impact by maintenance immunotherapy, for instance. There are some small trials related to radiotherapy. There’s some pharma studies, which have been recruiting or have recruited. So we’ll have to see whether they make a difference. There’s some interesting data of not doing radiotherapy in people with CRs with neoadjuvant therapy. But that’s the minority patient. So this is something that can be used for virtually anybody.

I should add about half the patients in the study had neoadjuvant chemotherapy. Seventy percent of the patients had concomitant therapy. So not everybody in the study had concomitant treatment for a variety of reasons. Again, that rate varied a lot from center to center.

Sam Chang: Sure. Sure.

Robert Huddart: Some instances, where virtually everybody got some sort of concomitant therapy. I’m pleased to say it included my center. But there are other centers where the rate of concomitant therapy was much lower. And so that’s one of the things we’re going to have a look at is actually how do these control rates vary according to what concomitant therapy people had, because there was a variation. The commonest concomitant therapy in the trial was 5-FU mitomycin.

But the subgroups of patients had gemcitabine, and a subgroup had the carbogen schedule, which is a hypoxic sensitizer. Very, very few patients had cisplatin. It reflected, I think, UK and Australasia practice that very few people had concomitant cisplatin, which I think is different to how it is in the US.

Sam Chang: Yes. No. I think looking at—getting more granular regarding the combination of systemic therapy, either neoadjuvant or concomitant, I think, will be quite telling as we integrate as surgeons and as, obviously, radiation oncologists with systemic therapies that are taking off in terms of clearly showing benefit for a significant number of patients and then being able to help determine who is actually responding, who not, to different types of various biomarkers, I think, will be incredibly important.

So, Professor Huddart, thank you so much for reference, your team’s efforts, and it’s really exciting work because there is no question that there is a movement towards, obviously, understanding better consolidated localized therapy, especially when you consider the possible, obviously, long-term side effects of bladder removal and having different options and being able to offer them to our patients will become—obviously, continually become more important for different patients in different stages of disease. So thank you for spending some time with us, and we look forward to future contributions.

Robert Huddart: Just to emphasize that one of the points of this is that as surgical techniques have developed over 20 years, what we’re doing now with radiotherapy, with this sort of adaptive, image-guided approach is much superior to when we started doing the BC 2001 trial in 2000, where we didn’t know what we were irradiating because we couldn’t see it on a daily basis.

So we’ve got much more precise treatment, and the control rates are better. The toxicity rates are better than they were. The addition of concomitant chemotherapy is there. And I think the promising—and there’s a decent chance to offer the majority of patients a bladder-preserving approach. There are patients who aren’t suitable, but there’s a large proportion of patients who could be offered this approach and preserve their bladder with no detriment to their overall survival.

Sam Chang: Yeah, absolutely important points. So again, Professor Huddart, thank you so much and look forward to seeing you soon.

Robert Huddart: OK. Thank you.