Read the Full Video Transcript

Ashish Kamat: Hello and welcome to UroToday's Bladder Cancer Center of Excellence. I'm Ashish Kamat, Professor of Urologic Oncology at MD Anderson Cancer Center, and it's a distinct pleasure to welcome Professor Alex Zlotta, a dear friend and an expert in bladder cancer and all things to do with oncology, who has been on multiple forums in recent years sharing with us his insights into various things to deal with oncology and bladder cancer.

But today we have Dr. Zlotta here who is going to talk to us about his recent publication, recent study, which has garnered at a lot of attention in the press, amongst people at various journal clubs, where he will present to us the data on the radical cystectomy versus trimodal therapy for muscle-invasive bladder cancer.

Today's session promises to be very informative, not just for practicing urologists and radiation oncologists or all those who are taking care of bladder cancer, but especially patients who for many, many years have been wondering what's going on with this field and why is bladder preservation something that people are not talking much about?

Alex, congratulations on this publication, congratulations on this team effort, and welcome and take it away.

Alexandre Zlotta: Thanks, Ashish. It's always a pleasure to be with you. We've been together many, many times. So as I said, it's a pleasure to be here and to present on behalf of the different teams who participated to this publication, Jason Efstathiou and the team in Mass General Boston and Sia Daneshmand at USC and all my colleagues, medical oncologists, radiation oncologists.

As we all know, the gold standard for muscle-invasive bladder cancer has been usually historically radical cystectomy, but we have seen a paradigm shift in the treatment of many cancer types towards organ preservation that combines a limited resection and chemoradiation. In muscle-invasive bladder cancer, it's called trimodality therapy. You perform an extensive resection of the tumor and then this is followed by radiotherapy with a radio-sensitizing chemotherapy.

TMT has emerged as the most robust sparing approach for muscle-invasive bladder cancer, but this is usually for select patients which have characteristics of unifocal tumors less than five centimeters, no extensive carcinoma in situ, minimal unilateral maximum hydronephrosis.

And although there has been a growing body of evidence about TMT, even including in guidelines, we have to accept that it's often not really performed. It's often reserved only for patients who are not candidates for surgery.

Unfortunately, prior randomized trials that have tried to compare bladder sparing approaches with radical cystectomy failed to accrue. The most famous one is the SPARE trial in England. And therefore, we were scratching our heads in the absence of any randomized trial and none being foreseen, how can we actually manage to compare the patients who would have been eligible for both options, radical cystectomy and TMT, in addition in multiple institutions, so we can increase the generalizability.

And so in this publication we included 722 patients T2-T4 clinical N0 that would've been eligible both for TMT and radical cystectomy that were treated in Mass General, Princess Margaret, Toronto General Hospital, or USC, Los Angeles.

The inclusion criteria were fairly strict. Tumors solitary less than seven centimeters, slightly above the five centimeter cutoff, no or unilateral hydronephrosis, no multifocal carcinoma in situ. And we used propensity scores using logistic regression, and we also incorporated them in propensity score matching using 3:1 matching and another well-established method in the absence of randomized study, which is IPTW inverse probability treatment weighting.

We had multiple endpoint, not only one, including of course our overall cause of bladder cancer specific survival. Our primary point of interest was actually net-free survival and importantly, because I really think this is important, it was an intent to treat. So patients who were treated with TMT but ultimately had to undergo a salva-radical cystectomy were accounted as TMT patients.
These are the multiple covariates that we use to match and balance patients, and we also performed several sensitivity analysis.
Death without failure was also considered as a competing risk in the estimates, which is also important.

Finally, and that was already the key message of the first slide here, is that when we looked at eligible patients that were operated by radical cystectomy at the contributing institutions, this was about 30%, so we could use 30% that filled the eligible D criteria could be matched.

Now, the classical Table 1 shows that before matching the groups were, as expected, different. But once you match it, you can see the two curves overlap. In a nutshell, out of the 1100 plus patients that were included in the propensity score matching, the median age was classical at 75% of men, 10% of hydronephrosis. And interesting, close to 60% of patients in both either received neoadjuvant or adjuvant chemotherapy, and 90% were clinical T2, so really a majority of clinical T2.

When we look at the outcome, whether it was the adjusted net-free survival, both with IPTW and propensity score matching, the cancer-specific survival, the disease-specific survival, there was absolutely no difference between the treatment arms.

There was a difference in overall survival, both IPTW and PSM, and I know there's been a lot of chatter, but the reality on the ground is that we did account the perioperative mortality into the overall survival that already accounted for 2.5% and one of the series in the first year of TMT didn't have anyone dying, which is probably honestly due to pure, pure chance.

We did quite a lot of sensitivity analysis and we did 1:1 matching, 2:1 matching no difference. We did actually looked at only T4 versus the T2. There was no difference, no group was underperforming. But more importantly, one of the key question was here, if you compare and you try to compare apples with apples, you take radical cystectomy that got neoadjuvant chemotherapy, so everyone got chemotherapy, versus TMT where, by definition, everyone got chemotherapy although at sensitizing levels, there was again no difference, MFS, CSS or DFS.

Another thing is that about 13% of patients treated by TMT ultimately required, because of a invasive recurrence, a cystectomy, and are accounted as I mentioned, into these equivalent results. And if you pull the trigger quickly enough and you follow those patients, there was no difference between those patients who required radical cystectomy and those who didn't.

Then another really logical question, was there any difference whether you operated in USC or Toronto or you treated by radiotherapy in Boston or Toronto? Absolutely not. That's the net-free survival cancer specific was exactly, exactly the same.
So as I mentioned, 38, 13% of patients, require TMT. So that means that patients who are treated need to know that about 10-15% ultimately despite our best will, will require a cystectomy, another 20% will have a non-muscle invasive recurrence and that really requires a lifelong and stringent follow up that patient may or may not want to follow.

Where I think I would like to bring your attention to is that surgery did really, really well. The final pathology in this cohort was pT3 and 44% and 26% of node positive. And actually, the cancer specific survival five year despite that, was still 83%. And if you see what was logically expected that when you have a pT2 without hydro, you do better than pT2 with hydro than pT3 was exactly observed in the cohort.

Speaking to the quality, I think, of the results is the number of nodes removed, 39. Set learner would say that doesn't matter now that he presented, but at least it speaks to certain degree of quality. The positive surgical margin only 1%, and the perioperative mortality of 2.5%, which is more than acceptable.

So to conclude, we do believe that this large contemporary multi-institution cohort who had been eligible for more approaches shows that basically these results are robust. We used multiple endpoints, we had two experience statisticians who initially, believe it or not, did it completely independently and blindly. We only merged them when we found exactly, exactly the same results using some differences, minute differences in methodologies that were reassuring for me and Jason and the group, that this was robust. We had multiple, as I said, sensitivity and analysis.

And I think this is really one point that I would like to make. By no means the study doesn't support surgery as an excellent option, it is an excellent option with 83% of cancer-specific survival. This is possibly even slightly better. What reassured me from that study is that I do think that surgery has improved over time, as a group, as a profession we've been doing. The consistency between center or release was speaking about the generalizability of those results.

Like any non-randomized study, there are limitations. Select patients, no one is talking about multifocal large tumors, non with fully resected. You have to have that in centers who are doing comfortably post radiation cystectomy. It requires a rigorous follow-up and some degree of infrastructure.

We also know that clinical staging is notoriously imperfect and we need to do better maybe with MRI in the future. Our need to follow up was slightly short of five years, so there's still a little bit a possibility. And at the end of the day, the best propensity matched cohort will never ever replace RCT. There's still confounders there, there's no way about it.

But as I said, I think that what the study shows, it provides the best evidence possible. I would never ever say the best evidence. It's a randomized study that does that. That probably for select patients TMT and RC are performing probably the same. The oncological outcomes were demonstrated using two different statistical method. And we honestly believe, and we have as a Wednesday clinic, VT clinic, every week since such a long time, Boston as well. It doesn't mean that cystectomy is not a great option, it's an excellent option. It just means that select patients who have the criteria that we described should at least be offered the possibility to discuss various option, and not only patients who are unfit for surgery. That's truly what we believe and I think that that study brings to the table.

Ashish Kamat: Thank you so much, Alex, for taking the time and going through your paper and all the salient points and you brought up very, very important points in your presentation, which saves me from having to ask you some of those and we can focus in a few questions.

So Alex, for the audience that is listening that are clinicians and trainees and people all over the world, could you explain to them a little bit about the methodology and why it makes your study so much more relevant and valid?

Alexandre Zlotta: The available evidence so far have been indirect comparisons. So nationwide registry wide comparisons, data hospitals and those kind of things, and all these analysis are very, very interesting. But we know that they're always fraught with some limitations.

You don't have granularity about the dose, you don't have a good granularity about the sizes. In many of the studies in the past, even the percentage of salvage cystectomy was really not mentioned and sometimes was quite low, 2%, 3%, which means that many people were not offered. And some studies have shown as an improved survival radical cystectomy, but these were registry based studies.

The others that were directly comparing the two, to the best of my knowledge, were fairly smaller studies, limited number of patients and very rarely multi-institutional where you can increase the generalizability.

And finally, to the best of my knowledge, we are the only study where we use two different statistical methods, which are well accepted to balance groups in the absence of randomized. And that concur with exactly the same results. That's what I think brings a little bit of additional data compared to what is existing.

Ashish Kamat: And again, those are all the strengths of the study that I think are very important. And you alluded to the fact that of course a randomized study would be the absolute level one evidence, but we know that in this field, people have tried and people have failed with randomized studies. So we do need this sort of evidence because, I'm sure you too, we've talked about this offline many times, we're a little bit disappointed that there's not more open discussion with our patients about the use of bladder sparing therapies. Radical cystectomy is great treatment, but bladder sparing is a good treatment too.

And with this data that's come out and with your expertise and understanding of the field, how would you counsel a patient or how would you recommend that someone that's seeing a bladder cancer patient counsels him or her while sitting in the office?

Alexandre Zlotta: So I think the way that our clinic works, and I know that Jason Efstathiou, as you know, who has truly been co-leading the study with me. We have a clinic where patients come in, then we together with the radiation oncologist, the medical oncologist, we're looking into our chart, we review the pathology, we review the imaging. We scope patients together and we come up with a treatment plan and we decide whether patients are eligible for one, eligible for both, eligible for something else.

And then after having decided as a group, whoever patients say eligible for radical cystectomy of TMT would come in the room where the patients are usually anxiously, logically waiting for an answer. And with the families, and we have a one-stop shop so to speak, where we are going to propose both options, highlight the pros and the cons.

For instance, if we say TMT, we will mention some of the side effects. The fact that 10-15% ultimately will require a radical cystectomy, the lifelong stringent follow up every three months or two years. And that some people may or may not want to do. The fact as well that you can do a neobladder upfront, but after radiotherapy, although technically it's feasible, we have to face it, the outcome will be worse in terms of having an irradiated pelvis.

And so I think these are things that we're trying to guide and then depending on some of the specifics, our colleagues from radiotherapy will highlight this and this. And after having this extensive discussion, we usually leave the patient decide with the family and then they come up with a decision.

I have to say, honestly speaking, that most of the time when patients are offered the possibility to keep their bladder, they usually take that. There is a small percentage who would prefer probably a neobladder upfront. And if they know that they face, if there's a 10-15% failure and they will prefer then the neobladder. These are the kind of discussion that we have.

Ashish Kamat: Great points and great practical points that everyone should be considering when dealing with these sorts of discussions we have with our patients.

Are there certain contraindications or certain tumor characteristics that you would say, based on not just this paper, but of course your vast experience, you would say, "Well, let's not consider this patient for trimodal therapy."

Alexandre Zlotta: The first thing is that in order to keep a bladder, your bladder has to function. So if someone has a crippled bladder and then truly already goes every five minutes and then has... Those kind of inflamed bladder with women without CIS, it's obvious that there's no point in keeping a bladder that is not functioning. Radiotherapy will not improve that. That's the first point.

The second point is that if you have multifocal CIS, which is a surrogate for genetic instability in the bladder where the chances of having another tumor popping out in the near future is high, again, that's not a great...

We don't knows the sweet spot, how far we can push it. We have been possibly a little bit more conservative than many people in England. We really restrict it to one tumor and to some certain size. I know that from Nick James and any other more liberal in the inclusion criteria, which respectfully bring to the table, that I do think that some of the results seem to be slightly better and therefore, I think, the next frontier will be what's a sweet spot? Is it one tumor? Can we push it to something which is fully resected or do we keep it at seven centimeter? That's not crystal clear.

We know that when patient have a majority, if not pure variant histology, I think it's the unknown territory. And that if you have a variant among the high grade, they do reasonably the same. Jason's group of Jan Cologne just showed it. So these are the kind of features that we keep in our decision making. It's obvious that when people have five tumors, low grade, high grade, CIS and/or bilateral obstruction or huge T3, where you perfectly know that there's no way that you're going to be resecting those tumors. These are just not great candidates for TMT.

Ashish Kamat: Yeah, no, again, I agree with you completely. I've chatted with Nick James and Annette multiple times and for some reason it seems in the UK they don't even need a TURBT. Those patients do phenomenally well with radiation therapy. But here in Houston, and I believe in Toronto, we do believe that at least debulking as much as you can and selecting the right candidate is still at least the way we would counsel our patients. So I'm glad you brought that up.

Alexandre Zlotta: And I completely agree. I can tell you, just going along what you say that, like you, we receive a reasonable amount of course of referral from outside and then anytime when there's microscopic or doubt of tumor left, Peter Chung, Alejandro Berlin, Srinivas Raman in the past, of course Rob Bristow would ask us to go back and resect, because the data show that if you fully resect you do better. So I agree with you, I'm a little bit skeptical that leaving those large tumor in place is the way to go.

Ashish Kamat: Alex, again, we could chat on this topic for a long, long time, but in the interest of time, let me hand you the stage back and then any closing remarks for the audience before we wrap up?

Alexandre Zlotta: First of all, I think it's important, as you mentioned, for patients, families, physicians, to know that there's another option out there. I think the next frontier probably will be how to predict pT0. So you basically don't even have to give anything in addition, reserving, maybe dry modality for people who have a small recurrence. And I think we're pushing the envelope more and more. We're not near that, but I think if at least the 20-30%, maybe slightly more patients who can be offered both options were offered and not denied and people would say, "You can't be operated. So basically nothing happens." It would be a giant leap forward for our professional and for patients.

Ashish Kamat: Once again, Alex, thank you for taking the time. Always a pleasure to chat with you and learn from you and this was great. So thanks again.

Alexandre Zlotta: Likewise. Thanks, Ashish. See you soon.