Read the Full Video Transcript

Ashish Kamat: Well, hello and welcome again 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 pleasure to have again with us Dr. Seth Lerner, who really needs no introduction. Seth is an exceptional urologic oncologist and educator. He has conducted numerous trials. And congratulations, Seth, on now being appointed chair of the SWOG GU committee.

But today, this trial that you spearheaded, led, and then presented, and the embargo was lifted at ASCO this year, has really garnered a lot of attention. And a lot of the UroToday audience from overseas who can't attend our US conferences really are looking forward to learning from you and hearing about that data. So with that said, welcome once again, and take it away.

Seth Lerner: Great, thanks a lot. So what I'm presenting is the results of SWOG S1011. So this was the trial that was led by SWOG, but we conducted it in sites across the NCTN. So this was a Phase III surgical trial to evaluate the anatomic extent of a node dissection performed at the time of radical cystectomy for curable muscle invasive bladder cancer. Clearly a team effort, and lots of shout-outs to leadership in SWOG. This was really Ian Thompson's idea a long time ago, and I always like to give credit to Bernie Bochner who actually had the idea first, and building on Bernie's thoughts about this we have conducted it, now we report the final results of this clinical trial.

The primary objective of the trial was to compare disease-free survival in patients undergoing cystectomy, comparing an extended node dissection and I'll just go over that briefly, compared to a cystectomy and bilateral standard pelvic lymphadenectomy. We defined DFS as the first documentation of relapse or recurrence or death due to any cause. We had a number of secondary objectives, and I'll be reporting where we reported on overall survival and some aspects of postoperative morbidity, namely serious adverse events and mortality.

This is the study design. So again, patients with curable, muscle-invasive bladder cancer. They could have clinical N1 or N2 disease, but, as the audience knows, that's confined to the true pelvis, which would be within the standard template. We had a number of stratification factors. Most notably patients were allowed to have neoadjuvant chemotherapy, and for those that did, we stratified between cisplatin, carboplatin, other regimens versus no neoadjuvant chemotherapy stratified by clinical stage T2 versus T3, T4a and performance status 0-1 versus 2.

So all patients got a standard node dissection and then it was an intraoperative randomization. So this really required the surgeons to do a thorough exploration to rule out disease in the extended template. And if there was a suspicion of positive nodes, they were required to do a frozen section and that frozen section had to be negative where the patient could not be randomized. They also had to rule out T4b which would be unresectable disease. And then postoperatively if they had pathologic T3, T4, or any positive nodes, there was a recommendation but not a requirement for adjuvant chemotherapy. The assumptions were based on a review, really an extensive review, of the literature of about 8,000 patients and an estimate of 55% three-year DFS in the control arm or the standard arm, and an 85% power to detect a 10-12% improvement in three-year DFS in the extended arm.

Why was that? If we were going to look for a smaller effect size, this was going to be over a thousand-patient trial and we really didn't think, at the time, that we could accomplish that. So this resulted in a sample size you can see there, 282 per arm, 10% intraoperative in eligibility, and a total accrual target of about 620 patients.

This is the CONSORT diagram. You'll notice that the first patient was enrolled in 2011, the last in March of 2017. We finished this trial on time as we had described it and designed it in the protocol. A total of 650 patients were enrolled and 618 intraoperatively randomized, 315 in the standard arm, 303 in the extended arm, and a number of ineligibilities, actually a small number of ineligible patients, resulting in 300 in the intent to treat in the standard arm, and 292 in the extended arm and a slightly smaller safety core cohort, which I'll describe shortly. And the median follow-up was six-plus years in both arms. So good, good, long-term follow-up.
So these are the pre-randomization clinical covariates, and you'll recognize that the majority of patients were T2. This is quite common for a cystectomy series. About a quarter of the patients had hydronephrosis, which is typically associated with more locally advanced disease, and the expected amount of variant histology. I should remind the audience that this is predominant urothelial.

Now what's interesting is that 57% of the patients in both arms had neoadjuvant chemotherapy. Nobody had come close to reporting that at the time that we did this study. So this was really an extraordinary effort by the multidisciplinary teams to incorporate Level 1 evidence into this clinical trial. 88% of the patients had, or 87% of the patients had, cisplatin-based neoadjuvant chemotherapy.

So these are the pathologic tumor and node stage based upon the cystectomy pathology. A slightly higher than expected number of patients were pT0 N0, I think normally you'd expect maybe about 15%. But the positive node rate was about 25%. Very similar between the two arms. Very similar pN1 or pN2. Again, this is confined to the true pelvis. Remember that all patients had a standard node dissection, so you would expect this to be similar. And then N3 disease, which is in the common iliac lymph nodes, you should see only in the extended template. These four patients had an extended node dissection, though they were randomized to the standard arm. The total number of lymph nodes removed, you would expect to be much higher in the extended arm, 39 versus 24, a slightly higher number of positive nodes as you might expect. And then these numbers here should be similar between the two arms. And then the estimated blood loss was slightly higher in the extended arm compared to the standard arm.

This is the take home message. So there was no difference in disease-free survival. I think anybody can tell that these are overlapping curves and non-significant hazard ratio. And again, I want to just call everybody's attention that the median follow-up was six years. So this is the timeframe within which you're going to see any recurrence events. And obviously these curves continued to go down as far as overall survival, which I'll show in the next slide. And this too was non-significant. So no difference in the primary endpoint of DFS and no difference in the secondary endpoint of overall survival.

We did a comparison according to the pre-specified stratification factors and pathologic stage, really no differences in DFS with receipt of neoadjuvant chemotherapy. Clinical stage, there was a, and I have to caution that this is hypothesis generating, that there looked like there was a difference in patients who had locally advanced but node negative disease. As far as DFS goes we saw something quite similar for overall survival, but this is hypothesis generating.

And then the adverse events. I've described here, grade 3, grade 4, obviously grade 5, which are fatal events. And you can see a slightly higher rate of sepsis in the extended arm, a slightly higher rate of grade 3, grade 4 wound complications, thrombotic events. And then the 30-day mortality, which is this first line here, 90-day mortality, both of those were slightly higher in the extended arm versus the standard arm 30 and 90 days. I will tell you that when looking at 90 days, there were some patients in here who progressed. These are all-cause mortality. We simply are not able to get a precise enough picture of attribution to the node dissection. So we just, essentially, took all comers as far as grade 3, grade 4, and 5 adverse events.

So in conclusion, we find that patients undergoing a cystectomy and an extended node dissection, they did have increased node yield, very similar pathologic tumor stage. There was a higher median number of positive nodes among patients that were node positive. There simply was no indication of a benefit in DFS or OS for the extended arm. There was a signal of slightly higher morbidity and mortality. And I didn't present the data, but the audience is likely familiar with a similar study that was performed in Germany, under the leadership of Juergen Gschwend, that also did not show a benefit to the extended arm. And I think that these two trials together really established that the standard of care is a bilateral standard node dissection for patients with predominant urothelial cancer and potentially curable muscle invasive disease. So thank you very much.

Ashish Kamat: Thanks so much Seth, and thanks again for taking the time and joining us today for the benefit of our audience. Just a few questions for you. With all your experience and expertise in the field and all the review of the literature and all the data that you know, how would you say that this trial, which again, congratulations, and it's a really well done trial with really important data, but what would you say to those that would say, "Well, yeah, the trial is great, but I still am going to do an extended lymph node dissection?"

Seth Lerner: So urologists are famous for ignoring Level 1 evidence, and you and I are probably guilty of the same, but there just really isn't a benefit. And what I tell people is that just understand the entry criteria for the study, right, and it's that patient population that doesn't appear to benefit. Now look, if I get in there and I see nodes in the extended template, I'm going to do an extended node dissection, but those patients were not included in this trial. I'm going to think about it in patients with locally advanced disease, although we would be cautious that there is a signal of increased morbidity and potential mortality. So it's always a risk benefit assessment, but I think, for the most part, these patients don't really benefit from an extended node dissection.

Ashish Kamat: Anything about the trial results that surprised you, Seth? Anything you weren't expecting to see?

Seth Lerner: Well, we were hoping to see that the extended node dissection would provide a benefit, but if you look historically, and I put this in the ASCO presentation, there were previous large Phase III trials in esophageal cancer, pancreatic cancer, a very famous study that was done in gastric cancer from Japan, and all of them failed to show a benefit to a more extensive node dissection. And the gastric trial actually showed that there was a higher risk of toxicity. So bladder cancer now falls in line with what we've seen with other major solid organ site cancer.

So what I'd like the audience to think about is this idea of deescalation or deintensifying some of the therapy and, in this case, the surgery that we do. But I think it's really important to convey the message that a standard node dissection is really important because this was shown back in the '40s and '50s, as you know, that it dramatically reduced the local pelvic recurrence rate. Why was that? Because we were removing microscopic or invisible nodal disease, which accounts for a lot of a local pelvic recurrence. So it really is important to both identify those patients who have positive nodes and to control the pelvis.

Ashish Kamat: Yeah, and I'm glad you made that point because I think that's a very important distinction, especially for those of us that travel the world and talk to folks overseas, there's this push from some of the prostate cancer folks to go to no node dissection in the pelvis. It's a different beast, right?

Seth Lerner: Absolutely.

Ashish Kamat: I mean, bladder cancer is a different beast. And along those lines set, for some of the trainees that might be listening, could you tell them specifically what you mean by a standard lymph node dissection and what the boundary should be?

Seth Lerner: Yeah, so it's pure anatomy. So it's external iliac, internal iliac, and obturator nodes, lateral limit, the generative femoral nerve. But it is a complete extirpation of all the potential lymph node bearing tissue in the obturator fossa from the pelvic sidewall to the bladder, and obviously up to the common iliac bifurcation. And I actually think that that may be one reason we didn't see a difference, as Dr. Kamat knows because he was a surgeon in the trial, is that we all submitted operative reports, pathology reports and photos from every single patient that we did. So sometimes knowing that someone, namely me, is going to be looking over the quality of your operation, and myself included, I think all of us were perhaps meticulous to a fault in making sure that we did a really thorough node dissection. And I think that's the other message is, we're going there to try to help these people get cured from their cancer, maybe with some help from chemotherapy or adjuvant immunotherapy, so it's really important to do a thorough cleanup.

Ashish Kamat: Yeah, again, a very important point. And I think the quality of the surgery is extremely important when it comes to radical cystectomy. Our patients get no second chance, there's no dress rehearsal with the radical cystectomy.

Seth Lerner: Right.

Ashish Kamat: And again, I don't want to put words in your mouth, but would you tell the folks that are doing this purely robotic or purely open, that this message holds true for everybody that deals with bladder cancer?

Seth Lerner: It absolutely does. I've always believed, and it's nice to see the contemporary data showing that an expert hands robotic assisted radical cystectomy can achieve most of the same goals, and we're now beginning to see some longer term follow up from an oncology standpoint. But this is an anatomic operation, whether you do it open, whether you do it robotically, the same applies. And we specifically did not allow robotic, and the audience should think about the timeframe when we started this. There just simply wasn't equipoise for robotic cystectomy. But this is anatomy. This is an anatomic operation regardless of how you approach it. And I think the same applies.

Ashish Kamat: Seth, once again, congratulations on this trial, and thank you so much for taking the time and always being willing to share your knowledge and experience and shedding lights on critical issues, not only when I call you and we chat as friends, but also with our audience that loves to always hear you talk. So thanks again.

Seth Lerner: Yeah, thank you. Appreciate it very much.