Intravesical Therapy for Non-Muscle Invasive Disease - Michael O’Donnell

December 16, 2021

In this 2021 LUGPA CME presentation, Michael O’Donnell discusses intravesical therapies for non-muscle invasive disease. He covers the background of indications for intravesical therapy, managing recurrent non-muscle invasive bladder cancer (NMIBC) including BCG failures and BCG unresponsiveness.


Michael A. O’Donnell, MD, Richard D. Williams Professor & Director of Urologic Oncology, University of Iowa Carver College of Medicine, Iowa City, Iowa

Read the Full Video Transcript

Michael O'Donnell: So today, we're going to talk about intravesical therapies for non-muscle invasive disease. And like Noah, it's hard to cover all this in 15 minutes, but I think I can give you a good taste of everything. These are my disclosures. Nothing particularly relevant, relative to today's discussion.

There are several indications for intravesical therapy. We're all familiar with the single postoperative use of chemotherapy to reduce papillary recurrences. And this is really only indicated for apparent low-grade papillary disease. In fact, it works best for stage Ta low-grade. Unfortunately, about a third of the lesions that look like they're low-grade actually happen to be high-grade. There's no harm in giving this treatment. Neither is there harm in giving it for carcinoma in situ, but the effectiveness is diminished and has not been shown to be statistically significant. There have been several agents that have been confirmed to be effective. For the longest time we gave mitomycin, or even adriamycin also known as Doxorubicin for this.

And then a important study came out of Dr. Messing's work, and was presented in JAMA in 2018. And as you see in the mid graph, the control arm was saline and the treatment arm was gemcitabine. It was two grams and 100 cc given within three hours of the original TURBT. And if you looked at the best-case scenario, which is pathologically confirmed Ta low-grade, there was a 20% differential in favor of non-recurrence for gemcitabine. If you mix the whole group together, it was about 12%, but that again contained a third of the patients that had apparent Ta low-grade, but were actually Ta high-grade. Now there are differences in cost and in toxicity that may make you want to choose one agent or another. In terms of cost, the cost for mitomycin now is over a thousand dollars per 40 milligrams.

Gemcitabine, the acquisition costs are somewhere between $17 and $35. For docetaxel, it's about $35 to $70 for a single dose capsule or vial. The toxicity also is in. I'm sorry, that was for doxorubicin, sorry. For doxorubicin it's only $10 per treatment. The toxicity definitely favors gemcitabine. In fact, it's so much in favor that you can use it, or if you have unrecognized perforation, you're not going to run into trouble with gemcitabine. Because it is what oncologists often refer to as a non-vesicant. It can extravasate from a vein and not cause significant tissue damage. Whereas there have been actually deaths linked to the use of mitomycin or doxorubicin if there was an unexpected bladder perforation. There's also other considerations. For instance, we all know that mitomycin can cause a significant polymer and genital rash, and it can lead to these calcifications in the bladder. In the worst case, frank bladder contractions.

So the bladder becomes nonfunctional. The effectiveness? There haven't really been head to head studies, but it would appear that Gem and Mito are similar, and probably a little bit better than doxorubicin. But again, this is somewhat conjectural. That's why I'm putting the question marks. The second major indication, of course, is for treatment or prevention prophylaxis after resection of intermediate-risk non-invasive disease. We're really talking about Ta low-grade with risk factors. That is large tumors, multifocality or frequent recurrences.

And the agents that are most used for this are mitomycin, gemcitabine and BCG. Again, cost considerations, mitomycin more expensive, and BCG, more expensive than gemcitabine. Effectiveness. It's hard to really compare that, but there was one trial, and I featured that on the far right by Addeo. Who compared gemcitabine at two grams in 100 cc versus mitomycin, 40 milligrams in 50 cc, for five treatments. And six in the gemcitabine group with monthly maintenance for one year. This was at intermediate risk group, contained some high-risk, but only about 25% of them were high-grade. There was no CIS allowed in this study and there was often a BCG failure, although it was usually just a failure of one course of therapy. But as you can see here, Gemcitabine appeared to outperform Mitomycin by an absolute 20% differential, especially after 20 months.

The next group that we're commonly used to using therapy for are our high-risk patients. Those with any high-grade disease, T1 disease or CIS. And of course, BCG with maintenance therapy has been the mainstay of therapy. Although Mitomycin has been used, it's about half as effective in used in this format than BCG. Now I'm going to show you some new data on upfront use of Gem/Doce, which is going to show that it is at least as good as BCG. So stay tuned for that. Now these next two categories are somewhat new, because as we now have recognized the problem with BCG failures and BCG unresponsiveness, we're now talking about therapy for that. And so I will show you data that shows that Gem/Doce is probably better than Gem/Mito, better than single agent Docetaxel, better than single agent Gemcitabine, which is about equivalent to Pembro, which is extremely expensive. On the order of $10,000 per dose. Greater than Valrubicin, which is about $4,000 per dose. And the fifth category, which I've made.

Because we have some data on that now, is what do you do when a patient fails BCG, fails rescue therapy, then what do you offer? I'm going to show you some very intriguing data showing that the combination of sequential Valrubicin, followed by Docetaxel is actually quite effective. Although it does come with a little bit of a price tag. So we are facing a interesting situation now with a BCG shortage. And it's due to the fact that the prior manufacturer of BCG, Connaught went out of business in about 2017. That left Merck, since it acquired the TICE strain as being the main supplier, not only for the US, but for 70 countries worldwide. As you all know, in about the early part of 2019, we faced a second major shortage. And it's predicted not to end for at least five to seven years, as new production facilities are under construction.

So what can we do in the meantime? Well, the first thing we can do is to make most out of our current BCG supply. That means a triage-type evaluation of prioritization. Such that for instance, patients with very high-risk disease, let's say T1 high-grade plus CIS would certainly take precedent over Ta high-grade, which would take precedence over multifocal Ta low-grade. The other possibilities are reducing the BCG dose. And a one-third reduction is probably the best you can do and still maintain a high degree of effectiveness. And to try to group patients together, so that one vial can in fact, feed three different people. This has been allowed by insurance as a billable charge. The other thing to do is especially consider that even if you need high-dose BCG or regular dose, you can always do reduced doses during maintenance therapy. And for patients perhaps with intermediate risk to see. The third point is interferon alpha actually allows you to use lower doses of BCG.

In fact, our large two studies on BCG plus interferon alpha demonstrated that the two-year disease-free rate is identical with using one-third dose BCG, plus 50 million units of interferon as a full dose of BCG using the SWOG protocol. So you certainly can get by with less. In fact, we actually reduce the dose during maintenance so that in any given three-week maintenance cycle, the first dose is one-third. The second and third dose are BCG at one-tenth the amount, allowing you to extend the BCG supply during maintenance therapy. The final point is that we've been reducing our maintenance program.

To the first set of three at three months, the second set of three, six months later, and the third set of three at 15 months. And our results again, combined with interferon are identical to the SWOG program of 36 months. The second major way of dealing with the crisis is to use alternative agents. And this has been proposed and championed by the SUO, the AUA and multiple other bladder cancer organizations, such as BCAN. And that is the idea of using an optimized form of mitomycin, 40 milligrams in 20 cc instead of 40 milligrams in 40 cc, will almost double the effectiveness of mitomycin. Making it almost equivalent to BCG for intermediate-risk patients. The other point is that single agent gemcitabine may be even better than single agent mitomycin. At least the standard form, not the optimized form. And I showed you the Addeo trial.

And finally, I'll show you the new data showing that sequential Gem/Doce is equivalent to BCG, even for high-risk patients. So this is our data from the University of Iowa. We have now treated 107 patients. This has not been published, it's been submitted for publication. But 107 patients, 52% had T1 high-grade, 44% had CIS. Our median follow-up is 15 months, and our high-grade recurrence-free survival. That is no high-grade disease at 24 months, is 84%. Our recurrence-free survival that includes low-grade is still 82%. So only a small percentage had Ta low-grade recurrence. By comparison, the historical value for BCG is about 60 to 65% at two years. In our group of patients treated with gemcitabine, there was no progression. Not even stage and grade progression from for instance, CIS to T1 high-grade. There was only one cystectomy for a patient with CIS who could tolerate only two treatments, and basically was a bladder cripple coming into the study.

On the right you see the toxicity profile. Almost 60% of patients had no side effects whatsoever. 96% were able to receive full therapy, at least five of six of the treatments. And treatments delays only occurred in 8% of patients. The kinds of toxicity were mild and easily manageable that we're very familiar with. Frequency, urgency, hematuria, some dysuria, retention. And a small amount of nausea associated with small amount of absorption, easily controlled with a Zofran given preemptively. Let's go now to BCG "unresponsive" disease. And we've defined this based on studies actually that you see here, which were performed in the BCG interferon trial about 20 years ago. Basically, patients were allowed to have any form of BCG failure and go on the study. And what we discovered was, there were two groups of patients that clearly did poorly. And you see that in the bottom chart. The green patients are those with CIS, the blue patients are those with Ta or T1 disease.

And they have all failed at least two courses of therapy. And it's not that they don't respond at all to BCG, but they respond very poorly. Such that at one year, only about 25% are disease-free. By two years, it's about 18%. And by three years, it's rare to have a patient that still is disease-free. Now there has been a second major type of what we call unresponsive. That has been promoted by the experts based on no data. And that is that T1 high-grade disease at first evaluation is also BCG-unresponsive. That simply isn't true. Our data show very clearly that 40% of T1 high-grade disease given a second course of this case, BCG plus interferon, are disease-free at two years. I think the worry was that progression could ensue during that time. Therefore we should classify them as unresponsive, because we don't know that they wouldn't progress. And they should go to cystectomy early. But there is no data to show that, at this point in time.

Okay. Now here are the early results of some single and doublet therapies for BCG failure. In cases, BCG-unresponsive is a very specific definition, so I won't use it here. But I will show you that later. If we look at single agent gemcitabine, that people were hoping would have a major impact. We look at the trial that was run by SWOG, by Eila Skinner are published in 2013 in Journal of Urology. What we can see is that of the 47 patients, only 28% were disease-free at 12 months, and only 21% at 24 months. Not much different than if you had given another course of BCG. Single agent docetaxel, by contrast, led by McKiernan's group at NYU Columbia showed that they had a very good response at 12 months, 40% disease-free. And with maintenance therapy continued on 32% at 24 months.

We started our first doublet of Gemcitabine-mitomycin about seven or eight years ago, and published that we got a 48% one-year response and a 38% at two years. This was reproduced by the Mayo Group, who showed in their group of 27 patients, 37% no evidence of disease. The problem with this is that you're still getting the mitomycin effect in terms of the bladder irritability. And finally, what happened was, we ran out of mitomycin through a shortage. I substituted docetaxel for mitomycin. We started to see some very interesting results. On the bottom panel, in the blue at the top of it, you saw our BCG-naive patients, and we had six of them. And basically they all responded. And that gave us the impetus to do the later BCG-naive study. But those that had failed once or twice still did very well with 54% disease-free at 12 months, and 34% at 24 months.

This was recurrence-free survival without specifying high-grade recurrence. Dr. Bivalacqua's group at Hopkins then substantiated it. So in a 56% at one-year and 42% at two years. That led to this large study that was published in May of last year, which was a multi-institutional retrospective study of 10 different institutions, all using Gem/Doce for BCG failures. In the bottom corner on the bottom part below, you'll see the actual numbers for the Kaplan Meier curves. Bottom line is at 22 months follow-up, those with high-grade disease that had failed BCG of any type, 52% were disease-free at 24 months, and almost 80% were disease-free at the first time of the evaluation at six months. For those that were BGC-unresponsive. And there were 71 that were CIS, and 34 that were high-grade papillary.

The results were very similar, with a two-year disease-free rate of 50% at 24 months and 58% for the papillary. If you look at progression-free survival, 93%. Cancer-specific survival, 96%, and overall survival of 87%. Of this group of 276 patients, 97 of them came from Iowa. So I can now give you long-term follow-up with 49 months median survival. And as you can see here, our results are very similar. 74% of the patients were disease-free at three months, and then there was a slow gradual decrease. But by five years, 31% of all high-grade recurrences were disease-free.

33% of the BCG-unresponsive were disease-free. 75% were cystectomy-free. And the progression-free survival was 68%. And the cancer-specific survival was 91%. If we look at what the current agents that are being studied for BCG-unresponsive disease sit, you're seeing here in the blue oval for the approximate durable, complete response for Gemcitabine alone. Pembrolizumab, Oportuzumab, and Nadofaragene are all at the 15 to 25% level between one and two years. Showing a marked difference using this combination therapy. Of course, with the caveat, this is still retrospective data.

Finally, I want to show you some very new data on what to do when Gem/Doce fails. And this comes out of a story of a patient who refused to have a cystectomy after failing Gem/Doce, and said, "Come up with something." So I thought, "Well, I've got Valrubicin, but that only gives me an 8% two-year disease-free rate." I thought Valrubicin, docetaxel, both very lipid-soluble, perhaps the two would interact in some crazy way to work. And I was surprised it worked so well. So here are 75 patients. There were 12 that had Ta low-grade. None of them had a high-grade recurrence, and 60% were disease-free at 24 months. That's the very upper line. We had 63 patients that had failed GemDoce, and almost all of them had failed BCG as well. See, these are double failures who now got Val/Doce.

And as you can see here, 60% of them had a complete response at the first three months' surveillance. And two-thirds of them were, approximately 40% remained cancer-free up to 24 months. And then with the curves leveling off at that point. But our median follow-up is still 21 months, so I don't want to say too much about that. So here are my major take home points. Number one, continue to use single-dose postoperative chemotherapy for what apparently looks like as a low-grade papillary disease. And I prefer Gem over Mito over Doxorubicin. Prefer chemo for upfront intermediate-risk disease, with gemcitabine or optimized mitomycin being probably equivalent. The gemcitabine looks to be superior to mitomycin alone. Docetaxel single therapy has never been tried for this, but is probably worthy of some investigation. It has the least toxicity of any single agent we use intravesicularly.

The options for recurrent intermediate-risk, BCG could be used as the rescue program if it's available. Otherwise consider some of the doublets such as Gem/Doce, or certainly single agent docetaxel. For the high-risk patients, if BCG is available, use it sparingly, use it strategically. If it is not, consider using Gem/Doce with maintenance therapy. The maintenance therapy is given for two years once a month, after the initial complete response is validated. Continue to ration BCG rationally, use dose reductions and spread out the maintenance. As I indicated, I consider Gem/Doce as first-line therapy for BCG-unresponsive high-grade disease. And an alternative might be Gem/Mito or single agent docetaxel, but not nearly having the degree of confidence we have now with the Gem/Doce doublet. And finally, Valrubicin-docetaxel appears to be a very useful second-line rescue therapy for Gem/Doce plus BCG failures.

However, if you've only failed Gem/Doce alone, you can also just use BCG if you have it available to use that. And it has been proven to be almost as effective as giving BCG upfront. Few little pearls around the way, is that be very wary as you enter this BCG-unresponsive state. That there are extravesical sites of recurrence that occur at about 20% of the upper tracts, and about 10% of the prostatic urethra. You must look for that.

Otherwise you will leave a potential significant area untreated, and allow it to grow and thrive while you're treating the bladder. For this reason, I do restaging procedures for all high-grade failures. That includes not only bladder cystoscopy, but I use blue light. I biopsy lesions that look suspicious. I biopsy random bladder. I do bilateral upper tract washes, and I do prosthetic urethra biopsies, looking for these sanctuary sites where the high-grade cancer, especially CIS hides. Consider doing at least twice a year CT urograms for the first two years, then probably yearly thereafter. And clearly document your discussions of cystectomy whenever high-grade failures occur. Thank you very much. I think you'll find these points very useful for busy practices such as yourselves.

email news signup