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. It's a pleasure to welcome once again to this forum, someone who really needs no introduction, Professor Sia Daneshmand from USC. Sia, you've done a lot of work when it comes to enhanced cystoscopy, better detection of tumors, resection, surgical technique, etc. And you were part of the pivotal study that led to the approval of Cysview and flexible cystoscopy blue light in the clinic. Now you have followed it up, of course, with this work here. Tell us a little bit, and share some insights into the racial differences in the detection rate with blue light cystoscopy.

Siamak Daneshmand: Sure. Thanks, Ashish. Always a pleasure to talk to you, and it's really nice to talk about blue light cystoscopy and discuss how we're enhancing care for patients using this technology.

This came about because we started thinking about the different races and whether there's any difference. We know there's a big predilection for males, for instance, in bladder cancer, but we're not quite sure whether there are racial differences in the detection of tumors and the biology of the tumors in different patients. So, we have this fantastic database, and we thought it’d be an opportunity to look back and see whether there are any racial differences in detection rates using blue light cystoscopy.

I'll just go through some background and aims of the study, looking at methods, results, and conclusions, sort of taking you through the paper a little bit.

The current standard, of course, for diagnosis is white light and urine cytology. And as you know, and you've contributed a lot to this field yourself, we see improved detection with blue light cystoscopy, in multiple studies. There are reduced recurrence rates, and possibly even reduced progression rates, although we haven’t really been able to prove that yet when we use blue light.

Most clinical trials have been conducted in Europe and North America, so there's really not a whole lot of study on Asian and Hispanic populations, for instance. So again, this was an opportunity to look, at least in the subset of our database, to see whether potentially there are any differences, or if blue light is as efficacious across the racial groups. So again, looking at heterogeneity and the characteristics of blue light for detection of these malignant lesions among the various racial groups.

And we broke it down into four racial groups. Of course, you have the White. And unfortunately, the majority of the database are White Caucasian males, but we do have 7% African-Americans, 4% Asians, and 4% Hispanics. So the numbers are small, but this is what we have. The registry is large, right? This is over a seven-year period. We have 1,290 patients in it. There are 2,300 lesions. So it is a powerful database. And we're looking at the usual outcomes of sensitivity, and the negative and positive predictive values for white and blue light cystoscopies.

Now, this is a multi-center prospective cohort, there are 13 centers, so it does have good representation from centers around the country. And so we thought, again, this is probably a good way to look at this.

So here are the results for the different races. As you can see, there are the Caucasians, the African-Americans, and the different groups here. You can see the age is your typical bladder cancer age. The majority of them are men, and most are smokers. And you can see the stage distribution there, with most of the patients having high-grade disease; in the Asian population, it was 80%. So, more or less, in that 40 to 50% are having high-grade, there are lots of lesions that were benign because we are taking patients for biopsies, and of course, they turn out to be benign. So, well distributed across the grades and stages of disease here.

So here's the sort of money slide. These are the detection rates, the positive and negative predictive values of the lesions using white light and blue light. So you can see in the first column you have white light cystoscopy only. The sensitivity is about 87 to 95% across, the average is about 89%. When you look at blue light only, sensitivity is higher, in the high 90 range. And of course, when you take either white light or blue light, it's of course, close to 100%, because that's how we're detecting the tumors, either by white light or blue light. The reason it's not 100% is because occasionally there are some patients who undergo random directed biopsies, which are sort of neither white light nor blue light positive; they're random. And it turns out to be a CIS that's detected by neither white nor blue light. So that's what it is.

And you can see that detection. The increase in detection rate by blue light is these numbers that we're used to, right? We're always used to seeing that 10% increase in detection rate, and the positive predictive values in the 80 range. And you can see that for Caucasians it's about 11%, in African-Americans, about 6%. Asians, that is actually higher. Now again, the numbers are small. These were statistically significant for Asians, for instance, in terms of increase in detection rate. But at least you can make that conclusion that it is working in this patient population. Interestingly, in the Hispanic population, the blue light wasn't as effective in picking up lesions. Again, the numbers are fairly small here. So you can see the positive predictive value here that we're again used to seeing, and the negative predictive value for blue light cystoscopy as well.

So here are the results seen in table format for the different types of lesions. Starting up top on the left side, the Asians, you can see again for Ta, T1 lesions, and CIS, the detection rates with white light and blue light were for Ta lesions. Blue light offers a little bit of an advantage. And for T1, 93 versus 96. But it's really for CIS, which is what we always sort of know for Caucasians. For African-Americans, similar sorts of differences in the distribution of CIS, 76% with white light, you increase that to 95% for CIS lesions. Whereas for Ta and T1, blue light is not as effective in detecting more cases. But you can see for the Asians, we did pick up more Ta lesions in the patients. As well as, of course, a vast difference in the CIS lesions. And again, that holds true for Hispanics as well, where the CIS detection is quite higher.

So fortunately, this all sort of matches up with what we know, but I think the gist of the study really is to show that this does work in all races. There may be some slight racial differences in the detection rates, and that may be biological, anatomical, or maybe just that we may not have enough cases opposed to the controls, the Caucasians in our database.

So in terms of sensitivity and the odds ratio, when you're looking at logistic regression analysis, you can see that for Caucasians, the odds ratio of blue light versus white light in terms of sensitivity is three times for Caucasians, 2.7 for African-Americans, and 16 times for Asians. For Hispanics, it was not quite as high. And you could see which ones are significant; for Caucasians and Asians, it was the most significant there.

So the take-home message, again, I think we can say that blue light cystoscopy has increased the detection rate of bladder cancer across all races, particularly among White Caucasians and Asian patients. It is significantly higher in the detection of CIS in all the groups, as we've always known.

So I'll stop there, and thanks for letting me present this data. This was just recently published as well, so you can see all the details of this study.

Ashish Kamat: Thanks so much, Sia. Let me ask you, obviously, you are presenting the results from a large registry, and the results speak for themselves. But was there ever any concern among you or others that the metabolism or the porphyrin pathway was different in different races? Was there ever that concern? Because personally, when I've used blue light, I haven't felt the need to be concerned about different populations. Share with us a little bit of your thoughts on that.

Siamak Daneshmand: Yeah, it's a great point, and I think that really points to the biology. There may be differences. Of course, we see different hematological disorders in patients, specifically African-Americans, for instance. So there may be differences in metabolism. To be honest with you, I don't know. The point of this study was to sort of see if there are potentially any signals and differences. It would be really neat to see a study where we have a different concentration, not so much concentration, but different times. We know that you need to put this in for at least 30 minutes up to an hour. Those studies were done a long time ago, and certainly not among the racial subgroups. So whether this gets metabolized the same way, we're not sure. But it is comforting to know that for CIS detection, you do see it across all races, that increase in detection rate. But it's a great thought, and the potential explanation for the differences is interesting.

Ashish Kamat: Yeah. After I saw this report, I was thinking about it, and again, it's obviously recall bias, but I get the sense that I'm recalling that the signal is higher in the Asian population. It looks more pink to me, and it may be just my imagination. Does this registry have quantification of the signal, or don't you track that?

Siamak Daneshmand: Great question. In the very beginning, especially when we were doing the flex blue light study, we were trying to quantify whether it's faint or highly blue light positive. Unfortunately, that data is just not available, and it's highly subjective too. There was one study we did, we didn't publish, but there were four of us experts watching the same videos on the flex blue light in the clinic, and sort of grading the positivity, negativity, and the intensity of the fluorescence. Unfortunately, the inter-observer variability was too high, and so we couldn't really make definitive conclusions. And we have therefore not included that in the database where it can be so subjective. Even, as you know, the white light positive, blue light positive designation is somewhat subjective; we're comforted by the fact that we have so many patients in the registry. And the people who entered the data, I mean, the investigators are experts. These are not novice surgeons. So when they call something blue light positive, we're fairly certain.

But as you know, they can be somewhat equivocal. And I do have a field just for myself and my dictations, where I'll call it blue light equivocal, because I'm not really sure. It's sort of a little bit pink, or I'll call it faint fluorescence. But I don't know of any study that has quantified the fluorescence along with the tumor grade, stage, or anything like that. In fact, I can tell you just anecdotally, having done so many of these, that sometimes the low-grade tumors light up, as you know, right? And they're sort of super bright on fluorescence, and you have some high-grade tumors that don't. So I'm not sure what the explanation is behind that, and I'm not sure it's known.

Ashish Kamat: Yeah, no. So I used to do the same thing. I used to have mild, moderate, and high levels of intensity in my op-notes, and it didn't correlate, at least in my series.

Siamak Daneshmand: Yeah.

Ashish Kamat: Because some of these more aggressive fleshy tumors don't have that surface metabolism of the substrate, and they don't shine quite as bright. But the other low-grade ones shine really, really bright pink. I think really the purpose of this study is very much to reassure people that you can use blue light to detect more tumors, high-grade tumors, across the board, right?

Siamak Daneshmand: Yeah. Yeah.

Ashish Kamat: We don't need to worry about it. Along those lines, I'm going to put you on the spot for one last question, not related to your study. But we've all heard about the PHOTO trial. It was published in the New England Journal, even though it was the open-access part of it, and it got a lot of credibility. And people say, "Well, maybe it's the UK population, maybe it's the North American population." I don't think it's true. I've had a discussion with Rakesh here about it, and of course, the endpoints were different.

Siamak Daneshmand: Yeah.

Ashish Kamat: But just in closing, if you could share with the audience your thoughts on the reliability of blue light in international populations, and maybe some thoughts about that study.

Siamak Daneshmand: Yeah, just briefly. I mean, we could talk for an hour probably on that study, and really sort of tease it apart and go through the details. But one of the major things was that I don’t think it's the population. This was done in the UK at many centers. So that's one of the strengths, actually, that multi-center study. And it's good. They've got a good number of patients. However, when you really break it down and go into the supplemental tables, there are a number of centers that did one or two blue lights. Now we know, you and I know from very early on experience, that experience matters in what you call blue light positive or negative. So that's part of the issue is that you don’t have expert centers contributing the data.

The other thing is, you do see a separation of the curve, and this was a single-time use of blue light. You use blue light at the beginning, and then you're looking at outcomes 15, 16, 18 months later. Well, whether or not you picked up additional tumors at the beginning, that's going to get diluted out if you don't continue to use blue light throughout the study, throughout the surveillance, for instance. And so yes, you're going to see some decreased recurrences, and you do see that in the first year in the PHOTO study. But later on, the curve comes together, and the recurrence rates seem to be the same. But to me, that's the biology of the tumor. Just like when you use BCG, you reduce recurrence rates. But when you use maintenance therapy, that's when you really see the differences long-term, right? You see the difference. You see the decrease in progression rates as well when you use maintenance therapy.

So just brief comments about that study. Great study, and I think it's thought-provoking, but the vast, vast majority of the data, from prospective studies, randomized prospective studies done a long time ago, and repeated studies, always show the same thing. There's increased detection. It's been difficult to show that progression rates have decreased. You did a great study looking at progression, and it seems to be decreasing in a retrospective look. And then, of course, the ultimate outcome, are we making a difference in progression to muscle-invasive disease? It's really hard to tease that out. But we're working on it. We're working on some good comparisons in the near future on very large databases, where we can compare white and blue light together. So hopefully, that data is forthcoming.

Ashish Kamat: And then we'll have you back on this forum again to talk about that, right?

Siamak Daneshmand: I hope so. Thank you.

Ashish Kamat: Sia, thanks for taking the time and I look forward to seeing you.

Siamak Daneshmand: Likewise. Thanks so much as you share it. Take care.