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 and Cancer Research at MD Anderson Cancer Center in Houston, and it's a pleasure today to be joined by Dr. Patrick Lewicki, who is a PGY 5 Resident at Cornell in New York. Dr. Lewicki is the primary author and has put together the study that looks at the underutilization of blue light cystoscopy for bladder cancer in the United States, and he is going to share with us the salient features from his paper and his thoughts as well. Patrick, the stage is yours.

Patrick Lewicki: Fantastic. Thanks very much for the introduction. Hi, this is Patrick Lewicki, and I just want to start off by thanking Dr. Kamat and the folks at UroToday for the opportunity to promote our work. Today, I'll be sharing some of our recently published data reporting on a widespread underutilization of blue light cystoscopy during TURBT using data from a large national cohort. I have no relevant disclosures.

Just a brief introduction, and I'm sure the savvy audience of UroToday is already familiar with most of this, but blue light cystoscopy using hexaminolevulinic acid, which goes by the trade name Cysview here in the United States, is FDA approved for use during either rigid, or now also flexible cystoscopy, and acts as an adjunct enhancement in the endoscopic detection of bladder cancer. Hexaminolevulinic acid is instilled into the bladder preoperatively and interacts with the heme biosynthetic pathway leading to the intracellular accumulation of photoactive porphyrins, which fluoresce pink, as shown here when excited with blue light between 375 and 440 nanometers in wavelength. Photoactive porphyrins accumulate preferentially in rapidly proliferating neoplastic cells, allowing for the visual distinction between the normal bladder and abnormal lesions on cystoscopy.

Blue light cystoscopy has been extensively studied both for the detection of bladder cancer, and via superior detection, decreased recurrence. I don't need to belabor the evidence base here, as numerous multicenter phase three trials have supported blue light cystoscopy's efficacy for these endpoints, but I just wanted to show one example of an off-sited high-quality meta-analysis, which used patient-level data for 1,300 patients to estimate the detection rates of bladder cancer and recurrence up to 1 year in a population undergoing TURBT. As was largely seen for these individual trials, blue light cystoscopy here significantly increased the detection of papillary bladder tumors, and particularly of CIS lesions. In fact, amongst all patients diagnosed with CIS across these studies, roughly one-quarter were detected only via blue light cystoscopy. When pooling studies with recurrence endpoints, there was a significant decrease in recurrence at up to 1 year following index TURBT. And while this difference of 10% may seem modest, I think it's important to keep in mind the large scale of non-muscle invasive bladder cancer on a national level and the significant burden of recurrent disease in these populations.

On the basis of this and other evidence, enhanced cystoscopy, and in particular blue light cystoscopy, has received guideline recommendations from the AUA and the EAU. Both guidelines statements make mention that these adjuncts should be used during TURBT, provided they are available with a moderate recommendation on grade B evidence from the AUA and a quote-unquote, weak level of evidence from the EAU. Both guidelines statements make mention of a certain false-positive rate. Although, of course, this should be interpreted in the context of a significant, and probably similar number of false positives that may be seen in white light cystoscopy. Also hampering the evidence grades here is a lack of large randomized trials, although more on this towards the end.

So, surely everyone is using enhanced cystoscopy, right? Oh, I think it's important to keep in mind that TURBT, being an endoscopic case, is something that is performed by a wide variety of urologists from a range of different backgrounds, training, practice patterns, et cetera, and given how many different surgeons perform TURBT nationwide, there is a significant window for care quality variation to arise. And we should care about quality variation here, because the index TURBT surgeon is really a gatekeeper who sets up subsequent endoscopic management, can establish muscle-invasive disease, and send somebody for a cystectomy, or even in cases where patients undergo chemoradiation, a high-quality TURBT is a cornerstone of whatever disease control can be achieved there.

We know from existing literature on post-TURBT intravesical chemotherapy, that sometimes urologists either do not follow, do not know about, or are not convinced by evidence and guidelines statements. So we wanted to see what the uptake is like in the broader urologic community of blue light cystoscopy to see what's the opportunity for improvement, or maybe, alternatively, another way of thinking about it, how convinced are urologists at large by the existing evidence and recommendations for enhanced cystoscopy?

To answer our question about adoption, we looked at the use of blue light cystoscopy during TURBT in the Premier Healthcare Database. Just a few words about this data source, since I think it is always helpful to remember what sort of patient sample we're looking at in a given study. Premiere is a hospital-based, all-payor cohort that includes encounter-level claims data, plus a few other non-claims items for a diverse group of hospitals across all census regions of the United States. The capture for hospital-based care is actually quite impressive, and it was estimated that up to 20% of hospital encounters nationwide were captured in 2017, for example.

Premier included 158,870 indexed TURBTs via the listed CPT codes between the first use of blue light cystoscopy in our cohort and the most recent update of the data that we have in roughly 9 years of time. As we mentioned, we used billing data to determine which patients received hexaminolevulinic acid or alternatives, and deemed these patients as having undergone blue light cystoscopy for our study.

The primary endpoint we had looked at was the proportion of patients undergoing TURBT who also underwent blue light cystoscopy, of course, and we looked at trends on a patient, surgeon, and hospital level. We also constructed some models to try and identify factors that may influence or inform the receipt or use of blue light cystoscopy. And for these purposes, it made the most sense for us, when looking at surgeon-level use of blue light, to look at patterns within hospitals where the technology was clearly available since this then reflects the guidelines statements, which recommend use, providing technology is available.

Here's a time plot showing the percentage of index TURBT during which blue light was used. The number is surprisingly low. The larger graph y-axis ranges from 0% to 100% while the inset rather shows 0% to roughly 3%. We presented like this to emphasize just how infrequently, on a population level, blue light cystoscopy is being used. The dotted line here corresponds to the publication of the AUA non-muscle invasive bladder cancer guideline, which is where blue light was recommended for the first time. And there is definitely an increase in use after that point. But when we looked at segmented linear regression estimates of the rate of change, there is actually stagnation in adoption after roughly mid-2018. When looking at the whole cohort, 1.8% of TURBTs were performed with blue light cystoscopy by our study's end, but things are a bit rosier, perhaps, amongst centers where the technology is actually available, with roughly 1 in 10 TURBTs performed at these centers involving blue light.

Here's a time plot looking at adoption by hospitals. The x-axis is adjusted here, looking at just the post-TURBT guidelines time period. To generate this figure, we looked at the proportion of hospitals performing TURBTs that had any blue light cystoscopy performed during a given month. So it's not just a rolling accumulation of hospitals, it's actually showing an increasing number of unique hospitals on a month-by-month basis using blue light cystoscopy. This is encouraging because, obviously, the first step when performing enhanced cystoscopy is to be doing it in cases at a hospital or a center where the technology is available. But what was surprising, and perhaps somewhat troubling, is that within hospitals with blue light cystoscopy capability, the number of surgeons using the technology actually decreased on a monthly basis towards the end of the study period. Unfortunately, from this graph, we can't account for potential causes of this phenomenon, whether that be expansion by hospitals, outpacing provider-level adoption, or if it is rather individuals who tried out blue light cystoscopy and subsequently abandoned it. A little bit difficult to know, but this was a concerning trend nonetheless.

We used mixed-effect logistic regression models to look at variables correlated with blue cystoscopy received from an individual TURBT perspective. Variables were chosen as a priority based on factors suspected to modify TURBT quality, and where dichotomized volume cutoffs were selected based on a comparison of pseudo-R-squared across multiple cutoffs. Everything I've shown here reached our predetermined level of statistical significance. And I've included odds ratios and confidence intervals here so you can infer the effect size of these various correlates, but largely things you might expect being correlated with blue light cystoscopy used on a hospital-level, like academic affiliation, care at high-volume centers, so on and so forth. On a patient level, male patients were more likely than females to undergo blue light, and self-pay or Medicaid patients were less likely to receive blue light, although this does not control or change the Medicare reimbursement, which occurred sometime in 2018. Some clinical pathologic factors were also associated with blue cystoscopies, such as CIS coded at the time of TURBT, along with larger tumor size.

Finally, we constructed a model to identify surgeon-level variables correlated with a complete omission of blue light cystoscopy, that is, those never using blue light, despite operating at a hospital with the technology available. Low-volume TURBTs and low-volume cystectomies, as well as those who have never used mitomycin C, [inaudible 00:08:48] where our model treated any blue light cystoscopy was used as the dependent variables, so the low odds ratios seen here actually suggest that these types of providers were more likely to completely omit blue light.

So, what can we take away from a brief retrospective look at practice patterns? Overall, the use of blue light cystoscopy was surprisingly low, with the hospital-level acquisition of technology outpacing surgeon-level adoption. Additionally, an apparent clustering of evidence-based TURBT and its inverse may reveal the potential targets for intervention. Some limitations include the non-capture of alternative enhanced cystoscopic technologies, such as narrow-band imaging. Furthermore, pathology features are lacking, which may adjudicate appropriate omission of blue light. And finally, this is claims data and it is largely hypothesis-generating in nature, and we can't know, just on what I've shown here, what really drives adoption, or lack thereof, and it could be any number of factors.

Where to go from here? Just some brief thoughts. An optimist's view about this whole situation is maybe that urologists at large are just waiting for more evidence to come out. And this might come in the form of the PHOTO trial, which is a randomized pragmatic trial that is ongoing in the United Kingdom that evaluates the clinical and cost-effectiveness of blue light cystoscopy. Perhaps additional indications for blue light cystoscopy, such as its use during surveillance flexible cysto in the office will increase adoption. Or maybe we need to turn to the world of implementation science, which has shown success in other aspects of TURBT care. Or finally, maybe we should be encouraging internal quality control via checklists in order to nudge urologists into considering the use of blue light.

And with that, I'd just like to thank my co-authors for their support in getting this work published, and thanks for the opportunity once again to share our work.

Ashish Kamat: Thanks so much, Patrick, for that succinct presentation. Let me ask you a couple of questions. What was your hypothesis going into doing the study? Did you have some clues from the referring patterns in New York or from the institutions that you've rotated through, for example?

Patrick Lewicki: Yeah, totally, totally. As a resident, I've gotten to spend a fair amount of time working at a number of different hospitals and I am exposed to a lot of different urologists, and I actually suspected that the use of blue light cystoscopy was going to be fairly low. And I'll even be a little bit more specific to say that some of the things that I wanted to include in our models, obviously these are not variables that are encoded in a claims database, but you know, one of the nice things about having a huge data set is you can use these sorts of things to figure out what providers have done in the past. So for instance, mitomycin C use or any intravesical post-TURBT chemotherapy.  I just had a suspicion that the people that were using blue light cystoscopy also seemed to be doing a lot of other evidence-based TURBTs, and a lot of other evidence-based urologic care, and I was maybe a little bit disappointed, but not wholly unsurprised to see an overall, quite a low rate of use of blue light cystoscopy

Ashish Kamat: And with keeping that in mind since you're obviously a resident, but you are interested in urologic oncology as a long-term career, could you share with us some of your insights as to, from a trainee's perspective, how useful do you actually find blue light for you specifically as you've gone through the years? And let me just ask you, do you plan to use it when you get out there and start practicing after your fellowship, of course?

Patrick Lewicki: Sure. I guess to answer the first part of the question, I think, and this is something that's been described previously, there is definitely a little bit of a learning curve with the use of blue light cystoscopy. I think your first time doing TURBT with blue light, you're saying, "Oh, this is pink. Oh, that's pink. Am I just looking at the tissue obliquely?" Or whatever.

But I do think that it's something that is useful, clearly, and getting to follow up on the pathology on some of our patients that we resect has definitely validated that from a personal experience. I don't have, for instance, any tabulated data from non-blue light and blue light TURBTs that I've done in the past, but it's definitely a helpful tool and something that I would say, more so on the basis of the evidence than rather on how it flows operationally, I think the evidence is there. This tends to be in populations that have to have so many repeat cystos and repeat resections and that sort of thing. If you can find even a marginal benefit to the patients for something that really represents one additional preoperative catheterization and some dwell time, but really is a well-tolerated medication, I think that certainly is something that I hope to use in the future

Ashish Kamat: Amongst the different reasons that we'll hear people talk about at various conferences or thought groups that we put together, think tanks, one is that it's very hard for it to be adopted into the flow processes. Again, from your perspective, as someone that's in training right now, any tips for other trainees on how the flow process could be improved?

Patrick Lewicki: Yeah. I think we are fortunate, many of the hospitals that I've rotated at, that the nurses who are onboarding these patients preoperatively are able to catheterize the patients and instill the hexaminolevulinic acid. I think maybe there are rare instances where they have to call one of the PAs or the NPs. But I think from the resident, and obviously, especially from the attending physician perspective, the workflow, at least at the hospitals that I've spent time at, is pretty easy. It's pretty straightforward. And it's a straight catheterization, which is something that's done by providers of all sorts of different skill levels. I wish maybe for the sake of answering that question, that it was something that we had changed or improved on. But I think based on my firsthand experience, things seemed to be relatively straightforward. Although I could imagine at institutions where, for instance, the people that are onboarding these patients or getting them checked in who are not able or not willing to catheterize patients, then I could see that being a significant deterrent to using Cysview. But I have not personally encountered any of those difficulties.

Ashish Kamat: Great. Well, thank you so much for taking the time and clarifying and going over your paper with us today.

Patrick Lewicki: Yes.

Ashish Kamat: Clearly a very important piece of data for the literature when it comes to blue light technology. Stay safe and stay well.

Patrick Lewicki: Thanks very much, Dr. Kamat. I appreciate the opportunity to share our work.