The Association of Smoking Status with Recurrence in Patients with Predominantly High-Risk Recurrent NMIBC - Richard Matulewicz

October 1, 2021

Richard Matulewicz joins Ashish Kamat presenting findings on the association of smoking status on recurrence of non-muscle invasive bladder cancer (NMIBC) in a contemporary cohort of patients with predominantly high-risk, recurrent NMIBC managed with photodynamic enhanced cystoscopy. A retrospective study was conducted in a multi-institutional registry. Smoking has a strong causal association with bladder cancer but the relationship with recurrence is not well established. The pair discuss the findings of patients with predominantly high-risk recurrent NMIBC managed with photodynamic enhanced cystoscopy.


Richard S. Matulewicz, MD, MS, Urologist and assistant professor, Department of urology at Memorial Sloan Kettering Cancer Center, New York City

Ashish Kamat, MD, MBBS, Professor, Department of Urology, Division of Surgery, University of Texas MD Anderson Cancer Center, President, International Bladder Cancer Group (IBCG), Houston, Texas

Read the Full Video Transcript

Ashish Kamat: Hello, and welcome to Uro Today's Bladder Cancer Center of Excellence. I'm Ashish Kamat, professor of urologic oncology and cancer research at MB Anderson Cancer Center. And it's a pleasure to welcome today, assistant attending Richard Matulewicz, who is now with Memorial Sloan Kettering, to present to us his data and his presentation on the association of smoking status and recurrence of non-muscle invasive bladder cancer among patients managed with blue light cystoscopy. Rich, the stage is yours.

Richard Matulewicz: All right, thank you so much, Dr. Kamat, and Uro Today, for inviting me. It's a pleasure to share some of this data, and this is something certainly near and dear to my heart. And something that I believe we can all improve the care of our patients by recognizing, and I'll get started.

So, as you said, I'm going to talk today, about a recently published study that assesses the association of smoking status and recurrence in non-muscle invasive bladder cancer patients in a select cohort managed blue light cystoscopy. And this is certainly the work of the entire group, as you can see the whole authors here, but it's my pleasure to present on their behalf.

So as we all should know, smoking causes bladder cancer, and the population attributable risk of smoking in bladder cancer is about 50%, making it one of the most smoking related cancers out there. The odds of developing bladder cancer, amongst current former smokers, are several orders of magnitude, or several odds higher, rather, than never smokers, and this is certainly seen among both male and females. We do know that smoking intensity is related to risk, and that the risk is actually increasing with time. And whether this is due to changes in cigarette smoking, changes in smoking patterns, is currently not known. But what we do see is that actually, the increase, the risk of developing bladder cancer over time, is getting worse.

Something also that we know, from a few nicely done studies, are that continued smoking after a bladder cancer diagnosis, portents a higher risk in several domains. One is that you see worse surgical outcomes related to cardiopulmonary side effects, and higher anesthetic risk. Another is that you see a greater risk of systemic therapy adverse effects, and actually an attenuated benefit of neoadjuvant chemotherapy in smokers. This is something that was nicely to described in the last year or two. And even among patients treated with radiation, that there's a greater risk of adverse effects, and a attenuated therapeutic benefit as well. However, it's been a tough egg to crack, to really figure out the relationship between smoking and how this changes, increases or decreases, the risk of recurrence in non-muscle invasive bladder cancer. And this is something we sought to figure out.

So the way we did this is, we used the Cysview Registry, which is a collaborative effort from about 15 centers, and multiple urologic oncologists, mostly high volume centers, that contribute patients that are managed with blue light cystoscopy, with Cysview. This registry contains all patients that are undergoing this procedure with blue light cystoscopy, that have either non-US invasive bladder cancer, or suspicion for non-US invasive bladder cancer. And the cohort that we put together consisted of complete data, as of September, of 2020.

Our exposure of interest in this study was smoking history, and this was defined qualitatively as either never, former, or current, because we had some missing data on the intensity duration and pack years related. Additional measures that we included in our analysis were, basic demographic factors, AUA non-muscle invasive risk category, as well as details on recurrence, or primary bladder cancer, as well as prior intravesical treatments. Primary outcome in this study was the first recurrence, and our secondary outcomes included, progression, as well as first high grade recurrence. And essentially, the analytic plan were several time to event analyses and multivariate logistic progression, to try to determine the independent association of smoking history with our primary outcome.

So a little bit about our cohort. As mentioned, the initial Cysview Registry contained almost 2000 patients. We excluded patients that had no history of malignancy. So these were patients who only a suspicion of having bladder cancer and underwent blue light cystoscopy, which left us with a cohort of about 1600. Excluding missing data, got us to a cohort of about 1000. And then, excluding the patients with an unknown smoking status, left us with our final cohort of 723 patients.

So this is just a quick table one, run down of the demographics of this table. It's important to understand that this particular cohort, as you can see here, based on the predominance of high risk AUA non-muscle invasive bladder cancer patients, as well as patients who had both recurrent disease and were heavily pretreated, with about two thirds having prior intravesical chemotherapy treatment. So this is a fairly high risk recurrent cohort of patients that we're dealing with here.

Additionally, our primary exposure here that we wanted to look at was smoking history. This is the distribution of smoking history in our cohort. About 13% were current smokers, about a third were former smokers, and about 50% were never smokers. And this mirrors, pretty much, the proportion of smokers that we see in our general practices. The table down here, what we're showing here is the intensity of smoking among the patients that we had pack or history data on. And as you can see here, in addition to being a higher risk bladder cancer cohort, these patients were actually fairly heavy smokers, as well, for the ones that we did have data on.

So getting to the results, the cohort had a median follow up of about 717 days, with some patients having far greater than three year follow up. 10% of the smokers, and I think this is an important secondary outcome that we looked at as well, is that from their first entry into the registry, on follow-up cystoscopy, six months later, only about 10% of current smokers quit. And if you really look at the intensity of smokers, this is actually, a fairly high number of patients who quit smoking. Unfortunately, we don't have any long term data on whether this was a sustained cessation effort, but I think this is something, that it was one of our notable findings in the study.

Regarding our primary outcome, which was recurrence, what we saw was, that recurrence occurred in about a third of patients, during the follow up period. And the one and three year probability of recurrence was 19% and 44%, respectively. And among these patients, we saw a rate of about 2.5% of patients who ended up progressing beyond non-muscle invasive bladder cancer, to either muscle invasive, or metastatic disease. None of these patients had anything beyond nodal disease at the time of cystectomy, however.

Looking at our unadjusted Kaplan Meier time to event analysis, what we can see here is our primary exposure and our primary outcome. A time series analysis, which demonstrates that we did not see any difference in our three level smoking variable, current, former, or never, as it related to a recurrence free probability over time. And you can see here, the overlapping Kaplan Meier curves, and this is something that, and I'll talk about this in a moment. This is something that's contrary to some of the prior studies, and we'll talk a little bit about why that may be.

Assessing some of the other variables of interest. What we did see though, is that there is a separation of our curves here, among AUA low risk criteria, when compared to intermediate and high risk, which are fairly overlapping. And this is, I can't say it was the first, it was probably one of the earlier studies, that do ultimately validate the recurrence differences among AUA low risk and higher risk criteria. So this is something that we did see a difference in. This is just a unadjusted Kaplan Meier for these.

Some other factors of interest are the gender. And we did not see any difference in gender with regard to recurrence-free probability. But interestingly, this is something that we were not necessarily expecting, but we did see a difference in the recurrence rate among patients who had prior blue light cystoscopy before entry into their registry. So this is something that prior use of blue light cystoscopy did ultimately decrease the risk of recurrence, and the underlying reasons for this are not able to be really teased out in this analysis. But I thought this was an interesting finding of our study, as well.

So I'm looking at a multivariable regression analysis, to look at what the independent risk of smoking status is on our primary outcome recurrence, what we found was that AUA risk criteria did have an independent increased risk, when compared to low risk disease, for both intermediate and high risk blue light cystoscopy usage, prior to enter entry into the cohort, also pretended a significant difference. And this is a lower risk of recurrence, or lower hazard for recurrence, among these patients. But, our primary exposure of interest, smoking status, did not have any sort of statistical or clinically significant difference in recurrence rates, when included in our multivariable regression analysis.

So what we found, and the primary outcome of our study here, was that was no significant difference in recurrence in our study, based on our three level smoking variable. And that is to be framed by the fact that this is a largely high risk recurrent non-muscle invasive bladder cancer cohort. Also, that these patients were managed with, what I would consider, the standard of care, which would be blue light cystoscopy, and the diagnosis and surveillance of non-muscle invasive bladder cancer. So this is probably the most contemporary, albeit high risk, cohort that's reported these outcomes to date.

We also saw that there's a lower risk of recurrence in patients who had been managed with blue light cystoscopy, and that we saw that there is a difference, at least between low, and intermediate, and high risk AUA categories, for the recurrence free probability.

There have been several nicely done prior studies, and unfortunately, this study did not end up being the definitive study on this topic, because these results still remain equivocal. But prior studies by Renck, Lammers, and van Osch, all have demonstrated differences, or different outcomes, rather, some of which have seen differences in smoking status, others have not. And the issue with these studies to date, is that the cohorts have been quite heterogeneous within studies and between studies. So unfortunately, the jury still remains out, on the actual relationship between smoking and recurrence risk of bladder cancer.

The good news is, there's more data forthcoming, some by my colleagues here at Memorial Sloan Kettering, called the DETER study, which is a prospective study, that has looked at several additional urinary biomarkers, as well as, in order to look at recurrence rates, with a specific look at smoking intensity and history.

What is necessary is, that as we move forward, especially in collaborative efforts, we do need better and more granular data on smoking intensity and history, as well as changes over time, to really get to the bottom of how this all fits together. And it's most likely that the true risk of recurrence, and even potentially, the risk of being diagnosed with bladder cancer, is actually a combination of smoking exposure, and some element of genetic predilection. And this is some work that's been done by several colleagues, including Josh Meeks at Northwestern, who's trying to get to the bottom of this.

Regardless, we do still believe that smoking cessation remains a critical component of bladder cancer survivorship, although it is not absolutely beneficial, as far as recurrence goes, at least not yet. What we do know is that the health benefits of smoking cessation likely extend far beyond the recurrence risk benefits. So we do recommend that all current smokers with bladder cancer get evidence based smoking cessation care.

So I'd like to thank the Cysview Registry group, for helping support this study, and also Photocure, for allowing data access to the registry, and providing the third party analytic support. And I think it's important to note that the company was not involved in any of these analyses, interpretation of the data, or the writing of the manuscript. This was truly an investigator initiated effort by myself and colleagues here.

Ashish Kamat: Thank you, Rich, for that presentation. I'm glad you ended with the statement that smoking cessation is still part, very much a part of bladder cancer survivorship program, because that was one of the questions I wanted to ask you. And I knew what you would say, but I wanted you to say it anyways, so I'm glad you said that.

With that in mind, and looking at this report, specifically, where you sought out to look at the association of smoking with recurrence in patients treated with a uniform standard of care, blue light cystoscopy, do you think, hypothetically, that there's anything to do with patient selection and motivation of the patients and the providers that might have influenced the results?

Richard Matulewicz:

Yeah. I think, there's so much that goes into, not only bladder cancer recurrence, but all of these continued carcinogenic exposures, that this is going to be a difficult question to unpack, certainly without really, really high quality, high numeric, and data. So to tackle that one step at a time, I do think that a bladder cancer diagnosis is a teachable moment. And I do think that patients will have a big opportunity for behavior change, if it is leveraged correctly.

So what we didn't see, or what we couldn't control for, or even analyze in this study, is what happened at that moment. Patients will react very differently to the diagnosis. Some will immediately be interested in quitting. Some will be able to do it cold turkey. Others really need to rely on their physicians or their family. So although this is usually a big time leverage for behavior change, without being able to understand, or see what happened with those particular patients, limits our ability to do that. Now, only about 10% of patients ended up quitting over the six month change from entry into the registry to the first follow up. And this actually may even influence outcomes.

There have been a few studies published, including one by van OSH in 2018, that actually looked at the effect of cessation at the time of diagnosis on outcomes. Although that they did not find an association between quitting at the time of diagnosis and recurrence risk, that study was limited by a very low event rate. So it would be quite difficult to find a difference just with that study design. However, we do know from several other studies as well, that health related quality of life, pain scores, cardiopulmonary tolerance, and just general health and mortality, are improved by smoking cessation. So, I think it's an absolutely critical part, but it's really difficult to figure out exactly how that all plays into this.

Ashish Kamat: When we looked at our series at MD Anderson, and even population based series that we've reported on, looking at genetic markers and SNPs analyses, smoking status almost always seemed to correlate with better outcomes. It may not always be better cancer specific, maybe overall survival, which obviously, could be related to cardiopulmonary status. But if you group patients into current smokers, former smokers, and never smokers, the current smokers almost always did worse than the never smokers, which would be understandable from overall cardiopulmonary health. But even the ever smokers, the ones that used to smoke and then quit, seemed to do better.

So again, I applaud you and the group for doing this work, and publishing your results, even though it was negative, according to your hypothesis. Right? But I also applaud you for sticking to your guns, when it comes to stating and recognizing that when we take care of patients with bladder cancer, even if there are some reports such as this, that show that, or suggest, that smoking may not be related to the effect on recurrence rates, it is still an important part of what we should be to talking to our patients about, and counseling on them.

Again, I just wanted to thank you for taking the time to share this with us. In wrapping up, are there any high level thoughts, or take home messages, that you want to share with the audience, related to your paper, but also related to the topic at large?

Richard Matulewicz: Yeah. This is something that I think urologists are going to begin to hopefully, embrace a little bit better. And the reason for that, I think, is that we're at a point with the electronic medical record, and a lot of what we have supporting us as clinicians, this is really ripe for improvement, and certain approaches that will allow us to integrate these types of beneficial survivorship programs into our routine clinical care. I think that a lot of research, and a lot of effort, is needed to figure out where this all fits in to optimize this within our workflow.

But I think it really behooves us to really help get patients, the evidence based care they need for tobacco treatment, because it's very likely that getting someone to quit smoking is far more beneficial than any of the other things we do. Whether it's intravesical chemotherapy, whether it's TRBT, whether it's neoadjuvant chemo, the most impactful health benefit we may have on patients is actually getting them to quit. I think that, although this is certainly not an area where urologists have traditionally been leaders, I think that it is a big opportunity for us to embrace, and really help one of the most smoking related cancers, to try to get patients to quit. So I think there's a big opportunity for us here.

Ashish Kamat: Absolutely. Once again, thank you for taking the time. Stay safe, stay well, and hopefully we'll get to see each other in person sometime soon.

Richard Matulewicz: Absolutely. Thank you so much for the invite.