Global meta-analysis examines urine microbiome for markers of bladder cancer

In this video, Laura Bukavina, MD, MPH, discusses the background for European Urology Focus paper, “Global Meta-analysis of Urine Microbiome: Colonization of Polycyclic Aromatic Hydrocarbon–degrading Bacteria Among Bladder Cancer Patients.” Bukavina is a urologic oncology fellow at Fox Chase Cancer Center in Philadelphia, Pennsylvania.

Transcription:

A lot of researchers have been using urine as a way to test for biomarkers for diagnosis of bladder cancer. They call it sort of a predictive or prognostic marker. But in the microbiome, we can also use it as a marker for prognosis of disease. It's also important to note that we can also use it to potentially look at exposure risks, such as your environment and how it changes your microbiome. We can look at genetics, to see if it changes your immune response, or if it changes any sort of mutations. So by us doing this study, we wanted to answer a few questions. The first question is that the microbiome across the globe, so not just the microbiome in 1 institution in the United States, looking at the urine microbiome in 1 institution that's isolated, within 1 state, we wanted to see if all bladder cancer globally has some sort of a signature that we could use that's reproducible, so if you get this signature in the US, it should be reproducible as a signature in Asia.

The second question I wanted to ask is about different collection methodology: Does voiding in a cup cause more contamination vs getting catheterized urine? Because if you think about it, hypothetically, if you're peeing in a cup, and you're a male, you're getting a whole lot of shedding from other parts, right? So you have the penile urethra, you have the prostatic urethra, and you're getting a lot of other contaminants. Females are also getting a lot of contamination in their urine.

The third and final question we wanted to see examine is, is there any kind of signature to where we can look at the microbiome, it's a type of bacteria in the urine and look at potential exposures that people have? So that was our hypothesis going in. We took samples from different countries all over the world. Microbiome studies are small, and that's a problem. Usually, you have 20 patients here, 30 patients there. It's not like a retrospective review where you have 1000s and 1000s of patients. So building up that cohort of patients to 150 or 200 patients is really impressive. We were able to get about 120 patients. And then on top of that, getting controls because you need the healthy non cancer controls to compare the group to. After we compiled the data, we did something different. Whenever someone does a meta-analysis, what they typically do is they take the data that's already analyzed, and they compile it, and then they make conclusions based on the data. So for example, in regular clinical research, when you do a meta-analysis, you use published data from tables, and you compile your own big table and you run a meta-analysis on it. We did something different because you have to be able to standardize how you analyze the data. So if we just took data that was presented already, and analyzed by someone else, and we said, based on how they analyzed the data, this is the table of distribution of the type of bacteria they have and then put it all of it together, that would be a wrong way to do it. Because everyone is analyzing it differently. Everyone has different thresholds. So we took raw data. We contacted institutions, or we were able to get it from a cloud service, so we were able to get the raw data, which means we got the sequencing data from all of these investigators. So it would decrease the amount of confounding that you have based on the analysis. So once we received all of these raw data, then we ran the analysis together. That way we're able to not only look at geography, but also minimize that risk of different principal investigators doing different things for the analysis.

This transcription was edited for clarity.