2018 PCRP Investigator Vignette

Title: Prostate Cancer Outcomes in VA Hospitals

Investigator: Bettina Drake, PhD, MPH; Washington University in St. Louis

My research here utilizes the VA Central Cancer Registry date in order to investigate prostate cancer disparities in prostate cancer outcomes.

Our ultimate goal is to identify modifiable factors that we can intervene on to help reduce the disparities in prostate cancer.

I’m really interested in prostate cancer outcomes within the VA hospitals, and so when I had an opportunity to use some data from the VA Central Cancer Registry, I thought that would be a great opportunity to look at this data because it is a hospital system where everyone has equal access to care, and access to care is usually one of those modifiers that shows differences between African Americans and whites or people in rural areas and urban areas.

We have 102,000 prostate cancer patients that were diagnosed across all VA hospitals within the United States between 1997 and 2009. And what we’ve done here is just kind of show you a few of the variables that we’ve cleaned and that we’ve looked at and the direction of the research that we’re going in for this study.

So, for example, age at diagnosis. We see African American men are diagnosed at a younger age, a mean age of 64 compared to a mean age of 67 in whites.

The interesting thing, though, is that, when we look at obesity, there have been some studies that have found that obesity is a risk factor for high-risk disease. And that African American men usually have higher rates of obesity, whereas in our VA cohort, we saw there really isn't a meaningful difference between the average BMI at time of diagnosis between blacks and whites, 28.2 versus 28.6—not a lot of difference.

Actually African Americans, a slightly higher percentage of them are diagnosed at a normal weight, 27.6 percent compared to whites at 22.4 percent.

And we also were able to look at BMI at 5 years prior to diagnosis as well as 10 years prior to diagnosis because gaining weight or losing weight could have different effects than just weight at one time point in your history.

Within the obese class one category, not a significant association with being diagnosed with more high-risk disease. It’s not until we get to this obese class two and three that we see a significant risk for being diagnosed with high-risk disease.

When we looked at weight change between 5 years prior to the time of diagnosis and your weight at the time of diagnosis, we saw that weight loss was significantly predictive for high-risk disease. So one of the things that we have been thinking about is that there might be other factors that might be contributing to a weight loss being predictive for being diagnosed for high-risk disease. Perhaps these are individuals that have multiple comorbidities or other kind of conditions that are also making them at higher risk for being diagnosed with high-risk disease. And so we wanted to stratify that to see what was going on within each of these categories.

Surprisingly, when we looked among people who were diagnosed, we didn’t see any significant impact of BMI. So, regardless of your BMI, it did not have a significant impact of having you being diagnosed at high-risk disease versus low-risk disease. Where we did see an impact of BMI, where BMI mattered, was among people that gained weight. So those individuals who were heavier, overweight, and obese one or obese two and three category, they were more likely to be diagnosed with high-risk prostate cancer disease.

So our takeaway is that perhaps there are some differences within the VA data that’s somewhat different than some of the clinical data. This is a cohort of men who at some point in their life, the vast majority of them, were probably very fit or had to pass the fitness test for the military and, even though we don’t have the data on, you know, their VOT Max or other measures of fitness, they should at some point in their life been able to be fit, and we see that weight gain has a significant impact on being diagnosed with high-risk disease.

So the VA Central Cancer Registry is a great resource in order to provide us with clinical information that we can use to help us understand disparities in prostate cancer. But often times we also need biological specimens to add an extra layer of information for us. And so the Prostate Cancer Biorepository Network is a great resource for me and for other investigators who are looking for biological specimens to answer questions about why certain individuals are at increased risk for prostate cancer or why certain disparities exist.

Washington University, the site that I lead, is the fifth and the newest site to join the network.

We provide fresh/frozen tissue samples. We really try to make sure that our samples have clinical data on all of the participants that have provided samples and as well as follow-up data. So we’re recruiting prostate cancer patients and contributing samples so that other researchers can assist with other prostate cancer endeavors.