CCTS Clinical and Translational Science (CATS) affiliate scholar, Dr. Charles Gaber, talks about his work using data science methods to predict individualized treatment effects for men with metastatic prostate cancer.
Charles Gaber, PhD
Assistant Professor, Department of Pharmacy Systems Outcomes and Policy
University of Illinois Chicago College of Pharmacy
Member, University of Illinois Cancer Center
Learn more about the CCTS's K Award program
00:00 Charles Gaber:
So much of my career to date, and going forward, is going to be about getting the most you can out of studies that didn't have randomization. And that requires careful thought about study design and your statistical analysis to get your least biased findings from those studies.
00:16 Voice Over (VO)
Welcome to Collaborative Endeavors, a podcast about how experts from different areas of research come together to tackle big health challenges, leading to better therapies and healthier communities.
In this episode, we meet one of CCTS’s newest affiliates in the Clinical and Translational Scholars – or CATS- program. Dr. Charles Gaber is an assistant professor in UIC’s department of pharmacy systems outcomes and policy. He is also a University of Illinois Cancer Center member. Dr. Gaber uses his background in epidemiology to study drug use, safety and effectiveness in the real world. This is known as pharmacoepidemiology, a big word with big implications on population health. Dr. Gaber talked to me about his current research project, his mentor team, and what he hopes to accomplish during his time as a CATS program affiliate scholar.
01:09 Gaber:
It kind of began as an undergraduate at Tulane University in New Orleans, where I majored in public health and economics. At the time it was one of the only institutions that offered an undergrad public health program, so that was a big draw for me. When I was there, I gained exposure to the idea of impacting health and medical outcomes through population level research and statistics. So by the time I wrapped things up at Tulane, I was pretty sold on becoming an epidemiologist. I went directly into a master's in public health program at the University of Michigan, which is where I met my wife Jen in the same program. After graduating from there, I worked for a couple of years at Arbor Research Collaborative for health before deciding to pursue my PhD In epidemiology. I just couldn't get enough of the coursework and the research. I did that at University of North Carolina and finished up in 2022. At UNC, I concentrated in pharmacoepidemiology and cancer outcomes research, which is where I developed this passion for conducting observational studies using large secondary data sets. And it was right after that that I joined as junior faculty in the department of pharmacy systems outcomes and policy at UIC. I got some momentum in the in the cancer world during my MPH at Michigan working on some projects in lung cancer. And then, I thought it was very interesting how each site of cancer can be very different in prognosis and the kinds of treatments available and patterns of care and all these things. It ended up being a really interesting and fruitful area of research for me that I've stayed in.
03:00 VO:
Dr. Gaber went on to talk about the value of observational studies and the importance of validating the findings of studies for which randomized trials were not an option.
03:12 Gaber:
There's a lot of interesting debate both on Twitter and at conferences, academic circles, and stuff about medical evidence, and it coming from randomized trials or observational studies. Randomized trials are known as the gold standard, but there's a lot of cases where there's not an industry incentive to do a trial of two existing therapies or certain patient-centered outcomes haven't been looked at yet, and you need to rely on observational studies. I think maybe it's too dichotomized as- oh, it wasn't a trial, so I can't use this or it's not valid or something. And there's kind of famous examples of biased observational studies, but there's also a lot of people who are thinking about how to improve observational studies and help them deliver important answers. So much of my career to date, and going forward, is going to be about getting the most you can out of studies that didn't have randomization. And that requires careful thought about study design and your statistical analysis to get your least biased findings from those studies.
4:30 VO:
I asked Dr. Gaber to describe the research project supported under the CATS Affiliate award.
4:38 Gaber:
My research project focuses on metastatic hormone-sensitive prostate cancer, and it has a couple of aims. The first aim, we're going to take these randomized clinical trials called TITAN and LATITUDE. They studied therapies for the treatment of metastatic prostate cancer. We're going to use these publicly available trial data and combine that with machine learning algorithms to derive more personalized treatment effects from the trials. As they stand, the trials kind of just report one global number that applies to, theoretically, everyone in the study population. Or maybe they'll look at subgroup analyses. But it's really hard to take the results of these trials and personalize their context for the patient in front of you. So, to get a better sense of that, we're going to be using these algorithms to derive predicted individualized treatment effects and then build a tool. The second aim allows people to enter person-level characteristics like demographics, tumor characteristics and then get an updated or personalized estimate from the trial. Pie in the sky, we would refine this tool over time after seeing how patients and clinicians both use it and seeing how to improve the model both statistically and its usability. And from there, it'd be really interesting to eventually do a trial to see how using this tool could impact care decisions and outcomes. So that's sort of long term. I do think it would be important for both patients and providers to use and talk about. That certainly wouldn't be a crystal ball or a way of absolutely selecting a therapy, but I think it's helpful to add context to trials where clinicians often have questions about, like, “Okay, you know, we've seen the main results of LATITUDE and TITAN, but how does that really come into play for the patient in front of me and thinking about tailoring treatment strategies?” So we hope that this tool is a small step towards more personalization.
7:04 VO:
Dr. Gaber then discussed his mentor team and ways these individuals help contribute to his development as a translational scientist.
7:13 Gaber:
I was fortunate to have individuals in my professional life already who were strong mentors and working in areas that were pertinent to this research, so it felt like a natural extension to ask them if they'd be on my mentoring team. And, graciously, they said yes. Rounding out the team is Dr. Lee from the College of Pharmacy. Dr. Todd Lee, Dr. Natalie Reizine from the College of Medicine, and Dr. Lisa Sharp from the College of Nursing. It's really exciting to have this strong interdisciplinary mentoring team because while they share a lot of strengths, they all have unique elements and perspectives and backgrounds that they're bringing to the mentorship as well. Dr. Lee, he's the head of my home department, and he's my faculty mentor. He's someone who I've had an ongoing research relationship with over the past couple of years, and he's always had a wealth of experience for me to be able to rely on when it comes to data analysis and doing observational studies and that kind of work. So, he's been just terrific. Dr. Reizine, I met a couple of years ago. I just kind of cold emailed her about doing some projects together, and then within 30 minutes she had emailed me back all enthusiastic about working together. And so, we've started to do some prostate cancer projects together in the past year. And then lastly, Dr. Lisa Sharp. She and I used to be office neighbors when she was in the College of Pharmacy, so that's how I got to know Dr. Sharp. And she, besides being just a total expert in research design and methods, she's also a very well-connected person at UIC and at other institutions, too. She's got great networking expertise, and then a lot of experience doing primary data collection, too, which I could see as a logical next step for future projects. So, I think she's going to be really helpful in in helping me get some experience and planning future projects too. Skills wise, I'm hoping to add quite a bit to the toolbox by doing some new predictive modeling work and learning how to do machine learning algorithms like random forests. That will be important for this project, but also for future work. Then on the clinical side, I don't have a clinical background. It's really important for me to have strong clinical mentors like Dr. Reizine, who can help me build the context around study findings and help me develop more expertise in this area be it through shadowing or attending tumor boards, programming at the cancer center. There's all sorts of opportunities at UIC that I think are going to be really important to sort of boosting my translational skills.
10:08 VO:
We are always interested in the translational science aspects of health studies. That is, what part of the project- a method, model, approach- could be generalized and reused by other investigators with adjacent research questions. Dr. Gaber shared his thoughts.
10:24 Gaber:
To date, most of the studies that I've done or been a part of kind of end with the peer reviewed, publication being the final result, and things stop there, right? The hope is that decision makers read your study, and they consider the evidence that you've presented in the study. And that's great, and that's sort of the traditional workflow. But with this project, I'm hoping to gain some experience through doing that second aim and developing something that can make the results interactive and more usable both for clinicians and patients. After we develop this tool, we're going to assess how it performs not just statistically, but it's clinical utility. How practical is this thing, and can we really use it in clinic to help improve shared decision-making and talk about different therapeutic options? I would say that the methods and the workflow would certainly generalize to clinical trials in any setting. The tool itself will be specific to understanding the decision making of these androgen receptor pathway inhibitors, discovering or estimating, at a personal level the expected benefit of these therapies. So, the tool will be for that specific therapeutic question, but these methods definitely generalize, and the idea comes from other disease sites, and conditions that have used these methods like in cardiovascular disease.
11:55 VO:
I asked Dr. Gaber to share his long-term career aspirations, as well as words of advice for other junior faculty or those who are starting to branch into translational research.
12:06 Gaber:
It means the world to me to have the support of the KL2 affiliate program and the UIC Cancer Center, too. Broadly, I want to grow as a multidisciplinary scholar whose work improves the lives of those diagnosed with prostate cancer. Be it through large observational studies- which has sort of been my bread and butter to date- or getting more involved in primary data collection and clinical trial data. I want to be part of a team that's making medicine more tailored to individuals and improves their outcomes, especially in metastatic prostate cancer. So, to do that, it takes a lot. And I think the goal is to develop an active program of research that's well funded, meet lots of people, and develop a network of collaborators based in the U.S. and internationally. Also important to me is taking these experiences into teaching and becoming an influential and dedicated educator in the field of pharmacoepidemiology more broadly. Find your niche area and run with it and have fun with the people that you're working with. I think that is a big component. And when you're building your professional network and your mentoring team, it really is a network. It's not just one person. I recently did an individual development plan or IDP for the CATS Affiliate program, and it made me think about- who do I go to when I have an epidemiology methods question? And thinking about those who are most closely aligned with your research to those farther out who you maybe want to meet and work with in the future. And so, thinking about these across different categories of teaching and research and different areas of your professional development. As a junior faculty, you do hear advice of like learn to say “no,” right? And I think that's definitely helpful. But, there's a bit of a tightrope that you're walking because you want to also meet new people and develop, maintain, strengthen your professional relationships. A lot of that is through working with people on projects and helping them on analyses, and they'll help you. I find that most people, even if you don't know them, are really responsive to just an email reaching out to them and expressing your common interests and seeing if there's ways that you can work together. I did that a lot during my PhD program and I continue to do that now. When I read an interesting paper, I just email the person and tell them that I liked it. And I think that means a lot to people. I was recently at the American Society of Clinical Oncology- or ASCO- Conference and there's a scholar there who I really respect and had known just from reading some individual studies and publications, very influential, and I just introduced myself to him. We talked about his research and stuff I'm doing, and at the end of it, he said he'd be happy to write like a letter of support for a grant that we're writing. You never know what just introducing yourself will lead to, and how a bunch of dominoes will fall after that. So put yourself out there, I think, is maybe my short version of my advice.
15:57 Voice Over:
Collaborative Endeavors is produced by me, Lauren Rieger, on behalf of the Center for Clinical and Translational Science (AKA the CCTS) at the University of Illinois Chicago. To learn more about Dr. Charles Gaber, CCTS’s KL2 CATS and CATS affiliate program, visit the links in our show notes.
The CCTS is supported by the National Institutes of Health’s National Center for Advancing Translational Science through their Clinical and Translational Science Award. Opinions expressed by guests of the show are their own and do not necessarily represent the views of myself, the CCTS or our funding agencies.
You can find more episodes of Collaborative Endeavors on Apple Podcasts, Spotify, Pandora, Amazon Music & YouTube. While we do not monetize this podcast, we do love positive feedback! If you like what you hear, go ahead and give us a 5 star rating to help spread the word.
To learn more about how you can work with the CCTS to make a positive impact on the health and wellbeing of our community, visit ccts.UIC.edu or follow us on X @UIC_CCTS.