Get to Know...David Swoboda, MD

With roots in both Florida and Tennessee, David Swoboda, MD, is a hematologist-oncologist at Tampa General Hospital specializing in acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS). Dr. Swoboda spoke with Blood Cancers Today about the family influences on his medical career, his role as a clinical investigator of the QuANTUM-Wild study, and his growing passion for artificial intelligence (AI) research and how it can transform oncology.

Where did you grow up, and when did you know that you wanted to be a hematologist-oncologist?

I grew up in Memphis, Tennessee. I have three brothers, and we moved from Florida to Tennessee for my dad’s job. We lived there my whole life, until I decided to move back to Florida for undergrad and medical school because my parents were moving back to Florida at the same time.

My mom, who was a computer programmer, decided to become a doctor of obstetrics and gynecology. While I was in middle and high school, she was going through medical school and then residency. Seeing the hard work and how much she enjoyed it, I started to learn from her and eventually work with her and obtain opportunities to work within the hospital system. That fueled my interest in becoming a physician, and I found working at cancer clinics to be extremely rewarding.

I particularly became a hematologist and a leukemia specialist when I was doing a sub-internship at the University of Chapel Hill. I went to Florida State for medical school, so this was my first true academic leukemia service. I loved meeting the patients, meeting their families, and seeing how challenging but rewarding it is to take care of this high-risk acute patient population. I grew a lot in that month, and that solidified my interest in not only being a doctor, but also being an oncologist and a leukemia specialist.

Were there any mentors who shaped your career path?

I’ve been lucky enough to have a lot of great mentors. While I knew I had a passion for leukemia, I actually did most of my research in melanoma during my residency at Georgetown. I found that a few of the melanoma providers, Drs. Michael Atkins and Geoffrey Gibney, fit my style of patient-centered care and helped me build a strong foundation of research.

When I moved to Moffitt Cancer Center, I wanted to find mentors who were similar to them in the leukemia space. I met Dr. Rami Komrokji, who was a big influence in my career. After talking to him and seeing his passion for MDS, I felt like he was someone I could shadow my career after and get mentorship from. That was a huge factor in moving to Moffitt and ultimately working under Dr. Komrokji.

Over time, Dr. David Sallman also took me under his wing and was a great mentor for me. He built out a lot of very strong projects, and I continue to think of Drs. Komrokji and Sallman as my main mentors. I still reach out to them all the time, even as a faculty now, to get career advice and continue to grow as a leukemia physician.

Can you further discuss your style of patient-centered care?

In my clinic, I try to put the patient first. I think about myself as if I were the patient. How do you create an environment that is extremely patient focused and referral provider focused, making it as easy as possible to navigate a system?

The reality is that getting diagnosed with leukemia or MDS is challenging enough, and if we can take some of the pressures of getting into the system and scheduling things off the patient’s plate, they can really focus on taking care of themselves and the disease.

At Tampa General Hospital, we’ve tried to streamline a lot of processes. We’ve tried to work with patients to make it as easy as possible for them to navigate their care overall. In our clinic specifically, we want to bring on physicians and staff that are great communicators. We want people to think about Tampa General Hospital as a place where they’re going to enjoy the staff, feel like they’re part of the family, and feel that we are real people who communicate in a real and understandable way. I hope patients leave our clinic with more knowledge, understanding, and comfort rather than confusion.

Are you working on any exciting clinical trials?

One clinical trial I’m excited about is the QuANTUM-Wild study of “7+3” quizartinib versus placebo in FLT3 wild-type patients. I was lucky enough to be at the center that enrolled the first patient on that clinical trial. It’s an interesting concept of using a targeted therapy for a specific mutation in patients that lack the mutation, because we know that a lot of patients have expression of FLT3 even though they don’t actually have that mutation.

I’m also the principal investigator of a menin inhibitor trial, which is another class of drugs that is rapidly evolving. Getting KMT2A patients—which is a very advanced and often times very challenging subset of AML—on these therapies that can hopefully have the potential to be lifesaving is really rewarding.

Aside from being a clinical investigator, my second and growing passion is artificial intelligence research. I’ve been working with Dr. Aziz Nazha at Jefferson Health on building out AI agent frameworks. The goal of our current research is how do we improve on these generalized large language models (LLM) like ChatGPT to be able to be utilized in real-life clinical scenarios?

If you put in a patient case in a general LLM, the output is not always accurate and can even generate false information. General LLMs are not ready for prime time in clinical practice. However, we’ve been able to use the technology behind these LLMs in a more specific way to a particular disease state and have found much better results.

In one of the most far along projects, we have built out a multiagent framework with five different AI agents. One of the AI agents is a moderator that takes a prompt, or in this case a clinical case. It can structure an unstructured clinical case; feed it to another agent that evaluates the diagnosis of a patient; then an agent that evaluates prognosis, treatment, and clinical trial recommendations; send it back to the moderator; and then provide the full recommendation. It is an autonomous MDS virtual tumor board that interacts with and provides information to the physician.

We found that using this multi-AI agent framework is significantly better than traditional generalized LLMs like ChatGPT. We’re expanding on the work and are having more investigators involved in validating this in a larger dataset and patient population. That’s my growing passion, and there’s a lot of different areas where we can use these AI agents. I think that AI is going to have a significant impact on healthcare and oncology, and I’m really excited to be a part of that.

What do you hope to see in the field over the next 10 years?

One of the biggest questions in AML is: “Do we need induction therapy for all patients, or can we start using a lower intensity–based therapy and a lower-intensity backbone for patients even if they’re not elderly and unfit?” Answering that question will open up a lot of opportunities for novel combinations with menin inhibitors, FLT3 inhibitors, or IDH1 and IDH2 inhibitors.

At Tampa General Hospital, we are not only looking at mutations and translocations, but the actual gene expression profiling through RNA. That allows you to think about some of these patients and clinical cases differently and potentially offer certain therapies to patients that we might not think would respond to therapy based on traditional molecular testing.

A good example is the proof of concept with QuANTUM-Wild. We think that patients with a FLT3-like gene expression profile will likely benefit the most in an FLT3 wild-type setting. We also know that patients most likely to benefit from menin inhibitors are those with KMT2A translocated, NPM1, or NUP98, but I think the common thread is that they have HOXA9 overexpression. There are other potential mutations and translocations that could have a similar increased expression and ultimately could be targeted by menin inhibitors.

In the short-term future, that’s one of the areas we’re going to learn a lot from. I think we’re going to start designing clinical trial therapies based on gene expression signatures as much as—or even more than—individual mutations. I think we’re going to have better outcomes doing it in that way.

What hobbies or activities do you enjoy outside of work?

I love the ocean. I love being on a boat fishing, surfing, or anything that is water involved. I spend almost every day after work in the pool with my two daughters, aged 4 and 7. When we have more time, we’re on the boat at the beach and spending as much time [as we can] enjoying the beautiful Florida weather.

Outside of that, another hobby of mine is playing golf. I love getting out in nature and trying to take my brain away from all the stresses and complications of being a physician. Hitting a golf ball is a great stress relief.

What is a fun fact most people would be surprised to learn about you?

I love to travel. I’ve been to a ton of different countries, including hiking the Great Wall of China in sandals. I’ve had lunch and a picnic on a glacier. My wife and I always try to find unique opportunities to experience other countries in ways that sometimes only the locals are able to do.