Skip to main content
15 minutes reading time (2999 words)

Expanding Access to Clinical Trials Through Decentralized Methods and Technology With Waqas Haque, MD, MPH, and Shaalan Beg, MD

In this installment of Oncology Data Advisor's Exploring AI in Oncology series, Dr. Waqas Haque speaks with Dr. Shaalan Beg, an Adjunct Associate Professor at the University of Texas (UT) Southwestern, who is spearheading efforts to decentralize clinical trials in oncology and expand access for eligible patients. Dr. Beg describes the tools that are currently in development for decentralizing trials, including the use of artificial intelligence (AI), natural language processing, and telehealth. Additionally, he shares his perspectives on transitioning to a career in industry.  

Waqas Haque, MD, MPH: Welcome to our latest episode of Oncology Data Advisor in our AI and technology series. My name is Waqas. I'm a third-year Internal Medicine Resident at New York University (NYU). I'll be starting Oncology Fellowship at the University of Chicago this year. I went to medical school at UT Southwestern, where I had the pleasure of meeting today's guest, Dr. Shaalan Beg, who has served as the Director of Gastrointestinal (GI) Medical Oncology and has been at UT Southwestern for over a decade. He has since been leading efforts in the decentralized clinical trial space.

Before we get started with today's episode, Dr. Beg, do you want to introduce yourself?

Shaalan Beg, MD: Yes, thank you. I've had the pleasure to work with Waqas for many years now. I work at the intersection of health information technology and clinical research, trying to see how we can use the technology tools that are available to improve evidence generation and how we perform clinical trials, focusing on technology-enabled clinical trial delivery both in the cancer treatment space as well as in the early diagnostic space. I'm super excited to be here.

Dr. Haque: Thanks so much. I want to get started by asking—since a lot of our audience are oncologists in the private practice setting, the community setting, and the academic setting—what's it been like for you to transition to the industry side? What do your day-to-day responsibilities look like on that end?

Dr. Beg: I view my role as not being the traditional industry job. I think when folks refer to industry, they think of pharma positions, either with medium-sized biotech or large pharma where you're actually working for the sponsor. The experience that I've had has been around working with sponsors as a clinical trial investigator, as a clinical trial site, to bring clinical trials to our patients. It has been a very different experience in performing and leading clinical trials in an academic setting versus developing it in a more private setting where there's a lot more emphasis on quality, speed, and delivery.

Again, with the focus that I've had on ways to reduce the burden for trial participation, there is less focus, I would say, on traditional academic metrics such as publications and grants. Now, it's interesting that they both can happen in different environments, but I think the focus and how much energy you dedicate to those different aspects is different depending on what kind of position someone's holding.

Dr. Haque: Got it, thanks for explaining that distinction. I saw that you had a recent viewpoint published in JAMA Oncology that talked about innovative clinical trials. For our audience, can you explain some of the different trial design methods that you talked about in the paper?

Dr. Beg: When we think about ways to improve clinical trial access, we've seen in precision oncology specifically that our protocol development process has really changed in the last 10 or 15 years. The leaders in the space recognize that we can't just keep opening individual protocols for every disease indication every time. We saw the development of basket studies, platform trials, and umbrella trials that are looking to test multiple hypotheses in the same protocol, and that really helps streamline clinical trial oversight and activation and organization. We've seen FDA approvals stem from a lot of those activities, and a lot of those have led to early signals of trials that spun off into other registrational studies.

But they didn't solve a problem that I feel is the most important problem plaguing precision oncology clinical trials and oncology clinical trials in general, and that's access to clinical trials. Most of the sites that were involved in those clinical trials were larger centers or centers that had established clinical trial programs. Most people with cancer are not seen at those centers. So, they were able to innovate and address a big chunk of administrative burden for clinical trials, but not the access issue.

When you see where we are right now due to changes that started before the COVID-19 pandemic, during the COVID-19 pandemic, and since the COVID-19 pandemic, we're seeing that it is very possible for us to incorporate decentralized clinical trial methods to bring clinical trial procedures where patients already are. That could mean letting someone continue seeing their primary oncologist wherever they may be, allowing them to be enrolled onto a clinical trial through a central process that allows all the clinical research and clinical trial activities to take place in a decentralized manner, including the investigator oversight, nursing support, drug shipment to the patient, and using local imaging for evaluating the patients.

This has been happening outside of oncology for a long time, and the oncology space was lagging, because, rightfully, people were worried that people with cancer are sicker and cancer drugs are more toxic than your traditional drug for other indications. But what we're seeing is that for a lot of clinical trials, particularly precision oncology clinical trials, that's not true. The patients are healthier, and the drugs don't have the same side effect profile that cancer therapeutics were notorious for in the '90s.

In this paper, we presented our vision on how decentralized clinical trial methods can be used to modernize how we are performing precision oncology trials and help address the issue of lack of access to clinical trials. We can harmonize the protocols as much as we want, and we can reduce the inclusion criteria or make them more inclusive as much as we can, but if the trials aren't available where the people are already present, then access is very limited. It's very naive for us to think that anybody who's eligible for a clinical trial has the resources, the capability, the strength, and the infrastructure to go to clinical trial centers to participate in clinical studies.

Then we broke down what decentralized methods could really look like in terms of identifying people for precision oncology clinical trials for shipping medications, using local imaging, and monitoring folks so that we can ensure that patient safety is maintained, and that all the while, the patient can continue maintaining their relationship with their primary oncologist and not have to move, which is a win-win.

Dr. Haque: I definitely agree, it's so important to address the lack of access. Only 6% of patients in this country who are eligible for oncology clinical trials end up in a trial. Even if you live 60 minutes outside of a cancer center, it's probably hard to access those trials that you mentioned. Would you like to talk a little bit about any pain points that you see with decentralized trials when it comes to patient selection or the kinds of treatments you use, and how that's being solved?

Dr. Beg: I talked a little bit about it when we were talking about the previous topic and the challenge is that not every oncology clinical trial can incorporate decentralized methods. There are always going to be medications that have a side effect profile that does not lend itself well to decentralized methods, or there's a patient population that will not lend itself well to decentralized methods. Inpatient centers are an example. There are early trials where we don't have the side effect profile; the toxicity profile of the agent is unknown, and that's another major example.

It's important that we don't spend energy in trying to apply these methods to the wrong trials and that we focus on the studies where we are ready to try and make that happen. It comes down to demonstrating that we can ensure participant safety so they're not at any increased risk by participating in a trial that's supported with decentralized methods, ensuring that we can collect the relevant end points in a compliant manner. For most oncology studies, we are able to prove both of those points.

During the COVID-19 pandemic, there were many studies that came out that evaluated emergency room (ER) visit incidence, quality of life, and toxicity for people who were just managed using tele-oncology versus those who were using brick and mortar services, and there was no difference. In some instances, it may actually have been better. That's not to say that in-person services are never required, but if I'm sitting in Dallas and I enroll somebody who lives two and a half hours away in West Texas, my oversight for them, while they're not under my direct purview, is very similar to the processes that we use when we incorporate decentralized methods to manage someone on a clinical trial.

Then when it comes to the endpoints, most of the oncology trial end points are survival-based and progression-based end points, quality of life surveys, performance status end points, or blood sample collections that we can very simply perform in a decentralized manner using home-based nursing, using telemedicine approaches, using local labs, and using local imaging facilities. That's why I believe this is going to be a large part of how clinical trials will be developed moving forward, both due to pressure from our sponsors, but also from the desire of cancer programs and cancer centers to better serve the people who are in their areas and to broaden their radius of influence for people who can enroll on clinical trials.

The FDA has released draft guidance, and we're waiting for what the final guidance will look like, but they've delivered draft guidance which essentially speaks to their support for utilizing these methods in situations where the investigators feel that it's allowable. I don't think that we should ever come up with a list of studies that can or cannot use decentralized methods. Rather, it's a mindset for us to think about. What are the procedures where we don't require people to come in? It could be the 15-hour pharmacokinetics (PK). It could be a survey. It could be electronic consent or prescreening activities using local facilities for scans, like I mentioned.

Dr. Haque: Thanks for sharing all that. Shifting gears a little bit, everyone talks about how AI is affecting this or that. I don't know if there's a clinical trial generative pretrained transformer (GPT) that's out yet. Would you like to talk a little bit about how machine learning and AI might impact some of the work being done to facilitate decentralized trials?

Dr. Beg: One of the areas where we're already seeing a lot of products being developed is in patient identification, recruitment, and ways to use natural language processing methods to find people who are eligible for clinical trials. They still have a long way to go in many instances. For example, one of the criteria that is hard to develop, especially before the AI awareness came about, is line of therapy. But with natural language processing method, that's something which can be inferred for a lot of patients and improve our ability to find patients.

We're still plagued by lack of interoperability between systems, so it's hard to screen people who don't belong to the health system where the trial is available. We are seeing really novel efforts using health information exchanges to get some of that data in a searchable format. There is a lot of development in patient identification and recruitment, which I'm really excited about. I think we're going to start seeing really good impact at centers even this year in terms of making it more feasible to find the right patient at the right time of their journey for the trials. The other aspect is being able to monitor patients using wearable devices or having risk screening tools in the background in the electronic medical records to predict who may be at high risk for developing immune toxicities or pneumonitis from newer antibody-drug conjugates. Those are going to start making their way into clinical trials as well.

On the industry and contract research organization (CRO) side of things, they are increasing their use of AI technologies to perform risk-based monitoring for patients to be able to look for new trends in toxicities and markers of efficacy, which is going to, I hope, turn on its head how data is collected and what type of data can be submitted to even regulatory agencies for either label expansion or to get their approvals when people try to glean more insights into the data that they've collected for their clinical trials.

Dr. Haque: Thanks for sharing that. I've actually seen some early-stage studies looking at remote monitoring for a lot of different cellular therapies where patients can still be at home and have a vital side monitor to look for early evidence of things like cytokine release syndrome and other kinds of neurotoxicities. Obviously, you can have risk and benefits of all those different things.

Very briefly, just to tie things back to the beginning, we were talking about what an industry job looks like and what it's like to be in a non–patient-facing role. Can you give any advice for physicians considering making a transition or any perspectives on that?

Dr. Beg: I think it's really important to know what excites you and what you want to learn and how you want to spend your time. I am seeing with current trainees and recent graduates that they are not necessarily viewing their clinical careers in the way that my cohort did or folks before me did. It's daunting, and it's also very liberating at the same time.

One of my mentors talked about having a framework where somebody looks to reinvent themselves every now and then and learns a new skill, something to keep them engaged, something to grow with internally. There's some debate on how often that should happen and what the right frequency for that should be. Most folks will say that it's around five to seven years where you see that people have plateaued in the work they're doing, and they're looking for the next sort of wave of growth and opportunities. So, I don't view industry positions or non-clinical positions as something to target. I think it's for the person to decide at the right time if they're looking for another growth opportunity, and if that's where they want to grow, I think it's important for them to consider.

There are also opportunities in nonprofit organizations, and there are tremendously increasing opportunities in the health tech space where they are looking for physicians. What I'm hoping is with the development of these newer AI-enabled tools, that physicians can start developing their own minimal, viable product for an idea that they have, even if they don't have coding experience and those skills. I'm hoping that people will start jumping more into that space and that we'll see much more innovative and practice-changing tools being developed by clinicians themselves.

I guess the answer to your question, in a roundabout way, is that it needs to come from inside. I don't think that trying to assign a certain label or external metric is sustainable or is the best way to think about it. Generally, history will say that it's the five- to seven-year mark where people are looking to reinvent themselves. A lot of folks are able to do that within the institution that they're currently working at, where they assume a different responsibility, develop a different program, take on a different administrative job, or have opportunities to consult and help other people with their work. I think even in academic programs, we're seeing more flexibility in the type of work that people are being involved in, that is being given credit in pharmacy and therapeutics (P&T) committees and is being recognized as meaningful and significant.

Dr. Haque: Thanks for sharing that. Well, that's all the time we have today for this Oncology Data Advisor episode. Thanks so much, Dr. Beg, again, for your time. I look forward to working with you and having you on again in the future.

Dr. Beg: Thanks for having me.

About Dr. Haque and Dr. Beg

Waqas Haque, MD, MPH, is a third-year Internal Medicine Resident at NYU in a Clinical Investigator Track. He is an incoming Oncology Fellow at the University of Chicago. Dr. Haque's research interests include innovative clinical trial design, value-based care delivery to cancer patients, utilizing technology in cancer care and research, and becoming an early-stage clinical investigator.

Shaalan Beg, MD, is an Adjunct Associate Professor of Medicine at UT Southwestern Medical Center, where he formerly served as Director of GI Medical Oncology. He is currently utilizing his wealth of experience as a clinical investigator by working in the industry field to lead efforts in decentralizing clinical trials and executing plans to modernize precision oncology. In addition, Dr. Beg serves as an Associate Editor for the ASCO Daily News and as an advisor to the White House Cancer Moonshot.

For More Information

Beg MS & Subbiah V (2024). Modernize precision oncology with decentralized trial tools. JAMA Oncol. [Epub ahead of print] DOI:10.1001/jamaoncol.2023.6786

US Food & Drug Administration (2023). Decentralized clinical trials for drugs, biological products, and devices: guidance for industry, investigators, and other stakeholders. Available at: https://www.fda.gov/media/167696/download?attachment

Transcript edited for clarity. Any views expressed above are the speaker's own and do not necessarily reflect those of Oncology Data Advisor. 




Related Posts