Mentors and Protégés: Conversations on Career and Craft With Riad Salem, MD
Alexander Logsdon, DO, and Adam Sucher, DO, talk with Dr. Salem about balancing research and clinical responsibilities, his journey in establishing Y-90 as a standard of care, and his advice to aspiring clinical researchers, among other topics.
Can you give us some background about yourself and where you did your training?
I am originally from Montreal, Quebec, Canada, and I graduated from McGill University in Montreal. I completed my residency training in Washington, DC, at George Washington University Hospital and simultaneously received my Master of Business Administration degree (finance) from George Washington University. Then, I pursued a fellowship at the University of Pennsylvania in Philadelphia, Pennsylvania. Once I fulfilled these training requirements, I spent about a year and a half at the University of Wisconsin in Madison, Wisconsin. That time was very exciting because I was able to fine-tune the skills I had developed during my residency and fellowship. After my time in Madison, I worked at Beaumont Health in Michigan for 3 years. From there, I moved to Northwestern Medicine in Chicago, Illinois, where I have been for the past 16 years.
How did you become involved in your research projects, such as radioembolization using yttrium-90 (Y-90)?
Research is something that physicians should evolve into and learn to appreciate. Sometimes physicians rush into research. Instead, when physicians complete their training, they should spend the first 2 or 3 years fine-tuning their technical skills. As an interventional radiologist, the skills and techniques required are extremely diverse in order to be able to perform the variety of tasks you may be presented with. The reality is, once your fellowship has concluded, you still aren’t done learning. The real-life experience begins! These experiences include things such as learning to manage acute decompensations of patients, fine-tuning hand-eye coordination, and increasing your knowledge of catheter/wire technique, among others. The first few years after your fellowship is where you have a great opportunity to fine-tune your craft. My recommendation is to focus on clinical expertise during those years. Then, once you’ve established that skill set, it's a lot of fun to be intellectually and academically curious and initiate a career in research.
I started using Y-90 in 2001 while I was treating patients with chemoembolization at University of Pennsylvania. I found Y-90 very intriguing because its side-effect profile was much lower, the response rate was higher, and people didn't have the pain, nausea, and vomiting that was associated with the use of transarterial chemoembolization (TACE); that’s how I became interested in it. With regard to my Y-90 research, I took a very systematic approach to working, learning, and developing Y-90 as a standard of care. I looked at technical aspects and completed analyses of who we should and should not treat, assessing the response, completing phase 2 studies, and elevating the dialogue of arterial liver-directed therapies. In the early 2000s, many in the interventional radiology (IR) community were heavily focused on TACE and dismissive of the impact of Y-90. But, then, randomized trials and survival data started accumulating, and techniques like segmentectomy and lobectomy were developed. I took a structured, rational, and meticulous approach to building the level of evidence. One thing our Northwestern group is very proud of culminated in a 2018 article in Hepatology, which essentially states that Y-90 has become our standard of care for arterial therapy after using it to treat 1,000 patients with hepatocellular carcinoma (HCC).1
What has the response to this publication in Hepatology been from your peers at other medical institutions?
For those who are already believers in Y-90, this is a culmination and recognition of the research. Of course, there are some who don't agree with that. However, for us, as I explain in the article, performing clinical trials in IR with overall survival as the endpoint using the randomized design without crossover between treatments is effectively impossible. If you recognize and agree that it's impossible to do it that way, then the only way to advance a standard of care is to evaluate other things like downstaging, radiologic-pathologic correlation analyses, time to progression, safety, cost, and response studies. That's what we did with this publication, and we found that no matter how you look at Y-90, it outperformed TACE.
What were some of your biggest challenges in your research, and how did you overcome them?
The first challenge I faced was the universal skepticism by colleagues about my research. TACE was, and actually still is in many areas, the universal standard of care. It does not help to be the lone wolf, but I was willing to try it because I saw something in how patients tolerated the treatment, and I was academically and intellectually curious about it. The skepticism from colleagues, whom you respect and are world authorities in the field, is not easy to deal with, but you have to believe in your conviction. My initial observations of the treatment have ultimately been validated because the major focus of the research of many skeptics today involves Y-90, not TACE.
Another challenge I have faced and still face is with regard to the FDA and approval. The FDA has not given full approval of these devices for the treatment of HCC, which is a problem because the FDA continues to evolve in its thinking when it comes to devices. Even with treatments that cure patients, prolong life, and are demonstrated to do great things for patients, overcoming the regulatory hurdles may be daunting. Long-term, this has not been something we’ve been very successful with.
Finally, industry can sometimes present a challenge, and it did with Y-90. At the time, many in industry weren’t sure what to make of Y-90. There were times when the future of Y-90 was uncertain because adoption was slow. However, once others started to believe in this therapy, growth was exponential, and it continues to be.
What advice would you give young professionals when it comes to starting and performing research?
First, as physicians, we are not trained to perform research. My suggestion would be to start learning and studying scientific and clinical trial methodology, and then try to adopt as much of that as you can into your IR community. Also, don’t be overly dogmatic; you can't have a one-size-fits-all approach to clinical trials. Read, be smart, and design something of interest where there is an unmet clinical need.
Second, be skeptical when reviewing the numbers, and always report outcomes in the most conservative manner. Sometimes, there are various ways to look at numbers and confidence intervals that deal with certain data points. From day one, what I have done is always report the worst possible scenario, because you know that it's only better from there. When other centers report data that are better than mine, that's good. Having been first out with Y-90, I didn't want to be accused of looking at things too optimistically; everything we report is with tremendous caution and conservatism, which sometimes makes our numbers look worse than they actually are. I suggest reporting data that way. As an example, when we first looked at toxicity from Y-90, we reported any biochemical toxicity that occurred subsequent to Y-90 as potentially related, no matter when it occurred. In our publications, liver toxicity 2 years after treatment was reported as possibly being related to Y-90, even if the patient received other locoregional or systemic therapies. In contradistinction, drug therapies often only report toxicities up to 30 days after infusion. Toxicity occurring 35 days after a drug infusion is often not reported even though it could be related. Therefore, report conservatively and then explain away a number that seems out of the ordinary.
My third recommendation is to engage statisticians and clinical trialists in order to learn about studies and how to evaluate data. It can be daunting, but it’s not that difficult if you commit to it. I'm self-taught when it comes to the basics of statistics and clinical trials. But for more advanced concepts, bring in the experts.
Finally, be intellectually curious. If you're interested in something but don't understand what it is, get your database going, look at it in every possible way, think critically about it, report the outcomes conservatively, and then look at it a bit more reasonably. If you apply these principles, it will always pay off.
The field of IR has grown exponentially over the past decade. Where do you see it fitting with other specialties such as vascular and neurosurgery?
Our clinical presence is not as robust as it should be compared to other clinical services, such as neurosurgery or vascular surgery. Hopefully, that will be dealt with by implementation of the newly combined IR/diagnostic radiology (DR) subspecialty. Because other specialties almost always have first access to patients, it's not uncommon that many of our techniques and procedures would be co-opted by other specialties; this can force us to be the perpetual innovators. Vascular intervention is a prime example—30 years ago, no surgeon believed in it, but today, no surgeon doesn't. Many of them are training to perform vascular procedures, so we had to evolve into interventional oncology, uterine fibroid embolization, prostate artery embolization, and all of these other areas. There's no doubt that we are always being pushed and challenged because our procedures are being imitated. Although that is one of the good things about IR, it also can be a drawback. The future will allow for more collaboration than there is now because there are skills that we can learn from each other. Collaborating with many other subspecialties is something all specialties need to improve on.
Which came first, your clinical interest or your research interest?
My research interests are derived from my clinical interests, which I think is a natural way of doing it. The reverse can be trickier.
What has contributed most to your success as a physician and a researcher?
I'm motivated, interested, and curious. If you put those things together and are able to find an environment that allows you to investigate and seek the answers to your curiosity, then you will thrive. Since coming to Northwestern in 2003, my mandate was clear: grow interventional oncology and expand this field as much as possible.
How do you balance the demand of research, clinical duties, and teaching?
I have a formulaic approach to that. Since 2003, I have been on clinical service Monday through Wednesday and performing research Thursday and Friday. That has been my week for the past 16 years. I know exactly what I'm doing on each day, which allows me to organize my schedule and travel responsibilities. As a result, this routine has helped me evolve Y-90 as a standard of care for treating HCC and on the international stage. By not having clinical duties on Thursdays and Fridays, I am able to travel and lecture around the world and be back by Sunday afternoon to fulfill my responsibilities.
What insights would offer to those who are training with hopes of pursuing a career as both a clinician and a researcher?
My recommendation would be to focus first on the basic technical aspects of IR if you want to do both clinical work and research. You cannot be a great interventionalist without great technical skill, and it's hard to be a great researcher without great technical skill. It gives you tremendous credibility when you are not only a good researcher but also a kick-ass interventionalist. Recognize that your training on the research side will always be suboptimal because you cannot fit everything into your fellowship training. But, with the advent of the IR/DR subspecialty, training that will also include research and trial design is not far away.
What advice do you have for physician-scientists to achieve work-life balance?
Work-life balance is very important. I am the first to admit that I probably worked too much early in my career, and I would recommend people focus on establishing a positive work-life balance early on. It is natural to want to work hard once you’re in the field, because you've been in school and in training all your life, and your identity is almost dominated by who you are as an interventionalist when you start your career. However, find a work-life balance early to enjoy the fruits of your labor with your family and children. You can indeed “have it all.”
1. Salem R, Gabr A, Riaz A, et al. Institutional decision to adopt Y90 as primary treatment for hepatocellular carcinoma informed by a 1,000-patient 15-year experience. Hepatology. 2018;68:1429-1440.