SGRT Implementation in a Large Free Standing Practice

My name is Scott Tannehill. I’m a physician with the Arizona Center for Cancer Care. We’re a freestanding, privately physician-owned, physician-led practice in the Phoenix area. Before we go too much further, I will add that I have no conflicts of interest. Vision RT, as with many of the speakers, covered my expenses, but I get nothing for this. Nada. Zero. Where’s Dan, the salesperson? Zero. No, it’s my pleasure to be here. I really believe in this technology, and it really is a pleasure to be here to talk about it.

So before we go too much further, how many of you here in the room are from institutions or practices considering an SGRT purchase? Okay, haven’t committed yet, but you’re considering it. And for those of you with your hands up, how many was that prompted by the billing changes in January? Just a couple. All right. Good. It’s a reality, and we have to face that, the seismic changes that we experience.

So how many of you are actively using the system in here for ongoing education, increased utilization, and so forth? Good, the majority of people. We’re all keeping up with current advances. How many of you are here because you scored a trip to Disney World? I see a few sheepish hands. Okay, good.

So my discussion’s going to be very casual because I’m a casual person—forthright, genuine, honest, warts and all, about everything. I’ll talk about the background and rationale for our implementation of SGRT, the process, some recommendations that I would make as it relates mainly to a practice of our design, and then some summary and personal experience. I was very impressed with Dr. Wang’s wonderful presentation this morning; I have my own personal experience with SGRT as well.

So we are a multidisciplinary group in the Phoenix area. We have all of these specialties: medical oncology, GYN, radiation oncology, breast surgery, surgical oncology, urology, dermatology, imaging. We have over 100 providers. Those show all of our sites. We have eight radiation oncology locations, 10 vaults total, four PET/CT scanners, four HDR units. I think a lot of people here are with hospital systems where you have a dominant central facility, and you may have multiple vaults, multiple CT sims. We are scattered because of the nature of our practice; we go to where the business was in the development, and so we don’t have a central hub. Each individual clinic has its own CT sim, and usually one, but sometimes two. We have a variety of equipment. We have an evolving category of machines—TrueBeams, several Halcyons, but we also have some legacy units, a VitalBeam, which is a TrueBeam light, and then we actually have a couple of older iXs that are sunsetting but are still being used for a variety of treatment categories. And so we have this broad repertoire of technology, which of course adds complexity. It’d be nice if we could just clean the slate and put in 10 TrueBeams, but I would be kicked out of the group if I proposed that.

So going back to the history of our implementation, the COVID-19 pandemic taught our practice a lot of lessons. It seems like it was just yesterday, but it seems like forever in some respects because it was six years ago. Hopefully, we never go through anything like that again, but it gave us a new appreciation of our vulnerability as a practice and a business. Indeed, medicine is a science of uncertainty and an art of probability. So uncertainty and probability is in our blood as providers; we have to be prepared for that. So we did a top-down practice review amongst the leadership, and our goal was to position ourselves to provide the practice the highest quality, competitive, and cost-effective edge for the community and to face unknown future challenges. You can say, well, you can’t predict the future, but you can plan for it. General Eisenhower’s comment that “no battle plan survives first contact with the enemy” is true, but the act of planning prepares you for these things.

What kind of challenges? Reimbursement cuts—this was back in 2021, 2022—increasing costs, and staffing shortages. These issues are just as important now as they were then. We’re still struggling with the consequences of the pandemic, the pipeline for therapists, for physicists. It’s a significant issue. And then there are these dubious solutions from conflicted entities. A hospital system that I used to work at—a colleague of mine, a former partner, said, “Oh, the hospital is worried about the future, so they’re thinking of buying an MR Linac.” I’m like, okay, that’s like a family running out of money saying, “Our plan is to buy a lottery ticket.” It’s not a solution; that’s just making the problem worse. That’s disguising the problem with a pseudo-solution. So I always call it “the shiny.” You must resist the shiny. The shiny is an MR Linac, a compact proton facility, a Zap, a Crush, a whizmo, a gidget, whatever it is that is a reaction rather than a plan. I’m not denigrating the gadget, the whizmo, the Zap, or the MR Linac—maybe I am a little bit—but those aren’t solutions to the problems that we’re talking about.

The future will require that we treat more patients in less time with fewer resources (budgets, staff) and with greater precision and quality, because that’s what our patients deserve. That’s what the specialty will expect, without compromise.

The specific catalyst for SGRT for us was, as always, an odd casual thing. I was going to ASTRO, and my partner said, “The Calypso support ended.” How many of you ever used Calypso? Thank you. It was neat. It was never really reimbursed or supported terribly well, but it had marketing value. That was the main value, and I was never a big fan of it. But my partner, who did a lot of prostate cancer, was impressed with it, and he was bemoaning the lack of Calypso support. The whole Calypso system was basically uninstalled and was sitting in the corner of our physics area for 18 months until someone threw it in the trash. I don’t know what happened to it. Then I went to ASTRO in 2022 to do my vendor floor recon. As you all know, ASTRO now has all the educational and research programs available online, so you can watch those later. When I was a resident, you had to try to find someone that went, get the copies, make copies, and share those copies—that was in 1993. Now, 100% of those presentations you can view. So when I go to ASTRO, I walk the vendor floor because 0% of those are available after you leave the meeting. I know it sounds unprofessional, but I can’t sit still. That’s why I was sitting in the back, because I have to run out sometimes. I can’t sit through eight hours of research and clinical education meetings, so I walk the floor. Our practice likes to send a partner to walk the floor and find out what’s new. You find a lot of interesting things that way.

This TruePose headrest that Standard Imaging has, which I’ll talk about shortly, is one of those items that I saw on that walk in 2022. It’s a 3DOF headrest. One can argue this is a solution looking for a problem, but we can talk about that. Another little simple thing, the Beekley Medical RTSpot marking wire. How many of you dosimetrists here were irritated when you had to go in and manually delete the wire from the contours? That arm went up really, really strong. This has a little cushion between the wire and the skin surface, and it can be effortlessly, automatically deleted. My chief physicist tells us it saved us an hour of dosimetry time per week—I don’t know if that was an exaggeration—but the point was we didn’t know about this until I walked the vendor floor at ASTRO. I walked up to every booth and said, “What do you sell? Okay, sell it to me. Tell me why I need this.” Just the act of that was very synergistic in developing ways to improve our efficiency.

Transparent bolus—Gary can show you the formable transparent bolus. There are many different vendor products for this. It saves you time setting up electron fields in the room. If you save five minutes trying to figure out if you’re on the target or not, these types of things can add up.

So I ran into the Vision RT folks. Also looked at the C-Rad and the Varian products. Clarity, of course, didn’t have a product back then, and we elected to pursue Vision RT. We selected Vision RT as our vendor. As we always do, I called everyone; it’s a small field, most of us know people. It’s like the seven ways of Kevin Bacon thing where someone knows somebody who knows everybody. I had enough conversations with people to decide that Vision RT was the best resource: recommendations of satisfied customers, an understanding of why some people were not satisfied, and good communication from them during the sales process.

At the time, since we’re a business, the G-code reimbursement obviously affected our favorable inclination. My personal recommendation was to start with one system and test it thoroughly. But at the presentation, one of my partners—the practice CMO, CEO, a very smart doctor, Dr. Christopher Biggs—at the end of the meeting, I don’t know if a couple bottles of wine had something to do with it or not, but he said, “No, let’s buy one for every vault.” In retrospect, that was very insightful. That commitment improves the implementation; you say, “We’re all in,” and there isn’t any easy way for folks to back out. It allowed us to daisy-chain the training process: one site was trained, and those people helped train the next site geographically. And then, of course, you get a better price at scale.

We started with one with local champions; the word “champion” and other words you’ll see in my talk were not because I colluded with other speakers. This is parallel evolution of information. We came to the same conclusions independently, and we’re even using the same language. We had the site visit by the vendor engineers and the prep of the facilities. The Vision RT folks came and said, “Hey, have your facilities people install the posts in your concrete lid.” My physicist and I looked at each other like the Spider-Man meme: “Facilities?” We ended up buying a hammer drill and actually drilled the mounts for the first two posts because we wanted to see what was involved. After I got a bunch of concrete dust out of my hair and eyes, we farmed that out to a contractor once we realized what was involved. Then, training on-site by the vendor, and then wash, rinse, repeat. The training was excellent.

In terms of the outcomes of the implementation, I’m going to give you a tale of two clinics. I’m at the Gilbert location; that was the first clinic where it was installed. The second clinic was nearby in our Tempe Chandler office. Clinic number one had a tight team that had been together for almost 20 years, firmly established routines, no experience with SGRT. I anointed someone to be the champion who was a manager but not normally present at that site—my first mistake. This clinic had a busy schedule, 50 to 60 patients a day—my second mistake. And we didn’t have an offline workstation available at the initial install—my third mistake.

That was contrasted with the next clinic at the Tempe office, a relatively new team with no “ruts” in their process. We had a newly hired lead RTT from another state that had extensive SGRT experience and was a fan. It was less busy, and we put an offline workstation there at the start. Not surprisingly, the first clinic where I was present didn’t go well, and the second clinic was almost effortlessly successful. We learned our lessons: have a champion, and have a champion who is respected by the team. My bias was to hire a young therapist who was tech-savvy, but I would say hire someone who has respect—more so than authority—and is a fan of the technology. That champion has more influence on the rapid success than any other variable.

At this point, we’ve implemented this at all of our sites. Most sites experienced some variation in that quality of implementation, from slow and painful to effortlessly successful. It was clear that people with prior experience helped. As the team understood the value, the implementation improved. I polled most of my leads, and all of them say, “Yeah, we would definitely miss this technology. We don’t want to not have this anymore.”

What’s an easy way to gently introduce the benefits of SGRT to potentially resistant staff? Go slow. What would you show them first? The postural video, right? Neat pictures, easy, helps fast. That gets your foot in the door; it helps change the mentality of “Why do we need this? This is more work.”

To recap: start at a slower site, add an offline workstation to allow the staff to import new patients from the simulation, draw their ROIs, etc., during the workday. Anoint a champion with experience who has the respect of the staff.

“Doomed to fail”—and these were all the unsatisfied customers I talked to—is a large bureaucracy-heavy hospital where SGRT was bundled with other purchases without input from anybody and then dumped on the department. If they’re just told, “Just do it,” it’s doomed to fail.

Regarding patient satisfaction, my example was a brain tumor patient I treated three years ago, still doing very well, who was a little high-strung. After their first treatment, he wanted to see me right away, almost in tears. He had to swallow hard twice during the arcs and was worried he moved. I showed him his treatment log, and I said, “You didn’t move a lick. Here it is right here.” Instant comfort. I think our industry underemphasizes the psychological value of this for reassuring patients.

We are now tattoo-free; we don’t have any skin marks as a routine part of any setup. We use the postural video for initial positioning, and we use it for all photon treatments almost exclusively, including DIBH for the lung, the breast, the upper abdomen. I replace IGRT for some treatments now. This is a slow, difficult process; we came from a culture of heavy IGRT use. But simple tangents and many palliative plans don’t need any imaging. In some respects, the coding and billing changes have released our conflict of trying to overuse image guidance, which we got paid for before. Now we don’t need to do a cone beam for that simple breast tangent; we can clearly see they’re lined up, and we’ve validated that. We are in the process of altering our technique for image guidance, using it on the non-SRS brain tumors, the head and necks, even the DIBH lung and breast patients, where we just use SGRT three to four days a week and then do a cone beam one or two days a week to look for specific anatomic changes. I can see someone who does not want to do that if you’re treating internal mammary nodes in a breast, I would understand.

But cone beams for the brain are still good because I do a lot of brain tumors, and I teach my staff that just because you’re setting them up doesn’t mean things haven’t changed. These calcifications that you’ll see in the brain—choroid plexus calcifications or calcifications along the dura—if you have a brain tumor patient on a cone beam, you can see those shift. It’s very common for you to see a little wiggle on the cone beam; that tells you there’s edema or mass effect that prompts you to consider a repeat simulation because the anatomy shifted. So there will always be value.

On the subject of using SGRT to improve efficiency: at one of our sites, even with the six-DOF couch, we were running out of pitch on some patients—head and neck, brain patients in the open-face mask, and pelvis patients. One could argue that that’s a process problem, but poor pitch correction led us to troubleshoot. 90% of issues are what I call R&R: real estate. You don’t have enough surface anatomy being seen by the system for an accurate picture, and your ROIs need to be adjusted. This is the highest learning curve; it’s as much art as science. Having your staff learn how to solve these problems with ROI adjustments and mask making is important. The solution is often upstream of the vault, in the CT simulation. The TruPose headrest, this 3DOF headrest, is actually useful for us; in concert with SGRT, it allows us to get the benefits of a 6DOF couch without having to take the machine down and spend $300,000. It’s not suitable for most brain SRS because of metallic elements, but it works for coplanar brains and head and neck.

To recap the timeline: 1993, skin tattoos and orthogonal X-ray simulations with grease pencils. Mid to late ’90s, 3D planning. Early 2000s, IMRT and early SGRT usage. 2000s, Tomotherapy—I know Tomotherapy really well; the first gantry put a hole in my office wall at the University of Wisconsin-Madison. Mid-2000s, integrated cone beam CT. 2026, we’re back to skin, except we use all of the skin.

Finally, my personal experience: in the middle of our implementation, I was diagnosed with localized prostate cancer. I elected not to be treated in my own center—too much pressure—and not at the competition. I chose an academic facility in Phoenix; they had SGRT and two beam-matched TrueBeam vaults. The therapist didn’t know I was a physician. I told my friends there, “I just want to be a patient.” Best decision I ever made. But one day, I’m in the room, they’re about to start, and I say, “Stop! Something’s wrong!” The therapist asked what was wrong, and I said, “Where’s the red light?”

I had been on a machine that had the red light—the Vision RT system—and I got used to it. There was a rhythm: lay on the table, position, cone beam, red light treatment, the buzz ended, I could relax. When I went to the machine next door that was newer and didn’t have the red light, I panicked. The therapist calmly explained they didn’t have a red light on that machine. I learned from that that with all I knew, I was still childlike in that moment. You’re still vulnerable as a patient. The SGRT advantage is that it “has your back,” as a patient and as a clinician. I tell my patients about the camera system; it’s very reassuring. We’re in a unique position to have that perspective as clinicians and as cancer patients. Thank you.