Christopher Crane, MD, discusses modern innovative techniques in GI cancers.
Christopher Crane, MD
Novel radiation technologies, such as MRI-guided radiation treatment and proton radiation therapy, are the way of the future for image-guided radiotherapy in gastrointestinal (GI) malignancies, according to Christopher Crane, MD.
“Both of these modalities enable us to overcome limitations that we have had in the past to deliver definitive curative doses of radiation,” he said.
During the 2017 OncLive® State of the Science SummitTM on Gastrointestinal Cancers, Crane, vice chair of the Department of Radiation Oncology at Memorial Sloan Kettering Cancer Center, discussed modern innovative techniques in GI cancers. In an interview during the meeting, Crane shed light on the progress of these technologies in pancreatic cancer and hepatocellular carcinoma.Crane: My talk focused on some of the newer technologies and the applications of those technologies in GI malignancies, and how that is producing better outcomes as definitive treatment. Specifically, there is proton therapy for liver tumors, and the image-guided therapies, specifically magnetic resonance linear (MR-Linac) accelerator that enables us to deliver more precise treatment to tumors that are near the GI tract.
Many tumors can be controlled and cured with high enough doses of radiation, but the problem is that the GI tract is very sensitive to radiation. If we are indiscriminate about where the radiation goes, we can only give a palliative dose. That is the way things have been for the past 40 years.
These new technologies have helped up overcomes those limitations, with pancreatic cancer as an example. We are giving 2 to 3 times the normal dose of radiation to these malignancies, and we have now—for the first time ever—a tail to the survival curve. This is something that is very difficult to do, and the limitations are that you have internal organ motion to deal with. Either it is respiratory motion or you have day-to-day organ motion, so you must have solutions for those challenges. These tools, specifically the image guidance, help us overcome those challenges. These are things that I have innovated over the past 10 years—using tools that are not as good as this.
The MR-Linac accelerator will make it possible for people who are practicing at smaller institutions, including community practices, to perform this treatment. For instance, I have worked at large institutions, such as The University of Texas MD Anderson Cancer Center and Memorial Sloan Kettering Cancer Center, where we have a huge physics team and sub-specialized physicists who can come together as a team to overcome these challenges. This machine makes it easier for people in smaller practices to replicate that. That makes this treatment scalable, whereas before it was something that only academic places could do.
In the near future, 5 years or so, you'll see other institutions reproducing what we have done, and some people have already started to do that with this tool and have results that are very similar. By that, I mean unprecedented results that have appeared to be better than surgery—all without an operation. The results from Washington University in St. Louis indicate that 80% of patients are alive at 20 months with locally advanced unresectable pancreatic cancer. Normally, you have about 20% alive, so it’s not even close. This isn't [based on] selection because there is no way to select patients like that. We have similar outcomes, but with a 9-year follow-up. The bottom line is that it gives people hope, it’s not a solution for all pancreatic cancer but it does give people hope to experience a similar outcome to what an operation would be—as this is for inoperable patients.
A similar thing is possible with liver tumors. We are able to ablate very large liver tumors, but proton therapy would be what we would select in that situation because it is able to spare the liver better. The liver is very sensitive to radiation, but small amounts of the liver can take very high doses. We want to protect a certain fraction of the liver, about one-third to one-half, similar to how a surgeon would. We want to protect that portion from radiation and that is what proton therapy does, it enters the body and stops at whatever point we want it to.
Those are 2 different technologies used for different reasons—to achieve the goal of a higher, what I call "ablative dose" of radiation—so that everyone understands what the intent is. It is not to slow the tumor down, it is to try to ablate it. The results have indicated that the local tumor is controlled on the order of 90% for as long as the patient lives. We still have the goals of distant metastatic disease in pancreatic cancer and similar for liver tumors, but people will live longer.It is a big educational challenge, and I don't know how that is going to happen, but it will probably have to happen though training programs. The people who will be leading the way after us will be academic institutions, so they are going to train people. There are some people who we have trained for junior faculty positions who are already well-versed in this technology. People who are trained in it and excited about it will start their own practices or bring it to others, potentially in the community.
Believe me, it is nerve-racking for people. However, the reality is that if we give the treatment over a course of 15 or 25 treatments, it is very forgiving. I have been able to mentor and co-manage patients with the junior faculty who are not specialized, and they have been able to pick it up and become more independent. I cannot imagine someone who has not been trained to do this doing it without someone holding their hand. For the next 10 years, it may just be an academic tool but, like any academic tool, people will be motivated to go to large practices and convince them to buy the tool. Once the results are well known and more believable to most people, it will spread that way. With the proton therapy, which has proliferated quite a bit throughout the country, we are pretty mature in the clinical trials that we have done from the GI standpoint. There are always questions to answer, and both are niche therapies. The MR-Linac accelerator makes a lot of sense for GI cancers because we are looking to give high doses near critical structure and we need image guidance. For other things where that is not an issue, it does not make sense.
We started the proton therapy trials in GI at The University of Texas MD Anderson Cancer Center 10 years ago. We have gone from phase I/II to phase III trials in the tumors where it made sense. Now we are doing a trial in liver cancer at the national level looking to prove that protons are better than X-rays. We also are doing randomized phase III trial in esophageal cancer looking at whether lung and heart toxicity can be reduced to the extent that it improves morbidity and possibly mortality.
Therefore, we think that the proton therapy will have less cardiac and less pulmonary morbidity and that will lead to fewer perioperative complications, less long-term heart mobility and mortality, and that will translate into a survival benefit. I have a hard time convincing some of my colleagues who do not understand it, but we know that late cardiac morbidity is a problem for GEJ cancers. We also know that perioperative acute respiratory distress syndrome is a big cause of morbidity and mortality in patients that undergo surgery after preoperative chemoradiation. Preliminary data indicate that there could even be a survival benefit, if not just a major morbidity benefit to the use of protons, just by confining the radiation away from the lung and heart better than we can with photons. We always should try to think about doing randomized phase III trials. We are at the phase I/II dose-escalation trial for pancreatic cancer, and the next steps are going to be getting enough people who feel comfortable doing that so that we can ask a phase III question. It is always a challenge with these new technologies because of the quality assurance of not only planning the radiation, but the delivery of it. We have to make sure everyone is doing that right.
It is a big challenge. It is something that we are working on and considering, but it is going to be hard to get phase III data with this kind of approach. There is no way to get a tail to the survival curve without effective local control. There are very few things that can do that; we can't pick patients who we know are going to be alive at 3 years. It just doesn't happen that often.