Christopher Thompson, MD, MSc, FASGE, FACG, AGAF, FJGES, director of endoscopy and codirector of the Center for Weight Management and Wellness at Brigham & Women’s Hospital, professor of medicine at Harvard Medical School, discusses duodenal jejunal bypass liner treatment and future innovations in gastroenterology.
Christopher Thompson, MD, MSc, FASGE, FACG, AGAF, FJGES, director of endoscopy and codirector of the Center for Weight Management and Wellness at Brigham & Women’s Hospital, professor of medicine at Harvard Medical School,discusses findings on treatment with duodenal jejunal bypass liner for patients experiencing type 2 diabetes and obesity. This interview took place during the recent annual meeting, Digestive Disease Week.
In this first part of a 2-part interview, Thompson discusses a study involving duodenal jejunal bypass liner (DJBL). This interview is edited lightly for clarity.
AJMC®: When treating patients with type 2 diabetes and obesity who were being treated with insulin, what are some patient experiences you saw firsthand?
Thompson: I'm a gastroenterologist and an endoscopist, so I spend most of my time doing procedures. I do remember taking care of patients with diabetes when I was a resident. I know that once they get on insulin, they start to have a decline in their health; insulin causes weight gain and causes all sorts of other problems—it's not well tolerated. So, I guess I would say this is kind of a unique niche for gastroenterologists to care for obesity. There's a growing number of us but since I've really started caring for obesity, I started to see these diabetic patients again, and nothing seemed to have improved. I mean, it was so many years ago, 20 years ago, when I was a resident—actually more than that probably—taking care of patients with diabetes and it seemed like it was very similar. There's more oral antidiabetic medications available now, but still, once they go on insulin, they don't seem to do very well. So, it's definitely time for a change, and whatever technology we can have to help these patients I think would be very welcomed.
AJMC®: Can you first explain how the duodenal jejunal bypass liner (DJBL) works, and what makes it different from other treatments for patients with T2D and obesity trying to achieve similar results (such as bariatric surgery)? Why is this new method needed?
Thompson: The DJBL is quite unique in that it uses the body's physiology and an understanding of the physiology to treat type 2 diabetes. What it's doing is sheltering the duodenum from seeing any food, and it's also preventing the mixing of food with bile or pancreatic juices. Well, I could start by saying the DJBL is a 60-centimeter-long sheath, like a sleeve, is anchored in the duodenal bulb and extends all the way to just about the jejunum. What it does is it shelters the duodenal from contact with food or chyme, and it prevents the mixing of that food or chyme with bile and/or pancreatic juices. So, it is getting this undigested food and bile to the distal small bowel in a state where it has a pretty dramatic effect on incretin-producing cells—we think—such as L-cells, so that we're exploiting the cells production of GLP-1 and probably other substances that then have some beneficial effects on type 2 diabetes. They're known to increase insulin production and decrease glucagon, and cross the blood-brain barrier as a neurotransmitter and signal that the patient may be full, also slows gastric emptying. These substances, these gut hormones, have several beneficial effects.
AJMC®: Can you summarize your research and findings into the utilization of the DJBL for this patient group?
Thompson: This was an FDA [approved] clinical trial, and they had some predefined end points. It was a randomized, sham-controlled trial with 2-to-1 randomization, so patients would be randomized. For every 2 patients, they got the procedure, 1 would get a sham procedure. The patients were blinded, as were the investigators that were following them. The predefined end points were both related to safety as well as efficacy. The efficacy end point involved hemoglobin A1C (A1C), and they wanted to see a change—an improvement—in A1C of at least .4% above the sham. Then, the safety end point was that they wanted the rate of serious adverse event-related device removal to be less than 15%.
So, the study did meet both of those end points, those primary end points. … the treatment group was 1.1% drop in A1C. The sham was .3%. Clearly, it was well above that .4% margin, as an efficacy end point. The safety end point, I think it was just above 9% removals due to SAEs [serious adverse event-related device removal] and that's well below the 15%. So, it did meet both of those end points. However, what did also occur in this study was that the hepatic abscess rate was unexpectedly higher than they thought. This was not something they anticipated. All the patients did well, all hepatic abscesses resolved with antibiotics and some cases some percutaneous drainage. But that was just something that caught them off guard, so the company stopped the trial due to that.1
AJMC®: Is there another study taking place as a result of these findings?
Thompson: There’s another registry trial run by a Dr Bob Ryder out of England. He's an endocrinologist. He has over 1000 patients in that and the hepatic abscess rate is only 1.1%, I believe. That's a lower rate than we were seeing in our study, but it has much more patients. It turns out that in our protocol, the PPI [proton pump inhibitor], we had maximum-dose PPIs; we were using very high-dose PPIs, and the rest of these patients the registry, they weren't. It makes sense that PPIs are certainly—they’re an independent risk factor for hepatic acid formation, and very high doses in diabetic population certainly could be the cause.
We started another trial. Now, we do not have the patients on PPIs. We’re using H2-receptor antagonists, and they're doing really quite well.2 We've not seen any recurrence of hepatic abscesses so far. We've also instituted some measures to make sure that we would catch anything early as well. The patients have Wifi-enabled thermometers, and they take their temperatures every day, and they get monthly ultrasounds to look at the liver. We've not seen any further hepatic abscesses since we stopped PPIs, so we're cautiously optimistic
AJMC®: How did this study ensure a diverse population? How do the findings reflect the needs of patients in various demographics?
Thompson: It is an FDA [approved] clinical trial, so they have in there—kind of baked into it—they make sure you have a certain amount of diversity. What you try to do is find centers that are in different locations that would make sure to pull in a nice, wide, diverse population. We did, I think, have quite good diversity. Over 50% were women—of the patients—and we had a good breakdown from underrepresented minority status as well.
AJMC®: What are the long-term implications of DJBLs? This study had a 12-month follow-up; are there any implications for a longer follow-up?
Thompson: Some other folks have looked at this. Ours—you know—the analysis was done at 12 months because the sham patients were then offered an open-label placement of the device. That's really all you could do as far as the analysis goes, but other studies have looked at it more longitudinally. It seems like the results you see during the treatment period persist well after device removal, and it's not clear exactly why that's happening. It might be due to some kind of remodeling of the duodenal mucosa. There are other companies are looking at just ablating the mucosa, and they're getting good anti-diabetic effects as well. So, it might be that the duodenum itself changes over time, and that's what leads to the durability of the results.
AJMC®: What other lifestyle changes do patients with DJBLs need to make?
Thompson: I think lifestyle is hugely important in achievement of obesity and diabetes, and keep in mind this population was a population of patients with obesity and diabetes, so it wasn't thin type 2 diabetics, which is a different population. No one had type 1 diabetes. Additionally, no one in this study was on insulin. That’s probably another important point to make, that these were people that you were trying to prevent them from going on insulin, right? It's not to say that it might not help get people off insulin, but I think that just the focus of the trial was to have a uniform population in a sense, and give it enough power and be able to understand the results. We had limited to people that were kind of poorly controlled diabetics, so their A1C had to be above 7.5% to 10%, somewhere in that range, and kind of maximal therapy, if you would, just short of insulin. We were trying to kind of look at that population specifically.