Latest / The Staple Line / Vol. 2, No. 14: GLP-1s for weight regain, would you do it again, and an AI that reads the abdominal wall

Transcript

Vol. 2, No. 14: GLP-1s for weight regain, would you do it again, and an AI that reads the abdominal wall · The Staple Line

• 0:00
  [on-hold music] Hey, welcome to The Staple Line, the weekly audio edition of the MBS Digest. I'm John Scott.
• 0:18
  This is our first full episode covering volume two, issue fourteen. This week, can a GLP-1 help when weight creeps back after surgery? Twelve years later, would patients choose surgery again?
• 0:30
  A new phase III drug that took off nearly twenty percent of body weight, and an AI that's learning to read the abdominal wall. Let's get into it.
• 0:40
  Let's start in surgery research with a question that every bariatric program deals with: What do you do when the weight comes back?
• 0:46
  A new study from the One Florida Research Network looked at adults who had regained at least ten percent from their lowest weight after surgery.
• 0:55
  They compared people who started with a GLP-1 drug after similar people who didn't. After adjusting, that was about seventy-two users and a hundred and thirty-nine non-users.
• 1:06
  Semaglutide was the most common drug, just over half. At one year, the GLP-1 users had lost about fifteen percent of their regained weight.
• 1:16
  The non-users, about eight percent, so nearly double, and no significant difference in the overall rate of GI and post-surgical complications. Abdominal pain was actually less common in the users.
• 1:28
  Now, the catch, and it's two things. First, that percentage is of the regained weight, not the total body weight.
• 1:35
  Just to put it in perspective, if someone regained twenty pounds, that's roughly three pounds back off versus under two. Real, but modest. Second, nobody was randomized.
• 1:47
  Patients and doctors chose the treatment, so there may be some differences between the groups that the analysis can't see.
• 1:54
  So the takeaway, this adds real-world support to earlier liraglutide trials that showed the benefit in this setting. But it's an association and a modest one.
• 2:05
  My read, worth discussing with patients with realistic expectations. [on-hold music] The second story, and this one is more about patients than data.
• 2:16
  The Hindsight twenty/twenty study sent a survey to nearly sixteen thousand people who had bariatric surgery at a large academic center, an average of about twelve years earlier.
• 2:28
  About fifteen percent answered, twenty-three hundred and eighty-five people. The headline, about three in four, seventy-four percent, said they'd choose surgery again.
• 2:40
  Seven percent weren't sure, and about eighteen and a half percent said no. The people that said yes pointed to overall health and weight loss.
• 2:49
  For those who were unsure or said no, the top reasons were weight regain at fifty-six percent, then complications outside the operation itself, like low blood sugar and nutrient deficiencies at forty-nine percent, and not losing enough weight at forty-one percent.
• 3:04
  About a quarter of the patients said they'd picked a different operation. The caveat, only fifteen percent responded, and people who answer surveys like this aren't always representative.
• 3:15
  So treat those percentages as a good signal, not a precise rate. My takeaway, the problems people name, regain and nutrition, are exactly the ones follow-up care can work on.
• 3:26
  [on-hold music] Over in obesity science, a name you might not know yet, lotorepotide. It's a weekly injection from the Chinese company Hanso, a dual GLP-1 and GIP drug like tirzepatide.
• 3:42
  Hanso reported phase III results at the European Diabetes Meeting, the LIGHTEN trial. Six hundred and four adults in China with overweight or obesity randomized, double-blinded, placebo-controlled for forty-eight weeks.
• 3:56
  At the top dose, fifteen milligrams, the average weight loss was nineteen point three percent versus two point seven percent on a placebo.
• 4:05
  About seventy-two percent of the people on the top dose lost at least fifteen percent of their body weight. Nausea was about eight percent versus under three percent on the placebo.
• 4:15
  And nobody stopped treatment because of drug-related side effects. Now the fine print. These are the company's numbers from a press release and a conference presentation. They are not peer-reviewed yet.
• 4:27
  The nineteen point three percent assumes everyone stayed on the treatment. Counting everyone as randomized, it was eighteen point seven, so very close.
• 4:36
  The company says its nausea rates are lower than the approved drugs, but that's different across different trials, which doesn't prove much, and it's a Chinese population only.
• 4:46
  A regulatory application has been accepted in China, and Regeneron holds overseas rights. My read is this: very promising but very early. Not something to counsel patients on yet.
• 4:57
  [on-hold music] In technology, the lead is a small study with a big idea. When you enter the abdomen with a direct optical trocar, you're watching your way through the layers of the abdominal wall.
• 5:11
  The authors note that entry is the phase of nearly half of major laparoscopic complications. So this tr- team trained an AI model to label those layers from video: muscle, sheath, and peritoneum.
• 5:24
  They used nearly a thousand video frames from ten patients. The score they report is called DICE, where one is a perfect match with an expert's outline. Overall, point eight eight. Best for muscle, point nine two.
• 5:39
  Lowest for peritoneum, point eight two. And that's the catch. The peritoneum is the last layer you cross, so it's where you want the help the most, and it's the layer the model did the worst on.
• 5:52
  Also, this measured agreement with expert labels, not whether patients were actually injured less. Ten patients is a proof of concept.
• 6:04
  The American Society for Metabolic and Bariatric Surgery published a white paper with eight recommendations for artificial intelligence across bariatric care.
• 6:12
  Keep it clinician-led, validated on bariatric patients specifically, audit its performance across patient groups, and record AI use in the chart and consent forms.
• 6:24
  Disclosure, board member Jerry Deng is a co-author of that paper and suggested it for inclusion in the digest and recused himself from the vote on it.
• 6:35
  And from South America in Colombia, a three-patient report of the magnetic duodenoileostomy after sleeve for weight regain or poor response.
• 6:45
  All three patients went home on the same day, and the connection was open at three to four months, and six-month weight loss ranged from about 5% to 20%.
• 6:54
  The authors say that three patients can't support conclusions about the safety or effectiveness of the device, and the first author has been a speaker for the device company.
• 7:04
  Felipe Giron, our Spanish edition editor, is also a co-author on that report.
• 7:13
  And in the marketplace, Ascensia Diabetes Care, which sells in more than 90 countries, announced on October 9th that it led a funding round for a startup called Sava and took exclusive rights to sell its sensor starting in Europe.
• 7:28
  Sava is developing a small wearable that sits just beneath the skin, starting with glucose. Why would a surgeon care? Because it's aimed at non-adjunctive use, meaning treatment decisions without a fingerstick check.
• 7:40
  That puts accuracy at low glucose levels front and center, which matters for monitoring post-bariatric hypoglycemia. But it's still in development, no investment amount was disclosed, and there's no accuracy data yet.
• 7:53
  A pivotal study is planned for this year, and the company is targeting European approval and launch in 18 to 24 months. And those are company targets, not promises. Watch for the pivotal data. Some quick hits.
• 8:11
  Modeling a two-stage SADI-S, sleeve first and conversion later, estimated 30-day complications of 11.5% versus 6.7% in one stage.
• 8:22
  Readmission and reoperation were higher too, but deaths, leaks, and bleedings were all similar.
• 8:28
  That two-stage number comes from stitching together separate groups of patients, so it's an estimate of the pathway, not a follow-up of the same people.
• 8:36
  A study of 105 patients in Taiwan found that higher levels of a circulating protein called GDF-15 before sleeve went with something less than weight loss.
• 8:46
  A modest link, tangled with age and kidney function, and no cutoff you could use in the clinic. That is the staple line for issue 14 of the MBS Digest.
• 8:57
  Every source is linked in the show notes, and the full issue is free at mbs-digest.com. This is a summary of the published evidence, not medical advice. Once again, I'm John Scott.
• 9:09
  Thank you for listening, and I'll see you next Wednesday.