September was light on headlines and heavy on numbers. Some of them are worth writing on the inside of your kitchen cupboard. One of them is a reason to be patient for a few more months.
Here’s what I read, and what I’d do with it.
A second triglyceride drug cut attacks in two big trials
On 30 August, at the European Society of Cardiology congress in Munich, Arrowhead presented the full results of SHASTA-3 and SHASTA-4. Both tested plozasiran, a quarterly injection, in 757 adults with severe hypertriglyceridemia (triglycerides above 500 mg/dL, which is 5.65 mmol/L).
Triglycerides fell by a median 79 and 81 percent at 12 months. In the pooled analysis, acute pancreatitis attacks dropped 78 percent against placebo. The company puts the number needed to treat at 24 over a year for the whole group.
The part that matters most to us: in people with high triglycerides who had already had an attack, attacks dropped 91 percent, and the number needed to treat was 3.
Two honest caveats. These figures come from the company’s press release and the congress presentation, and I haven’t seen a full peer-reviewed paper yet. And plozasiran is approved today only for the rare inherited form, familial chylomicronemia syndrome (as Redemplo). Arrowhead says it will file for the broader group before the end of 2026.
If your attack was a triglyceride attack, this is the second drug this summer with real pancreatitis data behind it. I wrote about the triglyceride number I check now when the first one, olezarsen, was approved in June.
EPI: about 40 percent, depending on the test
A meta-analysis in the Journal of Pancreatology pooled 14 studies and 7,663 people with chronic pancreatitis. About 40 percent had exocrine pancreatic insufficiency, EPI, where the pancreas no longer makes enough enzymes to digest fat.
The catch is in the test. Studies that used the stool elastase test found 52 percent. Breath tests found 32 percent. The old fecal fat collection found 21 percent. Smoking was linked to 51 percent higher odds of EPI.
That spread deserves its own post, and it gets one later this month. Short version: ask which test you had, and ask for the number.
Creon in the UK: one month at a time, into early 2027
Neuroendocrine Cancer UK posted an update on 29 September. The UK’s Serious Shortage Protocols for Creon were still in place, last extended to 2 October 2026, and pharmacists could hand out only one month’s supply per prescription. Ask your pharmacy whether that limit still applies when you next order.
Nutrizym 22 and Pancrex V exist, but there isn’t enough of them, so they go first to people who can’t take Creon. The new Abbott and Viatris plant is expected to finish its production phase by the end of 2026, with new stock in distribution in early 2027.
So the habits from my enzyme shortage post still apply. Order early. Don’t ration quietly. If you’re in the US, I couldn’t find an official update this month, so your pharmacist is the best source.
GLP-1 drugs: a big reassuring study that doesn’t answer our question
A study in the Journal of Personalized Medicine compared 291,152 adults with type 2 diabetes who started a GLP-1 drug against the same number who started an SGLT2 inhibitor. GLP-1 drugs showed no clear rise in acute pancreatitis (hazard ratio 1.10, with a range that included no difference at all). There was a small signal for gallstone pancreatitis in the main comparison that didn’t hold up in the other analyses.
Here’s the line that matters for us. The researchers excluded everyone with an existing pancreas condition. So this tells you about the average person with diabetes. It says nothing about someone who has already had pancreatitis. My GLP-1 post goes through that gap in more detail.
A computer that reads CT scans for early warning (research only)
Researchers in Shanghai built a model called ORACLE that combines CT images with routine clinical data to predict organ failure in acute pancreatitis. It was trained and tested on 2,746 patients. It flagged organ failure a median 3.5 hours before it became clinically obvious, and when it called someone low risk, it was right 97.2 percent of the time.
This is early. It was built on patients treated at Chinese hospitals between 2011 and 2024, and it isn’t in your local emergency room. I mention it because 3.5 hours is a long time when you’re the one on the trolley.
Who gets pancreatitis in the US
A new Global Burden of Disease analysis counted 235,610 new cases of pancreatitis in the US in 2023, and 5,717 deaths. The age-adjusted rate of new cases has actually fallen since 1990. The total grew because there are more, older people. A journal piece on what those numbers say about who gets this disease follows next week.
Your own cells in a dish
Mission:Cure published a plain-language explainer on 4 September about a Salk Institute study. Researchers grew mini pancreas ducts from 37 patients’ own tissue. About half of the ones tested had poorly working CFTR, including people with no CFTR mutation. This is lab work, years from a prescription. Later this month I’ll go through what it might mean if you’ve been told your pancreatitis is “idiopathic”.
Two things from the National Pancreas Foundation
If you’re near Chicago, the NPF Chicagoland Pancreatitis Alliance meets on 3 October at the University of Chicago, free and in person. And the NPF posted a recorded webinar on managing acute and chronic pancreatic pain across the lifespan on its YouTube channel. Pain gets less airtime than diet, so I’d watch it.
Stepan’s Rule: a trial result is a promise until a regulator signs it and a pharmacy stocks it. I read it, I write it down, and I change nothing until my own doctor does.
I’m a patient, not a doctor. Nothing here is medical advice. Take the numbers to the person who knows your case.
We pull these roundups apart in the PancreasZen community, where people bring their own pharmacy problems and their own lab results. Join us and tell me what I missed.