Hey readers! Here is something that flips the usual script on protein stress signals: a protein often filed under "cellular stress and death" turns out to be the thing your beta cells need to respond to GLP-1 drugs. This week we lead with that finding, then look at a GLP-1/GIP drug actually tested in type 1 diabetes, the case for measuring what your own cells still make, and a round of device and research updates worth your attention.
🧬 The JNK3 story: a helper hiding in a stress pathway
Unlocking a New Way to Protect Insulin-Producing Cells reports that researchers from the Diabetes Research Institute and the Endocrinology Division identified JNK3 as a key protein that helps insulin-producing beta cells respond to GLP-1 receptor agonists. – Diabetes Research Institute Foundation
That matters because GLP-1 receptor agonists are everywhere in type 2 diabetes care and are increasingly being explored in type 1. They work by nudging beta cells to release insulin and by shielding them from damage. This study asks a quieter question: what inside the cell actually makes that response possible?
When JNK3 was absent or blocked, beta cells were less able to respond to these treatments, produced less insulin, and became more vulnerable to inflammation and cell death.
The JNK family of proteins is usually discussed in the context of cellular stress, so finding that JNK3 specifically supports survival pathways and the GLP-1 response is a useful wrinkle. In pre-clinical models, the team found that JNK3 helped protect beta cells, which led them to suggest that enhancing JNK3 activity could make existing therapies work better and help preserve whatever insulin-producing cells a person still has.
A few honest caveats: this is pre-clinical work, not a human trial, and "could improve existing therapies" is a hypothesis for future research, not a result you can act on in clinic. But it fits a theme we keep coming back to in this newsletter, which is that protecting remaining beta cells, rather than only replacing insulin, is where a lot of the interesting science is pointing.
💉 GLP-1 drugs meet type 1 diabetes, head on
If the JNK3 work is about why GLP-1 drugs help beta cells, the next story is about what one such drug actually did in people with type 1 diabetes. Acmopatide phase 2: weight loss in type 1 diabetes covers a phase 2, double-blind, randomized, placebo-controlled trial of acmopatide (CT-868), a once-daily dual GLP-1/GIP receptor agonist, published in Diabetes Care on October 28, 2026. – Peptide Review Desk
In 111 adults with type 1 diabetes and overweight or obesity, acmopatide reduced body weight by 3.3% to 6.7% over 16 weeks, versus a 0.3% gain on placebo. Insulin doses fell, and there were no episodes of diabetic ketoacidosis across the 16 weeks.
No GLP-1 drug is approved for type 1 diabetes, so this is one of the few placebo-controlled looks at the drug class in that group.
Worth reading closely, though. The glucose effects were modest and statistically fragile: HbA1c fell by 0.4 to 0.6 percentage points on acmopatide versus 0.3 on placebo, and only the middle dose reached what the authors call nominal significance (a placebo-adjusted difference of 0.3 points, 90% confidence interval 0.1 to 0.6). This was a small, 16-week dose-finding study, so the DKA signal and the longer-term weight picture both remain uncertain. Weight and insulin dose moved; A1c barely did.
🔬 What we measure, and when
Several of this week's items circle the same practical question: how do we track the beta-cell function the JNK3 and acmopatide stories are trying to protect?
Making the Case for C-Peptide in T1D Therapy Development is a perspective published in Diabetes Care on April 20, 2026 by Jay S. Skyler and colleagues (including G. Alexander Fleming and Chantal Mathieu), arguing that C-peptide should be recognized again as a primary measure of treatment success in type 1 diabetes. – Diabetes Research Institute Foundation
C-peptide is a natural marker of how much insulin your own beta cells can still make. The authors tie preserving those cells to better glucose control, fewer dangerous lows, and lower long-term complication risk, and they argue that treating C-peptide as a primary regulatory endpoint could shorten the time and cost of developing therapies aimed at protecting remaining cells. If regulators embrace that, it could change which disease-modifying treatments get funded and approved.
Can CGM detect early beta-cell dysfunction better than OGTT? reports a Diabetologia study from Yale and the University of Padova on a CGM-derived disposition index that strongly correlated with the conventional OGTT-based index (r = 0.90; p < 0.001). First author Hadija Marchiori says the goal is to "enable home-based monitoring of beta cell function using a CGM," while cautioning that the cohort was small and needs larger, longer validation. – EASD e-Learning
Mapping Type 1 Diabetes Before It Begins describes a Nature Communications study (February 11, 2026) that analyzed nearly 25,000 pancreatic islets and found early beta-cell stress, increased immune activity, and gradual loss of vulnerable islet populations before diagnosis, suggesting the disease is more widespread across the pancreas than once thought. – Diabetes Research Institute Foundation
New Research Reveals Type 1 Diabetes in Adults May Not Be One Single Disease covers a Germany-based Lancet Diabetes & Endocrinology study (October 30, 2026) that used machine learning on 600 adults to identify at least three distinct adult-onset subtypes differing in immune, genetic, and metabolic profiles. The authors stress it is observational and may not generalize beyond European populations. – Type1Cure
The through-line here: if we want therapies that preserve beta cells, we need endpoints and tools sensitive enough to see those cells change. C-peptide, CGM-derived indices, and islet mapping are all shots at that problem.
🩺 Testing gaps and new approvals
Fewer than half of newly diagnosed children with type 1 diabetes get autoantibody testing is the practical gut-check of the week. An Epic Research analysis of 68,439 children found 51.2% received no islet autoantibody testing in the 12 months around diagnosis. – Managed Healthcare Executive
The Epic Research authors said that the untested children "represent a meaningful gap in the diagnostic workup."
Among those tested, 70.1% had at least one positive result and 44.3% were positive for two or more. Confirming autoimmune type 1 diabetes isn't academic: treatment and monitoring differ by type, and screening-dependent therapies rely on catching earlier-stage disease. If half of newly diagnosed kids aren't tested, a lot of that opportunity is being missed.
On the device and drug front:
New To Insulin Pumps? Twiist Gets FDA-Clearance For Type 2 Diabetes reports that on October 16, 2026 the FDA cleared the twiist insulin pump (which uses the Tidepool loop algorithm) for adults 18 and older with insulin-dependent type 2 diabetes; in the pivotal T2D trial, time in target range rose from 57% to 73%. The pump has been available for type 1 in the U.S. since March 2026. – Daily Diabetes News
Once-Weekly Basal Insulin Awiqli Approved for Pediatric T2D notes the FDA expanded Awiqli (insulin icodec-abae) on October 29, 2026 to include children aged 10 and older with type 2 diabetes, backed by a pediatric PK study (NCT05790681). It is a once-weekly injection that binds serum albumin to form a slow-release depot. – Renal and Urology News
📈 Devices doing the daily work
Youths with type 1 diabetes cut HbA1c with iLet bionic pancreas describes a single-center real-world study of 71 young people aged 6 to 21 using the iLet bionic pancreas from Beta Bionics. – Healio
At three months, time in range rose 22.4 percentage points and HbA1c dropped 1.8 points (P < .0001 for both). At one year, among the 35 patients with data, HbA1c fell from 9.6% to 7.6% and time in range climbed from 27.7% to 55.8%. The largest gains went to kids who started with the highest A1c, which is encouraging for those who've struggled most. The authors call for larger, prospective, multicenter work to confirm it.
🧪 Research watch, with caveats
A couple of items are worth noting mainly for what they teach about the limits of early science.
A drug-free diabetes treatment? An implant lowers blood sugar in mice reports in Science (October 22, 2026) on a wireless under-the-skin implant that uses a gold-and-platinum catalyst to convert glucose into gluconate the kidneys can excrete. In mice, switching it on normalized high blood sugar within 60 to 90 minutes. Endocrinologist David Klonoff called it "a potentially promising method," while outside experts flagged risks like gluconate buildup and electrolyte disruption, plus the long road from mice to humans. – Science News
Svensk cellterapi missade målet i studie på barn med T1DM reports that NextCell Pharma's ProTrans-Young trial in 60 children and young people aged 7 to 21 did not meet its primary endpoint for preserved insulin production versus placebo. The youngest group (7 to 11) showed a numerically better pattern (64% vs 51% of insulin production retained at one year), but the difference was not statistically significant and 12-month data were missing for four of 15 treated children. – Dagens Diabetes
Both are reminders of why the C-peptide endpoint debate and the search for reliable beta-cell markers matter: trials rise or fall on whether they can measure what they claim to protect.
That's the issue. If you take one thing away, let it be this: the fight to preserve your own beta cells is getting better tools, better targets, and better questions, even when individual trials don't land. Take care of yourselves.