PANORAMA — bihormonal artificial pancreas after total pancreatectomy

What is being tested, and why it matters

When the whole pancreas is removed — usually because of pancreatic cancer or severe chronic pancreatitis — a person loses far more than their insulin. They also lose glucagon, the hormone that raises glucose when it falls too low. The result is a particularly unforgiving form of insulin-dependent diabetes: glucose swings hard in both directions, and there is no internal brake at all against a low.¹

That makes this group the sharpest possible test of a bihormonal artificial pancreas. The Inreda AP5 delivers both insulin and glucagon under fully automated algorithm control, with no carb counting and no meal announcements.¹ For someone with no pancreas, it is the closest thing currently available to a replacement organ: the two hormones the body can no longer make, dosed automatically from a glucose sensor.

PANORAMA (NCT06346366) is the largest and longest trial of this idea in this population. It is the direct successor to the 2022 JAMA Surgery trial (APPEL 5+), which tested the same approach in 12 patients for one week per arm — PANORAMA runs for three months per arm, and the registry explicitly allowed former APPEL 5+ participants to take part.²¹

Who it is for

Adults aged 18–75 with insulin-dependent diabetes at least three months after a total pancreatectomy, already using flash or continuous glucose monitoring and treated with an insulin pump or injections. People on chemotherapy, with a BMI over 35, an HbA1c above 11.0%, on oral glucose-lowering drugs, or taking paracetamol (acetaminophen) daily were excluded — the last of these because paracetamol interferes with some glucose sensors.²

How the trial was designed

PANORAMA is an open-label, randomized crossover trial sponsored by Amsterdam UMC, with recruitment through Dutch Pancreatic Cancer Group hospitals and funding from the Dutch Cancer Society (KWF).²¹ Each participant completes a 3-month period on the bihormonal AP5 and a 3-month period on their own current insulin care, in random order, separated by a washout of at least three months.²¹ The primary outcome is the percentage of time in range — glucose between 70 and 180 mg/dL (3.9–10.0 mmol/L) — measured across each treatment period.² The registry lists it as a phase 3 study.²

Key results — provisional

Results were presented at the ADA 86th Scientific Sessions in June 2026 and reach us through a single conference report. No ADA abstract, sponsor press release, or peer-reviewed publication has been located, so everything below should be read as provisional rather than established.³

As reported, in 25 adults analysed: time in range rose from 57% to 81%, time below range fell from 1.5% to 0.9%, and HbA1c fell from 8.0% to 7.2% on the bihormonal system versus usual care.³ If those numbers hold, they are a very large effect — roughly six extra hours a day in range — and consistent in direction with what the same device has shown in people with type 1 diabetes.¹

The report also carries an important negative signal: 7 of the 25 participants stopped using the device, citing its size, its alarms, and general usability, and skin reactions at the glucagon infusion site were more frequent.³ That is a substantial dropout rate, and it echoes what users of the Inreda AP have said elsewhere — the glucose control is good; the box and the second infusion set are the problem.

Caveats worth holding onto

Three things temper this. First, the source: one conference write-up is thin ground for numbers this striking, and we will revise this record when a peer-reviewed report appears. Second, the numbers do not line up cleanly — planned enrollment was 40 in the registry and at least 30 in the published protocol, against 25 reported as analysed; we do not yet know how that gap arose.²¹³ Third, the registry has not been updated: it still shows status "unknown", an estimated primary completion of April 2024, and no posted results.² We call the trial completed on the strength of the ADA presentation, not on a registry update.

What it means and what is next

People without a pancreas are the group with the least to lose from a second hormone and the most to gain, and PANORAMA is the strongest test yet of whether a two-hormone loop helps them over months rather than days. The reported glucose numbers argue that it does. The dropout numbers argue that the hardware, not the algorithm, is now the limiting factor — which is exactly the case for a stable, pre-fillable glucagon and a smaller device. The next milestone is a peer-reviewed publication of these results.

References

1. Leseman CA, van Veldhuisen CL, Rompen IF, et al. Bihormonal artificial pancreas versus current diabetes care after total pancreatectomy (PANORAMA): study protocol for a randomized crossover trial. BJS Open (2026);10(1):zraf151. https://academic.oup.com/bjsopen/article/10/1/zraf151/8431533 ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

2. Amsterdam UMC. Glucose Control Through a Bihormonal Artificial Pancreas in Patients After Total Pancreatectomy (PANORAMA). ClinicalTrials.gov (NCT06346366; design, eligibility, endpoints, sponsor; status "unknown", last updated April 2024). https://clinicaltrials.gov/study/NCT06346366 ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

3. Closed-Loop Updates from ADA 2026 — Inreda AP5 PANORAMA results. Diabetotech (conference report of the ADA 86th Scientific Sessions, June 2026; sole source for the outcome figures; not peer-reviewed). https://www.diabetotech.com/blog/closed-loop-updates-from-ada-2026 ↩ ↩² ↩³ ↩⁴