Type 2 and GLP-1: a plain summary

Where these drugs came from

It is worth remembering that this whole class began in diabetes care. GLP-1 receptor agonists were developed, licensed and used for years to lower blood glucose in type 2 diabetes before anyone was prescribing them for weight. The weight effect was noticed as a consistent, welcome side finding, and only later became the target of its own trials.

That history explains a lot of what our type-2 circle finds frustrating. Members with diabetes sometimes feel like they have wandered into a conversation that has been repurposed around them — the language online is now overwhelmingly about weight, the shortages have been driven by demand from outside diabetes care, and the questions that matter most to them get comparatively little airtime.

So this guide is written for people with type 2 diabetes first, and it stays on that ground.

How they lower glucose

Four things happen, and the first two are the main event.

Glucose-dependent insulin release. The drug increases insulin secretion, but only in proportion to how high the glucose is. When glucose is normal, the stimulus fades. This is why these medicines on their own rarely cause hypoglycaemia — a genuinely important property and a real difference from insulin and sulfonylureas.

Glucagon suppression. Glucagon tells the liver to release stored glucose. Damping it reduces the glucose your own liver adds, which matters most between meals and overnight.

Slower gastric emptying. Food arrives in the small intestine more gradually, so the post-meal glucose rise is flatter. This is also the source of most of the gastrointestinal side effects.

Appetite signalling. Acting on brainstem and hypothalamic pathways, which for many people quietens what members here call food noise.

Tirzepatide adds activity at the GIP receptor, a second incretin pathway, which appears to contribute to both glucose and weight effects. What these medicines actually do has the longer version.

What the glucose trials found

The reductions in A1c are substantial and consistent. In SURPASS-2 (Frías, New England Journal of Medicine, 2021), a 40-week head-to-head trial in people with type 2 diabetes, all three tirzepatide doses lowered A1c by roughly two percentage points or a little more and each beat semaglutide 1 mg, which itself achieved a reduction close to two points. In SUSTAIN-6 (Marso, NEJM, 2016), semaglutide lowered A1c well below placebo across two years.

Two cautions our reading group always attaches. Trial participants start at a particular average A1c, and how much room there is to move strongly affects how large a reduction looks. And these were people receiving trial-grade support and monitoring, which is not what most people get.

What is not in dispute is the direction and the size of the effect. For glucose lowering, this is one of the more robust bodies of evidence in the field — considerably firmer than most of what gets argued about online. More on the test itself at A1c, and what it is not.

The outcome trials, which matter more

Lowering a number is only worth something if it changes what happens to people. Several trials have looked at exactly that in type 2 diabetes.

LEADER (Marso, NEJM, 2016) followed 9,340 people with type 2 diabetes and high cardiovascular risk on liraglutide or placebo for a median of about four years. Major cardiovascular events occurred in 13.0 per cent on liraglutide against 14.9 per cent on placebo, a hazard ratio of 0.87, with cardiovascular death also reduced.

SUSTAIN-6 tested injectable semaglutide over two years in a similar population and found fewer major cardiovascular events, with a hazard ratio of 0.74 — a trial designed primarily to demonstrate safety that ended up suggesting benefit.

FLOW (Perkovic, NEJM, 2024) tested semaglutide 1.0 mg in people with type 2 diabetes and chronic kidney disease, and found slower progression of kidney disease, with a hazard ratio of 0.76. It was stopped early for benefit. There is a full guide at kidney numbers.

Taken together, these are the reason many clinicians now think about drug choice in type 2 diabetes in terms of organ protection rather than glucose alone.

Two specific things to raise with your team

Hypoglycaemia, if you take insulin or a sulfonylurea. A GLP-1 medicine alone rarely causes hypos because its insulin effect is glucose-dependent. Combined with insulin or a sulfonylurea — gliclazide and its relatives — the risk is real, and it is highest when you start, at every dose increase, and during the weeks when you are eating noticeably less. Many members report having their other doses reduced at the outset. This is a conversation to have before your first injection, not after your first hypo, and it is one of the strongest reasons not to obtain these drugs outside a prescribing relationship if you take those medicines.

Retinopathy. In SUSTAIN-6, complications of diabetic retinopathy occurred more often in the semaglutide group than in the placebo group. The usual explanation is that rapid improvement in glucose control can transiently worsen existing retinopathy — a phenomenon described long before these drugs, seen when control improves quickly by any means. Participants with pre-existing retinopathy at baseline accounted for much of the effect. What follows in practice is that if you have retinopathy, or have not had a recent eye screening, that is worth raising before or around starting. Gavin had his screening brought forward for exactly this reason and described it as the least dramatic useful thing anyone did for him that year.

The everyday practicalities

Things our type-2 members consistently say they wish had been covered at the start.

• Sick-day rules. If you cannot keep fluids down, some diabetes medicines are usually paused temporarily. Ask which of yours, get it written down, and keep it somewhere findable. This is the single most valuable piece of paper in the circle.
• Monitoring changes. If you use a sensor or finger-pricks, expect the picture to shift over the first months, particularly around meals.
• Driving. If you take insulin or a sulfonylurea, licensing rules about hypoglycaemia apply and do not change because you have added another drug. Check your own country’s requirements.
• Eye screening and foot checks continue. Improving glucose does not retire them.
• Eating enough is still a thing. Appetite suppression plus glucose-lowering medication needs care. Eating enough when nothing appeals was written partly for this circle.
• Ask what happens if supply fails. Shortages have hit diabetes care hardest and having a fallback discussed in advance beats improvising. See shortages and substitutions.