What 100 years of insulin teach us about the GLP-1 wave
Insulin was the first life-saving peptide drug in 1921. GLP-1 analogues in 2026 are the fastest pharmacological diffusion of a peptide active principle since insulin. What can be derived from 100 years of pancreatic hormone history — about patents, supply, generics and expectations.
A recurring structure
If you slice the history of insulin into four phases — discovery (1921), first iterations (1930s–40s, protamine zinc, NPH), recombinant production (1980s, Genentech/Eli Lilly), analogues and pen systems (1990s–2010s) — a structure emerges that repeats in later peptide drugs: first the fundamental discovery, then a pharmacokinetic improvement, then production scale-up, then iteration on comfort and compliance.
The GLP-1 wave recognisably follows the same pattern, only in much shorter time. Discovery of the active principle: exendin-4 in 1992, approval of exenatide in 2005. Half-life optimisation: liraglutide daily in 2010, semaglutide weekly in 2017. Scale-up: supply shortages from 2022. Iteration: tirzepatide as dual GLP-1/GIP in 2022, retatrutide as triple GLP-1/GIP/glucagon in phase 3. What took 60 years for insulin is happening here in 20.
What insulin teaches us about patents and generics
Frederick Banting and Charles Best sold the insulin patent in 1923 for a symbolic dollar to the University of Toronto, arguing that a life-saving medicine should not be private property. This gesture was morally impressive — but it did not prevent insulin from remaining a high-priced medicine for nearly a century. The reason: each new iteration (recombinant instead of animal-derived, analogues instead of human, pen instead of syringe) was protected by new patents. Letting the original patent expire is not the same as making a medicine generic-eligible, when parallel patents protect the currently used formulation.
The GLP-1 wave will most likely develop along the same axis. The exenatide patent has already expired; liraglutide patents expire 2024/2025 in several markets. Semaglutide patents are protected into the early 2030s. So the first generics discussions become relevant for the first two substances — but exactly when tirzepatide and retatrutide already represent the next generation. That is not a failure of the patent system; that is its expected mode of operation.
What insulin teaches us about supply
Insulin faced supply problems from the beginning. In the 1920s it was extracted from bovine and porcine pancreas; production volume limited the number of patients who could be treated. Only recombinant production from 1982 (Genentech/Eli Lilly Humulin) decoupled insulin supply from animal raw materials. The transition took years — and global insulin supply, even 100 years after discovery, is not self-evident in many countries.
The Wegovy/Ozempic shortages 2022–2024 are not an anomaly. They are the expected friction when global demand for a peptide medicine grows faster than production capacity. Unlike small molecules, peptide synthesis is capital-intensive: solid-phase peptide synthesis is an iterative process requiring several reaction steps per amino acid; purification and quality control are demanding. New production capacity takes years, not months.
What insulin teaches us about expectations
Insulin transformed type-1 diabetes from a rapidly fatal to a chronically manageable disease. It changed the course of the disease — but it did not eliminate it. Lifelong substitution, constant glucose monitoring, acute hypoglycaemia risks and long-term complications have remained reality. Even a life-saving medicine does not solve the underlying biological problem.
GLP-1 analogues substantially reduce endpoint-relevant parameters (weight, HbA1c, cardiovascular events) in obesity and type-2 diabetes. They are not a cure. What happens at discontinuation (weight rebound), which long-term safety signals emerge over 20+ years, and how diffusion in lower-income settings is financed are open questions. Transferring the insulin experience suggests: expect success, but expect it as a chronic therapy with its own side-effect and supply logistics, not as the end of the disease story.
„A really good therapy makes a disease manageable. A really good therapy does not make it disappear — it reveals what needs to be solved next."
What can be derived from the comparison
- Patent expiry does not automatically lead to price erosion when parallel patents protect the current formulation.
- Production capacity is a structural bottleneck for peptides that is closed over years, not months.
- A successful pharmacological innovation opens a new market — but rarely solves all the problems it addresses.
- Cultural perception of a substance (insulin as 'miracle', GLP-1 as 'diet miracle') typically precedes scientific consolidation of long-term data.