GLP-1 analogs are a class of medication intended to lower blood sugar levels and promote weight loss. These molecules, which mimic the action of glucagon-like peptides that are normally produced in the gut, have in recent years seen significant growth as a class of drugs. With several GLP-1 drugs on the market today, GLP-1 medications are expected to reach at least $100 billion by 2030. This year, GLP-1 drug sales are projected to surpass PD-1 inhibitors, which help the immune system fight cancer cells, as the top-selling drug class. And maybe Pfizer missed an opportunity to set foot in the GLP-1 market because of a decision it made back in the ’90s.
While the first GLP-1 drug was FDA-approved in 2004, this class of drugs has a long, little-known history dating back to the 1980s. A detailed retelling of GLP-1 development, which served as the basis of this article, can be found in an article published in Perspectives in Biology and Medicine by Jeffrey Flier, who was involved in the early GLP-1 work. “People don’t write about things that failed for various reasons,” Flier told Labiotech. “I figured this bit of history, if I didn’t tell it, would be lost.”
Table of contents
How a small startup formed to study metabolic diseases
In 1982, an endocrinologist from UCSF named John Baxter founded a company. This company, California Biotechnology (or CalBio for short), was set on developing a drug to treat acute congestive heart failure.
During this time three Harvard Medical School colleagues – Jeffrey Flier, Alan Moses, and Martin Carey – were exploring a side project unrelated to the research of their labs. The trio was interested in finding a way to deliver insulin nasally in hopes of creating a nasal spray insulin formulation that could be used during mealtime to control blood glucose levels. The team sought industry support to help develop the drug and landed support from CalBio. A collaboration with Eli Lilly soon began but a few years later Lilly had pulled out of the project.
A few years later, Baxter enlisted Flier with a new task: to start a new company focused on diabetes and obesity. The company– which was called Metabolic Biosystems (MetaBio) would exist under the CalBio umbrella. Flier recruited two HMS colleagues – Rob Kahn and Bruce Spiegelman – for this task. CalBio quickly sought investment from large pharma companies, seeking both financial support and expertise, to back their efforts. Pfizer was up for the challenge and agreed to provide $30 million over the course of five years. Pfizer would get exclusive rights to the program and would spearhead any clinical development that came out of the work. The partnership launched in 1988.
The rise and fall of an early GLP-1 collaboration
At the onset of the MetaBio/Pfizer collaboration, the team focused on identifying insulin analogs, molecules that enhanced insulin’s actions, secreted factors from adipocytes, and gut factors that affect metabolism.
The project that quickly took off was the one that studied gut factors affecting metabolism. In 1987, Flier attended a talk given by another Harvard endocrinologist, Joel Habener. Habener reported the discovery of GLP-1 and that in islet cells and perfused rat pancreas, GLP-1 promoted insulin secretion dependent on glucose concentration. The higher the glucose level, the more GLP-1 enhanced insulin secretion.
While Habener had patented this discovery, it wasn’t licensed to a company at the time. Weeks later, MetaBio obtained an exclusive worldwide license to GLP-1 as a therapy for diabetes.
The MetaBio team soon pursued human data, finding that GLP-1 infused into people with and without type 2 diabetes increased insulin secretion in both groups. In those with type 2 diabetes, GLP-1 could lower blood glucose levels over the course of days. They also found that GLP-1 slowed gastric emptying and decreased hunger cues. The data seems promising.
However, the half-life of GLP-1 was mere minutes. This led MetaBio on a search for the protease that degraded GLP-1. The idea here was that by inhibiting the protease, they would increase GLP-1’s half-life. At the same time, the team also took another approach: designing a protease-resistant GLP-1 analog.
However, Pfizer pulled support from this program before MetaBio could identify the protease. This protease turned out to be dipeptidyl-peptidase IV (DPP-4), which was identified in 1998 by scientists at the University of Copenhagen. The FDA approved the first DPP-4 called sitagliptin, which was developed by Merck & Co., for treating diabetes in 2006. Now, this class of drug includes several more drugs approved in various countries and generates about $11 billion per year in the US.
Pfizer’s exit and repercussions
With the backing of Pfizer and the promising early human data, MetaBio could have been a big player in developing a new therapeutic for diabetes. So where did things go wrong for GLP-1 analogs?
Despite the positive data from human GLP-1 infusion, Pfizer came to the conclusion that there wouldn’t be another injectable besides insulin to treat diabetes. “It was an opinion about the marketability of a certain route of administration,” Flier told Labiotech. “In the late 80s and early 90s, there were a lot of discussions about alternate routes of delivery. There weren’t that many medications that were delivered by injection on a regular basis.” Combined with the unfamiliarity of injectable drugs, and the fact that there was a lot of work going on at the time on developing oral drugs to treat type 2 diabetes, it seemed that Pfizer did not think another injectable would be worth pursuing.
Pfizer pushed for a transnasal or transcutaneous route of administration for GLP-1 but the team knew that would be unlikely given the short half-life of GLP-1 of just a few minutes. These efforts were unsuccessful and three years into the partnership, Pfizer announced they would withdraw from the collaboration after four years into the program.
With the exit of Pfizer, the MetaBio founders pushed CalBio to seek additional fundings for this work. CalBio, however, had its eyes set on developing its acute congestive heart failure drug, Natrecor. This drug eventually became FDA-approved in 2001 and the company was acquired by Johnson & Johnson in 2003 for $2.4 billion.
Without finances from Pfizer, CalBio could not simultaneously support the GLP-1 work. Rich Casey, then CEO of CalBio, did try to garner support from other companies but he recalled companies like Eli Lilly and Novo Nordisk weren’t interested. It took several years after various publications on the benefits of GLP-1 in humans from different academic labs for pharmaceutical companies, including the above-mentioned Novo Nordisk, to become interested in it. “Stated another way, Pfizer was not alone in failing to appreciate the potential therapeutic value of intervening in this pathway,” Flier wrote in Perspectives in Biology and Medicine.
At the same time, the MetaBio founders weren’t interested in leaving academia to focus on GLP-1. CalBio bought out the founders, thus ending the work on GLP-1 for this small startup.
GLP-1 development in the 1990s – Novo Nordisk a key player
With MetaBio/Pfizer out of the picture, the story now turns to Novo Nordisk, a company that soon acquired rights to Haberner’s patents. At the time, Novo Nordisk was known as an insulin company, and now, Novo Nordisk has the largest share of the GLP-1 market, according to Flier. Novo Nordisk’s first attempts to create protease-resistant GLP-1 analogs were unsuccessful, but in 1995, they produced an analog with an acylated fatty acid to extend GLP-1’s half-life called liraglutide. In 2008 and 2009, they published phase 3 clinical trial data and by 2009, liraglutide was approved for diabetes in the EU and in the US by 2010. This drug was then approved for obesity in 2014. Liraglutide, with a half-life of 13 hours, was the first approved GLP-1 analog for once-daily dosing.
Now, Novo Nordisk’s Ozempic and Wegovy – both once a week injectables – have made Novo Nordisk Europe’s most valuable company. It currently has the only oral GLP-1 drug on the market.
Can Pfizer catch up?
Pfizer didn’t jump back into the GLP-1 game until decades after they dropped MetaBio. Its current candidate, danuglipron, is currently undergoing pharmacokinetics and safety studies. Designed as an oral, once-daily pill for obesity, danuglipron has thwarted Pfizer’s previous efforts in developing the drug. Previously, Pfizer designed the drug as a twice-daily pill but it faced discontinuation rates of over 50%, possibly due to side effects. To move danuglipron forward, Pfizer still needs to undergo clinical trials before seeking FDA approval.
But can it compete with its competitors already on the market?
“If approved, it will certainly be a case of profoundly delayed gratification,” Flier recently wrote in STAT News.
Hindsight is 20/20
Pharmaceutical companies can’t predict the future and hindsight can only reveal such mistakes in decision making. In a recent article in Science, chemist Derek Lowe points out two other instances of missed opportunities: (1) the head of research at Schering-Plough, which now manufactures Claritin, once said that there aren’t any nonsedating antihistamines and (2) when a large pharma company decided not to develop metformin to treat type 2 diabetes. According to Lowe, metformin is one of the most commonly prescribed drugs across the world.
“In hindsight it can be hard to believe the magnitude of such mistakes, and partly because institutions tend to bury such decisions under layers of rubble, for obvious business and psychological reasons,” Lowe wrote in Science.
While Pfizer ended its partnership with MetaBio one year early, would the outcome be different had they stuck through an additional year? Would Pfizer have made enough progress on a GLP-1 analog? Could they have gotten a GLP-1 analog onto the market sooner? “I can’t say with any certainty how that would have worked out, but I am optimistic that we could have done it,” Flier told Labiotech.
Partnering 2030: The Biotech Perspective 2023
