Until now the U.S. insulin market has been effectively a monopoly. With an 81 percent share, Indianapolis-based Eli Lilly & Co. (NYSE:LLY) dominates this $936 million business, according to 1998 data from IMS Health Inc. The only other player here has been Novo Nordisk A/S (NYSE:NVO) of Bagsværd, Denmark.
Hoechst will…take market share from Lilly and Novo.
Worldwide Lilly has a 50 percent share of the $2.4 billion insulin market with 1998 sales of $1.2 billion. Novo Nordisk follows with 42 percent on sales of about $1 billion.
But worldwide there is a third competitor in the insulin market, Hoechst Marion Roussel AG, a wholly owned subsidiary of Hoechst AG (NYSE:HOE), based in Frankfurt, Germany. It reported insulin sales last year of $165 million.
While small in insulin, this gigantic company has both the marketing muscle and—even more important—the hot new product to turn the insulin market both here and abroad upside down.
Hoechst is expected to merge with Rhône-Poulenc S.A. (NYSE:RP), headquartered in Paris, France, by November. The corporate headquarters of the new company, Aventis S.A., will be in Strasbourg, France. It will be the world's largest drug company.
Hoechst plans on cracking the U.S. insulin market with a new insulin analog that has no peaks or valleys and a very long duration. An insulin analog is one created through genetic engineering. Hoechst created this insulin analog by changing the order of the amino acids that make up human insulin.
The company completed its reporting for Phase III clinical trials of 2,870 patients, including 1,600 in the U.S., in January. In April it submitted the insulin for approval to both the U.S. Food and Drug Administration and the European Medicinal Evaluations Agency. Hoechst hopes to have FDA approval early next year and to have the insulin on the market by June 2000.
This insulin analog was known as HOE901 in its early trials. The World Health Organization gave it the generic name insulin glargine. Hoechst has submitted a tentative brand name to the FDA, but can't use it pending agency approval.
Whatever brand name it gets, people still generally know it as HOE901. Under that name Hoechst has been developing this insulin analog for years. According to a Hoechst spokesperson and outside medical consultants it appears to have few negatives.
Hoechst didn't report any contraindications to the FDA, says Eberhard Draeger, global project leader and senior director, Hoechst Marion Roussel Inc. in Bridgewater, New Jersey. But a disadvantage of the clinical trials is that they were open label—both the patient and the doctor know the kind of insulin the patient got.
"We couldn't blind the studies because HOE901 is a clear solution that is always homogenous and can't be mixed," Draeger says. "NPH [neutral protamine Hagedorn] insulin and the other long-acting insulins are in suspension, and the patient has to carefully roll them to get a homogeneous distribution of the crystals."
Draeger says even covering the pens wouldn't make it a blind study, because patients would know if the insulin they injected were in suspension just by looking at the drop that primes the needle. When they used syringes, patients would see the insulin in the syringe when measuring the dose.
Because of these trial limitations, Hoechst concentrated on objectively measurable hypoglycemia, or low blood sugar. The two biggest advantages of HOE901, Draeger says, is that it reduces nocturnal hypoglycemia in general and severe hypoglycemic, where the help of another person is needed, in particular.
"These are the two most important advantages," he says, because hypoglycemia prevents the patient from going closer to a normal blood glucose level." He admits that the trials did not show lower Hemoglobin A1c level, which measures blood glucose over a period of a month or two.
"The explanation is that the patient titrated to the same therapy," Draeger believes. "They should have given themselves more insulin, but didn't dare. They had used NPH insulin before, and knew from their experience that if they took more they would have had hypoglycemia."
Will Hoechst make HOE901 available for insulin pumps as well as for syringes and pens? "No," Draeger replies. "It doesn't make sense in pumps, because there you give a basal insulin infusion, and that is actually what we are mimicking once a day."
HOE901 is injected once a day, usually at bedtime, and covers the basal insulin for a whole day. It lasts at least 24 hours with a flat profile. In the trials patients did not have to reduce their dose over time, which would have been necessary if they the drug accumulated, Draeger says.
Hoechst is targeting both type 1 and type 2 patients. "It would be suitable for all type 1 diabetes patients, unless they need a pump," Draeger says.
How many is that? Probably no more than 75,000 Americans use pumps, almost all of them type 1, according to statistics from the two pump manufacturers. About 300,000 to 500,000 Americans had type 1 diabetes in the early 1990s, and about 30,000 new cases are reported each year, according to "Prevalence and Incidence of Insulin-Dependent Diabetes," chapter 3 of Diabetes in America, second edition (1994). The lead author of that chapter, Dr. Ronald E. LaPorte, professor of epidemiology at the University of Pittsburgh's Graduate School of Public Health, says that data is the most recent.
About 10 million more Americans have diagnosed type 2 diabetes, according to government statistics. How many of these are in Hoechst's sights?
Probably not as great a proportion as in Europe, Draeger admits. That's because type 2 patients there are much more likely to be on intensified therapy where they need true basal insulin as well as short-acting insulin to cover their meals.
Hoechst will target two groups of type 2 patients here. The first, Draeger says, is those who are using insulin but are not adequately controlled, i.e. with a Hemoglobin A1c level of 8 percent or more.
The second group are those who have failed on oral medications and need insulin support. "HOE901 would be the entrance to insulin therapy for them by giving one shot per day in addition to their oral medications," Draeger says. "If this were not sufficient you would slowly add other insulin to cover their meals."
This is a huge market, says Dr. John Gerich, professor of medicine at the University of Rochester who participated in clinical trials of HOE901. "It will replace NPH."
Another physician excited by HOE901 is Dr. Nancy Bohannon, a San Francisco endocrinologist. "I've done three trials with it," she says, "and a significant proportion of the people who used it really like it and are eagerly awaiting its approval. I don't see any reason in the world why it's not going to zip right through the FDA. As far as I know it's a no-brainer."
If and when the FDA approves, HOE901 will be Hoechst's lead insulin in a five-pronged strategy to break into the U.S. market. The second prong is short acting insulin that Hoechst is working on that would compete with Lilly's lispro (Humalog).
But besides HOE901, Draeger says, "all other insulins from all companies are basically the same. They are generic in the sense that all companies have them."
The third prong is highly concentrated U400 insulin that Hoechst has developed for Mini-Med pumps. The fourth prong is a disposable pen that Hoechst will introduce quickly after the introduction of HOE901.
The final prong is inhaled insulin, which Hoechst and Pfizer Inc. (NYSE:PFE) will manufacture in a jointly owned plant in Frankfurt for Nektar Therapeutics, formerly Inhale Therapeutic Systems Inc. (NASDAQ:INHL) of San Carlos, California. Inhale currently has Phase III clinical trials underway at 117 sites.
The agreement with Pfizer is key to Hoechst's marketing strategy. Both companies sell sulfonylureas—older oral diabetes medications—here. Hoechst markets Amaryl and Diaßeta, and Pfizer sells Diabinese, Glucotrol, and Glucotrol XL. Together, they could be stronger.
"We will have a co-promotion with Pfizer with separate sales forces," Draeger says. "Both will sell HOE901."
He says that Hoechst will certainly seek to take market share from Lilly and Novo. If so, will that lead to a price war?
"I would assume there would not be a price war, because there is nothing on the market that matches the profile of HOE901," Draeger replies. "Therefore we are in a better position than Novo was when it entered the U.S. market with a generic type of insulin."
Neither does Thane Wettig, Lilly's marketing director for Humulin and Humalog in the U.S., think that a price war will play out. "Hopefully there is enough differentiation with our product line that the competition will absolutely be a factor, but much less of a factor than it would have been two years ago."
Wettig admits that HOE901 will shake up the insulin market here "a little bit." He thinks, however, that the market will be smaller "than some people think."
Why? First, because HOE901 would be most appropriate for people who are on two to three shots a day, he replies. Only a relatively small group—5-10 percent of patients using insulin—are on three or more shots a day, he says.
He says that 40 percent of the insulin used today is a 70/30 mixture, which usually calls for two shots a day. It's 70 percent NPH and 30 percent regular insulin. Probably another 20 percent are on just one shot of NPH and another 20 percent are on two shots of NPH and regular mixed in the same syringe.
There is another reason why he minimizes the importance of HOE901. "While it does have a nice smooth profile, one of the major drawbacks is that the people who are taking three and four shots a day are mixing NPH and lispro in the morning and NPH and lispro in the evening. But HOE901 cannot be mixed in the same syringe with another insulin. Will patients trade off two shots?"
Wettig also doesn't think that diabetics who do not have good control will benefit as much from HOE901 as from focusing on insulin to cover meals. "That's where the surges in blood glucose take place."
He says that Lilly will probably launch a new premixed insulin in the first quarter of next year to address that need. Called Humalog Mixed 25, it will be like 70/30, but instead of the 30 percent being regular insulin it will be 25 percent Humalog.
Does Lilly have any very long acting analogs in its pipeline? "We do have an analog program that is very similar to Novo's NN304," Wettig replies.
Novo hopes to get NN304 on the market "somewhere out there in 2001 or 2002," says Peter Hansen, director of investor relations for Novo Nordisk of North America. While it will be a long-acting insulin, Novo doesn't know how long it lasts until it completes on-going Phase II clinical trials.
"In terms of medical potential basically everybody could be switched to long-acting insulin," Hansen says. Currently 50 percent of the insulin sold worldwide is rather long acting—mostly NPH—20 percent is short acting, and 30 percent is mixtures, he says.
With the once stable U.S. insulin market potentially in flux, Hansen sees the development of long-acting insulins like NN304 and HOE901 as an opportunity for Novo Nordisk. "It is an interesting opportunity to upgrade the market, and it's exciting to be able to offer this to patients."
But for two reasons he thinks Lilly should be concerned. "Lilly holds 80 percent of the U.S. market in volume terms, so they are most exposed to a new drug coming in. The second reason is that they have only a very early pre-clinical development of a long-acting analog."
The big problem for Hoechst, in Hansen's view, is that "diabetes is a very conservative market." Once patients are well controlled, they are unlikely to switch insulins, he thinks.
"Even lispro, Lilly's short acting insulin, after three years in the market place, has gained only a 15-20 percent market share in the short-acting segment they are operating in," Hansen maintains. "That's relatively little in three years. I think you will see the same with the long acting. Even though it has some medical advantages, you are going to see that conservatism."
Go back to Home Page
Go back to Diabetes Directory