Biotech Values

[DewDiligence]

NVS, MRK Maneuver for DPP-4 Hegemony

[This is an excerpt from the Oct 30 printed issue of Forbes.]

http://www.forbes.com/free_forbes/2006/1030/160.html

>>
By Robert Langreth and Matthew Herper
10.30.06

…The most intriguing new drugs are smart therapies that lower blood sugar only when patients need it, averting the dangerous drops that can occur with many other drugs. The first of this new breed is Byetta, a twice-daily shot from Amylin Pharmaceuticals (nasdaq: AMLN ) and Eli Lilly (nyse: LLY ) that both lowers blood sugar and takes off a few pounds. It was introduced in June 2005 with modest expectations but has become a hit among diabetics (including Friedman), who like the weight loss. Sales are likely to hit $500 million this year. Sales so exceeded Lilly's own projections that this summer the companies could barely keep up with demand; Lilly and Amylin had to tell doctors to stop prescribing it to new patients until a new plant came online this September. One worry is a handful of reports that Byetta may inflame the pancreas.

Both Novartis (nyse: NVS ) and Merck (nyse: MRK ) are nearing FDA approval for once-a-day pills that work in a fashion similar to Byetta, although without the promise of weight loss. While not a breakthrough, the pills appear to avoid the excessively low blood sugar and weight gain seen with some older classes of diabetes drugs. They work by inhibiting an enzyme (dipeptidyl peptidase) that gets in the way of the sugar-control process inside the body. (This convoluted chemistry is standard fare in medicine: To gain a desired effect, you inhibit something that inhibits that effect.) Merck and Novartis are likely to tout them as safer medicines that can be readily combined with other diabetes drugs such as GlaxoSmithkline's $3 billion (sales) Avandia. Novartis' entry, Galvus, "has a fantastic side-effect profile, much better than other diabetes drugs at this point in development," says Vasella, who thinks that the class could supplant older drugs. John Amatruda, head of Merck's diabetes clinical research, says his firm is "extremely excited" about how its Januvia pill stacks up against existing agents. Bristol-Myers Squibb (nyse: BMY ) and GlaxoSmithkline are testing similar drugs.

This could be the most competitive mass-market drug launch since the COX-2 anti-inflammatories Celebrex and Vioxx debuted in 1999. "Both companies have quite a bit riding on this. You can be sure they are going to go whole hog to make as big of a marketing push as possible," says Donny Wong of drug market research firm Decision Resources.

But both companies have signaled they will make this a fair fight, a restraint learned from the scandals of the overhyped COX-2 category. "They realize there is a benefit to not beating each other up. If there is a problem with the class, doctors aren't going to use either," says NYU Medical Center physician James Underberg. So far little clinical trial data for either drug have been published in peer-reviewed scientific journals, making it hard to evaluate the drugmakers' claims. Reported side effects to date include headache and coldlike symptoms for both drugs, diarrhea for Merck's entry and dizziness for Novartis'.

The new drug frenzy results from decades-old breakthroughs in understanding blood-sugar regulation. In the late 1970s Harvard Medical School researcher Joel Habener serendipitously discovered a hormone in monkfish called GLP-1 (glucagon-like peptide). It stimulated the body to secrete just the amount of insulin needed to control blood sugar, but not too much. (A relatively small overdose of insulin can put you in a coma.) Habener remembers thinking GLP-1 could be a "super-duper" drug.

Drug companies, including insulin maker Eli Lilly, thought so, too, but GLP-1 turned out to be an impractical medicine: Once injected into the body, it degrades quickly. Pharma firms also tried to make pills that mimicked its effects, with little success.

In the early 1990s endocrinologist John Eng at the VA Medical Center in the Bronx found a chemical in venomous Gila monster saliva that mimicked the effects of GLP-1. When he injected it into diabetic mice, it controlled their blood sugar all night long. He patented the compound with his own money, but the big drug companies weren't interested. When scientists from Amylin Pharmaceuticals saw Eng present his data at a 1996 diabetes conference, they ran for the phones and convinced their bosses to sign a licensing agreement. The Gila monster finding led to Byetta, a synthetic version of what is found in the lizards.

At about the same time, endocrinologist Jens Juul Holst of the University of Copenhagen figured out why GLP-1 dissolved so quickly in the body. In 1995 he and colleague Carolyn Deacon proved that the obscure dipeptidyl peptidase enzyme was the main culprit and in a paper published in the journal Diabetes suggested that drugs that blocked this enzyme would make good treatments.

The study immediately caught the eye of chemist Edwin Villhauer and his colleagues at Novartis in East Hanover, N.J. "Other companies were afraid. We jumped right in," he recalls. Villhauer took advantage of a new rapid-fire method called "combinatorial chemistry," which allowed him to brew hundreds of compounds simultaneously instead of one at a time. He quickly made 1,300 compounds, and by mid-1996 found one that blocked the enzyme. An early clinical trial found that the drug lowered blood sugar with no ugly side effects. That version needed to be given several times a day; it took Villhauer until 1998 to come up with a once-daily version.

By then the field was heating up. Merck in 2000 entered into a deal with the tiny German firm Probiodrug that was pursuing a drug to block the peptidase enzyme. But it sent lab animals into fatal multiorgan failure. "It was nasty," says Merck biologist Nancy Thornberry. Merck killed the molecule a year later. By early 2002 Thornberry had figured out that the problem was that the compound was blocking two other crucial enzymes. It turned out that her chemist colleague Ann Weber had already devised a compound called sitagliptin that spared these enzymes. They rushed it into clinical trials of 60 people in 2002. The initial safety tests showed a reassuringly clean profile. The next step would normally have been to perform a slightly larger trial on a few hundred people to confirm the drug's efficacy. But knowing it was behind Novartis, Merck gambled and moved immediately into large-scale clinical tests on more than 1,000 patients. The gamble paid off, as Merck was able to file a new-drug application six weeks before Novartis.

The next step for Merck and Novartis is to get their drugs through the FDA. That hasn't been easy lately. Last year around this time another diabetes compound, from Bristol-Myers and Merck, was on the verge of approval when cardiologists published data hinting that it caused heart attacks; the compound was dropped. Earlier this year another diabetes compound, this time from AstraZeneca (nyse: AZN ), also failed in late-stage trials because of signs of kidney toxicity. The FDA could render a verdict on Merck's Januvia by mid-October and Novartis' Galvus by December.

Analysts predict each of the drugs could reach peak annual sales of $2.5 billion. Byetta and similar injectables now in trials could bring in another $3 billion annually. But everything depends on whether the drugs' promised safety advantages hold up. After all, it wasn't too long ago that Vioxx was being pushed on the basis of its (apparently) wonderful side-effect profile.
<<