Needle Work

They were costly failures but important salvos in a new war on drugs. In November the Food & Drug Administration nixed an attempt by Novartis and Momenta, a small biotech outfit, to launch a copycat of Sanofi-Aventis' $4 billion clot-buster Lovenox. Reason: The knockoff artists hadn't proved safety and efficacy for their medicine. That same month Roche got FDA approval for a drug similar to Amgen's Epogen, an anemia drug, but was blocked by a judge who ruled Roche was infringing on Amgen's patent.

Generic drug makers have been pushing at the gates of a $30-billion-a-year market for protein drugs. But now big pharma is getting into the act, trying to produce generic versions, the very medicines (to treat multiple sclerosis, diabetes, anemia, deficiencies in growth hormone and the like) that started the biotech revolution 30 years ago and gave rise to Genentech (nyse: DNA - news - people ), Amgen (nasdaq: AMGN - news - people ) and Biogen Idec (nasdaq: BIIB - news - people ). There's too much at stake to let small setbacks get in the way. The giants are starved for new blockbusters--last year had the fewest new drug approvals in two decades--and one-third or so of new products in development are protein drugs. "If the marketplace is open," says Ken C. Cacciatore, a managing director at Cowen & Co., "they will participate." Large pharmaceuticals, he says, "have the capabilities at their disposal now." On another front, as part of their approaches to controlling health care costs, presidential hopefuls John McCain and Hillary Clinton have plans to make biogenerics a reality. Congress has five different bills, too.

But will the follow-ons be cheaper than the originals? To some degree that depends on Capitol Hill and whether a resulting law requires limited or extensive clinical trials. What can't be legislated away is the high cost of producing biologics.

Protein drugs require living cells and a delicate and very expensive manufacturing process that pushes the cost of developing a new drug, even a look-alike, to $200 million to $300 million, according to Cowen. That's less than the average $900 million you need to launch an original drug. But it's way more than what it costs to copy a simple molecule like the one in Zocor or Zantac.

Simple pills get an FDA pass based on clinical trials involving only blood tests to ensure the active drug is present at the right levels. A new knockoff of an existing heartburn treatment might cost $2 million. Competition among other generics will drive down consumer prices, creating a 30% drop when the first rivals step in and an 80% discount once 20 or more competitors pile on, according to researchers at Duke University. That's why small-molecule generics are so much cheaper in America than they are in Canada and western Europe.

It's a different story with biogenerics. Steep costs, as well as high barriers to entering this business, will restrict competition to a small contest among the titans of the industry. Some version of clinical trials, not just blood work, will be required to prove safety and efficacy. Most generics companies simply can't afford the hassle. All this is likely to keep prices high.

Already companies like Pfizer (nyse: PFE - news - people ), Merck (nyse: MRK - news - people ) and Bristol-Myers Squibb (nyse: BMY - news - people ) are quietly drawing new battle lines. They are spending hundreds of millions of dollars to build manufacturing plants--giant bio-vats that hold millions of gallons of cultured mammal cells--in order to develop new protein drugs to treat cancer, rheumatoid arthritis and Alzheimer's. Biogenerics will be reliable moneymakers, too. Big pharma needs as many cash machines as it can get.

Merck has a killer app when it comes to making copycat proteins. In 2006 it dropped $400 million on privately held GlycoFi to get its hands on a technology that can produce biotech drugs cheaply in yeast, instead of in complex and finicky mammalian cells. Merck wanted the tools for its own drug development. But one of GlycoFi's proofs of concept was a scientific paper showing its method could be used to create a copy of Amgen's Epogen.

By the Numbers

Drug Money

Genzyme's Cerezyme, which treats Gaucher disease, is a blockbuster, treating few patients at a very high price.

$1.1 billion Cerezyme's sales last year.

5,000 The number of people who have received Cerezyme prescriptions.

30% Expected discount from the first copycat drug when Cerezyme goes off patent.