Human Genome Sciences Inc said on Monday that its experimental lupus drug succeeded in a late-stage clinical trial, shocking many who had written the product off, and sending the stock up more than 200 percent.

The results showed patients who took the drug, Benlysta, demonstrated a statistically significant improvement in the symptoms of their disease compared with those taking a placebo.

"This is a classic binary event in biotech," said Ian Somaiya, an analyst at Thomas Weisel Partners. "You have a disease that's difficult to treat, nothing's ever worked and you have a unique scenario in that the trial was actually successful."

The outcome makes Human Genome the latest in a series of biotech companies to have caught investors flat-footed, reinforcing the volatile nature of a sector where success and failure depend largely on clinical trial data and on decisions by the U.S. Food and Drug Administration.

Sometimes, reading the FDA is like trying to read smoke signals. Earlier this year, the agency astounded Wall Street by approving Vanda Pharmaceuticals Inc's schizophrenia drug, pulling the company back from the brink of collapse after even its most bullish supporters had thrown in the towel.

A wrong-way bet on a single trial can leave investors impaled.

As of Friday, about 10 percent of Human Genome's publicly traded shares, or 13 million, had been sold short, meaning those investors were betting the trial would fail.

On the other hand, for those who held on, the rewards were spectacular.

As of March 31, when Human Genome's shares closed at 83 cents a share, the company's four biggest shareholders -- Taube Hodson Stonex Partners, Fidelity, Barclays Global Investors and T. Rowe Price -- held about 23 percent of the company's outstanding shares, according to SEC filings.

Those who held their shares through this morning's opening, generated a return of 1,200 percent.

"It really does take a healthcare specialist and devoted biotech investor to get involved in these names," said Thomas Weisel's Somaiya.

The odds of Human Genome's drug succeeding were, according to most analysts, low. Results from an earlier trial had been disappointing and other drugs, including Biogen Idec Inc and Roche Holding AG's Rituxan, and La Jolla Pharmaceutical Co's Riquent -- both failed in late-stage lupus trials.

Yet, results of Human Genome's 52-week trial -- the first of two requested by U.S. regulators -- showed 57.6 percent of patients taking a high dose of Benlysta experienced an improvement in their symptoms, compared with 43.6 percent who took a placebo.

"Benlysta could be the first true disease-modifying therapy for lupus patients -- a blockbuster opportunity," said Joseph Schwartz, an analyst at Leerink Swann.

Of patients who took a low dose of the drug, which is administered once a month by IV infusion, 51.7 percent showed improvement in their symptoms, a figure that was also statistically significant.

Benlysta's success moved the shares of other companies trying to develop lupus drugs, such as Immunomedics Inc and ZymoGenetics Inc.

Shares of Immunomedics rose 12.5 percent to $3.43, while shares of ZymoGenetics, which is developing a drug in combination with Merck Serono, a unit of Germany's Merck KGaA, rose more than 13 percent to $4.94.

Lupus is a complex disease that causes the immune system to attack the body's own tissue and organs, including the joints, kidneys, heart, lungs, brain, blood or skin. Symptoms include achy joints, fever, arthritis, kidney damage, chest pain and skin rash.

Data from the 867-person trial, known as BLISS-52, take the company one step closer to being the first to have a new lupus drug approved in 50 years. Multiple drugs are approved for other indications and used to treat lupus, but none has been approved specifically for the disease in decades.

The disease affects an estimated 1.5 million people in the United States and 5 million worldwide, according to the Lupus Foundation of America.

Roughly 150,000 patients in the United States stand to benefit from the drug, if it is approved, according to Thomas Watkins, the company's chief executive.

While that may represent a small portion of the total number affected by lupus, it is nonetheless an important advance, and Watkins said that from a corporate perspective, it represents a substantial revenue opportunity.

"A drug like this, with this kind of promise, has the potential to be a blockbuster drug," he said.

Drugs are typically referred to as blockbusters when they generate $1 billion or more in revenue. Profit would be split between Human Genome and its partner, GlaxoSmithKline Plc.

The second of the two late-stage trials is due to be reported in November and the company aims to file for approval of the drug by early 2010, according to David Stump, the company's head of research and development.

If the agency gives it a priority review, the drug could be on the market by the end of next year. Priority review is given to drugs that meet unmet medical needs. Without priority review, it could be on the market by early 2011.

Human Genome shares soared to an intra-day high of $11.29 on Nasdaq. Glaxo shares were up 2.5 percent in London.