Human Genome Sciences Inc. said yesterday that its potentially lucrative new lupus drug, eagerly awaited by investors and patients, had fallen short of expectations in an important round of human testing.

The news drove the company's stock down 29 percent. Although patient advocates, company officials and analysts argued that tests of the drug, called LymphoStat-B, actually contained some good news, investors were awaiting an unequivocal endorsement of the Rockville biotechnology firm's first drug.

The announcement casts a cloud over the drug's development and may lengthen the time it will take to reach the market. That would be a serious setback for lupus patients, who haven't had a new drug in 40 years, and for the 13-year-old company, which like other biotech firms has faced unstinting pressure from investors to deliver a product.

In this case, that product could be worth more than $1 billion a year.

Human Genome Sciences shares closed at $9.87, down $4.10. The shares had briefly fallen to $8.60, their lowest price in more than a year. The downbeat sentiment underlying the stock drop could affect the company should it seek to raise more money -- and large drug studies can cost several hundred million dollars.

Company executives held an hour-long conference call with investors and analysts attempting to mitigate the fallout of the announcement, painting the study results as a success, just not in the terms it had laid out with the Food and Drug Administration. They also indicated they would explore the possibility of carrying out more advanced tests after consulting with the FDA and their corporate partner on the drug, GlaxoSmithKline PLC.

Human Genome Sciences and the FDA had set two clinical goals for the drug: to show that it was twice as good as a placebo, sometimes coupled with standard care, at reducing the signs and symptoms of lupus after 24 weeks, and that it lengthened the time between painful lupus flare-ups. LymphoStat-B did neither for many test patients.

However, there was a large group of patients for whom the drug was successful, company officials said. Patients who had blood tests showing that lupus was actively attacking their body reported significant decreases in their symptoms, but after 52 weeks, not 24. Signs of the disease in the blood also decreased, the company said.

Advocates for lupus patients, company officials, analysts, and doctors who have treated patients with the drug stressed that those subset results actually provided a glimmer of good news.

"There is no doubt in my mind that this is an effective drug for lupus," said Sandra Raymond, president of the Lupus Foundation of America. "I am holding out promise for this drug and lupus patients are as well. I don't think we ought to be as negative about this drug as we might be."

Arthur Weinstein, a Washington Hospital Center physician who had several patients participate in the study, said: "Maybe this drug works in a different period of time than they set out with. It warrants further study. I don't think this drug is moribund or on death's door."

About 1.4 million people -- most of them women -- suffer from lupus, a potentially fatal disease known medically as systemic lupus erythematosus that develops when the body's immune system attacks the kidneys, heart, lungs, brain, blood or skin. Lupus strikes differently from patient to patient, making it difficult for drug companies to demonstrate widespread therapeutic effects required by the FDA.

That has left lupus patients with few options, including high doses of steroids and cancer drugs that have toxic side effects and leave bones dangerously brittle. Anti-malarial drugs are also used.

LymphoStat-B is designed to fight at least part of the biological process that researchers think causes the disease. The company began developing the drug in the late 1990s after discovering a protein called BLyS, which is required for disease-fighting cells to mature and produce antibodies that attack viruses and bacteria.

But sometimes, those disease-fighting cells instead produce cells that attack the body's tissue. Most people's immune systems kill those cells. Some do not, causing autoimmune diseases such as lupus and rheumatoid arthritis.

Research studies by the firm's scientists and others have shown that too much BLyS causes so much cell activity that the vicious antibodies thrive and start attacking the body. LymphoStat-B attacks BLyS and limits the production of those damaging cells, a finding confirmed by the results released yesterday.

David C. Stump, the company's executive vice president of drug development, said that even though the clinical goals in the study were not met, there is now a "clear path forward" to develop the drug, focusing specifically on patients with active signs of the disease in their blood. Those patients made up 75 percent of the 449 participants.

"I see a target population," he said. "It's meaningful. It's a population with need."

It was unclear yesterday how much longer it would now take to develop the drug.

The company is facing stiff competitive pressure from biotech giant Genentech Inc., which is currently enrolling patients in late-stage human tests of Rituxan, a cancer drug that has shown promise in treating lupus patients. The drug takes a more aggressive approach to the disease, depleting the damaging cells rather than moderating them.

Analysts said that the more aggressive approach might mean that Rituxan will work faster than LymphoStat-B. However, Rituxan may cause more side effects and infections because it depletes cells needed to fight infections. LymphoStat-B studies have shown few side effects.

"We may see Rituxan as a drug for flare-ups but maybe its side effect profile is too severe so that it won't be given on a chronic basis," said Edward Tenthoff, an analyst with Piper Jaffray & Co. "But we have to wait and see."