Human Genome Sciences Inc. has a treasure trove of cash, a highly respected management team and as many drug prospects as some of the world's top pharmaceutical companies.

But you would never know it from what Wall Street has been doing to it lately.

Over the past 11 months, the company's shares have plummeted nearly 68 percent, from $32.79 in January to $10.65 on Friday.

So why is the Rockville-based company being hammered?

The problem largely relates to cash. That may seem incongruous since the company has a stunning $1.55 billion, most of it left over from stock and debt offerings two years ago at the height of investor excitement over the mapping of the human genome.

But the company has turned heads lately by racking up financial obligations with the gusto of a profit-making powerhouse.

Instead, Human Genome Sciences is unprofitable and has no products on the market. Yet it is in the middle of a building binge.

Since last year, the company has initiated $526 million in acquisition and construction projects on three separate campuses, including a glimmering $250 million headquarters.

And while other biotechnology companies are slashing staffs and research programs because they can't raise money from investors who increasingly consider them too risky, Human Genome has been adding employees. It now has 1,100 workers, more than all but two other biotechnology companies in Maryland.

It also has pushed eight of its experimental drugs into expensive clinical trials.

By the end of next year, ThinkEquity Partners analyst Edward A. Tenthoff recently wrote, the number of Human Genome Sciences drugs in trials should grow to as many as 14 - a "pipeline [that] rivals most big pharma and biotech companies."

Still, some analysts question whether all the drugs HGS currently spends $200 million a year to develop are the right ones, given that some are targeted at helping people with rare diseases - "niches" - that don't always offer huge financial rewards.

But perhaps the biggest reason for Human Genome's recent stock-market pummeling has been the way it is paying for the building expansion: keeping the costs off the books, which allows the company to preserve cash for drug development now but which potentially comes at a greater cost later.

Human Genome has gotten others to borrow money to build or buy the buildings, costing the drug developer only what its pays to lease them.

Although the arrangements have always been disclosed to the Securities and Exchange Commission, they are receiving new scrutiny from some investors in the post-Enron environment.

"In a market like this, all you need is a little noise, and people will be piling on," said Christopher Raymond, a Robert W. Baird & Co. analyst who said Human Genome's accounting is proper. "All you have to do is raise the issue: Anyone has off-balance sheet [financing] of any kind, and people get crazy."

Human Genome's first off-the-books construction project came in 1998 when the company faced a financial challenge common to many young biopharmaceutical companies. It wanted to build a manufacturing plant, but it didn't have enough money to do that and continue to develop its drugs.

Biopharmaceutical companies can hire a contract manufacturer to do the work for them, but risk not being able to book space in the contract manufacturer's plant when they need it - leading to delays that might give advantage to a competing drug.

They can build a plant themselves, but risk being left with an expensive albatross that could sink the company if their first drugs aren't marketable.

"You're always betting on the come" in biotech, said A. Gregory Kelly Jr., a KPMG accountant and senior manager of its technology and life sciences practice. "There's not much you can do about it."

Several years ago, Human Genome's Chief Financial Officer Steven C. Mayer came to the conclusion that it was a better bet to build than to rely solely on contract manufacturers. But how? At the time, the company had only about $180 million in cash - and it needed the vast majority of that to spend on its drugs.

Mayer, a Washington lawyer's son who grew up perfecting self-made go-carts and - later - expertly speeding down Utah's mountain ski slopes, had an entrepreneur's penchant for taking calculated risk.

A Stanford University MBA who joined Human Genome in 1996, he already had spent several years overseeing finances at small biotechs that were often short on cash. He didn't want a repeat experience. At Human Genome, cash preservation would be king.

Mayer was in luck as he started exploring Human Genome's options. The state of Maryland badly wanted to prove it was a place where drugs were not only discovered, but made.

So when Human Genome contacted the state's Department of Business and Economic Development, officials there excitedly referred the company to Hans F. Mayer (no relation)-the head of a nonprofit charged with facilitating in-state economic development.

"Here was a company that not only wanted to develop but also to manufacture," said Hans Mayer, executive director of the nonprofit Maryland Economic Development Corp., or MEDCO. "This was an opportunity to make a statement."

MEDCO borrowed $23 million from a bank to build the plant, getting around the reluctance of traditional banks to loan money to risky biotechs. MEDCO then leased the plant to Human Genome Sciences Inc. for 20 years. The state and Montgomery County threw in an additional $12.7 million in loans, incentives and improvements. All Human Genome Sciences had to pay was a lease.

By the time the off-the-books lease-financing was complete, Human Genome had up to $45 million in plant and equipment with which to fine-tune the making of its experimental drugs. It also had a fund-raising model on which to build.

HGS announced the deal's completion and disclosed it in filings with the SEC. Mayer kept a spaghetti-like flow chart of its myriad components in a PowerPoint presentation, just to show it off. This wasn't the behavior of a company trying to keep a secret.

The off-the-books deal had a feature that later would be repeated on a far larger scale: It required Human Genome to set aside a small amount of its cash reserves as collateral. The money shows up as cash on the company's balance sheet and continues to earn interest, but can't be used for other purposes.

If all of the company's current projects are completed as planned, the restricted amount will balloon to $526 million in 2004 from $195.7 million,according to company filings with the Securities and Exchange Commission.

It is that restricted cash that is giving investors and analysts pause.

The worry is that Human Genome is expected to spend more and more on drug development in succeeding years as its drugs advance into ever-larger clinical trials. At the same time, the company will have to restrict more of its money as collateral as the construction projects advance.

"Having that much money in the bank and not being dependent on the capital markets for periodic funding doesn't necessarily lead one to choosing the most frugal ways of going about one's business," said T. Rowe Price analyst Dr. Jay Markowitz, who said the company should reduce spending.

The way investors perceive Human Genome Sciences is in part influenced by the reputation of William A. Haseltine, its chief executive officer.

It was Haseltine who was the chief spokesman for the company's strategy of genomics-based drug development. He often bragged that the methodology would more readily identify drugs that worked than traditional methods and speed them to market. He is often described by people as a "visionary."

The strategy sold big in the heady days of 2000 after the mapping of the human genome by Celera Genomics Group and a publicly funded project.

Human Genome's shares traded as high as $112.63 that year. Haseltine, with his professorial ability to explain complex sciences in everyday language, became a media darling.

"Human Genome Sciences was one of the original genomics companies, and it was sold to the public through the media and Bill Haseltine," said Dr. Ron Garren, chief biotechnology strategist for Diversified Biotech Group Inc. "This guy Haseltine is a salesman and he's good at it."

But Haseltine's claims have yet to be proven as some of the company's drugs have stumbled in human testing. Some investors began to see them as overblown - a perception fed by the sheer magnitude of its construction projects.

The projects include the purchase of the headquarters of publicly traded Life Technologies Inc. Human Genome now is expanding the complex, which came with three buildings.

Human Genome also has broken ground on a large drug manufacturing plant initially targeted to cost $200-plus million, though the company now says it may redesign the project. The third project, the corporate headquarters, is now being built by as many as 600 workers.

Human Genome Sciences said last year that it needed the office and lab space, given that its growing work force was spread across a dozen or so buildings. Mayer, the company's chief financial officer, didn't think the manufacturing plant would be an albatross because of the worldwide shortage of protein manufacturing space.

Still, the sheer size of the projects - especially the manufacturing plant - worries some analysts.

"The bigger question is, did they need to build that kind of capacity so soon?" said Raymond, the Robert W. Baird & Co. analyst. "That's a tricky question either way."

Mayer, however, still thinks the construction is a winner's bet because Human Genome has the option to buy or refinance the buildings - or walk away at the end of the lease.

The company is so confident in its decision to build that it has given banks this guarantee: The manufacturing plant, research space and headquarters will be worth $459.4 million at the end of seven-year leases covering them, or the company will pay the difference between that and the sale price if it has to walk away.

"What are the odds all these buildings are going to be worth zero at the end of seven years?" said Human Genome's Mayer. "That is the same probability that we'll have a $459.4 million liability."

The lease arrangements didn't become issues until October, when Florida investment bank Sterling Financial Investment Group Inc. highlighted them in a report that suggested investors sell or short the stock, causing shares to fall.

The Sterling report was widely dismissed by other Wall Street analysts. Nonetheless, Mayer was destined to spend much of the next several days denying Sterling's claim that Human Genome Sciences would have just $275 million to $300 million in net cash on its books by year's end - a one-year's supply.

Other analysts have put the company's supply of cash at about four years' worth, despite the company's $500 million in long term debt.

"We are in sound financial condition," Mayer repeatedly said. "We have more than adequate cash to finance our activities going forward."

A "tempest in a teapot" is what he called the concerns. The company was unfairly being swept up in investor worries over the kinds of hidden deals Enron used, he believed. In contrast, Human Genome's deals had always been fully disclosed - along with the risks.

Perhaps the bigger issue, suggested Garren, the Diversified Biotech Group Inc. strategist, is whether Human Genome Sciences is spending its cash developing the right drugs. Investors want blockbusters - drugs that potentially will bring in $500 million to $1 billion a year, he said.

"They can be better mousetraps or novel drugs with sizable markets," Garren said of the kinds of drugs he looks for. "When I see their pipeline, I don't see those drugs."

Human Genome's BLyS protein, for example, is being tested in clinical trials as a treatment for patients with common variable immunodeficiency - a disorder Tenthoff of ThinkEquity said affects just 3,000 to 3,500 Americans.

Its Albuferon drug - designed to be a longer-acting interferon treatment for chronic hepatitis C - is aimed at a much larger U.S. market of 4 million.

But it will be years before it's approved, if ever, and other long-acting treatments already exist. Among them are Roche's recently approved Pegasys and Schering-Plough's Peg-Intron.

However, Markowitz, the T. Rowe Price analyst, said Human Genome's next-in-line drugs would go through a more rigorous process to determine whether they are good bets for development.

That's because the company has beefed up its drug development team, including by hiring a respected former Genentech executive, Dr. David C. Stump, to head it. In the long run, he believes it's a good bet that at least one of Human Genome's drugs will pan out.

But the stocks of biopharmaceutical companies aren't trading up these days on what may happen years from now - unless investors are certain a company has the cash to get there.

"We've gone from a period of hubris to a period of stark reality," said Markowitz.

Human Genome Sciences, he said, has two things: "They have promise and they have cash. In periods of adverse sentiment, promise doesn't count for very much - if anything."

Human Genome Sciences Inc. at a glance:

Headquarters:Rockville

CEO:William A. Haseltine

Employees:1,100

3rd Q revenue:$1.6 million

3rd Q loss:($75.1 million) Includes one-time charge of $32.2 million

Stock symbol:HGSI

Drugs in clinical trials:Repifermin for wound healing; BLyS protein for common variable immunodeficiency; LymphoStat-B for lupus; LymphoRad-131 for multiple myeloma; Trail-R1 mAb for cancer; Albuferon alpha for chronic hepatitis C; Albutropin for human growth hormone deficiency; Albuleukin for solid tumors.