(FORTUNE Magazine) – The last time it won marketing approval for a product from the FDA, Genentech Inc. lit up the night sky over San Francisco Bay with a spectacular fireworks display. Employees, families, and friends gathered under a huge tent erected in the company's parking lot on Point San Bruno, a small promontory that a bay breeze cools like natural air conditioning. As the crowd stepped from under the tent to cheer the fireworks, neighboring San Francisco International Airport held incoming and outgoing flights during the spectacular finish to the festivities. The fireworks, on October 26, 1985, celebrated approval of genetically engineered growth hormone, which helps increase the stature of abnormally short children. Those were the days at Genentech. According to its many boosters, it was well on its way to becoming biotechnology's first runaway success. Yet, a few weeks ago gloom hung over Genentech's seven low-slung gray buildings, along with unaccustomed 90-degree heat. Biotechnology's famous flagship had just been hit by the equivalent of an Exocet missile and was taking on water. Late in May an FDA advisory committee recommended against immediate approval of Genentech's blood clot-busting drug, t-PA, or tissue plasminogen activator, pending further clinical studies. The drug was supposed ! to become as potent a weapon in fighting heart attacks as antibiotics are in suppressing infections. Boyish-looking Robert A. Swanson, 39, Genentech's co- founder and C.E.O., who had gambled the company's future on t-PA, likes to refer to himself as ''the captain of this ship.'' He insisted gamely in an interview with FORTUNE: ''Nothing has changed. We're still the same company. We're steaming ahead.'' PRACTICALLY everyone else, however, feels that a whole lot has changed -- perhaps irrevocably. Genentech's chairman is Thomas J. Perkins, a partner in the hugely successful San Francisco high-tech venture capital firm Kleiner Perkins Caufield & Byers. Tracked down vacationing in England, he described the advisory committee's action as ''a very serious matter for Genentech.'' At the company, another executive concedes, ''morale is horrible.'' Genentech's seemingly insurmountable problem is that -- even more than the rest of the biotechnology industry -- it has soared on promise, not products. According to its mesmerized following on Wall Street and in much of the business press, the company was moving smartly from one milestone to another in accordance with a brilliantly executed plan. T-PA alone, enthusiasts said, would bring Genentech $1 billion a year. Chief executive Swanson, a bright MIT MBA and a former Kleiner Perkins junior partner, introduced to the industry a number of clever financing plans, including limited partnerships, that underwrote the development of Genentech's three products: human insulin (licensed to Eli Lilly), alpha interferon (licensed to Hoffmann-La Roche), and human growth hormone, the only product Genentech markets on its own, which provided revenues of $43.6 million last year. In 1986, Genentech received an estimated $10 million in royalties from Lilly and Hoffmann-La Roche, and $73 million from research contracts. The company earned a further $6.6 million in interest on the $84 million in capital it has on hand. Besides cleverness in cloning cash, luck has helped too. Genentech's growth hormone, for instance, wound up as the only such product on the market when the FDA banned its predecessor after it was found to be contaminated by a dangerous virus. Perkins's presence as chairman didn't hurt. He helped start the company in 1976 with $100,000 in Kleiner Perkins money; the following year the firm put another $100,000 into Genentech. At the stock's high of $64.75 on March 24, Kleiner Perkins's $200,000 investment would have been worth $360 million; the total market capitalization of Genentech was $5.3 billion, nearly that of a well-established pharmaceutical house like Schering-Plough. Genentech's influential investment banker, Hambrecht & Quist of San Francisco, was certain the company could do no wrong.

Today all kinds of people are suddenly seeing all kinds of things wrong with Genentech, from management skills to claims about t-PA's potential. A bulletin for investors published after the FDA fiasco by Casdin Associates, a New York biotech investment firm, talks of ''either arrogance or management failure or both,'' and adds, ''One thing is clear: Genentech's image of invincibility is shattered.'' The report suggests that Genentech's stock will soon skid to $20 a share. In mid-June it was $42. Robert Fildes, president of Cetus Corp., a major Genentech rival, says he always found billion-dollar forecasts for t-PA ''ridiculous.'' Unlike SmithKline Beckman's ulcer-soothing Tagamet, one of the few drugs that actually does constitute a billion-dollar annual business, Fildes says, ''you can't pop t-PA like a tablet.'' Much more serious to Genentech are recent reports from clinics that all clot-dissolving agents are about equally effective if given within three hours after a heart attack. Genentech had based its claim of t-PA's superiority over older drugs on an earlier study. To compound the injury, in the morning of the same day that the FDA medical panel turned down t-PA, it recommended approval of an older clot-dissolving drug, streptokinase, made by West Germany's Hoechst and Sweden's KabiVitrum, even though an early clinical test indicated it to be slower acting than t-PA. Streptokinase, an enzyme extracted from the streptococcus bacteria, had been in use as a drug injected by catheter directly into the coronary artery. The committee suggested that intravenous use, also proposed for t-PA, should be approved. At a projected cost to hospitals of below $200 a dose, compared with up to $3,000 for t-PA, streptokinase will be difficult to dislodge once established in the market. ''What's the big deal about t-PA at that price?'' asks Ronald G. Leonardi, vice president for regulatory and clinical affairs at the U.S. division of KabiVitrum in Alameda, California. Pending final approval by the FDA, Hoechst and KabiVitrum plan to get streptokinase to market within two months. Genentech, in the view of most analysts, is unlikely to get t-PA approved for another year. On top of that, about two dozen U.S., European, and Japanese companies are all rushing variations of t-PA, streptokinase, or still other similar drugs into production. The field includes such formidable players as West Germany's BASF and Britain's Beecham and Wellcome, as well as U.S. drug and chemical companies like SmithKline, Lilly, Monsanto, and Upjohn, many of them working with small biotech start-ups. In that crowd, Genentech's t-PA will more than likely be trampled, going from blockbuster to also-ran, no matter how soon the FDA reconsiders approval. ''Genentech will be just one entrant among many in the clot-dissolving derby,'' says David M. Paisley, who is president of a drug consulting firm in Urbana, Illinois. SINCE GENENTECH'S future was riding on t-PA, the company is now an ex- flagship. In the view of many, Genentech will be lucky to attain product sales of $250 million in the early 1990s, much less the projected $1 billion. Sobered analysts look at the Genentech product pipeline and see little. A previously hailed cancer-fighting substance called tumor necrosis factor, or TNF, another big Genentech hope, has failed to show much promise in clinical trials. The company's gamma interferon, also in clinical trials, is iffy because of its limited effectiveness. Established products are under attack. Growth hormone, the only product the company markets itself, is now threatened by FDA approval earlier this year of Lilly's improved version. Lilly's superior sales force is expected to batter Genentech's smaller and now- demoralized sales crew. Patent issues involving t-PA are far from settled and add to Genentech's risk. Questions about the vaunted Genentech management skills will continue to haunt the company. Heads may roll, possibly including that of President G. Kirk Raab, brought in from Abbott Laboratories in 1985 to deal with FDA matters, among other things. What went wrong at Genentech is perhaps best suggested by Bruce Pharriss, senior vice president of Collagen Corp., a Palo Alto biotech company but not a Genentech competitor: ''When you start believing your own bullshit -- that's lethal.''