Then Vivus, a US drug delivery company, found an alternative - packaging the same drug into pellets to be delicately inserted into the urethra - and took half the $60m world market in four months.
The next improvement could come from PowderJect, which will float on the London stock market this month and has a pain-free, needle-less injection. "If all men suffering from impotence had treatment, this would be a $500m market," says Paul Drayson, PowderJect's chairman.
As this example shows starkly, the way a drug is given can matter at least as much as what it does. This is the premise behind the drug delivery industry. The sector has come from nowhere 10 years ago, to capture a $12bn slice of the $150bn world-wide pharmaceutical market. Moreover drug delivery is growing fast - at 15-20 per cent a year, twice as fast as the wider drugs market.
The product that put drug delivery on the map was a heart pill called Procardia XL. The first version of Procardia, sold by Pfizer, had to be swallowed three times a day, restricting sales to $400m a year, and in 1989 was about to lose its patent.
In a well-timed move, Pfizer linked up with Alza - a then tiny drug delivery company - which came up with a once-a-day version, reviving Procardia's patent and tripling sales to $1.2bn. Felix Theeuwes, Alza's president of research and development, remembers the excitement: "We enormously expanded the market for angina drugs. Our technology gave Pfizer its first billion-dollar drug."
The range of delivery technologies today is vast and ingenious. It includes everything from simple skin patches, inhalers, gels and nasal sprays to implants, pills with microchips and contact lenses that release drugs into the eye.
SkyePharma's Geomatrix technology, for example, can release a drug in tablet form immediately, slowly over days or in prescribed bursts so that, say, someone at risk of heart attack could take a pill before bed and get regular doses throughout the night. Elan is developing a smart pill embedded with a microchip and Alza has a titanium implant, no bigger than a matchstick, that leaks out a cancer drug over a year.
The most exciting area in delivery, though, is driven by the biotechnology industry, which is developing protein and peptide-based compounds like insulin. These large and complex molecules have to be given by injection because they are digested if swallowed and are too big to pass normally through the skin.
Among the many companies working in this area, Cortecs is developing insulin and brittle bone pills, Inhale has an insulin dry powder inhaler and PowderJect a helium pump which can drive molecules through the skin at three times the speed of sound.
For both patients and healthcare payers, there are huge benefits in all this - the easier a drug is to take, the more compliant a patient will be. Alex Zisson of US broker Hambrecht & Quist says the cost benefits of improving compliance are central to the delivery sector: "In the US, 28 per cent of hospital admissions can be traced to people not taking their drugs properly."
Glen Travers, chairman of Cortecs, says that less than a third of women prescribed hormone patches for brittle bones stay on the treatment after two years: "The cost to the US government in hospital fees to treat broken bones is $10bn a year and growing."
For the big pharmaceutical companies, meanwhile, a clever delivery system can extend the patent life of their drugs, generating substantial extra revenues. The number of alliances with specialist delivery companies is growing.
Alza's Mr Theeuwes says: "It is financially unrewarding for pharma groups to spend millions on their own technologies to improve just one drug." Donal Geaney, president of Elan, says: "Many medicines stay on the shelf because they cannot be delivered properly and this is costly."
Rolf Stahel, chief executive of Shire, sees another reason for alliances: "Research directors in big pharma groups are not enthusiastic about delivery. It does not lead to Nobel prizewinners."
Investors, though, should be more excited. While the small numbers of UK delivery companies are normally lumped in with their biotechnology cousins, as Jo Walton of broker Lehman Brothers points out, their risk profiles are substantially different.
"Delivery companies work on existing chemicals, where all the data - animal tests, safety, which patients to target, which clinical trials to use - are already known. Times to market are shorter and the chances of success are higher." Mr Zisson agrees: "It takes a delivery company about three to four years to get a drug to market, compared to six years for a normal drug."
Financial risks are also lower. Robert Chess, chief executive of Inhale, estimates his research costs are a tenth those of an average biotech company and with three times as many products, risks are spread.
And as Richard Stewart, SkyePharma's finance director, points out, the development costs of reformulating an old drug are borne by clients, while established sales and marketing teams can push the new version at minimal cost: "All this means more in royalties," he says.
Though royalties can limit upside, companies working in protein delivery or on big drugs stand to gain substantial returns and make profits sooner than most biotechs. So far, their potential is more apparent in the US. According to Ms Walton: "Investors are more cynical in the UK, where hopefuls like Cortecs have not yet made it. In the US there are some huge and profitable companies." UK delivery companies may not be far behind.
Drug delivery contenders
Company Market value
Alza Corp (US) $2.5bn
Cortecs (UK) pounds 277m
Dura Pharmacls (US) $1.7bn
Elan Corp (Irish/US) pounds 2.5bn
Ethical Holdings (US/UK) $78m
Guilford Pharmacls $501m
Inhale (US) $307m
Liposome Company (US) $993m
PowderJect (UK) pounds 109m
RP Scherer $1.3bn
Shire Pharmacls (UK) pounds 198m
SkyePharma (UK) pounds 307m
TheraTech (US) $237m
Vivus (US) $13m