Partly as a result, it is also very expensive: according to figures from the Pharmaceutical Research and Manufacturers of America (PhRMA), the average cost of developing a new drug in 2000 was around $800 million. Generating sales, not to mention a profit, may take years, which can be a big problem for a start-up company that has no sources of income from existing products. Haseltine came up with a solution to this problem that was not only extremely clever but also became the benchmark for other genomic companies. It also probably played an important part in triggering the increasingly stratospheric valuations that were placed on these companies in the years to come.
“We chose an exclusive arrangement with SmithKline Beecham that gave us $125 million up front, plus 10 percent of royalties from drug sales, and the right to co-develop a product in 20 percent of the market,” he later explained. Aside from the sum involved—a $125 million first payment to a company that had existed for barely a year and which was employing leading-edge technologies whose ultimate applicability in the clinical context was completely unproven—there are two other notable elements.
First, although the deal was exclusive—only SmithKline Beecham would be able to exploit whatever leads came out of Venter’s work for the term of the deal—HGS retained the right to develop possible drug candidates in a fifth of the overall market. Second, for any drugs that SmithKline Beecham brought to market, HGS would receive a percentage of the royalties. It was a brilliant solution, because it provided immediate funds to tide the company over while it ramped up its development efforts and guaranteed that it would share in any long-term successes that resulted from its research.
Soon Venter and Haseltine had 80 sequencing machines working on cDNAs in 248 different kinds of tissues taken from 120 human bodies in emergency rooms and pathology labs. By 1994, Haseltine claimed that they had produced 35,000 ESTs, and would be able to identify 80 percent of all major human genes. The financial side seemed to be flourishing, too: in November 1993, when HGS stock began trading, Venter’s 766,612 shares were worth $9.2 million, and had gone up another 50 percent a few months later
Whatever their apparent scientific and financial successes were, however, the relationship between Venter and Haseltine—dubbed the “Gene Kings” by Business Week in 1995, but the “Genomics’ Odd Couple” by Science in 1997— had begun to sour almost immediately. The sticking point, as would so often be the case in the world of digital genomics, was data access. Venter, who clearly regarded himself as a scientist first and foremost, intended to publish his EST results just as he had done in the past. Haseltine, as head of a fledgling start-up whose only assets were the ones that Venter’s sequencing machines were churning out, wanted them kept out of the public domain. After all, why would SmithKline Beecham pay $125 million for information that was freely available? Haseltine’s solution was surprising. As early as 1993, just a few months after the twin-headed venture first got off the ground, HGS launched its own internal EST program. These results could then be kept proprietary and delivered to SmithKline Beecham in accordance with the agreement.
In fact, all the data were later made publicly available, but with conditions attached. HGS would have first option on the commercial rights to any genes that were discovered through using the ESTs, and it would be notified up to 60 days in advance before any data derived from them were published. Many scientists found themselves in something of a quandary: HGS’s database clearly contained a wealth of information that would help drive forward their research, but the conditions imposed were unacceptable for many who saw science and commerce as distinct domains.
Nonetheless, it was commerce that helped scientists resolve this dilemma. On September 28, 1994, the pharmaceutical giant Merck announced that it would be funding an EST project known as the Merck Gene Index, which would produce 300,000 human gene sequences over a two-year period. The driving force behind the idea, Merck’s vice president for research strategy worldwide, Alan Williamson, explained the official thinking behind this unexpected generosity. “Merck’s approach is the most efficient way to encourage progress in genomics research and its commercial applications: by giving all research workers unrestricted access to the resources of the Merck Gene Index, the probability of discovery will increase.”
Digital Revolution and Bioinformatics
