The Antidote: Inside the World of New Pharma

by Barry Werth

  Vertex issued hefty stock options to most of the original scientists to keep them from defecting, although with the labs starting to click, and intriguing new targets sprouting across the spectrum of diseases, none of them had any real mind to leave, not soon. Boger deliberately hired people who, like him, craved the chance to compete at the forefront. Yet when the first researchers arrived in Cambridge, they discovered an atmosphere of almost willful anarchy, an antiorganization—“chaos,” says Sato, whose job it became to bring order. For two years, the scientists had no offices, lugging their backpacks and briefcases from one communal desk to another to use the phone. There was little rank or hierarchy; decisions were made by the project councils. During its initial public offering, launched during a speculative bubble, the raw, head-snapping speed and intensity of the chase and the relentless pressure to do important science while cutting corners overwhelmed the principles of an idealistic young crystallographer, who smashed a chair in the lunchroom in a fit, screaming, “You will all be stricken down!” He left the company soon after to attend medical school.

  Boger’s “social experiment” was designed to encourage self-selected leaders who not only would do excellent science but also grasped viscerally that the decline of Big Pharma was due less to cluelessness at the bench or fecklessness on the executive floors and in the boardroom than to the immobilizing sludge of middle management in between, which even at Merck had led to project heads prioritizing how many compounds a group made over whether or not they did anything useful. Boger wanted champions, people who would passionately disagree with each other and with him, who in the end would push groups and projects ahead because they had smarter ideas, worked harder, generated more compelling data, and persisted when others would quit.

  Mainly through their association with HIV protease, Murcko, Thomson, and Tung emerged from the founding group to become the company’s rising stars. Statistically, most pharmaceutical researchers work their entire careers without helping to produce a drug that makes it all the way to market. That meant that even if Boger was right and Vertex could improve its odds from 1 in 30 to 1 in 10, most Vertex scientists were subject to the same sobering reality, a career-long string of failures where you best found your job satisfaction in something other than success. For most, it became the daunting challenge itself. VX-478 was not a drug, but it looked as if it would be, and those who led in bringing it out of the lab enjoyed a surge in influence, credibility, and prestige.

  Each, in his way, spread his wings. A computational chemist and molecular modeler by training, Murcko stood between the structural biologists, who churned out torrents of data about the smushy interdigitation of atoms, and the chemists, who designed and made inhibitors. He considered VX-478 a tipping point, absolute confirmation of Vertex’s superior strategy for discovering major drugs. “For me the bottom line was: five chemists, eighteen months, two hundred six compounds. It all played out according to script. It was the project from central casting.”

  Murcko’s intellectual curiosity rivaled Boger’s and he was often, in the project councils and at Friday afternoon beer hour, the first and most effective to challenge him; “Joshua the Indeflectable,” he once mused. Nearly a foot shorter than Boger, he was built like a catcher, mustachioed. Like Boger, he liked to tweak the mighty, and he thought as deeply about the iterations of the discovery process and the interactivity of people and ideas as he did about the forces of atoms. Murcko had come to biochemistry through high-speed computing. On his first day at Vertex, he flew to Boston from Philadelphia rather than waste the time driving and worked that night until three in the morning, hands flying over his keyboard. In the early days, he often “pinned” the company’s supercomputer with his experiments; asked how much computing power he could use to do simulations, he deadpanned, “Infinite.”

  Awaiting a crystal structure for ICE, Murcko started transforming his research group into a nascent skunk works, hiring several new people who knew both biology and computer modeling and encouraging them to broaden their thinking about what more was possible. He authorized his scientists to invent new technologies and write their own code if they couldn’t find outside collaborators or buy what they needed. The spark of innovation is asking questions; Murcko questioned everything.

  “The software at the time didn’t take into consideration any of the downstream physical properties of drugs,” he recalls. “How soluble is a compound? If it’s not soluble, it can’t get into cells. Could you use computers to predict not just the lock and key—how does the drug bind to the active site—but go a step further, to see how one molecule might be better than another because it has better physical properties? We ran computer simulations on hundreds of associated molecules and asked, ‘If I was just going to change one atom, could I increase its solubility?’ For VX-478, one small change predicted to be at least a hundredfold more soluble turned out to be even better: five hundredfold more soluble.”

  In January biophysicist Keith Wilson finished the crystal structure of ICE. It was a big protein for the day, a complicated piece of architecture with two domains, and it would receive much glowing press when it was published in the journal Nature. Murcko and his modelers went to work. Sitting at aging Silicon Graphics workstations in a darkened room, they wore clunky wraparound 3-D glasses. Chugging Diet Coke, unshaven, they resembled cave-dwelling ancestors of the LeVar Burton character in Star Trek: The Next Generation. On the screen, stick diagrams of hundreds of connected atoms in brilliant reds, purples, and blues rotated gently, like hair-thin Tinkertoys, in a fathomless black sea. Within hours, the scientists recognized that although the overall folding was different from other protein structures, “down deep” ICE resembled a familiar type of protease with a similar cleaving mechanism. Fortified by what they trusted was an original conceptual breakthrough, since no one else had the structure, they started pulling hundreds of scaffolds from chemical catalogs and Vertex’s relatively tiny compound library and docking them in the binding pocket. Five weeks later to the day, running round-the-clock simulations, they chose a core for a new class of molecules that, once synthesized, would prove so much more potent than any other inhibitor that it leapt instantly to the lead in the combined drug discovery efforts with Vertex’s European partner.

  Tung took the AIDS compound VX-478 public. As lead inventor on the patent application, he became the face of what Boger called Vertex’s first “real major scientific publicity blitz.” Describing himself as “a sort of first-and-a-half generation, mixed background, American-Japanese and -Chinese,” Tung was thirty-four, deep voiced and serious. His sobering intensity masked a fiery ambition to aim high, discover drugs, and avoid ever becoming subordinate on a project. In his second year at Vertex, toiling for several months on a grueling synthesis, often until midnight, he started to sprout grey hairs, singly at first and then in clusters. After leaving the lab, he and the other chemists drank most nights until they closed the bars in Central Square—a rite at several local start-ups. Unlike Murcko he wasn’t yet entirely sold on the value of structure-based design, though this could also be attributed to a near-universal skepticism of bold claims based on partial data, even when the claims were his. “I live to be proven wrong,” he said.

  A year earlier in Berlin, Tung was taken to task for not disclosing the molecule’s chemical structure, and bravura gimmicks like embarrassing Merck at the Washington retroviral conference, while satisfying, only invited more doubts about what Vertex must be concealing. The issue was its patent position. The delay in filing its US application put the company two weeks behind Searle, which had made a molecule with a similar core that appeared to have, like VX-478, exceptional bioavailability.

  More to the point, Searle filed a so-called Markush structure. In the 1920s, chemists, to protect their inventions, sought a way to avoid having to patent individually each member of a class of compounds that would have a similar function. Eugene Markush, a dye manufacturer, filed for and won a patent for a class of compounds
with a replacement group of atoms at several positions, meaning it covered potentially thousands of molecules. Based on the alluring but false premise that combinations of different groups of substituents around a common core generate molecules that have the same activity and biological properties, Markush structures amount, legally speaking, to hurling kegs of nails off the back of a moving truck.

  Boger delayed revealing VX-478’s structure until after the company’s European patent application was made public. Merck, meantime, stumbled. At the retroviral conference in December, attendees had crammed into a presentation in the Washington Sheraton Hotel to hear the company report that patients getting the Merck drug had a 42 percent drop in HIV after just two days of treatment, compared with 1 percent for those on AZT. “We were beside ourselves,” Scolnick would recall. “We thought we had the cure for AIDS.” Then, six weeks later, a molecular biologist examining blood samples from trial participants discovered that the virus had mutated, building resistance to the drug. Another test indicated the virus hadn’t developed resistance. Scolnick called Anthony Fauci, the government’s top AIDS researcher, to get his read. “You’ve got resistance,” Fauci told him. Scolnick protested, bringing up the conflicting data. “I don’t care,” Fauci said.

  “You’ve got resistance.”

  Boger, Sato, and Tung conferred throughout the spring. Formulation and other problems had slowed the work at Wellcome to a crawl. There were so many treatments being tested in humans that it had become difficult to enroll patients for new trials, and it now looked as if Vertex’s molecule wouldn’t be in the clinic until the end of 1994. Pressured to show their hand, they opted to disclose the chemical structure at the Third International Conference on the Prevention of Infection in Nice, France, in June. The audience seemed skeptical from the outset. Despite hoping for a breakthrough, many were eager to see Vertex come down a notch. The compound was a sulfonamide, one of the class of molecules that were the first antibiotics—the sulfas—and now were also sold as anticonvulsants and diuretics. Novel functional groups extended from a core that shared common features with other protease inhibitors.

  Reactions ranged from qualified optimism to mild disappointment. No one felt that Vertex had been blowing smoke, but neither was anyone immediately convinced that VX-478 was all that the company had claimed. There remained major questions: Was it small enough to get inside the brain, something Searle’s compound had not been able to do? What of possible allergic reactions? VX-478 had chemical groups similar to those thought to cause some people to react violently to the antibiotic Bactrim. Carl Dieffenbach, the NIH point man for assessing new AIDS drugs, mused to a reporter, “I’m not sure their patent is as secure as they think it is.”

  Boger weighed these questions only to brush them aside, especially the last. As part of its due diligence, Wellcome had satisfied itself that Vertex’s patent claim was clean. Its lawyers had won—and held on to—the rights to AZT through ten years of intense legal strife, providing an intimidating ally. A few months later, during a November conference call with Wall Street analysts, Searle announced it was stopping development of its protease inhibitor. Two clinical trials had showed no indication of antiviral activity, and Searle researchers believed that the compound was soaked up by a blood protein and removed by the liver. Boger was relieved to learn that the problem was unique to Searle’s molecule, and he considered himself fortunate to have the threat of a patent war suddenly diminished.

  Murcko, as usual, thought well beyond the problem at hand. With the genetic material from Charlie Rice, the main scientific challenge was to discover how the hepatitis C protease worked and how to disable it. But Murcko had a secondary question, one more central to Boger’s mission: How do you evaluate new targets without knowing their structures? Could you predict, say, degree of difficulty? When was it wise to invest in new projects and when wasn’t it? “Up to now at Vertex, we said we either want to have a crystal structure available or we want to know that we can get there first,” Murcko says. “But sometimes maybe you wouldn’t be able to get the crystal structure as quickly as you’d like. Maybe if you had a model it could help steer you away from certain projects. Or say this one isn’t going to appear so easy.”

  Murcko did the experiment. Having no new positions but recognizing the uncommon gifts of a recent Harvard postdoc named Paul Caron, he hired Caron as a temp. Caron was advanced in thinking about gene sequencing and physical similarities among proteins, which fold into spirals and loops and cascading sheets according to the ordering of the amino acid residues out of which they’re made, following the text encoded in their DNA. In other words, if you had the genetic code for a target you might be able to model its active site prospectively, by mapping it against other known proteins. Sitting at a workstation in the modeling room, with a second PC at his side so he could calculate atomic charges and distances, Caron lined up the structures of the few other viral and mammalian proteases that were available and quickly noted that HCV lacked the usual cleft that caused the binding site to be buried in a pocket. “What we saw was virtually a bowling ball,” he said. “Very smooth. All the big loops that come around and make the channel weren’t there.” The target was going to be far more difficult than anything they’d done before.

  One of the company’s advisors, Harvard structural biologist Steven Harrison, categorically dismissed the model, saying it couldn’t be true. Others confessed equal doubts. No one wanted to stop the project, but Murcko and Caron, knowing that the featureless active site presented a steep challenge to design, wanted people to be realistic. “The question at the time, once we had this model, was, ‘Do we continue?’ ” Caron recalls: “It’s going to take a relatively large inhibitor to get enough binding energy, and it’s not going to be an easy thing. We said, ‘This isn’t going to be HIV again. This is going to be a long, hard project.’ ”

  Thomson and his protein group also were learning that HCV protease might not be as tractable as the company’s earlier targets. “We knew enough about the polyprotein that gets made by the virus that it was not a dead ringer for HIV,” he says. “It had some funny new funky ways for creating proteins, including this little cofactor that was essential. It was very conspicuous that there were two adjacent regions that interacted with one another very significantly, and then another third piece that was key to the activity of the protease. And it was a puzzle as to what the architecture was going to look like in the end. More particularly, it gave us this chicken-or-the-egg problem of: we couldn’t make the protein, and we didn’t have a reliable assay to test whether we had made protein. That was really the bind that kept everyone on the planet anchored in the early days.”

  Throughout 1995, the pressure mounted to solve the structure. Merck, Roche, structure-based rival Agouron, and many others were pouring major resources into the race. Meanwhile, public health officials were reaching the conclusion that many more people were infected with HCV than previously predicted, perhaps twice as many as with HIV. Vertex’s attempts at protein production stalled, cycling blindly in a cul-de-sac; the scientists didn’t know what material they had made and couldn’t test it to determine what to do next. With a reliable assay they could try hundreds of subtle changes in conditions to isolate more protein, but that wasn’t an option. “The Dark Ages,” Thomson called it.

  “Everyone along the way had some different cross to bear,” biophysical chemist Ted Fox, who led the lab effort, recalls. “Those project councils were really tough. You got a lot of really energetic people, excited passionate scientists, and you’re getting beaten over the head week after week. I remember at one point, we finally had some active enzyme—doesn’t look good, not very much; maybe we have an assay—and Josh saying, ‘My God. This thing wouldn’t survive if it were this inefficient in the real world. Guys, you’ve got to work harder, or you’ve got to do more.’ ”

  It was Thomson again who broke the logjam, though others at Vertex were thinking along the same lines. He asked Rice’s g
roup to calculate the space between two regions by mapping the interceding genes, then asked the chemists to synthesize chains of amino acids that could be whittled, atom by atom, roughly to that same length. The chemists fashioned tiny artificial spanners, ten or twelve residues across, to form the most intimate molecular “embrace.” The exercise worked. “You took your protein from the bacteria that were engineering it, then you mix it with this little synthetic peptide, and bingo, you’ve got active protease,” he says.

  As Murcko anticipated, some targets are much harder to isolate and purify than others. Nearly two years after receiving Rice’s clones, Fox’s group delivered the first Thomson Unit of HCV protease to the crystallographers.

  In February 1995, a year later than Boger had predicted to Wall Street, VX-478 was given to eighteen AIDS patients in a Phase I clinical trial, a dosing study of oral availability, pharmacokinetics, and tolerability—not its effect on HIV. Less than three weeks later, US regulators cleared a merger between the Burroughs Wellcome Company and Glaxo, creating the world’s largest prescription drugmaker. During the next fourteen months, Aldrich and Boger’s brother Ken, the company’s lead outside counsel, joined with Glaxo Wellcome’s lawyers to try to convince Searle to come to a reasonable solution over its patent claim.

  “Searle had this guy, total jerk, totally irrational about this stuff, but by virtue of that, he ended up being quite effective,” Aldrich recalls. Early in the negotiations, he and Boger flew to Chicago to meet with a group from Searle. Never one to hide his contempt for the practice among drugmakers of aggressively pursuing patent claims for molecules in areas where they already had shut down their own projects, Boger ridiculed Searle’s Markush structure. Searle hadn’t made a sulfonamide and had no experimental data on whether it would work. Aldrich told them Vertex wanted to “work something out and get this important drug to patients.” The discussions dragged on for months afterward, until Glaxo Wellcome’s lawyers also took an unsuccessful stab at it.