In the weeks before the Glaxo visit, when not only Thomson but all of Vertex’s scientists were working feverishly, Boger often sat at his desk cranelike and unmoving for long periods, his eyes fixed on a computer screen, a look about him of preternatural composure. Occasionally he unfurled several long fingers to type something or reached without looking into a foot-high mound of plastic slide envelopes stacked precariously at his left. Extracting a slide, he examined it against the light, then put it in a small box or returned it to its slot, his visage unchanged. He stopped eating during these times, causing Roger Tung, one of the chemists, to suggest that he was an autotroph, an organism of such exquisite metabolism that it supplies its own food and can live on air. Compared with the researchers’ own anxious exertions and the towering unknowns that swirled before them, Boger’s calm and assiduous attention to his slides seemed to many of them—dangerously, a few argued—aloof.

Boger thought to convince them otherwise was “an unpersuadable issue,” and so he didn’t try. He was in full “sell mode”—he was thinking about money—and there was no complication in his mind about what Vertex had to do, what he had to do, or what the scientists had to do. He let them think about him whatever was necessary to make a better drug than FK-506. Boger was now acting on first principles: Long before Vertex could design and sell its first drug, it had to design and sell itself. Small companies that are years away from making a profit are called story stocks on Wall Street because their value is based not on products or sales, but information—information about, and generally supplied by, the company. Usually the term connotes a flagrant volatility: Company A’s value soars upon winning FDA approval to test a drug in humans, then plummets as Company B files a patent interference. But there is also an element of illusion and seduction, of hype. Among investors, stories generate heat and, if one is fortunate, lust. Far from incidental, the possibility for such arousal is oxygen to emerging drug companies, because it enables them to raise hundreds of millions of dollars in the face of what appear to be wanton, savings and loan-sized losses. Thus Boger’s attention to his slides. They comprised a montage of the story he was scripting for Vertex, a narrative that would have to stand up, with or without actual progress in the labs, indefinitely and against other stories that were equally compelling.

Boger knew that stories have to be accessible and that what investors want most from them is affirmation, so he molded Vertex’s slide show not as a disquisition on science or business strategy, but as a quest. The grail—the object of the quest—was structure-based design and its transcendent prize of safer, smarter, more profitable drugs. The impetus, as always in such stories, was a combination of righteousness and greed; Vertex had a better way to discover drugs than screening and biotechnology (both of which, Boger would say, were terminally limited) and was intent on capturing the spoils of its victory whole. The rationale for the quest was the company’s unique melding of disciplines and technologies, which he represented as a kind of circular flying wedge, and its scientists, who, he noted, all came from the world’s most powerful research institutions. Harvard, naturally, was a key supporting element, as was Merck, and on the financial side, Benno Schmidt. FK-506 and immunosuppression were the story’s set pieces, meant to illustrate its correctness.

That was the text. There also were subtexts that Boger didn’t mention, the most intriguing being about himself. Boger never referred in his slides to his relationship with Merck, but he was seldom introduced anywhere without it being mentioned. To listeners with a knowledge of the drug industry, his defection was the most tantalizing part of Vertex’s story, introducing, as it did, a whiff of patricidal intent, of vengeance. Here was Boger, a scion of America’s Most Admired Corporation, the most productive drug company in history, Wall Street’s gold standard, rejecting all that it had to offer because he thought he could do better. It didn’t take a rereading of Genesis: Boger’s saga of defiant departure was as old as Adam.

Because of his central place in the story, Boger believed he also had to be the one to tell it; as Aldrich put it, “People want to kick the tires. They want to see Josh.” The decision complicated Boger’s position within the company. As Vertex’s chief scientist and CEO, he was responsible equally for its research and its business. But he had not stopped traveling in a year, and the rigors of selling had drawn him increasingly away from the labs. His grasp of the intricacies of each lab’s work and of their interrelationship remained unequaled within the company; chemist Jeff Saunders, far from dismayed by his absences, was astonished by his “omniscience.” But some of the scientists began to doubt that he could competently manage both roles for long. A few missed his confident presence and were floundering.

Boger saw no evidence that he or the company was seriously faltering, and so he dismissed their concerns. He believed he could run both the science and the business as long as he preferred, provided that everyone else did what he expected. Typically, as whenever others raised worries about events that hadn’t occurred yet, he decisively deferred action. If it got to a point where he couldn’t do everything, he told them, he would do something about it then. Now, his job was to sell, which in Boger’s view meant a calculated attempt to master and improve the contradictory art of selling.

“Anytime you talk to people—especially a group of people—you’re by definition projecting an illusion,” he said some months later. “You’re projecting the illusion that you’re talking to each one of them individually. So you’ve got to picture in your mind what the conversation would be if it was just with one person. You’ve got to make each individual feel like they’re witnessing a play in a drawing room and that they’re sitting in a comfortable chair listening to lines that are intended just for them.”

Boger used a variety of stage tricks to elicit such feigned intimacy—lowering his voice to make a point; speaking to the back row, which gave those in forward rows the impression that he was speaking to them. His main device, though, was a strenuously rehearsed spontaneity. As he spoke, Boger listened intently to his own words, then quickly anticipated the questions of those in the audience and tried to answer them matter-of-factly in the next sentence or two. Thinking that this was what anyone who wanted something from another person did unconsciously, as a matter of course, he was surprised that others didn’t try to systematize it. And yet he did it so well that it allowed him to give the same talk over and over, never once the same, without the slightest trace of boredom.

“You can’t fake excitement and you can’t fake sincerity,” he said. “It can’t be done . . . and to do it properly you have to practice.”

Practicing now, Boger believed he knew exactly what it would take to sell Vertex’s story to Glaxo. He also knew that Vertex didn’t have it; no company did. In the process of cross-examining himself, he heard questions that the big drugmaker would surely want fully resolved and for which there were no firm answers. What role did cyclophilin and FKBP have in suppressing the immune system? How did they function? Why should Glaxo put up a dime to design molecules for a target that, though it bound to FK-506 chemically, had no demonstrable link to the drug’s activity?

Boger treated the question of FKBP’s biological relevance with a studied pragmatism; it mattered, he would argue, but not nearly as much as generally believed. Because FKBP bound so tightly to FK-506, a drug of rare and proven efficacy, it either had to be the right target or was close enough in molecular structure to serve as an interim template for drug design. It was a chemist’s view, to be sure, and Vertex’s biologists worried especially that Boger gave the question short shrift. They suspected him of whistling past the graveyard. Boger, however, saw no immediate downside and so persisted.

Appealing to Glaxo on strategic, rather than scientific, grounds was for Boger a no-brainer. Glaxo, thoroughly and cantankerously British, had developed the world’s biggest-selling drug, the antiulcer agent Zantac, through a traditional combination of screening and medicinal chemistry; it had a well-earned and lordly disdain for unproven methodologies like structure-based design. Yet in Boger’s reckoning, the company couldn’t fail to be deeply interested in Vertex. Best-selling drugs seldom retain their position for more than a few years either because their patents expire or new blockbusters come along and push them aside. The treatment of autoimmune diseases with immunosuppressants, meanwhile, represented a virgin market of perhaps $4 billion per year. Implicit in Vertex’s story was that FKBP, discovered by Harding and controlled by Schreiber, was the best target for immunosuppression yet found and that Vertex knew more about it than anyone else, including Merck. Glaxo, needing to stuff its pipeline in its product war with Merck, could not but be extremely attentive.

Of course, there were at the time other stories about other small companies also working in immunosuppression that Boger, though he disdained most of them, was forced to take into account. For instance, a California company named Cytel had recently signed a $30 million deal with Sandoz to develop drugs for rheumatoid arthritis, juvenile diabetes, and multiple sclerosis—the same autoimmune diseases targeted by Vertex and practically every other new company working in the area. Cytel planned to design molecules that blocked receptors on the surface of T cells that falsely recognized the body’s own tissue as alien. Seattle-based Icos, the most talked-about biotech story of the year, had raised $33 million while still on paper by saying that it would block “cell-adhesion molecules,” which act like miniature Velcro tabs. The company intended to gum up the molecules so that they couldn’t snare errant white blood cells and divert them to areas where they result in inflammation.

Boger was unimpressed. Perhaps these molecules were legitimate targets, but whether anyone could block them with drugs remained unproven. On the other hand, FKBP bound to a molecule of clear therapeutic value. To Boger’s mind, the distinction was immense. Cyclosporine and FK-506 had already vouchsafed Vertex’s approach. Cell-adhesion blockers? Boger thought perhaps in a decade the world would know better whether such molecules had even a chance of working.

And yet Boger couldn’t deny that companies like Cytel and Icos made great stories. Preparing his slide show, he factored them in now. He would not resort to their extreme speculation, but if Glaxo wanted biology, biology he would give them. If it wanted proof that FKBP was the optimal target for immunosuppression, he would make a powerful case. If the company wasn’t, as Chugai was, interested in structure-based drug design, rejecting Vertex’s very reason for being, he would still find a way to represent a deal between the two companies as being in the best interest of both. Stories, Boger knew, could be edited to suit a variety of audiences. It was the key to successful selling.

•  •  •

Unlike the meeting with Chugai, which was informally arranged and anointed by Schmidt, Glaxo’s visit to Vertex’s labs was steeped in protocol and strained by nuance. Lab coats were cleaned and pressed and, for the first time in many cases, worn. Every penny of the $4 million Boger had spent on equipment was ceremoniously engaged, every machine and computer “pinned” for maximum display value. Shepherding Glaxo’s six-man delegation was its North American licensing chief, Rick Hammill, who had done the advance work for the visit and stood to gain most if it presaged a deal. But it was a staid Briton, Leslie Hudson, that Boger knew Vertex must impress. Hudson was Glaxo’s chief of immunology, which by definition meant he would lose by a deal with Vertex. Not only would it imply a failure within Hudson’s department, but he’d have to pay for it out of his own budget. Boger detected no light behind Hudson’s visage, nor did he expect to.

As Jeremy Knowles had cautioned at the October SAB meeting, impressing a potential scientific collaborator without revealing so much that you become dispensable is often impossible. Thus Vertex began at a disadvantage. Hudson, speaking first, predictably said little beyond the obvious: that Glaxo was launching a screening program in immunosuppression and was looking for targets to screen against. But Vertex couldn’t afford such generality. Boger had warned the scientists that they had to risk telling Glaxo all they knew about FKBP in order to create a strong negotiating position. Now, as they began detailing the work of the past several months, a few of them felt, despite the decorum, that they were about to be reamed.

David Livingston would recall that ten minutes into his slide show, two or three members of the Glaxo team were scribbling so furiously in their notebooks that the table shook. Others noted the same breathless interest in the comments of Debra Peattie. Peattie, a tall, thin biologist whom Boger had hired away from Harvard Medical School, suggested a way to improve the use of FKBP to screen compounds, based on an unpublished observation. Sensing Glaxo’s keenness and Aldrich’s horror—“Rich was turning white,” she says—Peattie quickly returned to the lab to record the idea in her lab book, having her assistant, Judy Lippke, cosign the entry.

Hudson was politely enthusiastic about Thomson’s sample of pure bovine FKBP and about Manual Navia’s disclosure that he had coaxed from it what appeared to be the first known crystals of the protein. Privately Navia called the slender, needlelike crystals Mickey Mouse, but the ability to grow them just three months after the opening of the Vertex labs supported his and Boger’s boast that Vertex would generate and use structural information faster and better than anyone else. Reluctant to take credit for so preliminary an accomplishment, Navia acknowleged his debt to Thomson for providing him protein of exceptional purity, a blandishment that Thomson underscored by working again through the night.

Several times Hudson importuned for more data, but otherwise he seemed impatient: It was Schreiber he’d flown across the ocean to see and hear. Boger hadn’t misled Glaxo about the delays with Harvard, nor had he told Schreiber of his internal exile. He wasn’t dissembling. But his plan was to exclude Schreiber from any discussion of chemistry or biophysics, then cede him the major part of the meeting to talk about biology.

Schreiber performed flawlessly. He made a compelling case for his theory that FK-506 consisted of two distinct regions—binding and affector—and that the necessity of both for immunosuppression suggested that the molecule worked by imbedding itself in a host, then reaching out and hooking onto something else, most likely another protein. Though Schreiber didn’t claim to know what this partner was and although his hypothesis implied that it, not FKBP, was the drug’s real target, the pluses in his scenario strongly outfavored the minuses in supporting Vertex’s view of FKBP’s critical importance. Boger was pleased. Vertex’s bases were well covered. Even if there was such a molecular partner to FKBP, the clear implication was that Vertex and Schreiber’s group at Harvard were already collaborating to find it.

Of distinctly more interest to Vertex than Glaxo was Schreiber’s announcement that his lab had now produced 400 milligrams of pure, recombinent FKBP. Boger, sitting across from him, noted to himself the amount. The only conceivable reason for directing a highly promising graduate student or postdoc into the career abyss of grinding out that much protein was to try to solve its structure. Schreiber would not want to sit on such a treasure for long if a collaboration with Navia couldn’t be arranged soon. Boger also noted Schreiber’s pleasure at speaking with Glaxo. “If he gets stroked any more he’ll be too shiny to look at,” he said privately after the meeting.

Hudson listened noncommitally. Boger was uncharacteristically quiet, preferring to let the story unfold by itself. Neither was undiplomatic, and yet as the morning went on both were clearly becoming more and more irritated. Hudson was not hearing the proof he wanted that FKBP was the target Glaxo should be pursuing, and Boger was beginning to sense that Glaxo’s disdain for rational drug design was more than academic. It was about money. “I could see they were trying to figure out how they were going to fold everything else but biology into overhead,” he says. “We were going to be expensive for them, and I didn’t need to have my research controlled by someone who didn’t believe in my basic philosophy. I’ve been there before.”

Finally, breaking the standstill, Hudson asked Boger for his own view on how immunosuppression worked. The room drew silent, a scuffling of chairs quenching in its own echo. Tartly put—“belligerent,” Livingston recalled it—the question seemed to have only two answers, equally unacceptable. Either Boger had to concede he didn’t know, deflating Vertex’s plumped-up claims for FKBP, or else he had an explanation but insufficient data to support it. Either way, he seemed to be left little room to avoid being embarrassed, much less to salvage a deal.

Boger leaned forward. He told Hudson insouciantly that it was too early to have such a view nor was it important to have one at the present time. He would have one, he said, when the information warranted.

Hudson stared. He said nothing. It wasn’t until after, as the two groups of scientists stood around talking, that he remarked offhandedly that he agreed.

Whether Boger had anticipated the question and rehearsed his answer or was simply reasserting his view that decisions are only incorrect when they’re premature or neglect the most advanced data, Glaxo’s tone suddenly now resumed its earlier friendliness. Hudson, meeting privately with Boger and Aldrich in Aldrich’s office, suggested that Glaxo might be able to fold some of Vertex’s structural research into a deal after all. In the next days, a flurry of follow-up calls from Rick Hammill confirmed Glaxo’s interest. In the postmortems at Vertex, a consensus now emerged that the company, with barely three months of experiments behind it, had held up under direct fire. After the months of misgivings about Schreiber and Harvard, the aura of triumph, of invincibility, of big things about to happen, returned to the lunchroom. Aldrich, generally pessimistic in any negotiation, now placed the odds of doing a deal with Glaxo at 50-50. Once again, Boger seemed charmed.

Only Livingston, formerly a senior scientist with a once-promising biotech company that had run out of money and was sold amid layoffs and other misfortunes, debunked the good feeling. “I think all this hype about this being Vertex’s deal to lose is a crock,” he snapped. “Glaxo did exactly what they should have done. They took back everything that we gave them, and they’re keeping us waiting as they evaluate it and see where it fits into their plans.”

•  •  •

Events hurtled disproportionately, quickly now. Two days after the visit from Glaxo, on the afternoon before Boger and Aldrich were to leave on a long-scheduled eleven-day door-knocking mission to Japan, Chugai’s director of research and several other top company officials arrived at Vertex as a follow-up to the initial contact in New York. To Aldrich, it was a sign of Chugai’s exceptional urgency that the company would not be put off until after his and Boger’s Japan trip—a perception that grew decisively when its haggard delegation, arriving from Europe, took a cab directly from the airport to Vertex and remained cloistered in scientific discussions for six hours before finally going to their hotel. At 7 P.M., with the normal time frame for negotiations telescoping wildly, Aldrich went to the University Club for a swim, then returned to Vertex until midnight to write a formal business proposal to Chugai. A week before, the two companies had barely known of each other.

He and Boger boarded a 7 A.M. flight to San Francisco, where they would make connections for Tokyo. Aldrich called these trips “death marches,” though it was during them that his status rose. At Vertex he was often excluded by the scientists, but here, in the air next to Boger, he was the company’s vanguard, its elite. Aldrich called Vertex from the cabin, only to discover that Glaxo, apparently having second thoughts, was also now pressing insistently for another meeting. The company had called that morning, inviting Boger to fly to London at once for discussions with its senior officers. Caucusing in front of the first-class lavatory, Aldrich and Boger reviewed their options. “It wasn’t going to be a straightforward exercise,” Aldrich recalled.

Glaxo and Chugai didn’t know about each other, nor could they know except in the most general terms. On the other hand, the attractiveness of using two ardent suitors as leverage against each other was an obvious no-brainer. Boger and Aldrich decided to stall Glaxo, at the risk of losing its interest, at least until they returned to Cambridge.

It was gratifying, this sudden surge of attention, and as Boger would say, wholly anticipated. And yet it also pointed up sharply the strengths and weaknesses of his position. He had assembled all the elements of a luminous story—so luminous that it had enticed two companies that saw within it prospects that couldn’t be more different and represented for Boger, and for Vertex, starkly divergent futures. Glaxo was the more formidable development partner: richer, more aggressive, more capable of bringing a major new drug to market. But Glaxo, like Merck, saw structure-based design as a sideline to the main business of screening for billion-dollar drugs. It would command and take a larger share of whatever Vertex made and possibly treat the company as a biological service in return, perhaps even discarding it after Vertex came up with the information it required. Glaxo would overshadow Vertex in any collaboration. It might end up using Vertex’s molecular design work in its annual reports but not in its drugs.

Alternatively, Chugai saw structure-based design as key to the future of drug discovery and wanted to buy into Vertex while it was still affordable. It would want not only drugs but technology, not only data, but proof of concept. As a classically Japanese company, it intended to swallow Vertex’s example whole. Thus it would want Boger to do everything he set out to do, including ascend to the leading place in science he seemed destined for. It was the best way to ensure Chugai’s own success.

Boger’s ego wasn’t immune to those who would aggrandize him, but he realized, in considering which company’s money he’d rather take, that there was another reason he might need Glaxo’s strength: as a counterweight to Merck. It had been slightly more than a year since he had left Rahway, but Boger was not easily let go. Merck, both as a competitor and as the parent of his ambitions, still dominated his thinking. It defined him in more ways than he liked. If he was to succeed, it was going have to be in spite of and measured against Merck’s own singular success. He sometimes jokingly called the company “Mother Merck,” aware of the implications. But in the manner that he viewed it and the broad streak of history that now funneled narrowly into his thrust for independence and keen ambivalence toward the company, the firm was much more like a father to him. In that context, Glaxo looked like an optimal partner, even if its surrogacy threatened to undercut Boger’s control over his science as much as Merck had, even if, as he understood, it dismissed everything he stood for.

•  •  •

Manuel Navia was also a renegade son of Merck and, having solved three important protein structures by age forty-three, unique for Vertex: He was already famous when he arrived. His hiring was a coup for Boger, important enough for Aldrich to write a press release that was printed in the New York Times. And yet his presence as an unofficial first-among-equals rankled those who believed Boger’s social experiment barred such favoritism. Navia, an outgoing, fastidious, first-generation Cuban-American and the only Vertex scientist—including Boger—to wear a tie to work each day, overcame the discrepancy initially by downplaying it. He made comic self-flagellating gestures with his tie, muttering “Ka-chunk . . . ka-chunk” as he alternately flailed his shoulders, suggesting a hairshirt as prickly as everyone else’s. But everyone else didn’t fly to New York to meet with Chugai, or have his experiments performed on the space shuttle, or serve as company spokesman while Boger was in Japan. Clearly, Navia, formidable and articulate, was to have a larger role to play, though typically Boger had been careful not to specify what it might be.

Navia was distinct from Vertex’s scientists in other ways. He seldom drank, after-hours drinking being something of a company team sport. With his early-1960s haircut and attention to grooming, he looked like a divinity student at his kid brother’s frat party, gripped by no vice save perhaps ambition. He was also worldly and likable and had a keen sense of the absurd, as when describing his stint in the army: He had spent a year during the Vietnam War firing laser beams at pigs, helping the service defend itself from wrongful injury suits by tank drivers who’d made the mistake of staring into their nightscopes.

An only child, Navia arrived with his parents in New York City in 1953 after an overnight flight from Havana and immediately became his parents’ interpreter and the sole, dutiful object of their rising expectations. In the 1960s, he attended Xavier High School, an all-boy Jesuit military school in Manhattan, and then New York University, because they were close to home. He later received his Ph.D. at the University of Chicago and won a fellowship at NIH, where he solved the first structure of a human antibody. As courteous and well spoken in public as he is sardonic informally, Navia professes to be motivated only by a desire to design drugs. Despite his fame, he claims not to care about scientific competition and its reward system, insisting, “I’m not an academic.” This is partly true, but no research career as visible and respected as Navia’s occurs haphazardly. Navia knew an opportunity when he saw one and attacked it with a ferocious temperament. Surprised by a small dog that began barking at him once from behind a chain-link fence on Allston Street, he exploded at the hapless animal. It was not the only time that others at Vertex would see him snap.

With Boger and Aldrich in Japan and without enough protein from either Thomson or Schreiber to begin major work on FKBP, Navia used the time after the Glaxo visit to plan out experiments and catch up on some scientific literature. Along with Boger and Livingston, he tended to read more outside of his field than others at Vertex and had begun thinking broadly about other projects besides immunophilins. One article in particular quickly absorbed his attention. It was in Nature and was about a possible new drug for AIDS. A well-known Belgian chemist named Paul Janssen had synthesized a new group of inhibitors that in the test tube were more active against the AIDS human immunodeficiency virus (HIV) than any other known group of compounds, including azidothymidine (AZT), at the time the only drug approved for treating the disease.

Janssen’s molecules were benzodiazepines, derived from the same basic chemical structure as the best-selling tranquilizers Valium and Librium. He speculated that they worked by blocking reverse transcriptase, an enzyme that enables HIV to hijack healthy human cells and turn them into factories for making new virus. The ability to convert the body’s sentinels into assassins is what makes HIV such an insidious killer, but reverse transcriptase is highly mutable and thus a poor target for drugs. Because of the extreme toxicity of AZT, Janssen’s more specific molecules conceivably represented a major breakthrough.

Years earlier, Navia had studied benzodiazepines closely and had committed their wedgelike triple-ring structure to memory. At Merck, he had solved the structure of HIV protease, another enzyme crucial to the proliferation of new virus. Now, reading Janssen’s paper, he was thunderstruck. “The two compartments of my brain immediately folded together,” he recalled, snapping his hands shut like an alligator’s jaw. “I took the structure of HIV protease, and I built a model of one of Janssen’s compounds, minimized it, and matched them up. And there it was. The overlays were just incredible.”

Believing that Janssen’s compounds inhibited not reverse transcriptase but HIV protease thrust Navia into a searing personal and professional dilemma, one that had vivid implications not only for him and Vertex but, Navia thought, the world. Navia saw AIDS as a disease of “planetary significance.” He thought that unless the virus was checked it could eventually kill off the human race. These were not hysterical claims, and yet they were so remote from the terms in which AIDS was generally discussed within the drug industry as to be considered irrelevant and bizarre. There, universally, diseases are not treated first and foremost as diseases, but as markets. And the industry consensus in the winter of 1990 was that on the bottom line AIDS was going to be a loser for all but a few companies. The coldness of this calculation was reinforced by Boger, who swore practically as a matter of principle that Vertex would never work in AIDS. He was not unsympathetic, but given the odds—the large number of other companies already in the field, Vertex’s embryonic lab effort and onerous finances, what was known scientifically about the virus—Vertex appeared to have little to offer and everything to lose by taking on HIV.

Navia had agreed with Boger, but now he was transformed. He couldn’t prove his hypothesis—not without supplies of protease and Janssen’s compounds and not without crystallizing them together to show that they indeed bound as he envisioned. But in a series of experiments using high-powered computers and three-dimensional imaging software, he was able to create visual models that gave his “hallucination” striking credibility and form.

HIV, like all viruses, consists solely of a snippet of genetic information inside a protein casing—so low on the order of life that it can’t reproduce outside a host and so small, as writer Fred Hapgood has observed, that compared to a human cell it’s like a basketball against the World Trade Center. HIV protease is one of a well-known class of enzymes found up and down the evolutionary scale that cleaves protein into smaller subunits. In HIV, it works during the assembly stage. Slowly, indolently, the enzyme draws fragments of protein into it maw, where two scissor-like groups of atoms cut them to the specific length and chemical composition needed for the construction of new virus particles. Theoretically, a well-placed inhibitor would fill the tunnel-like cavity where the pincers are located and occupy them, like chucks driven into lobster claws. Without raw material, replication would slow. No new virus would be produced, and the spread of infection would be checked. No such inhibitor had yet been tested in humans, but many researchers now considered HIV protease the most promising drug target in AIDS.

Well aware of Boger’s admonitions, yet stirred by the prospect of an important breakthrough, Navia now plunged into designing HIV protease inhibitors based on Janssen’s compounds. For several days he sat in a darkened room next to the X-ray lab at a highspeed graphics workstation that had previously been used mainly by some of the chemists to simulate aerial dogfights, maneuvering benzodiazepine derivatives into the active site of HIV protease. One after another, he positioned and repositioned them inside the enzyme. And as he imagined, and with an excitement that quickly became uncontainable, he found he could fill the contours of the site, occupy the pincers, and satisfy a striking number of the known chemical requirements for protease inhibition. Boger believed that Vertex’s primitive facility for designing drugs was insufficient to fighting AIDS. But Navia’s models were so convincing that he quickly came to believe in them. This computer modeling—speculative to be sure—was the germ of structure-based drug design. Bursting, Navia felt he had to act quickly toward the next step: seeing the molecules made and tested.

“I didn’t think we were ever going to work on HIV so it was a moral issue,” he said. “My thought was, we’ve got to do one of two things; either we get somebody else to work on it or we have to publish it—put it out and see who picks it up.”

Concerned that no one would pursue his observations if he simply published them, Navia decided he would rather offer them to another drug company as a freebie—a simple matter, it would seem, in an area where the stakes were humán extinction. And yet scientific companies are filled with their own researchers brimming with their own ideas. They’re organized in such a way that at every level, from technician to senior vice president, they must compete, often brutally, for the resources to see their most promising observations acted upon. No self-respecting drug company welcomes over-the-transom proposals, especially from someone who can’t get his own company to commit the time, money, effort, and faith to pursue them. Navia needed a company that had both the understanding and ability to make use of his concept and regarded him highly enough to take his suggestions seriously. Only one came to mind.

“The dilemma was,” he would later say, “could I trust Merck to run with the ball? I still had a certain credibility there, and I felt as if at Merck I at least knew where all the handles were. I knew where the buttons were. I thought I could engineer it in such a way that the thing would get a hearing.”

During the first week in March, with Boger still in Japan, Navia wrote a four-page letter to Merck’s director of research, Edward Scolnick, detailing his work on HIV. As the company’s number two scientist, Scolnick was an obvious choice: He had the authority to act unilaterally, yet if he didn’t, Navia still had room to appeal to CEO Roy Vagelos. Both men had personally beseeched Navia to stay at Merck after he’d decided to leave.

The choice of Scolnick had an unintended irony. Navia had been hired at Merck in 1980 to help the company design drugs, but his assistance often was not welcome by those who were skeptical of his efforts. His departure was precipitated by one such incident. Navia had assumed that his structure of HIV protease, which had been greeted in the world at large with front-page coverage in the Wall Street Journal and by senior company officials with lavish praise—would become a vital part of Merck’s drug discovery. But when he asked for routine information about the company’s compounds so that he could do what he was now doing at Vertex—trying to predetermine their biological activity—the project’s head chemist refused. Incensed, Scolnick overruled the decision, but Navia, appalled that it had been made in the first place, withdrew from the project in angry protest. Now, as a competitor, he was asking Scolnick to have Merck do what it had failed to do when he worked there.

With his letter to Scolnick in hand, Navia had only to apprise Boger before sending it off. He faxed a copy to Chugai’s headquarters in Japan, expecting quick approval. Boger received the fax just as he and Aldrich were about to meet with the company’s executive committee, which had taken the unusual step of canceling its own meeting to receive them. He and Aldrich talked most of the night about the matter, then again the next day while touring the Tokyo Stock Exchange. That night, Boger ordered Navia in a confidential fax not to send Scolnick the letter.

If there was something to Navia’s hypothesis—and Boger believed there could be—he wasn’t about to give it away. Contradicting his own sworn judgment and with barely enough science and scientists to warrant what he was spending on immunophilins, Boger authorized Navia to begin a second “protoproject”: an exploratory program in AIDS.

Scolnick never learned of the letter or its contents. But two days later, on March 5, he unintentionally answered Navia’s question during a speech at Harvard Medical School. Asked by Schreiber, who was now concerned about such matters, whether having the crystal structure of HIV protease had helped Merck design new inhibitors, Scolnick answered: “It gives you some ideas you might not otherwise have.” Pressed, he added, “It’s given us some help, not dramatic help.” But by then Vertex had set out to prove him wrong.

•  •  •

All Boger’s ghosts came ablaze at the decision. Ever since high school he had been on a seesaw course between doing better than anyone else and waging guerrilla attacks against those whose authority he believed kept him from doing still better. In tenth grade, he harshly criticized his chemistry teacher for making mistakes, then taught the class himself from October to June. At Wesleyan University, in the early 1970s, he founded CRAW (Committee for Reform of Academics at Wesleyan), which fought to keep the school from compromising its mission of “learning for learning’s sake.” Largely successful, its call for what Boger called “savage rage” was tolerated, in part, because of the intellectual rigor and studious presentation of his arguments and because he was the top student in his class. The situation repeated itself at Merck, where his relentless attacks on screening and on the company’s bureaucracy were balanced by brilliant feats of science and a scrupulous ability to get things done. “You almost can’t do better than Josh did at Merck,” says former vice president of basic research Ralph Hirschmann, who first hired Boger out of graduate school. If Boger was insurgent, he was always cautious to lead his rebellion from the front row after doing all his homework and scoring perfectly on all his exams.

But this was another order of challenge. Even more than the decision to leave Merck and go it alone, the attempt to design drugs against AIDS based on the same information that Scolnick said Merck had found only marginally useful set Boger irrevocably on a more oppositional path. He was rejecting the judgments of the world’s best drug company and of his own mentors. As with immunophilins, he was trying to reach beyond screening and medicinal chemistry to a more rational way of finding drugs, yet he was doing it in an area he swore he would never enter, against one of the most complex and insidious diseases in history and against every leading drug company in the world. He was outside the classroom now, outside the school, setting up alternative classes on the playground.

He was in repudiation mode, as he might say. “Joshua’s found a new religion,” observed Hirschmann. “I just hope he doesn’t forget that the old religion, in its way, was incredibly successful.”