Scientifically and financially, Vertex has three modes: project, protoproject, and what Boger calls “preprotoproject.” Immunophilin research had been by necessity in full project mode from the minute Boger conceived it because it was all that he had. Its larger shortcomings—notably the question of whether Vertex had the right target—were blindly forgiven as Boger needed to identify quickly a field in which he could claim to have a lead. Businesswise, the strategy worked. Boger had been able to sell the project at a premium. Now, as he weighed the imperative of a second project, he knew it wouldn’t be nearly so easy.

“This is the most important decision we’ll make in the next twelve months,” he told the scientists. “If we blow the second program, it could be fatal.”

Prone to hyperbole, this time Boger wasn’t exaggerating. Industrywide, fewer than one in ten programs yields a drug: No company, no matter how smart, can survive on a single project. Worse for start-ups, they are often marginalized to the most speculative, high-risk areas of business and research or sandwiched into niches. Small, emerging drug companies are like first-time horse players with gravid ambitions and dire means: They can’t bet enough on favorites to win big so they try to parley intriguing long shots. Boger had wagered almost everything he had on immunophilins. With the race still on, the windows were now closing again. He needed to place a bet.

Vertex’s one protoproject—AIDS—had failed to move him. Navia’s “hallucination,” which started it, ended dismally, a bad trip. Two chemists, Roger Tung and Dave Deininger, spent four months making compounds based on his computer models only to find that they were dead—inactive—against HIV protease, the enzyme Navia predicted they would block. Meanwhile, a compound first made by the chemists to inhibit FKPB showed some inadvertent promise but not enough. Unimpressed, Boger had begun to think more about what HIV protease could do for Vertex than what Vertex could do to the enzyme. “We’re not going to do a smoke and mirrors project again; it won’t fly,” he told the scientists at a meeting in September. “But there’s the possibility that we could use HIV protease to field a larger aspartyl proteinase program. That makes it interesting.”

That Vertex might use AIDS as a stalking horse to raise money to bootstrap other projects struck several of the scientists as cynical and disingenuous. Boger had no such qualms. Science is rarely, if ever, linear, and information isn’t necessarily most useful in answering those questions that elicit it in the first place. Indeed, the brightest facet of the drug industry’s on-again, off-again optimism about making AIDS drugs was its recent successes with related systems. Renin, the protein-cleaving enzyme on which Boger had done his career-making work at Merck, was a perfect example. No renin blocker had ever become a drug despite a decade of intensive research efforts at Merck and elsewhere. But trying to inhibit renin had prepared the industry for HIV protease, which is strikingly similar.

Boger: “One of the things Merck did by publishing and patenting early [with renin] was to greatly stimulate the search for variations of this model if only from a patent-busting point of view. Consequently, there’s an enormous literature on renin. Yet the ultimate payoff for all of this might turn out to be HIV protease. There are groups all over the world, in all kinds of laboratories, all set up with a huge amount of methodology and experience, ready for HIV protease. If renin inhibitors never get developed, it’ll be a perfect argument for how applied science is really basic science and how there’s no distinction between them.”

Boger’s interest in developing HIV into a project had another impetus: Nissin. The giant Japanese noodlemaker with the tomblike labs that Aldrich had likened to Dr. No’s fortress was keen to “make a play” in AIDS. The company had set up a lab to screen compounds across the river in the biomedical Manhattan surrounding Harvard Medical School and had invited Vertex to submit promising molecules. In the conventional wisdom, it was extremely late in the business development cycle of the disease. AIDS was already crowded with players. Those looking for strategic alliances had long ago found them, and Vertex was perhaps a year from having done enough science to approach a bona fide drugmaker. But if Nissin, which sold $1 billion a year in ramen noodles, a company whose TV pitchman in Japan is Arnold Schwarzenegger, was interested in HIV protease, Boger was more than willing to craft a compelling story.

“What would nail it up and make that a chipshot,” he told the scientists, was if Vertex could stop the proliferation of the virus in cells. “Of course, if we can do that, internally we’d all believe we’d have a renin inhibitor, too.”

Around this time, Boger flew to San Francisco for a week-long conference on aspartyl proteases, the class of protein-splitting molecular scissors that includes renin and HIV protease and which he calls by the more formal proteinase. He seldom went to scientific meetings anymore unless he was asked to speak, but Vertex’s interest in HIV promised to make the trip fruitful. “It’s everybody,” he told Aldrich. “I’ll get a real snapshot of where everybody is.”

“You going to do any business development schmoozing while you’re there?” Aldrich asked.

“I don’t have much to bring to the party.”

“That never stopped us before.”

“It just means I’ll have to be more mysterious. I’ll bring my veil.”

In fact, the meeting was surprisingly productive. The overwhelming obstacle to making drugs that were protease blockers remained the orthodoxy that the best inhibitors were, far and away, all peptides, those bulky chains of amino acids easily dismembered by the gut. However, two large drug companies, Roche and Abbott, now challenged that dogma. They had made compounds that blocked HIV protease in the lab but that were less than strictly peptidal—“funky,” Tung called them. Boger was intrigued. Calling Tung from the conference, he authorized him to begin making the molecules, to explore them as possible leads.

Even more tantalizing were reports of another protease that Boger saw as tailor-made for Vertex, cathepsin E. Researchers had just identified the enzyme as cleaving a key protein in the chemical pathway that causes high blood pressure. As Boger knew from long experience, hypertension was the world’s largest, richest, and most competitive market for drugs. That market included some of the best drugs ever made, and yet because of its incomparable demographics (older, insured, mostly male, chronic, at risk, and concentrated in industrialized countries) had a seemingly insatiable appetite for new ones.

Here, from a business standpoint, if not scientifically, was the precise FK-506-like scenario that Boger had told the scientists not to expect again. If Vertex could prove, with a small, self-sustained effort, that cathepsin E was a relevant target for drugs and if it could make nonpeptidal protease inhibitors based on Abbott’s and Roche’s compounds, it might suddenly have a lead in an important new area of research that matched its position in immunophilins a year earlier.

Upon such leads, real and imputed, Boger knew big money was to be raised. He began thinking that with the right sort of deal, he not only could support cathepsin E, but bootstrap a protoproject in renin as well. He might even be able to float HIV, which was beginning to cost $3,000 per day, until that, too, got on its feet. Boger had vowed, as he had with AIDS, to cede aspartyl protease research to the big drug companies, but now the logic of bundling a major effort in the area became all but irresistible. In his mind, as he flew back to Cambridge, he quietly upgraded cathepsin E to a preprotoproject. Quickly, he began formulating how to sell it to the scientists.

•  •  •

“That’s a religious question,” Boger said, hoisting his size 13 brown oxfords onto a chair and tugging at his beard in a pose of mock introspection. “Is it possible to go faster than the speed of light? Yes. Are we going to do it in the next few years with the money we have? No. It’s a question of resources. If you’re asking me whether I think we can design a pill that will inhibit an aspartyl proteinase in a limited amount of time with a limited amount of money, the answer is, yes, I think so.”

Murcko, the questioner, shrugged. He was exasperated. For months he had suspected Boger of dragging his feet on HIV. If Vertex was going to be working on it—and he was sure it would be—why wasn’t the company attacking, now, before it fell even further behind? From his one-hundred-hour weeks of modeling inhibitors at West Point, Murcko knew the kind of resources Merck and the other big companies were bringing to the problem. It was obvious to him what Vertex needed to do. But Boger had refused to commit more than five people full time to the project. Murcko thought Boger’s caution went deeper than mere prudence. He was trying to pin Boger down, excise from his calm demeanor some nodule of doubt that would explain his reluctance, get him to say that he thought the goal of the program was impossible and that he was looking for a nonautocratic way to end it. Boger, ever indeflectable, wasn’t giving.

Boger came to this meeting, several weeks after his trip to California, with an agenda that, if not hidden, was less transparent than Murcko and the rest of the scientists in the room would have liked. A regular Wednesday morning meeting of the Aspartyl Protease Project Council, Vertex’s working group on HIV, the session was intended to review the group’s work and discuss its now three potential targets. Like all Vertex project council meetings, it was closed-door, with the casually dressed scientists sprawled around a table in the conference room. The room, barely larger than the table and a dozen or so chairs, is decorated functionally with two blond credenzas, a small whiteboard, and Chugai’s Mount Fuji pictures, its telling gift. Boger, preferring “flat” management, created the councils to speed the flow of information while eliminating the need for middle managers, although Murcko, especially, suspected him of using them to handle the scientists.

Murcko thought Boger was trying to engineer the overthrow of HIV protease by cathepsin E. Boger would never admit it, he thought, but the AIDS project threatened him. Unlike FK-506, which Boger had initiated himself, HIV had been championed by the scientists over and above his resistance. Despite Boger’s constant invitations for dissent, Murcko thought he was incapable of ceding control on such a crucial matter as choosing a project.

Boger’s concerns, in fact, were less psychological. If he favored cathepsin E, it was only because Vertex’s progress in AIDS had been too inconclusive for him to try to sell to a prospective partner. Once Vertex announced a project, as opposed to a protoproject, the company was committed long term: perhaps $5 to $10 million a year for five years; more, possibly much more, after that if the effort generated a drug candidate. Boger desperately needed to launch a second project soon but had yet to see anything in HIV protease to justify that kind of investment. As always, the icy short-term pressure of raising money subsumed the vagaries of morality and social good. Vertex wouldn’t pick the project that would benefit the most people or target the most pressing need; no company would. It would pick the project with the best overall chance of success, which first meant having a good story to tell investors. In that arena, by late 1990, AIDS was a tough sell. Perhaps the toughest.

Murcko was not alone in trying to dislodge Boger. Roger Tung, another former Merck chemist, also now saw in HIV protease, if not a superb drug target for Vertex, a notable opportunity for himself. Tung, thirty-one, like Armistead, is unabashedly ambitious. Like him, he had jumped from a promising career track at Merck and was trying to regain, along the alien footing of Boger’s social experiment, a similar one at Vertex. A “sort of first and a half generation, mixed background, American-Japanese and -Chinese,” Tung also was motivated by a redemptive streak to equal Yamashita’s. His mother’s father, a Japanese trade liaison before World War II, lost everything and moved with his family to the United States; they were interned in Montana, where Tung’s mother, now a writer of technical books, spent part of her childhood. His father’s father was chairman of the board of the Bank of Hong Kong; his father the lead engineer on one of the most popular mainframe computers ever designed at IBM. As a boy growing up in upstate New York, Tung had watched his father falter in his own career because he wasn’t assertive and because he didn’t have a Ph.D. “He hit the proverbial glass ceiling,” Tung says. He was determined not to be so constrained himself.

Early on at Vertex Tung had come up against Armistead’s predominance in immunophilins and it had frustrated him. “I’m a very headstrong person,” he said. “I don’t like the idea of being subordinate.” Thus HIV became an escape hatch. For months he worked slavishly to make the Janssen molecules that had inspired Navia, not because he thought they’d work—he didn’t—but as a “way of getting myself into a different area, where I wouldn’t be jostling with Dave and where Dave was predestined to come out on top.” But Tung’s independence took its toll. Demanding and intense, he alienated most of the other chemists with his high-handedness. His coal black hair became filigreed with gray seemingly overnight as he labored over the bruising synthesis of the Abbott molecules. Though he considered fighting HIV “the most important problem right now,” he had to defend a position he had deep reservations about, namely, that Vertex could actually design an AIDS drug. Tung is a skeptic—“I live to be proven wrong,” he once said. Yet he had put himself in the contradictory position of saying he thought Vertex could do structure-based design in AIDS if only to advance his own ambitions. “I want to find out whether the science we espouse so freely actually works,” he would say. “I want to know if we tell the truth.”

Now, on the council, Tung joined Murcko in appealing to Boger’s vainglorious pride. He pointed out that the Abbott compounds, which Boger favored, suggested that nonpeptidal HIV inhibitors might work as oral drugs. In the company lore, that meant that Vertex ought to be able to assemble ones that were even better. Boger—disingenuously perhaps, for he derided big companies as much as anyone in the room—responded that Abbott was a large, successful drugmaker and that having a molecule of such potency gave it a compelling lead, a lead that Vertex might be wise to respect.

“But they’re not doing anything with it,” Tung said.

“You don’t know what they’re doing with it,” Boger snapped.

“But they’ve got a history, and it’s not good,” said Navia.

Cathepsin E, Boger pointed out alternatively, was virgin territory: Vertex would be starting out equal to the rest of the industry. Boger obviously favored such a competitive position and said so. But he was swiftly rebuffed by several of the scientists, who noted correctly that the reason there was no competition in cathepsin E was that no one had proven that it was a drug target. Whatever HIV protease’s difficulties, the prospect of becoming totally, embarrassingly, ruinously wrong wasn’t one of them.

“I don’t need to know the biological use of the enzyme to be excited about it as a project,” Boger said. “I just have to know we’re not chasing the rainbow.

“Listen, if I told you I had a harmless pill that, if you took it every day and had a heart attack, would reduce your chances of permanent heart damage by 40 percent, wouldn’t you take it?”

“The FDA’s never approved anything like that,” Navia said.

“Yes, it has,” Boger said. “Aspirin.”

Of course, aspirin had been in use for more than a century and no one yet knew that a cathepsin E blocker would be effective, much less safe. But that didn’t impede Boger’s argument. He liked these exchanges, was good at them, saw them as healthy. He also liked riling the scientists, who, he thought, had become complacent about the inevitability of working in AIDS. Stirring the pot further, he now mentioned that besides being implicated in blood pressure diseases, cathepsin E was found to have a role in regulating the immune system: In other words, an inhibitor might not only help prevent heart damage, but help stop organ rejection and cure autoimmune diseases.

“An immunosuppressive hypertension agent?” Navia said, chortling absurdly.

“Why not?” Boger said. “One-stop shopping.”

At this, Navia’s incredulity shifted into annoyance. As Vertex’s senior scientist and ranking voice within the pro-HIV group, he was the one person Boger deferred to and the one Boger often seemed the least willing to confront. Dismissing Boger’s fictive cathepsin E inhibitor as “a drug looking for a disease” and questioning whether the enzyme’s immune function was crucial or inadvertent, “a feature or a bug,” he denounced the idea out of hand.

“The FDA hates multiindication drugs,” he said. “One drug, one disease, that’s how they like to do it.”

Boger cut him off: “I don’t think the FDA cares about anything unless an animal falls over and dies. The FDA doesn’t sit with a biochemical flow chart, waiting to say, ‘This could happen.’ They’re not that interested.”

Not quite a stalemate, the meeting ended tersely: Vertex would bootstrap the two programs in parallel. Tung would continue working on the Abbott compounds, which they would send along with Vertex’s own molecules to Nissin for testing, while the biologists tried to produce enough enzyme for Navia to crystallize. Meanwhile, Dave Livingston’s enzymology group would begin the series of experiments to prove cathepsin E’s legitimacy as a drug target.

Boger was pleased. He had managed to get cathepsin E onto the table without permanently alienating the HIV group. At the same time, he had sent them a signal that they had to move fast if they wanted the project to go forward. Having set in motion a nonauthoritarian approach to decision making, he now found himself more and more having to channel the insurrectionary impulses of the scientists. It was a difficult act to balance. He still thought AIDS could be disastrous for the company. On the other hand, he had allowed some of his best scientists to imagine themselves in charge of a major structure-based project in HIV, a project that Merck had launched amid exorbitant hopes and international acclaim, but on which it had been making little visible progress. As Aldrich put it, “We can tell a great story. We’ve got the Merck first team here.”

“Six months from now,” Boger announced, “we have to know which horse we’re going to be riding.”

“What’s so magical about six months?” Navia asked, still piqued.

“Nothing’s so magical. Eighteen months from now we’re in trouble; six months we’re not. But in October 1991, we run out of money unless we go back to the board. That’s where the six months comes from. If we sit down in March with a prospective partner, there isn’t a prayer that we’ll have a deal closed even by November. But at least we’ll know we have a front-burner contingency to raise some venture capital.

“June,” he said ominously, “would leave us with an almost irrevocable decision either way.”

It was mid-October, and Boger still thought it wasn’t too late for Vertex to enter either area. Two months later, with Tung still struggling with the Abbott compounds and Livingston mired in cathepsin E, he was less sure.

“I have this nightmare that I’m going to read in Nature the results of an experiment that we could have done that definitely proves cathepsin E’s biological relevance,” he said at a meeting of the project council. “At that point I’m just going to writhe on the ground and gurgle.”

•  •  •

Late fall came and went with a depressing lack of progress, not only in chemistry and enzymology, but in all the labs. Failed experiments, equipment losses, recalcitrant vendors, unverifiable results, artifacts, squabbling, jealousy, frustration, impatience, rage—all took their toll. It was as if the company had a virus, a malaise, that worsened with each pass. Six months earlier Vertex had run a T-shirt design competition, won, amid suspicion that business had again hijacked science, by Aldrich. On the back, the shirts said: “We don’t leave success to chance.” Now, though some of the scientists still wore them to work, a darker sensibility prevailed. “I’d rather be lucky than smart,” moaned Navia, only half-joking and still stuck in the search for heavy atoms. “What’s our credo?” Thomson asked Moore, leading a familiar call and response. Grunted Moore, borrowing a response from Murcko: “Grief! Pain! Angst!”

They had all experienced slumps and dry spells before, but for a year Vertex had been so productive, so surpassingly successful, that a Bogerite exemption seemed to float down and lift them in its embrace. None of them was deluded into thinking they were doing rational drug design, not yet. But now the problems were harder. The scientists felt themselves moving farther into the unknown, which, as they knew, was unknown for a reason. More than a few feared it might be unknowable.

Boger, as always, led by example. He was upbeat, dauntless, equable, tireless, and expansive. Flipping between business and science, between existing work and planning ahead, he was in Zen field general mode, calmly and gregariously attending areas away from the main action that he was certain would be crucial later on. Like Tishler, he kept “all the threads in his own hands” but seldom pulled on the ones that the scientists expected or wanted him to. For instance, although he had told Chugai that Vertex would have molecules that were immunosuppressive in cells by the end of the year, he put no pressure on the chemistry group. He stayed out of the labs completely, concentrating instead on making new slides or micromanaging the growth of the project councils. “I don’t think there’s a lot of understanding about what I’m trying to do organizationally,” he said on a day when anxieties were racing about the general sluggishness in the labs. “I think everybody wants desperately for Mussolini to come in and run things. But I’m not going to do that. The big issue here is how I can evolve a radically different organizational structure into something that works. Believe me, if I turn out to be wrong, it’ll take forty-five minutes to fix.”

Part of this elusiveness was Boger’s determination not to interfere with researchers he believed to be at the top of their fields. Part, too, was an appreciation for the rhythms of science, rhythms that in almost any other endeavor would be maddening, months and years of unrelieved failure punctuated, if one was fortunate, by sudden leaps ahead, and if one wasn’t, by prospects of everlasting darkness. Patience and detachment were necessities. Boger had predicted clearly in September that Armistead’s group would have cellular activity. He had even called the type of breakthrough, saying, “It’ll go in a jump. It won’t creep down 5 percent at a time.” Having done so, he then stayed out of the way, keeping things, as he said, “indeterminate.” The emotional rigors of maintaining such a calm distance amid soaring stakes and widening failures unnerved some of the scientists, but Boger loved them, especially in periods like this one when nothing seemed to go right. “It’s got to be hard,” he said in December as the inertia in the labs seemed to thicken. “It’s got to be a little scary or I lose interest.”

Individually, the scientists would have been happier with a few breaks and some help. If Boger thought that by setting them wheeling together in the right motion Vertex would accomplish great things, most of them were up against deadlines and competitive pressures or driven by company and career goals that made them think only of themselves.

They were strapped, oversubscribed. Harding, for instance, was now mainly responsible for determining FKBP’s role in immunosuppression and whether or not it was the right target. It had long been assumed that FK-506, like cyclosporine and many other drugs, bound tightly to more than one protein within cells, that there were other FKBPs of varying sizes and shapes and that they, perhaps, not Harding’s original protein, triggered immunosuppression. The search for new immunophilins had attracted a formidable field, and Harding, as codiscoverer of cyclophilin and FKBP, had a major stake in maintaining his position within it. Not only would discovering another FKBP, or, better yet, a family, further his priority, it would establish his independence from Schreiber, for whom the stakes were just as high or higher. There was also the patent income. Harding earned thousands of dollars in royalties every time Yale and Harvard licensed FKBP to a drug company. The money had become an important stopgap during the past two years when he and his wife, Robin, had moved from New Haven to Boston, buying a house in the suburbs at the height of the New England real estate boom. Robin had given up her job as a financial officer in a hospital to make the move, and the couple had a new baby.

Harding devised a series of experiments to try to pull related proteins out of Thomson’s thymus extract with synthetic antibodies, but he found it nearly impossible to do them. He was busy screening compounds for cellular activity and helping Debra Peattie’s group search for the gene that encodes the enzyme—everyone’s work but his own. Typically, he was frustrated and angry with himself. Self-abrogating to a fault, he began behaving like a martyr, complaining, dithering at the bench. He was especially jealous of Schreiber, his erstwhile collaborator, who as an academic had no responsibilities other than to pursue his own interests, pander to his own goals.

“In biology, even in the best circumstances, if one day out of five is productive, you’re doing well,” he muttered. “I’m just not doing any science.”

•  •  •

Structure.

Jon Moore glimpsed it initially in snatches, like the first faint shadings of a photograph plunged into a developing bath. A crook here. A hairpin there. Suddenly a straight run of three- or four-billionths of a meter. After weeks of sitting morosely in a darkened office next to Boger’s or in an open area outside the modeling lab, wrestling with the positions of hydrogen nuclei, he began to develop a reasonable certainty about a prize sliver of spiral. Embedded in the blackness of his computer screen, it was, Moore thought, a stretch of helix near the beginning of FKBP’s amino acid chain.

It was a few days before Christmas. Moore had the wan, subhuman cast of a cave dweller. He also was wildly encouraged and straining to check his optimism. The architecture of FKBP was beginning to tumble out before his eyes, and no one else had published it yet. Schreiber’s minions still hadn’t beaten him.

“Fantastic!” Navia said, entering from the lunchroom. Mired in the search for heavy atoms, he had found it irresistible to visit on occasion: Macy keeping an eye on Gimbel.

“Three strands and a ‘bit o’ helix,’ ” Moore reported.

“You may pull the fat out of the fire for our end,” Navia said. “A low-resolution structure for us would be tremendously helpful. If you have a ribbon diagram [an overall map of a protein’s fold], we can plug it in and short-circuit this heavy-atom crap.”

Navia had begun to face the grim possibility that it might be years before he and Yamashita could solve the structure with the approach they were using. The prospect had made him consider other options. Recently, crystallographers at NIH and elsewhere had pioneered a new approach for orienting crystal data. Called molecular replacement, it involved using a piece of NMR structure of the same protein as a map, a guide. Navia, who would take any shortcut he could get at this point, saw in Moore’s emerging images his best and perhaps last hope not to be shut out with FKBP. His self-imposed deadline of Thanksgiving had come and gone. He, like Thomson, had stuck his neck out, and heads had begun to wag.

“If we can pull this off, it’ll be unique,” he said, “It’s never been done for real. It’ll put an end to this X-ray versus NMR crap.” Navia is as violently competitive as anyone at Vertex, though now—Moore wondered if it was for his benefit—he denounced the absurdity of scientific rivalry. “Two fucking methodologies that are absolutely complimentary, yet you hear people at meetings . . . they’re infants.”

Moore was noncommittal. He didn’t have enough structure yet to be of use to the crystallographers, and even if he had, he wasn’t sure whether what he was seeing on his screen was correct. He’d still located only about half the protons he needed for a complete diagram. He said nothing, and Navia quickly changed the subject and the tone.

He told Moore a story about being at a costume party and not being recognized.

“Gee,” he said in character, exaggeratedly scratching his head. “Too bad Manuel missed the party.”

Then, in another voice, he said, “Gee, Clark, Superman was here, but where were you? You missed him. How come whenever Superman’s here you’re not around?”

Navia loves to mock dumb wonderment. He was now hopping from foot to foot, flipping his tie and smiling widely. “Even as a kid, I thought, What’s the matter with these people? They’re so thick? Can’t they see everytime Superman appears, Clark vanishes?”

Moore laughed, enjoying the diversion. Working so closely with crystallographers who were racing to solve the same structure as he was a new situation, one he was unlikely to have encountered at a university. It was awkward. He was pushing ahead with work that could make his career. Yet if he could help Navia and Yamashita, Vertex would benefit, even if he didn’t. He decided to concentrate on his work and play it by ear.

Two days later, piecing together scraps of structure, Moore glimpsed for the first time a major piece of the molecule’s backbone. It looked like the back of a baseball glove—a web of five parallel strands—and comprised about 40 percent of the enzyme. It was enough perhaps to try a molecular replacement experiment, but he nonetheless withheld it from Navia, and Navia didn’t ask for it outright. Boger, assuming as always that the scientists would seek collaborations that were explicitly in their own interest, said nothing.

•  •  •

In her lab coat, worn most days over a dress with a Mickey Mouse pin on the lapel, Patsi Nelson looked like a mildly eccentric smalltown pediatrician. At thirty-nine and a mother of three, she brought an effusive maternal selflessness to Vertex’s craggy, egotistical emotional climate. She was warm in a place that’s usually raging hot or ice cold.

Nelson is an immunologist and, by culture and training, a Californian. She came to Vertex after a postdoc at Stanford, a stint at the Scripps Institute in La Jolla, and several years at Gene Labs, a West Coast biotech company, where she worked in AIDS research. The job brought her in contact with infected blood and instilled in her a keen respect for the dangers of her work. “I always treat everything as if it’s contaminated,” she says.

Nelson’s compassion was pronounced, and unlike Navia and the other men in the company, she didn’t try to exorcise her feelings about her work. Many women, competing in a male-dominated world like science, adapt with an underground humor of their own. “He had two arms of research and I was one of them,” Nobel laureate Gertrude Elion once said, speaking about co-prizewinner George Hitchens. Nelson took her science personally, felt its thrills and torments deeply, and assumed others did, too. “Oooh, I feel so bad for the chemists,” she would say when their compounds turned up dead in her cell assays. Watching Harding struggle, she said sympathetically, “Poor Matt. I wish I could help him out.”

Throughout the fall, the chemists delivered Nelson several new compounds a week, and, as Boger predicted, none was much better than the others. His promise to Chugai loomed ever larger, and Nelson worried for the chemists. She wanted desperately for their molecules to succeed and was clearly pained when they didn’t. One group of molecules, submitted in late November, particularly distressed her: They inhibited FKBP as well as FK-506, but when she and Harding put them in with human T cells, they clumped and fell out of solution. It reminded Harding of one of those watery Christmas scenes that you shook up and then watched as the snow swirled gently to the bottom. Nelson was dismayed: The molecules weren’t even getting inside the cells. Even if they were active, there was no way of knowing. Either way, they were useless.

In mid-December, Armistead submitted a compound he thought might overcome the gap between enzyme inhibition and cellular activity. By now he thought he had done all he could by making FK-506 smaller and was adding chains of atoms to those redesigned pieces of the core that were the company’s best blockers. He was trying to occupy the same space as FK-506 by attaching different atomic subgroups, an effort akin to a sculptor’s experimenting with different ligatures for the arms of a heroic statue. On one of the molecules—the 367th compound produced at Vertex and submitted for evaluation—he replaced the original crumpled horn of FK-506 with a symmetrical V shape. It was three carbon atoms long in each direction and had a ring of atoms at each tip. On paper, it looked like a crooked rabbit ears-type TV antenna or, as Boger elegantly suggested, a semaphore, one of those signals made, usually on the decks of ships, with outstretched flags. Armistead, eyeballing the distances, had gotten it out of the Aldrich catalog.

When Nelson got back the data, she was ecstatic. The molecule was one hundred times more potent than Vertex’s previous best inhibitor, exactly the gain that Boger had predicted to Chugai and an extraordinary improvement. Tested several ways, it was just as active as cyclosporine. Yet such a huge leap in potency innately worried her. It could be an artifact, an error in her cell-counting equipment, anything. Normally she would wait and retest the compound the following week, but the magnitude of the improvement demanded immediate confirmation.

Each week Nelson got a delivery of T cells from a blood bank at a hospital that culled them from whole marrow, but it was closed until after the holidays. She couldn’t wait. Showing Harding the data, he agreed that it was too important to sit on, and so he volunteered his own blood. “Poor Matt,” she recalled. “I bled him. The first time I went through his vein, but he wouldn’t let on how much it hurt. Finally, I got enough cells to test the compound again.”

The results were identical. Nelson immediately went to her computer and composed a Christmas card for Boger. On the overleaf were two graphs showing V-367 and cyclosporine spiking, then declining in similar fashion.

“It’s beautiful data,” Boger exulted, “and the best news is that there are more compounds behind these that’ll make 367 more active just by improving solubility. We’ve started a series that should enable the chemists to quickly bring up the potency by another factor of ten.”

Boger was soaring. Everything was as he’d foreseen. In a little more than a year Vertex had generated novel druglike compounds. In three months, it had gone from having poor cellular inhibitors to having a molecule ready to test for immunosuppression in animals. Another jump like the last one and it would have molecules as potent as FK-506. And that was before having structural information about FKBP, which, with Moore’s apparent success, would also now soon be available. It was all just as he had said it would be, all as he had sold it. Brandishing Nelson’s Christmas card, with its simple line drawings of Vertex’s compound and provocative data points, Boger said: “If this piece of paper fell into the wrong hands, I’d be worried for the next six months about somebody derivatizing our molecules and beating us. If we had structure data, I’d say, ‘Go ahead, try.’ We’d be so far ahead we’d be unstoppable.”

The year thus ended as it had begun, as Boger had said it would, triumphantly, though now the triumph was less a matter of Boger’s speculations and more of hard-won science and its attendant good luck. It had taken a “slugfest,” as Boger called it, in chemistry to get the company to where it was. Yet now that it was there, the aura of invincibility that had seemed to disappear with the Chugai party suddenly returned. There were those who were still wounded: Navia and Yamashita, for instance, who smiled tautly when Armistead and Saunders began calling the crystallography lab “the chemist’s smoking lounge.” But even Thomson was resurgent. Recovering from his siege with FKBP and his pique at the depredations of the Chugai episode, he arrived the following night at Vertex’s Christmas party in an outlandish double-breasted tuxedo and bearing a tray of elegantly prepared thymus fillets.

“Fuckin’ A!” announced Armistead.

“Cyclosporineland!” Saunders said, brushing jealousy aside.

“We’re cruising,” said Aldrich, who knew the reward for scientific success was more, and more expensive, science, and thus the need for more money, more business development, more and bolder storytelling.