He made his application to the chief medical officer at Dana-Farber, a man named David Livingston, who like Druker was both a doctor and scientist. He wasn’t the top leader at Dana-Farber, but he was the gateway to all faculty appointments. Livingston declined Druker’s request for his own lab. “He didn’t believe in either me or my work,” said Druker. “He didn’t think I had what it took to run my own lab.”
Livingston presented another option to Druker. Dana-Farber was opening a new molecular diagnostics lab, and he offered Druker the option of running the lab. As scientists and doctors learned of the various mutations cropping up in tumor sequencing, the idea of knowing a patient’s mutation status assumed importance. No one knew which abnormalities might be relevant to the progress of a cancer, but including their presence in a cancer patient’s medical records seemed a logical next step. Plus, inherited traits associated with cancer were increasingly coming to light (as in the case of Li-Fraumeni syndrome, a hereditary condition that predisposes people to cancer), so the notion of testing families for these genetic conditions was gaining ground. This lab was one of Dana-Farber’s first forays in that direction. Livingston thought that running the new operation would take up about half of Druker’s time. The other half could be spent pursuing his chosen research.
Druker was intrigued. If half his time was still his own, he reasoned, that could work. He could get somewhere with his kinase research. So he made a list of what he thought the diagnostics lab would need—the technology and other resources that were making this new discipline possible. “I never heard back,” said Druker. Apparently, Livingston was offended because Druker had asked for more than the company was willing to give. Poking that hornet’s nest made Druker realize that it would have been a fruitless pursuit. “I wasn’t going to do something that they’re not going to put any money into,” he says. “Why would I spend my time and energy on something that was set up to fail?” But, having offended a man so near the top, Druker needed to face the fact that he was not going to be offered a job at Dana-Farber. He had to make a choice: stay in the lab with Tom Roberts or leave one of the world’s most respected cancer research institutions and strike out on his own. He knew it had to be the latter.
While Druker was negotiating with Dana-Farber about the diagnostics lab, he’d also applied for a job at Beth Israel Hospital in Boston. They were setting up a special unit for studying signal transduction, the sequential cascades of signals that flow from one protein to the next within the cell. It was the exact work to which Druker had dedicated the last six years of his life. Druker was offered a position, just as he was waiting to hear back from Dana-Farber about the diagnostics lab. When Druker pressed to get something in writing, the chair of medicine who’d made the offer insisted it wasn’t necessary, that the position was his if he wanted it. Druker waited to hear back from Dana-Farber, still hoping he could stay there. When he finally called to accept the offer at Beth Israel, he was told the position was gone. The hospital had already hired someone else.
“So at that point, I’ve been kicked in the stomach twice,” says Druker. “Do I just disappear, or do I get up and decide what I really want to do and make something of this?” Quickly, 1993 was turning into a low point for Druker. His marriage had failed, and the city he’d made his home seemed to have no space for him as a scientist. As far as he could see, there was only one possibility. “I need to find a position,” he decided. “I need to get out of Boston.”
Opening a private oncology practice was an option, but it was never something Druker wanted. He didn’t want to give people chemotherapy. His dream was to fulfill his vision of bringing better treatments to patients. He wanted to make good on the promise he’d written in letters to the families of his patients who’d died. “I was going to go into the lab, and I wasn’t going to come out until I had something that was better than what we had to offer,” Druker had told them.
Now, that vision had taken on a very specific form: creating a drug for CML that targeted Bcr/Abl. “That was my goal. I knew that was what I wanted to do. If I were going to make a difference, if I were going to live up to the promise I had made my patients, that’s what I needed to do,” he said. Suddenly, it didn’t matter how illustrious the institution was; all that mattered was that he work toward this singular focus. “And it all became really clear.”
Colleagues asked Druker why he was taking such a hard road. Surely, they asserted, he would burn out from all the effort, the constant fight for funding, the endless struggle to create a drug that still seemed impossible. But Druker knew that it was the seemingly easy road that would burn him out the fastest. He knew the endless conversations informing patients that there was nothing more he could do would leave him spent. During his years at Nashoba, he was having that talk once every week or so, telling hopeful sufferers that the current drugs were no longer working and that it was time for him to focus on making them as comfortable as possible during their final days. “If I were in practice, that would be [happening] once a day,” he says. “There was no bike ride I could take that could get me through that.”
So he made a list. He wrote down all the academic institutions with small but growing oncology programs. He figured that a nascent program, rather than one that was fully established, would allow him to have a hand in its development and would give him the freedom to pursue his rogue dream of kinase inhibition. He also made a list of places where he’d like to live. Now that he was leaving, Druker had to admit he’d never felt at home in Boston. “I always felt like an outsider there,” he said. Letting go of the fixation on prestige so ingrained within academia made space for other considerations. He began to think about the outdoors. He missed the lifestyle he’d had in San Diego during his early medical studies, and he realized how important exercise had become to handling the stress of his career. He needed to be in a place where he could run and bike year round.
He looked into hospitals in New York, at the renowned Cold Spring Harbor laboratory, at the University of Iowa. Everything was interesting, but nothing was quite right. Then he made a visit to the Oregon Health and Science University, in Portland, to meet a man named Grover Bagby. “I saw a person who was committed to growing a cancer program, who believed that targeted therapies were going to be the future,” said Druker, “[and] who I thought I could trust.” His first interview happened to fall on a sunny day in January—“I think there’s two of them,” Druker said dryly—and the mountains and profusion of green were as attractive to him as the job for which he was interviewing. “I just fell in love with this place,” he says. So wooed was he that he didn’t even think to look at how OHSU was ranked academically. “I guess if I had done that, I would have thought twice,” he says. After all, he was coming from one of the world’s top academic research institutions. Yet his experience there had clarified that such status was no guarantee of personal success. When Bagby made him an offer soon after his visit, Druker readily, and happily, accepted.
Shortly before he left Dana-Farber, once he knew he was on his way to Portland, Druker called up Nick Lydon. Knowing he was leaving, Druker had no more patience for operating within the confines of the institutional agreement with Sandoz. He wanted to talk to Lydon about kinase inhibitors. He told Lydon his plans and asked if he had any inhibitors for the Bcr/Abl kinase. “As a matter of fact, we do,” Lydon told him.
He excitedly informed Druker that the chemists at Ciba-Geigy now had several candidates for kinase inhibition, including one with strong activity against Abl. Since Bcr/Abl existed only inside CML cells, the experimental molecule had been screened only against Abl, the enzyme naturally occurring inside normal cells, and not against the mutant fusion protein that resulted from the Philadelphia chromosome. They had to wait for someone like Druker to test the drug in actual CML cells to know whether it inhibited Bcr/Abl. That work could have been done by anyone, but Lydon had been hoping it would be Druker.
“Would you like to test them?” Lydon asked.
&
nbsp; “I can’t test them until I move,” Druker replied.
“Okay,” Lydon said. “Call me when you get to Oregon.”
By July 1993, Druker had moved and was setting up his lab. He rented an apartment in Lair Hill, in the southwestern part of the city. He walked to work each morning along a winding road bordered by towering moss-covered trees, their branches a braid of neon green, a view interrupted only by the sight of bikers and runners making their way uphill. Woodpeckers tapped a beat into the cool air of the Pacific Northwest as Druker strode to the top of Marquam Hill. There, the cluster of buildings that housed OHSU boasted views of Mount Hood and Mount St. Helens, their snow-covered sides glistening in shifting shades of pink and yellow on sunnier days.
By August, just a few weeks after his arrival in Portland, Druker had received several compounds from Ciba-Geigy and had begun the experiments he’d been dreaming about for several years.
21
_______
KILLING CELLS
In 1993, Ciba-Geigy’s lead compound in the kinase program, the one considered to be the most promising to bring to market, was not the one targeted against the Abl kinase. Rather, it was another compound, similar but with key differences that made it active against the PDGFR kinase. Even though Lydon believed that the anti-Abl kinase for CML was the company’s best shot at proving the principle behind kinase inhibition, the rarity of CML failed to generate enthusiasm among the higher-ups. Matter, Lydon, and the rest of the development team had been instructed to continue creating molecules to block kinases involved in more common cancers. That put the PDGFR compound—code-named CGP-53716—out in front. CGP-57148B, the development name for the molecule that blocked Abl so potently, was second in line. Lydon sent Druker these two compounds, two other candidates, and a dummy molecule that had no anti-kinase activity. When Druker received the molecules, they were labeled with the company code names, but there was no information about which kinases the different molecules blocked. He was told nothing about CGP-53716, CGP-57148B, or the other active molecules. He also wasn’t told which one was the control, the molecule that the Ciba-Geigy team knew did nothing. Like the sugar-pill placebo often used in human drug studies, the control was there to confirm that any changes seen with the experimental molecules were not also seen with a compound known to have no effect. Because he didn’t know the identity of each compound, Druker could conduct a blind study, which legitimized the results. Knowing which compound was active against which target would have introduced a potential for bias that could undermine his lab findings.
Before Druker had unpacked in his new home and settled into his new lab, he was testing Ciba-Geigy’s compounds. For the next year and a half or so, Druker, along with one of his two postdocs, focused almost exclusively on testing the Ciba-Geigy compounds. Another postdoc studied Bcr/Abl signaling, research that was more grant-friendly than the work he was doing with Ciba-Geigy. Druker was given a couple of graduate-level classes to teach, but his primary goal was exactly what he’d come to Portland to do: help usher in a new drug for CML that could, if successful, prove revolutionary for the treatment of cancer.
He experimented with CML cells derived from people and on white blood cells taken from mice that had been forcibly engineered to express Bcr/Abl. He also had human cells derived from other types of leukemia that did not contain the mutant kinase. Finally, he had samples of bone marrow from CML patients who’d undergone bone marrow transplants and people who had undergone the transplants for reasons other than CML. He laid out a protocol that would enable him to isolate the effects each experimental compound was having, if any. Much of the work followed the approach he and Jim Griffin had presented in their grant application to Sandoz back in 1990. Druker had to test whether any of the experimental compounds killed the cells, and, if that happened, confirm that the cause of death was the compound and nothing else.
As he had learned to do in Tom Roberts’s lab, Druker set up a tray of test tubes. In four of them, he mixed a minuscule amount of CGP-57148B (which, unknown to Druker, was the compound directed against Abl) with cells derived from patients with CML. In another group of tubes, he incubated those same cells with a higher amount of the experimental compound. Yet another group held cells only, with no compound present. He arranged a parallel tray for the mouse cells. Each day, for four days, Druker pulled a sample out of each tube and counted the number of cells. If CGP-57148B was working, the number of cancer cells present should decline markedly each day. Blocking the kinase responsible for the cancer should kill the cancer cell.
Each incubation had started with anywhere from 5,000 to 20,000 cells. At the end of those four days, many of the tubes were still rife with viable cells. In the tubes with human CML cells and nothing else, the number of cells was now around 800,000. The mouse cells had similarly continued to proliferate in the test tubes, and the lower dose of the compound had not appeared to thwart the activity of either type of cell.
But in the tubes where the cells derived from human CML had been incubated with CGP-57148B, something astonishing had occurred: the malignant cells were all dead.
In another group of test tubes, Druker incubated human-derived cells containing the src gene, instead of the bcr/abl fusion gene. This change enabled Druker to test whether the cell-killing activity of the compound was connected to the presence of Bcr/Abl. In those tubes, the cells kept multiplying. CGP-57148B was selective. In the mouse cells, the results were identical.
Druker then made sure to run a blind round of experiments, one in which a second researcher tested the candidates without knowing the results of Druker’s experiments. Even though Druker was already blind to the target of each compound supplied to him by Ciba-Geigy, he wanted to make sure he wasn’t inadvertently biasing the results. He dissolved the compounds, gave a postdoc four vials labeled A, B, C, and D, and asked him to test them. He knew that if his study was published, another lab would immediately set out to replicate his research. He also knew how easily error could creep in, for example by counting a cell as dead when it might not be. The blinded approach prevented the postdoc from manipulating the data just to please the boss. When the results revealed that the vial containing CGP-57148B undoubtedly killed CML cells, Druker was confident that anyone who got the compound would reproduce their findings.
“Do you want to join us? Brian has the first results,” Lydon asked Zimmermann, back at Ciba-Geigy in Basel. It was early 1994, and Lydon was heading to a conference call with Druker to hear the findings of his initial studies, about three months after Druker had gotten the experiments under way. Some of Druker’s experiments had been done weeks earlier, but with his postdoc’s verification now complete, Druker was finally ready to report all the results.
Zimmermann joined the table, where Lydon, Matter, Buchdunger, and a handful of others waited to hear the news from Portland.
“Thank you for the shipment,” Druker began. “The samples have arrived, I have tested them in my assays, and they seem to work.”
Zimmermann still remembers the moment vividly. “That was unbelievable,” he recalled.
Elisabeth Buchdunger was also amazed by the news. “When we got back the data that clearly showed that the compound was killing only the cells that had the Bcr/Abl kinase and did not affect normal blood cells, this was a really fantastic thing,” she said. “It clearly proved the selectivity.”
There was one other piece of information Druker had to share with the group in Switzerland. The compound also inhibited another kinase that had been introduced to the screening tests: Kit. This kinase phosphorylated tyrosine, just as Abl did. Druker had added Kit to the panel of kinases against which the molecules were screened, and CGP-57148B was active against this one, too.
The collaborative team—Druker, his postdoc, and the main chemists and biologists at Ciba-Geigy—wrote a paper describing the study that they’d soon submit to a journal. A publication was essential, they all knew, for staking their claim in the kinase territory. A prominent pape
r was crucial. It would help position them as leaders in the field, it would ensure that they would not be accused of following someone else’s work, and it would create buzz about the increasingly realistic possibility of creating a selective kinase inhibitor. For Druker, the report would be the first tangible testament that his long-held hunch about kinases and the vision of improving cancer care that had propelled him forward was now inching into reality.
Considering the striking results, the team of authors felt confident about placing their work in a prominent journal. They sent the manuscript to Science, but it was rejected. Undeterred, they submitted it to Nature, another top-tier publication. Again, it was rejected.
Druker found himself simultaneously inspired and frustrated, not only from the journal rejections. The findings from this first preclinical study were strong enough to warrant moving the compound to the next phase of development. “It was exciting, and we were on a pretty linear track to clinical trials,” says Druker. But everything was taking too long. He had expected the company to respond to his data by initiating further research immediately. He knew there were patients in need of new treatment options, and he knew this compound could potentially save their lives.
He also knew that researchers around the world were getting more interested in kinase inhibitors, and he didn’t want to miss the chance to be the first. “Part of it was I didn’t want to get beat,” Druker acknowledged. He had reason to be worried. In 1995, while the journal article he and the others had prepared was under review, Alex Levitzki, who’d discovered the anti-kinase activity of staurosporine (the antifungal agent that blocked too many kinases at once), published a report about a kinase-inhibiting compound based on that naturally occurring chemical. The researchers with whom Druker was working feared being scooped. “There were patients, there was competition,” he said. “I wanted us to get moving.”
The Philadelphia Chromosome Page 15
