The Philadelphia Chromosome

by Jessica Wapner

  The marketing team had to relent. After all, they knew the patent would expire and exclusivity would end in a few years, leaving the CML market wide open. Novartis had to have another drug waiting in the wings to fill its place. Vasella was also well aware of the pressure to keep creating new drugs. In his view, that was the whole purpose of patent expiration. “Unless you innovate, you’re dead,” he says, “which is a pretty good system.” Despite the fact that the Gleevec trials had just proved the principle of kinase inhibition, murmurs of skepticism about creating an even better inhibitor filled the company’s hallways. “Most people said no, we have no chance to beat Gleevec,” Vasella recalls. “But we did.” In 2002, the next-generation tyrosine kinase inhibitor, called nilotinib, was synthesized. It was a direct descendant of imatinib but with better binding capacity.

  In separate phase I clinical trials, dasatinib and nilotinib were given to groups of patients who had stopped responding to Gleevec. Just as with the first trial in 1998, dose-escalation studies were conducted for the sole purpose of finding the most effective dose at which the drugs could be given safely. But again, the data far exceeded that goal. Of twelve patients with chronic-stage CML, eleven had a complete hematologic remission with nilotinib. In the dasatinib study, that response was seen in thirty-seven of forty chronic CML patients. The drugs also worked in the more advanced stages of the disease. Patients with large numbers of blast cells in their marrow saw their blood counts return to normal. Cells carrying the mutant Philadelphia chromosome gene decreased in number. In 2006, the FDA approved dasatinib (the brand name is Sprycel) for the treatment of CML in patients for whom imatinib didn’t work. Both of the second-generation drugs had side effects, but as with their predecessor, nothing that was intolerable. A year later, nilotinib (brand name Tasigna) was approved for the same indication.

  With first- and second-generation tyrosine kinase inhibitors available for CML, the number of patients living with the disease continued to expand. “The prevalence, the number of people living with CML, keeps doubling” said Michael Mauro, the oncologist who’d joined Druker’s team in 2000. Before Gleevec, there were 25,000 to 30,000 people living with CML in the United States. “By mid-century, it will be a quarter of a million,” Mauro said. “Ten times greater [than before Gleevec].” That number will eventually plateau because the number of new annual diagnoses remains constant. But the expense of CML care will become even more formidable in the coming decades. “It is a good problem to have, but we have to be thoughtful about . . . the best way to manage a patient indefinitely.”

  The cost of treating CML for decades combined with the disappearance of the Philadelphia chromosome in many patients has led to speculation that some patients may be able to stop taking the drug after a prolonged period of time. That question is now being investigated, but many clinicians are skeptical and hesitant to take the risk. Some patients who had achieved a complete molecular remission—no detectable sign of the cancer, even with the most powerful disease-spotting equipment—began experimenting with stopping Gleevec after several years. In some of those patients, the cancer soon returned. People who stop taking Gleevec run the risk of the drug not working as well when they resume, though the second-generation drugs would then be an option.

  The long-term management of healthy patients wasn’t the only problem remaining. One group of CML patients did not respond to any of these drugs. These patients had a mutation known as T315 that was impervious to all of the tyrosine kinase inhibitors at any dose. By the end of the first decade of the twenty-first century, patients found to have this mutation were routinely told that a bone marrow transplant was their best shot at survival. In 2009, however, a compound designed specifically for patients with this mutation, ponatinib, entered clinical trials.

  Hans Loland is a CML patient from outside Seattle. After failing to respond to any drug treatment and watching his best friend, who also had CML, die after a bone marrow transplant, Loland was offered a spot in the ponatinib trial by Mauro, who was leading the trial at OHSU. Loland knew it was his last chance. Three months after starting on the drug, he had a complete cytogenetic response. Two years later and with a five-month-old son, Loland now has to stop himself from worrying that the response will disappear. “Before, I had everything to gain,” he said. “Now, I have everything to lose.”

  In late 2012, ponatinib (brand name Iclusig) was approved by the FDA for treatment of CML. There are now approved tyrosine kinase inhibitors for all known permutations of the disease.

  38

  _______

  A GLEEVEC FOR EVERY CANCER

  By 2011, the landscape of cancer research looked vastly different from when Druker started, and it seemed an alien planet compared to Nowell and Hungerford’s early days. The advent of Gleevec had let loose an avalanche of change for cancer research and treatment.

  Finding the driver mutations behind different types of cancer became a central thrust of research. “If you know it’s broken, you fix it,” is how Druker summarizes the pursuit. As the data on tyrosine kinase inhibition continued to mature, Druker turned his attention to tumor sequencing. By identifying the cluster of mutations that matter to a particular cancer, investigators could create treatment regimens tailored specifically to individual tumors. Currently in the thick of the search, Druker sees routine tumor sequencing as the near future for personalized medicine. Genome sequencing of an individual’s entire DNA, with the aim of creating a personal profile of susceptibilities, prognoses, and prescriptions, is, he said, the fifty-year vision.

  But tumor sequencing has been a steep mountain to climb. As researchers continued to profile cancer after cancer, reports were turning up as many as 200 mutations, and at the same time finding that only as many as five could be crucial to the disease. Among those five mutations, some may be easily drugged, and others not so much. At OHSU, Druker has continued to expand screening efforts to include more than 1,000 different mutations across 200 genes. Today, when intriguing mutations are identified, the storehouse of compounds accrued from years of academic and industry efforts can be screened for activity against them.

  Alongside searching for new drugs is a quest to create a smoother road for bringing them to market. The eighteen-year odyssey of moving the first tyrosine kinase inhibitor from the labs of Ciba-Geigy to doctors’ offices in the most remote US towns did more than change scientific thinking about cancer. It also transformed views on how drug development should best be done.

  In the wake of Gleevec’s approval, Druker and many other investigators began to think up new ways of testing experimental drugs in humans, “a completely reconfigured clinical trial process and FDA approval process.” In the clinical trials for STI-571, the first fifty chronic-phase CML patients had the same response rate as the last fifty. Studying the drug on this homogenous population, he said, is akin to “running 500 identical twins through that study.” Valuable as it was to have that much data, the same results would have been had if the study had stopped at fifty or a hundred enrollees.

  Some researchers, including Druker, say that a conditional approval based on those fifty or so patients would work just as well as a conditional approval after a phase II study. Full approval could be withheld until safety data from the full-scale trial of 1,000 patients is cleared, but insurers would begin covering expenses for the remaining 900 patients, enabling drug companies to start recouping their investment sooner, greatly decreasing the cost of bringing new drugs to market.

  Novartis and some other companies are already trying this approach as a way to prove concepts earlier in the development process. A small patient population with very similar disease characteristics is enough to show whether the rationale behind the design is correct. “Even if you have only twenty patients, if you then have a number who respond, you can say probably my hypothesis is right,” said Vasella. Whether the FDA will consider small populations for conditional approvals, however, remains uncertain.

  Druker and Vasella were also l
eft with a similar outlook on whether, and how, academia and industry can work together. In the aftermath of the clinical trials and in light of the evolving direction of cancer research, Druker could see that improved collaboration would be essential to the future of drug development. “The drug companies aren’t evil,” said Druker. “They make drugs, and we should help them.”

  Federal funding has remained flat in recent years, the revenue for hospitals attached to academic institutions will likely not be rising anytime soon, and philanthropic support has also faced challenging times economically, which leaves industry an indispensable source of funding. Plus, drug companies exist to make drugs and are well equipped to do so. “[Gleevec] was a successful collaboration between academic and industry,” Druker said. “Either side probably has some complaints about the other side, but it worked. We got a drug from a drug company to people, and it works, and that was an academic/industry collaboration.” Vasella also began wishing for more open collaboration with academia. In light of that need, the increased vigilance of bias among physicians who work with pharmaceutical companies seems, to him, detrimental. “I think you have to work together and you have to trust each other,” said Vasella. “And you can’t see always the evil in things, [and be] so focused on believing in rules and regulations and disclosure.”

  Druker could see now how the Bayh-Dole Act made life unexpectedly complicated for people interested in such collaborations. Passed in 1980, the act, also known as the Patent and Trademark Law Amendments Act, gave universities ownership of the intellectual property created there, even when that creation was made possible by federal grants. That change in law—before, any invention made using federal grant money was public domain—allowed for the biotech revolution.

  But an unintended consequence was runaway bureaucracy. Technology transfer offices sprang up in academic institutions where research labs were creating or testing substances that might advance medicine. The offices were there to manage the transfer agreements, but the process grew increasingly complex, making it difficult for academic investigators to work with companies. When Nick Lydon sent compounds to Druker to test in his lab, executing the contracts for that transfer took minutes. Nearly twenty years later, the process has become incredibly laborious, with academic institutions on hyperdrive about protecting any potentially lucrative additions to an experimental compound sent from a pharmaceutical company. “They’re fixed in their view of getting the best deal out of every deal,” said Druker, who feels that approach only wastes time. “I’m not willing to milk the last dollar out of every deal because it will just make it difficult to work with industry.”

  People involved in experimental cancer drug development also have become increasingly aware of the need to test combinations of experimental drugs in clinical trials. If cancers are driven by multiple mutations that can be targeted by multiple drugs, then being able to test two or more genetically targeted drugs at once is essential to advancing treatment. The problem is that the FDA generally does not allow two investigational drugs—meaning two drugs that have not been approved—to be tested in the same clinical trial. As Vasella continued to lead Novartis into creating drugs for new therapeutic areas, with an increased focus on rare diseases that has been extraordinarily successful for the company, he could see that these restrictions were holding up the pace of research. “One should be much more reasonable and flexible, and look at the rationale,” said Vasella, “and then decide from case to case based upon a concrete request.” At the very least, he said, companies should be able to collaborate or synthesize one another’s drugs for animal studies.

  Other pressing issues persist. In recent years, Michael Mauro, in addition to caring for leukemia patients at OHSU, mentored physicians in India, Africa, and Southeast Asia. Immediately Mauro was confronted with problems stemming directly from poverty and scant access to medical care. A physician in Africa told Mauro that even when he knows patients have CML, he often can’t get them the proper medicine because the diagnostic test to confirm the presence of Bcr/Abl requires resources that patients and hospitals don’t have. A couple of years ago, Mauro met a young woman from Mexico with two children. When she was diagnosed with CML, she was told she could have Gleevec for a while but that ultimately she would need a bone marrow transplant because a matching donor had been identified for her. Her insurer would not pay for continued medication and insisted she undergo an extremely dangerous procedure instead. “She knew [transplantation] was risky and would take her away from her family,” Mauro said. “So she came into the US illegally.” She found her way to OHSU, where she received emergency coverage and patient assistance from the drug manufacturer.

  Novartis has also become embroiled in a prolonged patent fight in India, where generic Gleevec has been made despite the company’s insistence that the drug is still patent protected. Vasella cheers the company’s patient-assistance program for being generous internationally, not only in the United States. “In India we have thousands of people who get [Gleevec] for free, and we have about 2,000 who pay for it,” he says. But the Indian government has accused the company of evergreening, extending the life of the patent by making ever-so-slight adjustments to the compound, altering it just enough to warrant patent extension without changing the underlying mechanism of the drug.

  As that fight has worn on, Novartis has continued to raise the price of Gleevec in the United States. The current cost for a one-month supply is $6,328. From 2001 to 2011, sales of Gleevec worldwide totaled $27.8 billion. The US patent for Gleevec, number 5,521,184, originally set to expire on May 28, 2013, has been extended to January 4, 2015.

  THE ADVENT OF Gleevec as a targeted drug has changed not only the course of cancer treatment, but medicine as a whole.

  “The revolution in cancer research can be summed up in a single sentence: cancer is, in essence, a genetic disease,” the eminent oncologist Bert Vogelstein is quoted in Siddhartha Mukherjee’s Pulitzer Prize–winning book on the history of cancer research, The Emperor of All Maladies. Today, stories about tumor sequencing and genetic drivers of cancer are top news headlines daily. Tyrosine kinase inhibitor programs are commonplace at large pharmaceutical companies, as are new, small biotechs created to develop a single targeted compound. When drug developers imagine the best-case scenario for their rationally designed drugs, they are imagining Gleevec for CML.

  In the decade-plus since Gleevec was approved, tyrosine kinase inhibitors have become a mainstay of cancer care. More than fifteen such drugs are now available. Among them are erlotinib for lung cancer, lapatinib for breast cancer, and sunitinib for kidney cancer as well as for GIST that hasn’t responded to imatinib. All were approved in the past decade, and all have improved the odds for cancer patients, extending survival time and offering a less harsh treatment option compared to chemotherapy alone, although many of these new drugs are given in combination with traditional chemotherapy. Tyrosine kinase inhibitors are a $15 billion-a-year market, an amount that is projected to double over the next ten to fifteen years. The fact that these drugs can often be taken at home adds to the way they have transformed cancer care. For many patients, treatment can be bent around their lives, rather than the other way around.

  The pharmaceutical industry has seized on the promise of tyrosine kinase inhibitors. Every large company has a kinase inhibitor pipeline: Novartis, AstraZeneca, Bayer, Johnson & Johnson, Merck, Pfizer, Eli Lilly, GlaxoSmithKline, and AstraZeneca, to name but a few. The number of potential kinase targets has become a vast landscape of technical research and endless abbreviations: PI3K, MEK, JAK1, JAK2, cyclic-dependent kinase inhibitors, the CAMK family, the TKL family, p38—again, to name but a few. All of these proteins are involved in one cancer or another, suspects in the search for its cause. All are encoded by genes that are, in some way, abnormal, proto-oncogenes that have transformed into oncogenes, just as abl is when it is translocated next to bcr, just as the src gene was altered when it was integrated into the Rous sarcoma virus. Today, more
than 500 kinase inhibitors are in development at more than 250 different companies, targeting more than 200 different proteins.

  Targeted therapy, the catchier name for rational drug design, extends far outside the realms of the kinase. The approaches divide into two main types: small molecule inhibitors that fit inside the cell, and monoclonal antibodies, which don’t. The success of Herceptin for breast cancer, approved just as the phase I trial of STI-571 was reaching an effective dose level, gave rise to a growing number of monoclonal antibodies. These drugs block targets outside the cell or on its surface using the body’s natural immune system. Among the small-molecule inhibitors, tactics beyond kinase inhibition include inducing cancer cell death, delivering radioactivity to cells containing cancer-causing molecules, cutting off the blood supply to tumors, and binding a structure called the proteasome as a way to kill cancer cells. Scientists are on a continual search for new pathways to target. Owen Witte, still researching in California, is currently trying to remove lymphocytes, part of the immune system, from the body, genetically engineer them to kill cancer, and then reinject them into the body to accomplish that goal.

  So far, targeted therapy has hardly lived up to the expectations set when Gleevec was approved. The benefits have, for the most part, been incremental. Patients’ lives are extended by months, a significant and important amount of time, but hardly the normal life span that people living with CML—many of whom don’t think of themselves as patients anymore—are now experiencing.

  The failure of tyrosine kinase inhibition or other targeted therapies to transform other cancers into tolerable chronic conditions has generated skepticism about the future of the approach. Some scientists point to genetic instability—the constant and unpredictable appearance of new mutations—as a guarantee that cancer cells will eventually become resistant to whatever therapy is thrown their way. “Therapies for Cancer Bring Hope and Failure,” ran the headline of a 2010 article in The New York Times by Andrew Pollack providing a where-we-are-now appraisal of the targeted-drug approach. “We’ve gone through a very rapid period of high expectations, maturation, and disappointments,” Dr. J. Leonard Lichtenfeld, deputy chief medical officer of the American Cancer Society, was quoted as saying in that article. “I think there was almost a naiveté that if we could find the target, we would have the cure.” Many targeted therapies have come on the market in the past ten or so years, and many are disappointments.