Finding therapies that take down only one biological molecule remains a challenge
Since 1990, Richard W. Hanson at the Case Western Reserve University School of Medicine has been living with cancer. An expert in metabolism and an opera aficionado with a weakness for Mozart, 77-year-old Hanson has chronic lymphocytic leukemia, the most common form of leukemia in adults. Over the past 12 years, Hanson has endured six rounds of chemotherapy, taking purine analogs, antibodies and other drugs to stop his mutant B cells from multiplying indiscriminately. Each round of chemotherapy has come faster and harder on the heels of the last. “If you have a disease like CLL, it becomes more than just an abstraction,” says Hanson. “It becomes a critical part of your life.”
But Hanson is a scientist to the core, so he has turned himself into an experiment. Over the past year and a half, he has participated in a clinical trial for a drug that specifically targets a tyrosine kinase in B cells. With the new drug, Hanson sees the promise of targeted therapies: “I can truthfully say that without this drug, I would be in that big lab in the sky.”
Targeted therapies are drugs designed with the knowledge of the target’s mechanism of action and biochemical role. In contrast, traditional drug discovery is more happenstance. Compounds, such as natural products, are screened against a number of targets or cells; the ones that show an effect are then studied to understand why selectivity exists.
Targeted therapies are considered to be a relatively new paradigm in drug discovery. The sequencing of the human genome and the following genomic revolution have dramatically increased the number of possible biological targets. As Vern Schramm at the Albert Einstein College of Medicine of Yeshiva University points out, drugs used to be found by screening combinatorial chemical or natural product libraries. This now means that medications in the U.S. Food and Drug Administration’s current approved list hit perhaps just 1 percent of the biological molecules that can be targets. “We’ve really only scratched the surface,” says Schramm.
Much of the excitement about targeted therapies is seen in the cancer world. Historically, a number of chemotherapeutic drugs, like the purine analogs that Hanson took, were little more than poisons that attacked rapidly dividing cells, both healthy and cancerous.
Therapies for other diseases, such as autoimmune, psychiatric and cardiovascular conditions, fared better. “You knew there was a molecule that was involved in a particular pathology, and you targeted it,” says John Kyriakis of Mercury Pharma. In-depth molecular knowledge was missing for many kinds of cancers for a long time.
But the notion of targeted therapies for cancer is a longstanding one. As Siddartha Mukherjee explained in his Pulitzer prize-winning book “The Emperor of All Maladies,” Paul Ehrlich, who won the 1908 Nobel Prize in medicine, dreamed of finding a magic bullet that could distinguish between malignant and normal cells. Stanley Farber and dozens of other physicians and scientists followed him, hunting for chemicals that could specifically seek out cancer cells.
Indeed, some experts note that “targeted therapies” can be a buzz term. Every drug is targeted in some sense. Furthermore, a number of targeted therapies, despite their name, inadvertently hit other molecules, much like traditional chemotherapeutic drugs.
How Gleevec shook up cancer therapy
Targeted therapeutics took the cancer world by storm more than a decade ago. In 1998, trastuzumab, better known by its trade name Herceptin, came out of Genentech. It was an antibody that bound to HER2/neu receptor on breast cancer cells. But in 2001 a molecule came along that made scientists and oncologists take note. It targeted a kinase.
“Ever since the discovery of kinases and signaling phosphorylation in the 1960s, the consensus was you’d never be able to make a specific kinase inhibitor because they all bind ATP in exactly the same way,” says David Stokoe of Genentech. Kinases are involved in signaling pathways that control critical functions, such as the cell cycle, protein expression and genome stability. As the first kinase structures appeared in the early 1990s and the Human Genome Project eventually showed that there were at least 500 of them, many scientists doubted if it would be possible to get an inhibitor specific enough to hit just one or two.
Imatinib, best known by its U.S. tradename Gleevec and marketed by Novartis, washed those doubts away. Developed in the late 1990s by Nicholas Lydon, formerly at Novartis and now at Blueprint Medicines, Brian Druker of Oregon Health and Science University, Janet Rowley at the University of Chicago and others, imatinib was the first drug to inhibit specifically the Bcr-Abl receptor tyrosine kinase that is the root cause of chronic myelogenous leukemia. The constitutively active Bcr-Abl kinase is produced by a reciprocal translocation between chromosomes 9 and 22. Ninety-five percent of CML patients have this mutation, which also shows up in several other cancers.
Unlike conventional chemotherapeutic drugs, imatinib targeted only the cancerous cells expressing the mutant kinase and left alone the cells lacking it. The FDA approved the drug in May 2001, and Time magazine put the drug on the cover as the magic bullet to cure cancer.
Although designed as an inhibitor of Bcr-Abl, the compound also inhibits the platelet-derived growth factor receptor, a cell-surface tyrosine kinase, and c-kit, a cytokine receptor on hematopoetic stem cells that is a tyrosine kinase. Mutant PDGF receptors are involved in chronic myelomonocytic leukemia and c-kit mutations are found in stomach cancers. For these reasons, the FDA expanded its approval for imatinib to treat 10 different cancers by 2011. In January, Lydon, Druker and Rowley were awarded the Japan Prize for their work.
Imatinib drove home the point that, although all kinases use ATP, “every enzyme is mechanistically different in atomic detail,” says Schramm. The drug changed the treatment of CML. Prior to imatinib, CML chemotherapies eventually forced patients to get bone-marrow transplants. Now more than 90 percent of CML patients are treated with a pill. Very few go on to have transplants, and the number of deaths caused directly by CML per year is less than 100 in the U.S., says John Byrd of Ohio State University. “The therapy has completely changed how the disease is managed.”
Designing targeted therapies
Imatinib set off the hunt for more targets in cancer. Targets can be defined in several ways. They can be genetic mutations, such as the one that produces Bcr-Abl. Another way to define a target is to pinpoint molecules that are essential in metabolic and signaling pathways. Once a target is identified, scientists also have to make sure that it’s accessible to drugs.
But as Stokoe explains, it’s often difficult to find genetic changes that are responsible for the cancer. Most cancers, especially solid tumors, “are just genomic carnage. All hell has broken loose,” he says. “I think the hard part is to try and find the genetic alterations that have actually benefited the tumor cell over all of the noise that has come along for the ride.”
The drug that has kept Hanson out of the “big lab in the sky” illustrates the principles of targeted therapy design. Before he took the drug, Hanson’s B cells were crowding out red blood cells and platelets and enlarging his lymph nodes to the point that his blood circulation was affected. Walking down a hallway became hard work. He sweated at night. With a weakened immune system, he was always worried about getting infections. After the sixth round of chemotherapy, Hanson’s lymphocyte level was 30 times higher than normal.
Because cancerous cells continue to mutate, “the clone you started with is not the clone that kills you,” says Hanson. He eventually experienced the common CLL mutation in which the short arm of chromosome 17, where the gene for p53 resides, was deleted. His B cells were free of the tumor suppressor. “That’s a true death sentence,” he says. The life expectancy for patients with that mutation is, on average, up to a year.
That’s when Hanson heard of a clinical trial at the Arthur G. James Cancer Hospital at the Ohio State University Comprehensive Cancer Center that was being run by John Byrd and his colleagues. The trial was testing a drug developed by Pharmacyclics that binds to a molecule called Bruton’s tyrosine kinase, or Btk for short. Hanson was allowed to enroll in the trial during its early phases.
Joseph Buggy at Pharmacyclics describes how the company knew to go after Btk. There was 10 years of scientific literature supporting the importance of the B-cell receptor signaling pathway in B-cell proliferation. Btk is an essential kinase in the signaling pathway downstream of the B-cell receptor. The pathway in which Btk is involved leads to the phosphorylation of several proteins that are antiapoptotic. When phosphorylated, these proteins prevent apoptosis.
But the most critical piece of information about Btk was genetic. There is a disease called X-linked agammaglobulinemia in which patients lack mature B cells. “That told us that if we could come up with a molecule that was selective enough for Btk, it shouldn’t affect other organs or tissues,” says Buggy.
Every expert interviewed for this article emphasized that genetic validation is the key to finding proper targets. Genetic validation “takes guesswork and the need to understand the biology almost out of it,” says Kevan Shokat at the University of California, San Francisco. “For every degree you get separated from the mutated human oncogene, the more biology is incumbent on you to figure out in order to be sure it’s going to be a satisfactory target.”
Pharmacyclics developed ibrutinib, an irreversible Btk inhibitor that binds to a cysteine found in only 10 or so kinases. When this drug blocks Btk, it induces apoptosis in cells that otherwise refuse to die (1). The irreversible binding of ibrutinib to the kinase meant that “you can durably inhibit the target, even if the drug is eliminated quickly” from the body when it’s not bound to the enzyme, says Buggy. Once bound, ibrutinib molecules cling onto Btk for as long as 24 hours. The drug is now going onto phase III clinical trials.
Hanson now pops a 420-mg pill of ibrutinib every day. He says the daily dose “has made me feel like I have a future. I am amazed by it. I should be dead.”
But it’s not a completely rosy picture with targeted therapies, and a number of challenges dog the hunt for targets. For instance, earlier this summer, a group of researchers, which included Bert Vogelstein of Johns Hopkins University and Kelly Oliner of Amgen, demonstrated why colorectal cancer patients eventually develop resistance to a targeted therapy called panitumumab after several months of treatment (2). They showed that well before these cancer patients started on the drug treatment they had a number of tumor cells with randomly mutated genes that evolved into providing resistance to the drug.
Side effects remain a concern, despite these drugs being so-called “targeted therapies.” “We are seeing toxicities,” says Thomas Force of Temple University, citing two examples – sunitinib, sold by Pfizer as Sutent to treat gastrointestinal stromal tumors, and dasatinib, marketed as Sprycel by Bristol-Myers-Squibb to treat certain adult leukemias.
As Force explains, one organ that is inadvertently affected by drugs that target kinases is the heart. The heart is an enormous consumer of energy. Any perturbations to the energy production system, either from dialing down or up kinase activity, could cause cardiac dysfunction. Other organs may be affected, but the heart, says Force, is “the first thing to go.” With more therapies targeted against kinases in the pipeline, Force anticipates that researchers will see cardiac abnormalities crop up in some patients taking some of the drugs over the long term.
Another challenge is knowing when the therapy will work. Vemurafenib is a drug developed by Plexxicon and Hoffmann–La Roche that received FDA approval for the treatment of late-stage melanoma last summer. It is an inhibitor of B-Raf oncogene, which is mutated in about 60 percent of malignant melanomas. Patients with a particular mutation, the V600E mutation in which the valine at position number 600 in the kinase is replaced with glutamic acid, respond well to vemurafenib. But if the drug is given to patients who don’t have the mutation, they develop a secondary cancer.
“It’s the prototype right now for a drug that can be given safely to only patients with a mutation” in B-Raf, says Shokat. “If you give it to a patient who does not have the mutation and may have a Ras mutation, they are going to get another cancer induced. It’s not life-threatening like the first one, but it’s not going to benefit them.”
Other stumbling blocks are emerging. For instance, researchers are beginning to appreciate that kinases are dynamic entities that can easily change their conformations. Vemurafenib causes secondary cancers in patients missing the V600E mutation because it causes B-Raf to dimerize. The dimerized kinase stimulates more signal transduction. “Now that we know that happens, we can screen for molecules that don’t disrupt the kinase conformation and don’t allow dimerization,” says Shokat. “But just until a few years ago we didn’t think that was all that important.”
Targets can be two-faced. One kinase that drug companies are pursuing hotly is TOR. Shokat and colleagues recently demonstrated that TOR can either help or hinder therapy. They showed that inhibition of the TOR kinase could inadvertently cause acceleration of the very signaling pathway the inhibitor was supposed to stall, because TOR is involved in both positive and negative regulation of the Raf/ERK pathway. Depending on whether it’s dialing the pathway up or down, TOR can be either a target or an antitarget (3). “The difficulty is TOR is not a target in every cancer setting, but that’s because cancer is not one disease,” says Shokat.
Another challenge is that the number of feasible kinase targets is limited. “We’re running out of good kinase targets,” says Stokoe. “There are probably five to 10 really well-validated kinases that are known to drive tumorigenesis in a significant number of tumors.” The restricted number of known good kinase targets also limits the ways in which signaling pathways can be disrupted, narrowing down possibilities for therapies.
But it’s important to note that kinases aren’t the only molecules that lend themselves to the endeavor of targeted therapies in cancer and other diseases. Molecules such as nuclear receptors, histone modifiers, poly ADP ribose polymerase and proteins involved in ubiquitination are all candidate targets.
Despite the obstacles, the optimism for targeted therapies is unabated. Experts urge continued support for fundamental research in molecular biology and biochemistry so that findings can be translated into the clinic in the form of targeted therapies. Byrd, who does both fundamental research and patient care, sees the bridge from bench to bedside every day. “When what you’re doing in the lab is touching patients like Dr. Hanson to put their disease into remission, that’s just very special,” he says.
- 1. Honigberg, L.A.; et al; Proc. Natl. Acad. Sci. USA 107, 13075 – 13080 (2010).
- 2. Diaz, Jr et al; Nature. DOI: 10.1038/nature11219 (2012).
- 3. Dar, A.C. et al; Nat. Chem. Bio. DOI: 10.1038/nature11127 (2012).
Rajendrani Mukhopadhyay (firstname.lastname@example.org) is the senior science writer for ASBMB Today and the technical editor for The Journal of Biological Chemistry. Follow her on Twitter at www.twitter.com/rajmukhop.
View a three-part video seminar by Brian Druker on targeted therapies, as prepared in 2010 for iBio Seminars: