What a difference an amino acid makes. Five years ago, researchers noticed that many patients with rare blood cancers called myeloproliferative neoplasms, or MPNs, had a phenylalanine in place of a valine residue at a specific position on JAK2, a protein kinase involved in blood cell development. The discovery set off a hunt for molecules that could block the errant enzyme and thereby quash the diseases.
Today, a handful of JAK2 inhibitors have reached early clinical trials, and the most advanced drug, Incyte Pharmaceuticals’ INCB18424, could hit the market late next year. So far, the JAK2 blockers appear to have a profound effect on the symptoms of MPNs, but it’s not yet clear whether they can send patients into remission. Several drug company scientists discussed the ongoing development of JAK2 inhibitors at a symposium held by the Division of Medicinal Chemistry at last month’s American Chemical Society national meeting in Boston.
By name alone, JAK2 wouldn’t appear to be the most auspicious of drug targets. Australian researcher Andrew F. Wilks discovered two intriguing protein kinases in 1989. After sequencing them two years later, he flippantly referred to each member of a family of four proteins as “just another kinase.”
Over time, the JAK family became understood to be part of a signaling pathway, called JAK-STAT, used by the immune system. For years, drug companies were primarily interested in JAK3, an attractive target because of its role in white blood cell mediation and because, unlike its ubiquitous family members, it is expressed only in immune cells.
Fast-forward to 2005, when researchers found that many patients with MPNs had a mutation at residue position V617F in JAK2. Roughly 80% of people with polycythemia vera, a disease marked by overproduction of red blood cells, and half of those with myelofibrosis, characterized by scarring of the bone marrow, have the errant gene. About 20% of those with essential thrombocythemia, in which excess platelets are made, also have the JAK2 mutation.
Though slow-moving, MPNs are still quite serious. All patients suffer from an enlarged spleen—so large that it can hang almost a foot below the ribcage—and a variety of other debilitating symptoms that depend on the type of MPN. Worse still, the MPNs can eventually develop into acute myelogenous leukemia, a cancer with a dismal prognosis.
But prior to 2005, there was little interest in tackling MPNs, owing to their unknown origins. News of the JAK2 mutation set off a land grab among companies looking to develop drugs against a validated target.
“Before the discovery of JAK2, there hadn’t been a really good, testable idea in a long time,” explains Ross L. Levine, a hematologist at New York City’s Memorial Sloan-Kettering Cancer Center who helped identify the mutations in MPNs while he was a fellow at Harvard Medical School. In the intervening years, “the field has grown dramatically,” he adds.
“In just five years, we’ve seen a number of compounds already advance quite far into the clinic,” Mark W. Ledeboer, a medicinal chemistry research fellow at Vertex Pharmaceuticals, told the audience in Boston.
Indeed, a survey of the patent filings for compounds targeting JAK2 confirms a sudden surge of corporate interest in the protein. According to data from Chemical Abstracts Service, the firms Vertex, Merck & Co., Novartis, and Incyte have been the most active filers of patents on JAK2 inhibitors.
Others have bought their way into the space. This summer, Sanofi-Aventis acquired San Diego-based TargeGen, whose JAK2 inhibitor, TG101348, has completed Phase I trials. Last December, Novartis paid $210 million up front to license rights outside the U.S. to Incyte’s INCB18424.
As the most advanced compound in development, INCB18424 is generating excitement as a possible first-ever treatment for myelofibrosis. Incyte began its chemistry campaign for the compound in 2004, on the basis of long-standing evidence that STAT, JAK2’s partner in its signaling pathway, could be oncogenic. Incyte’s goal was to find a compound that could block both JAK1 and JAK2 but wouldn’t hit JAK3. At the time, two known JAK inhibitors were disclosed: Pfizer’s CP690550, a JAK3 inhibitor with some activity for other JAKs, and a JAK1/2/3 blocker by Merck that featured a tetracyclic pyridone scaffold.
Incyte chemists took an empirical approach to finding their lead inhibitor, starting with the Merck scaffold and modifying it to shift selectivity to JAK1 and JAK2, recalls James D. Rodgers, the biotech firm’s executive director of medicinal chemistry. But the first compound they came up with, a selective tetracyclic analog, wasn’t sufficiently bioavailable, meaning it couldn’t be given as a pill.
Around that time, Pfizer reported that its JAK3 inhibitor, in clinical trials as an immunosuppressant, had good pharmacokinetic properties. Incyte changed tack and decided to develop a hybrid scaffold that would incorporate the selectivity of its tetracyclic compound and the oral availability of the Pfizer molecule. The result was a bicyclic scaffold that chemists diligently modified to increase its potency and selectivity. Incyte filed a patent for the compound class in late 2005, and by June 2007, it was in clinical studies.
Although the JAK2 mutation is present in the majority of patients with polycythemia vera, Incyte’s first human tests of INCB18424 have been on myelofibrosis. “We decided to go after myelofibrosis because those were the sickest patients with the greatest need,” says Richard S. Levy, Incyte’s chief drug development and medical officer. The first patient was suffering from a spleen so enlarged he was practically incapacitated. “Within two weeks, his spleen had shrunk by 50%, and within a month, he was playing golf,” Levy recalls.
That kind of dramatic reduction in symptoms is “unprecedented,” says Sloan-Kettering’s Levine, who has collaborated with Incyte and received funding from Novartis and AstraZeneca. “The spleen reductions are quite significant. No drug has ever done that.”
Incyte’s drug candidate appears to be well tolerated and has continued to have a major impact on the symptoms of myelofibrosis in Phase I/II trials. A paper highlighting the results of the study, published in the New England Journal of Medicine earlier this month, showed that INCB18424 was able to quickly reduce the spleen size of about 75% of patients studied, an improvement that lasted for more than a year on treatment (2010, 363, 1117).
The company will unveil the results from Phase III studies later this year and expects to file a New Drug Application with the Food & Drug Administration in the first half of 2011. Anticipating approval for treatment of a variety of MPNs, Cowen & Co. stock analyst Eric Schmidt forecasts sales of INCB18424 to reach $625 million in 2015.
Although Incyte had a head start on the development of JAK2 inhibitors, most other companies began their programs after scientists unearthed the genetic link between the protein and MPNs. As the therapeutic and financial potential for JAK2 inhibitors has risen, so has the interest in starting drug discovery programs around the target. Since 2005, some 24 companies have filed patents in the space.
Cytopia, an Australian firm founded by Wilks, the “father of JAK,” had its own advantage. In 2005, the biotech firm was focused on JAK3 but was able to revisit its library of JAK inhibitors when JAK2 was identified as an intriguing target. A class of compounds with a pyrimidine scaffold served as the starting point for the identification of CYT387, an aminopyrimidine derivative.
Like Incyte’s compound, Cytopia’s blocks both JAK1 and JAK2, an approach meant to improve the inhibition of the IL-6 family of immune system proteins, which use the JAK pathway for signaling. “We were likely to give ourselves a number of other opportunities outside of MPNs,” explains Christopher J. Burns, who led Cytopia’s medicinal chemistry campaign. For example, IL-6 proteins are implicated in prostate and breast cancer as well as multiple myeloma.
Early Phase I results show that CYT387 quickly reduces spleen size. This spring, YM Biosciences, which had just acquired Cytopia, said data were emerging that suggest CYT387 is as potent as the Incyte drug and is better at avoiding JAK3.
AstraZeneca also initiated a JAK2 program after hearing the compelling genetic story behind its role in MPNs, Stephanos Ioannidis, an AstraZeneca medicinal chemist, said in Boston. After a large screen was carried out that identified two promising compound series, the company homed in on a series of pyrazol-3-ylamino pyrazines, which, after modifications, led to AZD1480, its drug candidate. AZD1480, which is more potent for JAK2 but also blocks JAK1, is now in Phase I/II trials as a myelofibrosis treatment.
With the proliferation of JAK2-related chemistry campaigns, it’s getting harder to create unique intellectual property (IP), Vertex’ Ledeboer pointed out at the Boston meeting. Most of the molecules in the clinic seem inspired by earlier JAK inhibitors from Pfizer and Merck, he noted.
Companies interested in finding novel JAK2 inhibitors face a medicinal chemistry challenge. Vertex has spent the past several years attempting to develop an inhibitor that would stand out from the pack, but it has had a tough time finding a compound with the right mix of selectivity, safety, and oral availability.
“The IP landscape is getting crowded,” Ledeboer told the Boston audience. “The question for medicinal chemists is whether we can develop proprietary compounds.”
Vertex had already been working on JAK3 inhibitors when the correlation between JAK2 and MPNs was discovered, thus easing the move into JAK2. An initial screen of the company’s library of protein kinases yielded a few early leads: Of particular interest were 3,4-ring fused 7-azaindoles and deazapurines, both of which showed reasonable potency and had a low molecular weight.
The finding eventually led to the selection of a polycyclic azaindole-based derivative. But the compound had some stability issues, and Vertex wasn’t able to achieve sufficient selectivity for JAK2 over JAK3. The company eventually abandoned it because of concerns about liver toxicity, Ledeboer said.
Meanwhile, early successes in the clinic come with caveats. The current therapeutics are clearly good at alleviating the symptoms of the disease—a major victory for patients who can be sidelined by discomfort. But clinicians are skeptical that a JAK2 inhibitor can quash blood cancer on its own.
Incyte’s compound, for example, has yet to show a significant reduction in mutated JAK2 in patients taking the drug. The finding is a bit of a disappointment. News of the frequency of the JAK2 mutation in MPNs led many to hope that inhibitors of the protein would generate the kind of swift remissions seen in patients who take Novartis’ chronic myeloid leukemia (CML) drug, Gleevec, which targets a mutation in bcr-abl, a fusion protein that drives the disease.
“The initial disclosure got a lot of us excited because we thought it was another opportunity to affect disease in exactly the same way the bcr-abl inhibitors have been successful in CML,” YM Biosciences’ Burns says. “The reality is now we’re realizing that the MPNs are a much more complicated family of diseases.”
From the beginning, the fact that not every patient with an MPN has a JAK2 mutation meant that other processes are driving the disease. “In cancer, we will probably need to have a combination of a JAK1 and JAK2 inhibitor and some other therapeutic,” Burns adds.
Hematologists are anxiously awaiting results from the Phase III studies of INCB18424 to see whether long-term use of the drug prompts a drop in mutated JAK2. Academic researchers are also trying to unravel other proteins that might conspire with JAK2 in blood or other cancers, with the idea of combining therapies.
Earlier this month, Sloan-Kettering’s Levine and colleagues published a paper that proposes blocking both JAK2 and HSP90, a protein that assists in keeping other proteins properly folded (J. Clin. Invest., DOI: 10.1172/jci42442). “To our surprise, JAK2 is a very unstable protein,” Levine says. “There’s a constant battle in cells between ongoing synthesis and degradation of JAK2.” Knocking out the protein that helps keep it stable could tip the scales in favor of degradation.
Researchers also see more room for improving JAK2 inhibition—the protein must be hit hard to achieve a therapeutic effect—and note that the efficacy of existing drugs might be improved just by switching up how they’re administered. “Maybe the way we dose these drugs isn’t right, or maybe multiple agents are needed,” Levine says. “We’re earlier into the story than we might have presumed we would be five years after the discovery. There’s a lot to understand and a lot of this journey left to take.”