At the annual conference of the American Society of Clinical Oncology (ASCO), starting tomorrow in Chicago, drug giant Merck will be presenting no less than 100 abstracts detailing its efforts to develop new cancer drugs. The company’s oncology franchise head, Gary Gilliland—who left a 20-year faculty position at Harvard to join New Brunswick, NJ-based Merck in 2009—is excited about every one of them. This is, after all, why he left academia to work in the drug industry. “We have an increasingly sophisticated understanding of the genetics of cancer, but we need to effectively apply that so we can improve patient outcomes,” Gilliland says. “From my perspective, the best place to do that is at a company like Merck.”
At ASCO, Merck (NYSE: MRK) will spotlight several examples of the company’s evolving approach to cancer drug development. Under Gilliland’s tutelage, Merck has increased its focus on what he calls “biomarker-driven patient stratification”—the effort to identify the people who are most likely to respond to particular drugs based on the molecular characteristics of their tumors. The company has also doubled down on its efforts to form innovative partnerships in oncology research, and to make the entire development process more efficient, so the drugs with the highest likelihood of success will get to the market quickly.
Gilliland points to a couple of key drugs that will be presented at ASCO as examples of the early fruits of Merck’s oncology strategy. The first is MK-3475, which was designed to mobilize the patient’s own immune system to kill tumor cells. The molecule is part of an emerging class of drugs called programmed death-1 (PD-1) inhibitors. In some tumors, a particular type of PD-1 is over expressed, resulting in a breakdown of vital immune regulators called T-cells. In early human trials, Merck’s PD-1 inhibitor appears to re-activate T-cells without causing the high level of side effects seen in other drugs in the class.
Merck is currently assessing whether the positive results seen in the trials are enough to speed MK-3475 into late-stage trials. “When you have striking efficacy signals, it’s tempting to move forward at a more rapid pace,” Gilliland says. “Sometimes you don’t need as many patients to demonstrate the value.” He says his team is now determining how best to move the compound forward into the pivotal trials that will be required for FDA approval.
At ASCO, Merck will also present early-stage results trials of MK-2206, a drug that inhibits a cancer-causing enzyme called AKT. In some of the studies, Merck teamed up with AstraZeneca (NYSE: AZN) to test MK-2206 in combination with an AstraZeneca molecule that inhibits a different enzyme involved in cancer. “We anticipated that as a single agent, [MK-2206] was not likely to have dramatic therapeutic activity,” Gilliland says. “The reason for that is if you turn one pathway off, cancer cells are very clever and they simply turn on another one with enzymes that enhance their proliferation.”
Gilliland says Merck and AstraZeneca were the first two pharmaceutical companies to form a partnership designed to test a combination of drugs that have not yet been approved separately by the FDA. If future trials prove the theory that the drugs will work better together than they would as solo therapies, the companies intend to jointly file for FDA approval of the combination—a strategy that’s never been tried with two novel drugs, Gilliland says. “We’re having ongoing discussions with the FDA and they have been supportive,” he says. “They understand that in cancer you have to use combinations, because they improve efficacy and they help prevent resistance.”
Gilliland predicts that the strategy of partnering very early in the development process will catch on in the pharmaceutical industry. “We shouldn’t all be making the same drugs—we should be talking to potential partners,” he says. “Our argument is if you show synergy with a combination you should run with that, rather than putting patients on single therapies that are not likely to be successful.”
As for identifying patients who are most likely to respond to particular drugs, a group of researchers will be presenting a study on Tuesday that Gilliland says demonstrates Merck’s commitment to that strategy. The study involves ridaforolimus, a drug that inhibits an enzyme called mTor and that Merck is developing with Cambridge, MA-based Ariad Pharmaceuticals (NASDAQ: ARIA). The drug has hit some bumps, most recently on March 12, when an advisory panel to the FDA voted 13-to-1 against the approval of ridaforolimus for patients with some bone cancers. The FDA is expected to hand down its decision by June 5.
In the study being presented at ASCO, ridaforolimus was tested in patients with non-small cell lung cancer who have a specific genetic mutation called KRAS. The study shows that patients with the mutation who were put on the drug experienced significantly longer “progression-free survival”—the period of time during which their tumors did not grow—than did those who were given a placebo. “We don’t know yet if it will translate to overall survival, and it’s a small study, but it’s a big signal,” Gilliland says. “So we’re very excited about the opportunity for bringing it forward into a broader confirmatory trial in lung cancer.”
Gilliland says identifying patients most likely to respond positively to cancer therapies will alleviate a rampant problem in the pharmaceutical industry: drugs that fail in late-stage trials, after companies have spent several years and millions of dollars trying to prove they work. “Since 2010, there have been 16 [late-stage] failures in oncology across the industry,” he says. One of those involved Merck’s lymphoma drug vorinostat (Zolinza), which the company hoped to have approved to treat mesothelioma. But the trial failed last September. “We didn’t have a clear idea of what the [drug] might be doing to the cells,” Gilliland admits.
Now, says Gilliland, Merck is much more vigilant in early trials about choosing the right patients to test, and giving up quickly if they don’t respond as well as predicted. “The insight we’ve gained collectively with these failures is you have to understand what subset of patients is gong to respond to your drug and you have to have the biomarkers to enable that,” he says. “If we’re going to fail, we try to fail early with as few patients as possible.”
He hopes Merck’s high-profile presence at ASCO will show the company is following that mantra. “We’re looking for a very high degree of efficacy, and if we don’t see what we’re looking for, we stop,” he says. “We’ve set a high bar for futility.”