When David Schenkein left Genentech to lead Agios Pharmaceuticals (AGIO -0.44%) in 2009, part of the draw was the fact that the Cambridge, MA-based start-up was positioning itself as a trailblazer in a raw field of oncology—cancer metabolism. The idea, essentially, was to starve tumors by blocking the enzymes that help them gobble up nutrients.

Five years and many millions of dollars later, the company has produced the first evidence in humans that the idea might yield viable treatments for cancer—though there is, of course, still a long road ahead and many hurdles to clear before it will know for sure.

Agios today is announcing the first clinical data on its lead experimental cancer metabolism drug, AG-221, at the American Association for Cancer Research's annual meeting in San Diego. To be clear, this is an interim look at a small, early trial—a Phase 1 study that's designed mainly to assess the drug's safety and side effects and to determine the maximum dose that patients can handle. And Agios is only reporting data from seven patients in the study, each of whom has acute myeloid leukemia.

But six of those seven responded to AG-221—and three of them had no trace of cancer in their blood after 28 days of treatment. This is despite the fact that they only received the lowest two of four planned dosages of AG-221. Also encouraging is the fact that the company so far hasn't run into toxicities that would keep it from ramping up the dosage.

"It's unusual," says Eytan Stein, the study's lead investigator and an attending physician at Memorial Sloan-Kettering Cancer Center. "[But] it's extremely early. I am cautiously optimistic, certainly this seems like a drug that has great clinical potential, but there's always that possibility that the next 10 patients aren't going to respond at all."

AG-221 targets an enzyme known as isocitrate dehydrogenase-2, or IDH2, which is part of the molecular machinery that generates energy within cells. Researchers say that when IDH2 is mutated in cancer, it ends up producing a byproduct called 2-hydroxyglutarate, or 2HG, that flips a genetic switch in immature bone marrow cells turning them into cancer cells rather than the blood cells they were supposed to become.

By binding to the mutated IDH2, AG-221 is designed to stop this process from happening and allow the normal maturation of the marrow cells to resume. In other words, rather than killing cancer cells, the drug is theoretically stopping them from being born in the first place. "That's very different than the way we think about any other kind of AML but perhaps it could serve as a paradigm of what we really need to do to attack other genetic subsets of AML," Stein says. (About 10 to 15 percent of the patients with AML have the IDH2 mutation, he adds.)

This approach, if it proves successful, could offer a dramatically different treatment option for patients with AML, a fast-moving type of blood cancer. The typical treatment for the disease is chemotherapy, which essentially nukes the blood system, destroying cancer cells and healthy cells alike and leaving people weak, sick, and prone to infections.

"Certainly the hope is that [the Agios drug] won't have the kind of immunosuppressive effects that chemotherapy has," Stein says. "Theoretically what should happen is the patient should never feel worse than they felt before, they should just start feeling better. And that's actually what we've seen in the patients that responded to the drug. They gain weight, they have an appetite, they don't have infections. So that's very encouraging."

AML patients also often wind up needing bone marrow transplants, which carry their own set of risks. In an absolute best-case scenario, AG-221 would eliminate the need for transplants. But, Stein cautions, "I think that it's way premature to say that's what's going to happen."

Indeed, much is left to be proven. Here's more on what the study has shown so far:

Agios enrolled 22 patients who have either AML or myelodysplastic syndrome (when the bone marrow doesn't produce enough healthy blood cells) and an IDH2 mutation, and have failed one to four prior rounds of chemotherapy—very sick patients. These patients, who have a median age of 62.5, are receiving 30 mg, 50 mg, or 75 mg of AG-221, in pill form, twice a day, or a 100 mg dose once a day, for 28 days.

The data being presented today are on seven patients who were on 30 mg or 50 mg doses of AG-221. (Three additional patients were in the 30 mg group, but died of disease-related infections and couldn't complete a full cycle of treatment.)

Six of the seven patients responded to the drug, meaning their 2HG levels decreased by more than 90 percent, according to Agios. Three of them saw their cancer completely wiped out. Two others had a complete remission as well, but their platelet counts haven't returned to normal levels yet (Stein says investigators are hoping with further treatment they'll get to that point). A sixth patient had a partial response to the drug.

Stein says that the one person who didn't respond to the drug, who was on a 50 mg dose, is being evaluated to find out what was different.

"It may have something to do with other gene mutations the patient had that may be driving their leukemia," he says.

So what about side effects? While Agios said the drug has been well tolerated so far, there were two potential issues. One patient developed differentiation syndrome—a complication in which fluid leaks into the lungs and other tissues. While that condition can be dangerous, Stein says it's easily treated with steroids. Nonetheless: "That's something we're going to have to keep a very close eye on," he says.

A second patient developed confusion, though Stein says the patient who had it was "critically ill" and it's unclear whether it was related to the drug, and that no other patients in the study have experience such episodes.

Agios has already begun testing the 75 mg and 100 mg doses of AG-221. They haven't had any safety problems or dose-limiting toxicities, but the company doesn't have the efficacy data on those patients yet, according to Stein.

Agios still hasn't chosen which dose to study in its next trial. Once it does, Stein says the company will expand its current study, treat a few more patients at that dose, sit down and look at all the data, and decide what the best clinical path forward is.

Though uncertainties still abound, today's presentation marks an important milestone for Agios, which was launched in 2007 by Arch Venture Partners and Flagship Ventures. Third Rock Ventures joined Arch and Flagship to lead a $33 million Series A for Agios the next year. The VC firms then hired Schenkein, the former head of cancer drug development at Genentech, to steer the ship.

Schenkein took over with big dreams for Agios. "What I saw was the potential to move into a novel area of targets that could [hit] cancer from an orthogonal approach that we had just not been doing," he told me last year. "It had the feel of 'go big or go home.'"

And indeed, a lot is riding on AG-221. While Summit, NJ-based Celgene holds worldwide rights to the drug under a $130 million deal the two companies struck in 2009, and would thus gain the most financially from its success, Agios will earn milestone payments and royalties should it continue forward. But more importantly, the drug is a test case for Agios's approach of going after cellular metabolism targets. A second drug, AG-120—which Agios owns U.S. rights to—is designed to work very similarly to AG-221, for instance. Agios is developing AG-120 as a treatment for solid tumors with the IDH1 mutation. And the company is looking into a class of diseases known as inborn errors of metabolism after that.

Agios raised over $120 million in equity financing and another $141.2 million in partnership revenue before its IPO in July. It added another $111 million in net proceeds from the offering, and ended 2013 with about $194 million in cash on hand.

This article originally appeared on Xconomy, along with: