Array Biopharma Inc.'s Phase II flameout with ARRY-162 in rheumatoid arthritis means the company will quit pumping its own cash into the inflammation program, but a would-be partner might restart the push, as work with the small-molecule MEK inhibitor in cancer goes on.
CEO Robert E. Conway told investors during a conference call that Array may split indications with collaborators, since the dose would be much higher for cancer patients, but he added that it's unclear whether the RA effort would generate much interest now.
Kevin Koch, the company's chief scientific officer, noted that the latest outcome likely will make RA more of a challenge. Any partner would need to gather data "beyond signs and symptoms and work with the FDA to look at imaging response," he said. "I'm not saying [it's] difficult, but a more complex trial to run. The pre-TNF population is probably not viable at this point."
Response rates seemed encouraging overall. The drug given at 10 mg twice daily garnered a 60.4 percent ACR20 response rate at 12 weeks, and 20 mg at the same frequency gained 54 percent, while 40 mg given four times per day hit a rate of 61.2 percent. Anti-TNF drugs tend to give patients an ACR20 score in the range of mid-60 percent.
Still, the endpoint was missed, and Boulder, Colo.-based Array's shares (NASDAQ:ARRY) fell 23 percent Friday to close at $2.90, down 88 cents on word that preliminary data showed no treatment group in the 201-patient trial with ARRY-162 hit statistical significance in the ACR20 response rate.
The study included three dose groups and a placebo arm, testing patients not completely responsive to methotrexate. The overall placebo change was a 46.9 percent rise in the ACR20 response rate. In Eastern Europe, the number was 34.6 percent. But in South America, which enrolled 99 patients, placebo pulled 60.9 percent - a number that Christopher Raymond, analyst with Robert W. Baird & Co., called "shocking."
How the trial was conducted in South America may have had something to do with the huge difference, Raymond told BioWorld Today. "We don't have any of the details, but one that's telling is Array is giving up on the product [in RA]," he said.
Eastern Europe, with 101 patients, showed at least a trend toward efficacy as measured by DAS28-4 (p = 0.067), "good" EULAR response (p = 0.105) and ACR20 response rate (p = 0.115). Array plans to offer full results from the trial at a medical conference though probably will not have the assembled data in time for the American College of Rheumatology meeting, slated for late October in Philadelphia, Conway said.
Koch pointed out that ACR scores in the treatment arms were "still heading down at week 12, so there is a significant onset effect with this mechanism." But the jury is still out regarding whether there's "any need or any desire to move this into another RA trial." In any case, Conway said, Array won't be paying for it.
In mid-July, the company filed an investigational new drug application with the FDA to begin a Phase I trial in cancer.
The campaign with MEK inhibitors so far has yielded partnerships for ARRY-886 and ARRY-704, both the subject of a tie-up with London-based AstraZeneca plc. ARRY-886, dubbed AZD6244, has reached Phase II trials. Separate collaborations in other areas exist with South San Francisco-based Genentech Inc. (now Roche), Celgene Corp., of Summit, N.J., and InterMune Inc., of Brisbane, Calif.
Last month, Array disclosed positive top-line data from a Phase I, single ascending-dose study with ARRY-403, a small-molecule glucokinase activator, in Type II diabetes. Multiple ascending-dose data are due later this year, which could prompt an early out-licensing deal, in the view of Piper Jaffray analyst Edward A. Tenthoff.