Login to Your Account

Alkermes’ SPCD Clinical Magic: Right for Euthymics?


By Randy Osborne
Staff Writer

Alkermes plc’s plan, unveiled this summer, to “retool” its opioid drug R&D got its first vote of confidence from the FDA, which is letting the company forgo the customary end-of-Phase-II meeting and move directly into Phase III trials with Alks 5461 for refractory major depressive disorder (MDD).

The Dublin-based firm in September outlined its goals and methods for the late-stage experiments in a briefing document to the FDA. Included were the usual: primary endpoint, statistical methodology, strategies to meet preclinical and clinical requirements for the new drug application, and the confirmatory study plans.

But the most important aspect of the blueprint is the use of study designs that make use of sequential parallel comparison design (SPCD). Maurizio Fava and David Schoenfeld of Massachusetts General Hospital (MGH) came up with SPCD in 2003.

“Basically we’ve been running clinical trials the way they were designed decades ago, always in the same way, with great inefficiency,” Fava told BioWorld Today. “Too many patients entering trials were placebo responders,” he said.

The SPCD route boosts signal detection by incorporating at least two stages of treatment, and using the results data twice from those patients who are placebo non-responders, thus reducing placebo response. Fewer trial participants – 20 percent to 50 percent fewer – can be enrolled without reducing statistical power.

“Typically, the average resistant population [who respond to neither active drug nor placebo] is about 50 percent of the sample,” Fava said. “The average placebo response rate, in a population for example of depression, is 40 percent. So, if you put together the 50 percent resistant, and the 40 percent who respond to anything you give them, you’re left with 10 percent of informative patients.”

Not very good odds. On the other hand, with SPCD, “if you let the cards fall where they fall in the first phase, you can have a placebo response that can be excessive, just right or low, it doesn’t matter,” he said. “But then, you re-randomize patients who have not responded to placebo to either stay on placebo or go on active [drug]. At this point, the odds of responding to placebo if you failed placebo are much lower. So, even if you had 40 percent [placebo response] in the first phase, but then you had 10 percent in the second phase, your average placebo response rate will be 25 percent.”

Alkermes “has certainly done a great job in showing that the SPCD is a very sensitive measure,” said Anthony McKinney, president and CEO of Euthymics Bioscience Inc., the first to use SPCD in a pivotal trial. In the Phase II study by Alkermes, also an SPCD experiment, the overall placebo response rate added up to a measly 15 percent, which is “superb, considering that most depression trials have well over 50 percent placebo response,” he told BioWorld Today. “Somewhere between 35 and 50 percent is expected.”

‘The System Will Re-adjust’

Cambridge, Mass.-based Euthymics’ 342-patient, Phase IIb/IIIa study tested its lead candidate amitifadine (formerly EB-1010), a serotonin-preferring, triple-reuptake inhibitor for MDD. Called TRIADE (Triple Reuptake Inhibitor Anti-Depressant Effects), the trial randomized 342 MDD patients at 39 centers in the U.S. to compare amitifadine, paroxetine and placebo in MDD patients failing one full course of therapy with typically given drugs.

Top-line data reported in May showed amitifadine efficacy at 50-mg or 100-mg doses did not show a statistically significant difference from placebo in the primary endpoint of a change as measured by the Montgomery-Åsberg Depression Rating Scale). The active comparator arm showed the study itself did work, and the dose range for amitifadine may have been too low.

Euthymics is putting together another, higher-dose trial, which also will use the SPCD model. Meanwhile, a scientific journal will publish full results, which McKinney said were pleasing with regard to the drug’s lack of effect on sexual function, and tendency to reduce alcohol consumption in patients, probably because of dopamine-related action. “Obviously, we’re going to be taking every advantage of [these findings] in future trials,” he said.

MGH owns several patents related to SPCD. The hospital’s exclusive licensee, RCT Logic LLC, of Jacksonville Beach, Fla., offers licenses under the issued patents to companies planning studies. More than a dozen other SPCD-based trials – three of them funded by the National Institutes of Health – are finished, ongoing or starting shortly, according to RCT.

Alkermes’ drug would be an add-on to existing therapy, McKinney noted, whereas Euthymics’ therapy would be a switch. Alkermes said the new mechanism of action of Alks 5461, designed to be non-addictive, is based on a balanced combination of agonism and antagonism of opioid receptors. In July, the firm cited Alks 5461 – a fixed combination of buprenorphine, a partial agonist, and Alks 33, a mu-opioid antagonist – as an example of the way the company is “rethinking psychiatry.” (See BioWorld Today, July 24, 2013.)

But it’s the new mechanism of putting trials together that may end up getting just as much attention, even if contract research organizations might be less than enthused about putting together smaller, less lucrative studies.

“No one, the sponsor or the CRO or the FDA, benefits when the trial blows up,” McKinney said. “One out of two depression trials fail. No one’s happy in that situation.”

Clinical research, he noted, “is generally not the place where you use the most cutting-edge designs, because if you make a mistake, people lose their jobs. Once the FDA has agreed that the novel concept makes sense, that’s different. The endpoints, the statistical analyses, everything is buttoned up.” McKinney predicted the industry will see “a lot more people jumping” into the SPCD pool.

Fava, not speaking for RCT, MGH or any company, believes the same, and likened the situation to the wider acceptance of the hybrid, gas-saving car. “At the end of the day, the system will re-adjust,” he said.