Shares of Acorda Therapeutics Inc. (NASDAQ:ACOR) took a hit Wednesday, falling $11.20 to close at $17, on reports of potentially treatment-related adverse events in its tozadenant program for Parkinson's disease. While reports of agranulocytosis that led to cases of sepsis and death prompted analysts to drop the drug from their models, the Ardsley, N.Y.-based firm has no plans at this time to discontinue work on the phase III oral adenosine A2a receptor antagonist it acquired in last year's buyout of Biotie Therapies Oyj.

Acorda, however, has opted to suspend further enrollment in ongoing studies, while it awaits review from the FDA and the independent data safety monitoring board (DSMB). How that pause might affect the two long-term safety studies – the CL05 extension and the CL06 trial – remains to be seen. But the pivotal CL05 trial is near enough to completion that Acorda still anticipates a sufficient dataset when the study reads out in the first quarter of 2018.

There is only "a small minority of patients left in the phase III double-blind portion" at this point, according to President and CEO Ron Cohen. "Even if tomorrow all the people left [in the trial] dropped out [due to adverse event concerns], we'd still have a good dataset to deal with."

The CL05 study, which has enrolled about 450 patients, according to Cortellis, and is being conducted under an FDA special protocol assessment agreement, is designed to compare tozadenant vs. placebo as an adjunctive therapy in levodopa-treated patients with Parkinson's disease. The primary objective is to demonstrate the effectiveness of tozadenant in patients experiencing end-of-dose wearing-off, as based on the change from baseline to week 24 in the number of hours per day spent in the "off" state.

Should the trial yield positive data, the possibility of continuing toward regulatory submissions isn't out of the question, pending data from the safety studies. From a risk-reward perspective, Cohen told investors on a Wednesday morning conference call, tozadenant is designed to target a "desperately ill population," that continues to experience degeneration as currently available therapies lose their impact over time and for whom surgical options – "which have their own morbidity and mortality issues associated with them," – are often a last resort.

"To the extent you may have a drug that has a different mechanism to anything that's out there, that provides additional relief to patients even in the presence of all those other drugs ... and maybe can be an alternative before they have to consider surgery, that's still something that we believe is worth contemplating," Cohen said. More will become clear once the efficacy profile is available. "We have to see the actual data before we can talk in more specifics."

More work, too, will have to be done to determine whether treatment with tozadenant can be linked definitively to agranulocytosis, a drastic drop in white blood cell count. As analysts attempted to tease out details on Wednesday's call, Cohen was able to provide some additional information but said, as of now, "there's no clear pattern that has yet emerged."

To date, including the previously completed phase IIb program published in The Lancet in 2014, about 890 patients have been exposed to tozadenant in clinical trials; another 234 patients received placebo. That works out to roughly 300 patient years of tozadenant exposure vs. 75 patient years for placebo, Cohen said.

"We've seen seven cases of sepsis in all the tozadenant groups, five of which were fatal," he said. Four of those sepsis cases were associated with agranulocytosis; of the other three cases, two patients did not have white blood cell counts available at the time of sepsis and one had a high white blood cell count.

Three of the sepsis cases had been reported in the phase IIb trial, in which there were a total of six deaths across three of the four drug arms – Cohen noted that no deaths occurred for the 120-mg dose that was used in the CL05 study. "The DSMB did a full investigation and nobody came up with a unifying theme that could explain it all."

Only one of those patients also had agranulocytosis, which investigators at the time attributed to the sepsis diagnosis itself. "It's a chicken-and-egg thing because sepsis itself can lower white blood cell counts," Cohen explained.

In the phase III program, investigators have noted four cases of agranulocytosis, including one patient whose blood cell count was noted before becoming symptomatic.

"The only differentiating feature that is popping out at this point in terms of adverse events is the agranulocytosis and associated morbidity and mortality with respect to sepsis as you might expect," Cohen said. "Once you pull that out, there's nothing there that differentiates placebo from drug."

If the agranulocytosis adverse event is determined to be related to tozadenant treatment, further work will have to be done to determine the rate of incidence. Cohen noted, however, that, "to our knowledge," it was not an AE observed in other studies of similar candidates.

"At this point, we would have to call it idiosyncratic," he said.

Another adenosine A2a receptor antagonist, preladenant, developed by Merck & Co. Inc., was discontinued for lack of efficacy after three failed phase III studies in Parkinson's disease. Meanwhile, Kyowa Hakko Kirin Co. Ltd. is preparing to resubmit a regulatory filing seeking approval of its drug, istradefylline. The compound has been approved for treating motor complications in Parkinson's disease in Japan since 2013. (See BioWorld Today, May 28, 2013.)

Still ahead of the odds

Calling the safety snag a "major surprise," given how soon the study is expected to readout, Leerink Research analyst Paul Matteis dropped the drug from his firm's model for Acorda. "While the program may continue to the top-line efficacy readout in [the first quarter of 2018], we are now very concerned regarding the approvability of toz – even if it does work – given its positioning as a symptomatic therapy with a moderate level of benefit," he wrote in a research note.

Jefferies analyst Michael Yee agreed. "We assume the risk/benefit of this drug is hampered for approval at this point," he wrote.

In his view, JP Morgan analyst Cory Kasimov said his firm already had low expectations for tozadenant, though the latest news "substantially reduces the potential for upside optionality."

Acorda gained tozadenant through its purchase of Finnish company Biotie in a $363 million cash bid, a bargain price given the compound's advanced stage of development. That deal also brought on board SYN-120 a dual serotonin receptor antagonist in phase II development for Parkinson's disease dementia. (See BioWorld Today, Jan. 20, 2016.)

Also in its pipeline, Acorda has Inbrija, an inhaled levodopa candidate for improving motor function in Parkinson's patients. After receiving a surprise refuse-to-file letter in August, the company has been working to address outstanding issues and remains on track to resubmit the NDA this quarter. (See BioWorld, Aug. 30, 2017.)

"We're looking to a very significant drug here," Cohen said of Inbrija. "We've said greater than $500 million. We think it's at least the size of Ampyra."

Multiple sclerosis drug Ampyra (dalfampridine), which gained approval in early 2010, recorded full-year net revenue of $493 million for 2016. In March, the U.S. District Court for the District of Delaware invalidated four patents related to the drug, a decision Acorda has appealed. (See BioWorld Today, Jan. 25, 2010.)

The company also has its Arcus platform technology, which is designed to formulate therapies for inhaled delivery. Inbrija was the first from that program, which was acquired in the 2014 acquisition of Civitas Therapeutics Inc. (See BioWorld Today, Sept. 25, 2014.)

"We're already working on a migraine therapy" using the Arcus delivery, Cohen said. The company also has a "sizable grant" from the Bill & Melinda Gates Foundation to apply Arcus for an inhaled surfactant product to treat children and infants with respiratory problems in developing nations. Based on that work, there's also potential use for an inhaled surfactant in the developed world, he added.

Cohen said he estimates Acorda ending 2017 with more than $200 million on its balance sheet, even taking into account the tozadenant setback, and is set to remain in good financial shape even if the firm should not prevail in the Ampyra patent appeal and generics hit the market at the end of July 2018. "Even with that, worst-case scenario, we wind up with over $200 million at the end of next year," he said. "And that does not include the impact of the royalty sales and asset sales that we're pursuing right now."

Aiming to put the latest news into even larger context, Cohen reminded investors on the call that "over 90 percent of products that hit phase I ultimately fail in our industry, so, so far, even though we've had our setbacks and we've had a number of them, we're still ahead of those odds."