Aerie Pharmaceuticals Inc.'s stock took a dizzying hit on word of its phase III failure in Rocket 1 with once-daily eye drop Rhopressa for intra-ocular pressure (IOP) in glaucoma, while shares of another player in the space, Inotek Pharmaceuticals Inc., added value in Wall Street's eyes.

The miss could mean a change in protocol with ongoing Rocket 2. It's designed similarly, with a secondary endpoint that – if results line up with Rocket 1 – the trial should hit. "As you can imagine, we're going to be spending some time here, from a regulatory point of view, assessing whether it makes sense for us to go start talking to the agency about taking a look at switching the endpoints here, and using the secondary endpoint as our primary one," Aerie CEO Vicente Anido told investors during a conference call. "All we've reported so far is that we've completed enrollment of that clinical trial. We have not locked the database, and so we haven't seen any of the data at this point. If we're going to make some changes, we'll do so before all those things happen, and we'll do that in concert with the FDA. We're planning ahead."

Research Triangle Park, N.C.-based Aerie (NASDAQ:AERI) closed Friday at $12.87, down $22.52, or 63.6 percent after investors learned the study called Rocket 1 missed its primary efficacy endpoint of noninferiority in lowering IOP as compared to twice-daily timolol, the most widely used comparator in registration experiments for glaucoma therapies. Inotek, of Lexington, Mass. (NASDAQ:ITEK) ended the day at $5.61, up 86 cents, or 18 percent.

Officials at Aerie said Rhopressa, which inhibits Rho kinase (ROCK) and norepinephrine transporter, did show noninferiority compared to the beta-blocking timolol for patients in the study with IOP below 26 mm of mercury (mmHg) at all nine measured time points, and numerical superiority over timolol at the majority of measured time points. About 80 percent of glaucoma patients arrive in ophthalmologists' offices with IOP of 26 mmHg or less at the time of diagnosis, Aerie noted.

Specifically, Rhopressa did not meet the primary efficacy endpoint of demonstrating noninferiority of IOP lowering compared to twice-daily timolol, based upon IOP measurements at the end of the second and sixth weeks, and day 90. Rocket 1 included 182 patients in the Rhopressa once-daily arm and 188 patients in the timolol twice-daily arm, with baseline IOPs tested in the study ranging from above 20 mmHg to below 27 mmHg.

The results showed a slight loss of efficacy in the sixth week and day 90 measurements, Aerie said, and across the full study, 36 patients or about 20 percent, showed signs of loss of efficacy during the study. Hyperemia was the primary adverse event, experienced by about 35 percent of the Rhopressa patients, of which 80 percent was reported as mild.

Along with the finding related to patients with IOP of 26 mmHg or less, Rhopressa showed, in a pre-specified analysis, noninferiority to timolol; it showed numerical superiority over timolol at all nine measured time points for patients in the study with IOPs below 24 mmHg. What's more, the number of Rhopressa patients who experienced loss of efficacy over time was reduced to half for patients with IOPs less than 26 mmHg, and the frequency of such occurrence progressively decreased for Rhopressa patients with lower IOPs.

The secondary endpoint in Rocket 1 and Rocket 2 measures efficacy in patients above and below 24 mmHg, so that if the secondary is switched with the primary endpoint in Rocket 2, Aerie could be more likely to score a win. Rocket 2, which also includes a twice-daily treatment arm, is due for an efficacy readout in the third quarter of this year, with long-term safety data likely near the end of the calendar year. "We will continue to do quite a bit of work and make sure we understand all the data [from Rocket 1]," Anido said. "We won't get the final data for another week or two."

NON-COMPLIANT 'DRIFTERS'?

Promise appeared strong for Rhopressa, also known as AR-13324, the enzymatic conversion of which produces two separate molecules, one of which estimated 10 to 160 times more potent at inhibiting ROCK than others. The drug also works against a secondary target, protein kinase C, known to act in parallel with ROCK.

Inotek, which completed a $40 million IPO in February, has lead candidate trabodenoson (formerly INO-8875) poised to enter phase III development as a monotherapy in the middle of this year. The selective adenosine mimetic was rationally designed to lower IOP by restoring the eye's natural pressure control mechanism, selectively stimulating a particular adenosine subreceptor in the eye. As a result, the intrinsic function is augmentation of the eye's trabecular meshwork (TM), which regulates the pressure inside the eye and is the main outflow path for fluid inside the eye that often builds up pressure in patients with glaucoma. (See BioWorld Today, Nov. 6, 2014.)

By restoring the natural function of the TM and the outflow path, Inotek contended – rather than changing the fundamental dynamics of pressure regulation in the eye – trabodenoson's mechanism of action should result in a lower risk of unintended side effects and improve long-term safety. The way the drug works in TM also may complement the activity of existing glaucoma therapies, and Inotek is developing a fixed-dose combination of trabodenoson with much-prescribed prostaglandin analogue Xalatan (latanoprost, Pfizer Inc.).

Piper Jaffray analyst Joshua Schimmer said that "while many investors had believed that Rhopressa was high probability for success whereas Inotek's trabodenoson [has a] low probability for success, we've wondered whether the differences between the two drugs was as much a reflection of trial design as opposed to true efficacy. We still think there is a path forward for Rhopressa, in the same way that we think there is a path forward for trabodenoson." But, with Aerie's Rhopressa blowup, Inotek has gained traction "and can also learn from the Aerie failure to better optimize its own prospects for success," Schimmer wrote in research report.

"As investors are learning, these ophthalmology trials are fraught with noise and noncompliance, making it challenging to hit primary endpoints," Schimmer pointed out. "However, the FDA appears to recognize this and seems highly flexible with the drugs it will approve despite non-optimal efficacy findings." That, he said, is what happened with Rescula (unoprostone isopropyl ophthalmic solution) from Sucampo Pharmaceuticals Inc., of Bethesda, Md., which in late 2012 won FDA approval of an supplemental new drug application for lowering IOP in patients with open-angle glaucoma or ocular hypertension. Rescula is a BK (big potassium) channel activator, different from other IOP-lowering agents.

Aerie's Anido said that, in Rocket 1, some of the "drifters" – patients who ended up not showing benefit – "spiked tremendously, back up to nonmedicated levels," which suggests they might have been noncompliant but stayed in the study. "Some of these patients at the higher IOPs could actually have had glaucoma for quite a while," and thus had more fibrosis, making them tougher to treat, he said.

Anido cited a phase IIb trial in which Rhopressa did well, and another study that involved Aerie's quadruple-acting combination drug Roclatan for IOP in which Rhopressa "did even better." In Rocket 1, "it just didn't do quite what we expected it to do," he said. (See BioWorld Today, June 26, 2014.)

The New York law firm of Bernstein Liebhard LLP said Friday it's investigating whether Aerie issued materially false and misleading statements to investors in violation of the federal securities laws, or the board breached fiduciary duties.