乌鸦传媒

WASHINGTON -- An intraocular device for presbyopia met near-unanimous rejection from an FDA advisory panel.

By a 15-1 vote, the Ophthalmic Devices Panel of the Medical Devices Advisory Committee said the totality of evidence did not support a conclusion that the benefits of the VisAbility Micro Insert outweigh the risks.

No more than two panelists voted "yes" for any of three key questions related to the device's safety and efficacy. The panel found that the device, at best, provided modest benefit with a modest risk, all of which had to be considered within the context of an elective surgical procedure. Some panelists faulted the design of the pivotal trial for failing to provide meaningful data to assess the device's efficacy.

"There is a fundamental, fatal design problem," said Stephen D. McLeod, MD, of the University of California San Francisco. "You have essentially a study that is set up for improvement measurements in something that is unmasked, nonrandomized, and subjective. You're enrolling based on meeting the criteria and of not being able to read at a certain level, and then coming back for a whole series of visits, multiple visits over time, where you're either expecting to improve or to stay the same."

"The examiners have your prior data sitting in front of them," he said. "I mean, it's just that there's no credibility to it whatsoever."

At FDA request, the sponsor, Refocus Group, used pachymetry to identify patients who appeared to derive greater benefit from the device, said panelist Marian S. Macsai-Kaplan, MD, of the University of Chicago Pritzker School of Medicine. The data showed that the best results occurred within a small pachymetry range, but the sponsor did not specify that range in its proposed indication, she said.

"I have great concerns about efficacy, that when you move out of that range of pachymetry or a thicker, more stiff sclera, you may not get the same outcome. On that basis, I feel that the efficacy is in question," Macsai-Kaplan said.

Sami S. Dahr, MD, of the University of Oklahoma in Oklahoma City, identified himself as the lone panelist who voted in the affirmative for all three questions.

"I don't think a dataset is ever complete or perfect," he said. "The dilemma here is that surgical interventions for presbyopia are going to be low benefit. I think here the benefits were small, the risks were small. I gave a slight edge to the benefits."

Even though the pivotal trial failed to meet one of its co-primary endpoints (missing a confidence interval by 0.5%), the data were sufficient to sway Ronald D. Hays, PhD, of UCLA Fielding School of Public Health. He agreed with Dahr that the device's effectiveness had been demonstrated but voted "no" on the questions of "reasonable assurance of safety" and "totality of evidence."

Pivotal Clinical Trial

In the reports submitted to the Ophthalmic Devices Panel, Refocus and FDA staff described the VisAbility Micro Insert System as a bilateral scleral implant system to improve unaided near vision in patients ages 45 to 60. Candidates for the device should have a minimum near correction of at least +1.25 diopter reading add.

Refocus scientists hypothesize that the inserts "gently tent the scleral overlying the ciliary body," and that this action "expands the circumlenticular space, tightening the zonular fibers that have become lax due to the age-related increase in the diameter of the lens."

The system consists of a main body with two legs and a locking segment to prevent device displacement or migration. During implantation, four inserts are placed in a single eye.

Principal supporting data came from the prospective, multicenter VIS-2014 trial, which did not have a control group. The study involved a total of 396 patients, 336 of whom underwent implantation without randomization. The remaining 60 patients were randomized at three study sites to have either immediate or delayed surgery. Patients randomized to delayed surgery were followed for 6 months, at which time they were eligible for implantation of the device.

Eligibility criteria included distance-corrected near visual acuity (DCNVA) and uncorrected near visual acuity of 20/50, 20/63, or 20/80 in each eye. Additionally, eligible patients had preoperative manifest refraction spherical equivalent in each of -0.75 to +0.50 diopters and no more than 1.00 diopter of astigmatism.

The primary endpoint was DCNVA of 20/40 or better plus at least 10 letters improvement in DCNVA in the primary eye, defined as the dominant eye and implanted first.

For the trial to be considered successful, the results had to satisfy two statistical objectives related to the endpoint: both components of the endpoint were achieved in 75% of the primary eyes at 12 months, with a 75% lower limit for confidence intervals (CIs); and a statistically significant (P<0.025) difference in the primary endpoint in favor of immediate surgery.

Key Findings

The 12-month data showed that 79.1% of primary eyes met the primary endpoint of DCNVA 20/40 plus 10-letter improvement on a vision chart (responders). However, the lower boundary of the 95% CI was 74.5%, missing the prespecified value of 75%.

Additionally, 77.8% of fellow eyes met the response criteria, associated with a lower CI of 73%. When the results for the primary and fellow eyes were combined, the response rate was 78.5%, associated with a lower 95% CI of 74.2%.

The investigators performed a post-hoc analysis of a bilateral effectiveness cohort: results in all primary eyes of patients who underwent bilateral implantation as indicated for the procedure. The analysis included 340 primary eyes and showed a response rate of 80.9%, associated with a lower 95% CI of 76.3%.

During discussions with Refocus Group, the FDA requested a separate analysis of results at 6 months for the 60-patient randomized subgroup, with a specific method of imputation for patients who had missing data (co-primary endpoint). That analysis showed a significant difference in response rate in favor of immediate surgery, regardless of the method of imputation (P<0.001).

Follow-up in the trial continued for 24 months. A post-hoc analysis showed that 289 of 344 (84.1%) primary eyes met response criteria at 24 months. The lower limit of the CIs exceeded 75%, making the result statistically significant.

The trial did not have a specified primary safety endpoint. Investigators tracked and tabulated a variety of adverse events, including surgical complications. During follow-up, 23 patients had a total of 28 secondary surgical interventions involving 28 eyes. Two patients underwent implant removal the same day as surgery because of abnormal pupillary responses. Subsequently, an additional six patients had all implant segments removed, involving 11 eyes.

During discussions, several panelists raised questions about the improvement in response from 12 to 24 months. Some panelists noted that the favorable outcome appeared to be driven by exceptional results at a handful of participating centers, leading them to question the generalizability of the trial results to clinical practice.

Looking Ahead

Patients and ophthalmologists have few options for presbyopia in the absence of cataracts, said Sidney Gicheru, MD, of LaserCare Eye Center in Dallas, who was not involved in the FDA meeting.

"I think if the device can be shown to be safe, effective, and become FDA approved, it will be another tool to help patients over 40 with their presbyopic symptoms," Gicheru, an American Academy of Ophthalmology clinical spokesperson, told 乌鸦传媒 via email. "On a personal note, I am over the age of 40 and presbyopic, so the prospect of this device excites me."

Middle-age patients rarely have isolated presbyopia, which could be beneficial for the implants if they receive FDA approval, he noted.

"Unfortunately, most of the patients with presbyopia that we see also have myopia, hyperopia, and astigmatism," said Gicheru. "It is the rare patient who presents solely for correction of presbyopia and is naturally emmetropic. Those patients rarely show up to a refractive practice. Certainly, if the procedure could be paired with other laser vision correction procedures such as LASIK, SMILE, or PRK, the number of treatable patients would increase."

The FDA is not bound by advisory committee recommendations, but follows the recommendations in most cases.

Comment