Purpose
To evaluate the safety and intraocular pressure (IOP)-lowering efficacy of the ocular pressure adjusting pump in subjects with normal-tension glaucoma (NTG).
Design
Prospective, multicenter, masked, randomized, fellow-eye controlled trial.
Subjects, Participants, and/or Controls
Subjects with NTG with an IOP ≥12 mm Hg and ≤21 mm Hg were enrolled. One eye of each subject was randomized to receive negative pressure application; the fellow eye served as a control.
Methods
Subjects wore the device overnight for 1 year and the applied negative pressure was programmed by subtracting a reference IOP of 6 mm Hg from the baseline IOP.
Main Outcome Measures
The primary effectiveness endpoint was the proportion of eyes achieving an IOP reduction ≥20% at Week 52 during the day. The secondary endpoint was the proportion of eyes achieving a nocturnal IOP reduction ≥20% at Week 52. Exploratory endpoints included mean IOP reduction in clinic and in the sleep lab.
Results
A total of 186 eyes were randomized across 11 sites. 120 eyes successfully completed all visits across 52 weeks without protocol deviations. At Week 52, 88.3% ( n = 53) of study eyes vs 1.7% ( n = 1) of control eyes met the primary endpoint. For the secondary endpoint, 96.7% ( n = 58) of study eyes vs 5.0% ( n = 3) met the endpoint. For exploratory IOP analysis, the mean nocturnal IOP reduction at Week 52 was 8.0 mm Hg (39.1%) from a baseline of 20.4 ± 2.5 mm Hg to 12.4 ± 2.7 mm Hg. There were no serious adverse events. The most commonly reported adverse events were lid (11.8% study, 1.1% control) and periorbital edema (12.9%, 1.1%).
Conclusions
The ocular pressure adjusting pump safely and effectively lowers both daytime and nocturnal IOP in patients with NTG.
INTRODUCTION
Glaucoma is the leading cause of irreversible blindness. Intraocular pressure (IOP) remains the only clinically-validated and modifiable risk factor and thus, all current treatments aim to lower IOP to prevent further vision loss. Current treatment methods include topical medications, laser procedures, and a wide range of surgical treatments. All current FDA-approved therapies, particularly those that are less invasive, remain less effective with a lower baseline IOP and many of our treatment options do not effectively treat nocturnal IOP elevations. ,
Data from studies evaluating measurements over a 24-hour period demonstrate that greater than two-thirds of patients experience their peak IOP outside typical clinic hours with this largely occurring at night. Further, most commonly prescribed topical options have proven daytime efficacy but have minimal or reduced effect on nocturnal IOP. Recent studies evaluating selective laser trabeculoplasty have shown reduced IOP when averaged over 24 hours but the procedure does not impact the 24-hour rhythm and presence of nocturnal IOP peaks. , To date, the only treatment option that has been demonstrated to provide sustained and consistent 24-hour control of IOP is trabeculectomy. ,
Normal-tension glaucoma (NTG) is a common subset of OAG characterized by a measured IOP ≤21 mm Hg representing about 30% of glaucoma in the US and 70% to 90% of glaucoma in East Asia. Lowering of IOP in NTG is more challenging than OAG and often necessitates more aggressive subconjunctival surgery which is associated with high morbidity and high failure rates. , Further, nocturnal IOP elevations are more prevalent in patients with NTG. A recent joint paper by the American Glaucoma Society and American Society of Cataract and Refractive Surgeons emphasized the need for additional treatments stating that (1) 24-hour IOP monitoring/control and (2) noninvasive therapeutics that lower IOP and improve ocular blood flow were unmet needs, “especially in challenging patients who do not adequately respond to current therapies or those in whom IOP is already within the normal range.”
The FYSX ocular pressure adjusting pump (Balance Ophthalmics) is a novel treatment device that consists of a pair of goggles connected to a pressure-modulating pump and represents the first FDA-approved nonsurgical and nonpharmaceutical IOP-lowering therapy. , When the ocular pressure adjusting pump is worn, negative pressure (or vacuum) ranging from −5 to −20 mm Hg is applied over the eye to create a localized decrease in atmospheric pressure on the eye, leading to a corresponding decrease in IOP. When the device is worn with applied negative pressure, the IOP reduction is sustained throughout wear. The design of the ocular pressure adjusting pump allows for individualized and titratable negative pressure application to each eye. Multiple randomized, controlled studies and other clinical studies have been performed highlighting the safety, tolerability, and IOP-lowering ability of the device. , , ,
This present study was the pivotal trial designed to evaluate the safety and IOP-lowering effectiveness of the ocular pressure adjusting pump and achieved FDA approval.
METHODS
STUDY DESIGN
The HERCULES trial is a prospective, multicenter, randomized, controlled study and was performed at multiple sites across the US. The study procedures were in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standard. All subjects included in the study provided informed consent prior to the beginning of the study. This study was approved by the Advarra Institutional Review Board (IRB). This study is registered with the US National Library of Medicine ( http://www.clinicaltrials.gov identifier NCT04236869) and the study period was from January 2020 to October 2022. Subjects who met eligibility were randomized to receive application of negative pressure in one eye while the contralateral eye served as a control (0 mm Hg negative pressure). All research personnel assessing effectiveness were masked to treatment assignment and masked to IOP measurement values throughout the entire course of the study.
ELIGIBILITY CRITERIA
The key inclusion criteria were subjects aged ≥40, a confirmed diagnosis of NTG in both eyes confirmed by glaucomatous optic nerve head or retinal nerve fiber layer (RNFL) structural abnormalities and/or visual field (VF) abnormalities (based on threshold VFs performed within 60 days prior to Visit 1). Subjects were required to have no documented unmedicated IOP measurements >21 mm Hg in either eye and a baseline IOP ≥12 and ≤21 mm Hg. In subjects using ocular hypotensive medications, a minimum 30-day washout period was required to ensure that the unmedicated IOP was ≤21 mm Hg. Following the washout period, subjects resumed their medications. Subjects had to demonstrate the ability to achieve a seal and successfully average ≥3 hours of sleepwear with the device during at least 3 nights of a consecutive 7-day run-in period.
The key exclusion criteria included the history of an eye condition that could limit evaluation of the study results or compromise patient safety, retinal detachment history, fundus findings that could limit visualization of the posterior segment, eyelid edema, conjunctival chemosis, history of corneal transplant, allergy to silicone or history of a filtering procedure (eg, tube shunt or trabeculectomy).
STUDY DEVICE
The ocular pressure adjusting pump consists of a right and left pressure-sensing goggle and a programmable, pressure-modifying pump ( Figure 1 ). Each goggle chamber is connected via dual-lumen tubing to the negative pressure pump that allows for individualized pressure control over each eye and allows for instantaneous control of negative pressure within each goggle.
This study included a specially adapted version of the ocular pressure adjusting pump device ( Figure 2 ), known as the Excursion ocular pressure adjusting pump. The excursion iteration of the device is identical to the standard version but is adapted with an access port on each side to facilitate IOP measurements while the device is active with negative pressure application. This device and the corresponding IOP measurement method have been described in detail by multiple previous studies and validated as a form of reliable, repeatable, and accurate IOP measurement. ,
IOP MEASUREMENTS
Subjects underwent a series of IOP measurements at each visit and included both GAT and Model 30 pneumatonometry measurements prior to wear of the goggles. The IOP value measured with the M30 pneumatonometer served as the basis for negative pressure programming since the M30 pneumatonometer was also the device used as part of the excursion method of IOP measurement when the goggles were worn. The IOP measurement obtained was used to determine the negative pressure setting by subtracting a reference IOP of 6 mm Hg. For example, if a subject had a measured IOP of 16 mm Hg, the negative pressure programmed into the pump was −10 mm Hg (16-6 mm Hg = 10 mm Hg). This programmed negative pressure was used for subsequent home use after the initial visit. While programming was intended to remain constant throughout the course of the study, investigators were given discretion to adjust study eye negative pressure setting for subsequent home use based on subject feedback.
For each set of IOP measurements, the IOP value for the eye was calculated as the average of two measurements for that eye or, if the two measurements differed by more than 2 mm Hg, a third measurement was taken and the median of three measurements of the eye was used. At each visit, IOP measurements were obtained prior to and during wear of the goggles to determine a baseline IOP for each. Measurements were taken with the Excursion goggles on prior to negative pressure application. The goggles were then activated with negative pressure to the setting used during home use and measurements were repeated. All IOP measurements were obtained in an assessor-masked fashion to prevent bias. Further, the study personnel obtaining IOP measurements were masked to treatment assignment. Two study personnel obtained the IOP measurement with the Model 30 pneumatonometer. One obtained the measurement, while the other read the measurement from the Model 30 to eliminate the potential for bias in IOP measurements.
VISIT SCHEDULE
Following randomization, subjects were instructed to use the device nightly and returned for 5 in-clinic visits at weeks 6, 12, 26, 38, and 52 and 2 overnight sleep lab visits at Weeks 0 to 3 and Weeks 49 to 52. The device was programmed to apply negative pressure to the study eye and the control eye served as a control with no applied negative pressure (0 mm Hg). At each visit, at-home data usage from the ocular pressure adjusting pump was downloaded and IOP measurements and safety assessments were performed.
Within the first 21 days following randomization, subjects came in for the first 8-hour sleep lab visit to evaluate changes in nocturnal IOP during use of the ocular pressure adjusting pump. During the sleep lab visit, IOP measurements were obtained in the supine position and measurements occurred at 11:00 pm, 2:00 am, and 5:00 am. At the initial sleep lab, the baseline IOP was measured in the supine position; if the baseline IOP differed from the measured in-clinic IOP, the ocular pressure adjusting pump was re-programmed for subsequent use based on the supine IOP measurement Near the conclusion of the study, in the weeks preceding the 52-week in-clinic visit, subjects repeated the 8-hour sleep lab visit (Week 52 sleep lab) with the same approach.
SAFETY AND EFFECTIVENESS ENDPOINTS
The primary effectiveness endpoint was the proportion of study eyes with a Week 52 in-clinic IOP reduction ≥20% during negative pressure application in comparison to baseline. The secondary effectiveness endpoint was the proportion of study eyes with a Week 52 sleep lab IOP reduction ≥20% during application of negative pressure as compared with a baseline IOP measured in the supine position.
In addition to the primary and secondary endpoints, another prespecified, exploratory endpoint was the percentage change in IOP during negative pressure application both in-clinic and in the sleep lab in comparison with baseline IOP.
Prespecified safety outcomes consisted of ocular and periocular adverse events (AEs), nonocular AEs, changes in best-corrected distance visual acuity (BCDVA), and IOP elevations greater than 10 mm Hg after completion of negative pressure application. VF data were also evaluated which included changes in mean deviation (MD) and pattern standard deviation. Imaging evaluating changes in RNFL thickness using optical coherence tomography (OCT) was also evaluated as a safety parameter. VF and OCT data were not prespecified statistical endpoints consistent with the standards for other glaucoma device and drug trials. The University of Iowa VF reading center analyzed VFs and OCT imaging in a post-hoc, masked fashion.
STATISTICAL ANALYSIS
The sample size justification was based on a prior study and was based on the primary effectiveness endpoint and adjusted for the secondary effectiveness endpoint. Based on McNemar’s exact conditional test with a two-sided significance level of 0.05 for paired nominal data with a correlation of ≤0.45, a sample size of 50 subjects at 52 weeks post-treatment would provide a statistical power of at least 92% approximately to demonstrate superiority of treated eyes over control eyes that achieve ≥20% IOP reduction.
The McNemar test was used to compare the percent of eyes between the treatment and control groups with IOP reduction ≥20% for the in-clinic and sleep lab IOP outcomes separately. For the sleep lab, the McNemar test performed was to be concluded for the treatment effect only if the test for in-clinic outcomes was statistically significant. A significance level of 0.05 of was set. For continuous variables, the mean and standard deviation measurements are presented. Counts and percentages by treatment group are presented for categorical variables. All reported covariate analyses were prespecified.
RESULTS
BASELINE CHARACTERISTICS AND STUDY POPULATION
A total of 186 eyes (one study eye and the contralateral control eye of each subject) from 93 subjects were randomized across 11 investigational sites. A total of 33 subjects failed to complete all required study visits ( n = 31) or had a protocol deviation ( n = 2). The most frequent reason for study discontinuation was withdrawal of consent ( n = 20). Of the 20 that withdrew consent, 5 reported issues related to comfort of the goggles and 2 withdrew due to poor fit. The remaining withdrew consent due to miscellaneous reasons including but not limited to: time constraints, unrelated health conditions, and COVID-19 diagnoses. Separate from withdrawal of consent, 5 subjects discontinued due to noncompliance with at-home use requirements and 6 subjects discontinued due to additional miscellaneous reasons including closure of a study site’s sleep lab due to COVID-19, loss to follow-up, and 2 subjects with possible allergic reaction to silicone in the goggles. Overall, 120 eyes completed all visits across 52 weeks and had no protocol deviations. As mentioned, investigators were given discretion to adjust negative pressure settings in study eye based on subject feedback; 26 eyes were adjusted with 17 being adjusted for comfort.
For all 186 eyes from 93 subjects that were randomized, the mean age was 62.4 ± 10.7 years (range 40-85 years). Most subjects were female ( n = 63, 67.7%). For demographics, 68.8% ( n = 64) of patients were white, 14.0% ( n = 13) were African American and 16.1% ( n = 15) were Asian. The baseline IOP as measured via GAT was 14.7 ± 2.0 mm Hg in the study eye and 14.8 ± 2.2 in the control eye. For medication use, 55.9% of study eyes were on at least 1 hypotensive medication, and 53.8% of control eyes were on at least 1 hypotensive medication. Of note, combination medications (dorzolamide-timolol) were recorded as separate medications. There was no statistical difference between the study and control eyes at baseline and the baseline characteristics for the study and control eyes are outlined in Table 1 .
| Characteristic | Study Eye ( n = 93) | Control Eye ( n = 93) |
|---|---|---|
| Topical ocular hypotensive medications | n (%) | n (%) |
| 0 | 41 (44.1%) | 43 (46.2%) |
| 1 | 35 (37.6%) | 35 (37.6%) |
| 2 | 10 (10.8%) | 10 (10.8%) |
| 3 | 5 (5.4%) | 3 (3.2%) |
| 4 | 2 (2.2%) | 2 (2.2%) |
| BCDVA at baseline (LogMAR) | ||
| Mean (Snellen) | 0.06 (20/23.1) | 0.08 (20/23.8) |
| Standard deviation | 0.12 | 0.14 |
| Manifest refraction spherical equivalent | ||
| Mean | −1.0 | −1.4 |
| Standard deviation | 2.5 | 2.7 |
| Baseline IOP (GAT) (mm Hg) | ||
| Mean | 14.7 | 14.8 |
| Standard deviation | 2.0 | 2.2 |
| Vertical cup-to-disc ratio | ||
| Mean | 0.67 | 0.66 |
| Standard deviation | 0.15 | 0.16 |
| Visual field mean deviation (MD) (dB) | ||
| Mean | −4.03 | −3.67 |
| Standard deviation | 4.86 | 4.65 |
EFFECTIVENESS ENDPOINTS
All effectiveness endpoints were met. The primary effectiveness endpoint, which represented the proportion of eyes with an in-clinic IOP reduction ≥20% during negative pressure application in comparison with baseline IOP (measured prior to negative pressure application), was met both in the modified intent-to-treat population which included dropouts as failures, and in the per-protocol population who completed all study visits without protocol deviation. In the per-protocol population, the primary endpoint was met by 88.3% ( n = 53/60) of study eyes vs 1.7% (1/60) of control eyes, which amounted to an 86.7% between-group difference (95% CI: 73.7%, 94.1%, P < .001).
The secondary effectiveness endpoint, which represented the proportion of study eyes with a sleep lab IOP reduction ≥20% during application of negative pressure, was also met by both the modified intent-to-treat population and the per-protocol population. In the per-protocol population, the primary endpoint was met by 96.7% ( n = 58) of study eyes compared to 5.0% ( n = 3) of control eyes. The between-group difference was 91.7% (95% CI: 79.7%, 97.8%, P < .001). The primary and secondary effectiveness endpoint results are shown in Figure 3 .
A tipping point analysis was performed to evaluate the primary endpoint using all scenarios, including the worst-case scenario where all control eyes with missing data were imputed as responders and all study eyes with missing data were imputed as nonresponders. Even in this worst-possible-case scenario, the primary efficacy endpoint was met and remained statistically significant as demonstrated in Figure 4 .
EXPLORATORY IOP ANALYSES
For exploratory endpoints, the prespecified mean IOP reduction was statistically significant for both in-clinic and sleep lab settings at Day 0 (baseline), Week 26, and Week 52. These results are outlined in Table 2 .
| Day 0: In-Clinic | Week 26: In-Clinic | Week 52: In-Clinic | Initial Sleep Lab | Final Sleep Lab | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Study | Control | Study | Control | Study | Control | Study | Control | Study | Control | |
| IOP (excursion tonometry with NP OFF) | ||||||||||
| N | 93 | 93 | 68 | 68 | 61 | 61 | 80 | 80 | 61 | 61 |
| Mean | 16.8 | 16.8 | 17.2 | 17.2 | 18.0 | 17.4 | 20.1 | 18.6 | 20.4 | 19.4 |
| SD | 2.6 | 2.6 | 2.4 | 2.6 | 3.2 | 2.5 | 2.5 | 2.5 | 2.5 | 2.3 |
| IOP (excursion tonometry with NP ON) | ||||||||||
| N | 93 | 93 | 68 | 68 | 61 | 61 | 80 | 80 | 61 | 61 |
| Mean | 10.7 | 16.0 | 10.9 | 16.9 | 11.4 | 16.8 | 12.5 | 16.8 | 12.4 | 17.7 |
| SD | 2.4 | 2.9 | 2.4 | 2.9 | 3.0 | 3.0 | 2.3 | 2.7 | 2.7 | 2.4 |
| Change in IOP from NP OFF to NP ON a | ||||||||||
| Mean | −6.1 | −0.8 | −6.2 | −0.3 | −6.6 | −0.6 | −7.6 | −1.8 | −8.0 | −1.6 |
| SD | 2.5 | 1.7 | 2.5 | 1.9 | 3.1 | 1.6 | 2.2 | 1.6 | 2.5 | 1.4 |
| Percent change in IOP from NP OFF to NP ON a | ||||||||||
| Mean | −35.9% | −4.4% | −35.9% | −1.4% | −36.0% | −3.4% | −37.5% | −9.4% | −39.1% | −8.4% |
| SD | 12.7% | 10.8% | 12.8% | 11.1% | 13.9% | 9.3% | 9.0% | 8.5% | 11.1% | 7.3% |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
