Contemporary Clinical Trials 37 (2014) 294–300
Contents lists available at ScienceDirect
Contemporary Clinical Trials journal homepage: www.elsevier.com/locate/conclintrial
Randomized trial of the ForeseeHome monitoring device for early detection of neovascular age-related macular degeneration. The HOme Monitoring of the Eye (HOME) study design — HOME Study report number 1☆ Emily Y. Chew a,⁎, Traci E. Clemons b, Susan B. Bressler c, Michael J. Elman d, Ronald P. Danis e, Amitha Domalpally e, Jeffrey S. Heier f, Judy E. Kim g, Richard A. Garfinkel h, Appendix 1 for AREDS2-HOME Study Research Group a
National Eye Institute, Bethesda, MD, USA The EMMES Corporation, Rockville, MD, USA c Department of Ophthalmology, Wilmer Eye Institute, Johns Hopkins Hospital, Baltimore, MD, USA d Elman Retina Group, Baltimore, MD, USA e Department of Ophthalmology and Visual Sciences, University of Wisconsin–Madison School of Medicine and Public Health, USA f Ophthalmic Consultant of Boston, Boston, MA, USA g Department of Ophthalmology, Medical College of Wisconsin, Milwaukee, USA h Retina Group of Washington, Bethesda, MD, USA b
a r t i c l e
i n f o
Article history: Received 24 September 2013 Received in revised form 25 January 2014 Accepted 5 February 2014 Available online 13 February 2014 Keywords: Age-related macular degeneration Controlled clinical trial Home-monitoring Choroidal neovascularization Best-corrected visual acuity
a b s t r a c t Objective: To evaluate the effects of a home-monitoring device with tele-monitoring compared with standard care in detection of progression to choroidal neovascularization (CNV) associated with age-related macular degeneration (AMD), the leading cause of blindness in the US. Patients and methods: Participants, aged 55 to 90 years, at high risk of developing CNV associated with AMD were recruited to the HOme Monitoring of Eye (HOME) Study, an unmasked, multi-center, randomized trial of the ForeseeHome (FH) device plus standard care vs. standard care alone. The FH device utilizes preferential hyperacuity perimetry and tele-monitoring to detect changes in vision function associated with development of CNV, potentially prior to symptom and visual acuity loss. After establishing baseline measurements, subsequent changes on follow-up are detected by the device, causing the monitoring center to alert the clinical center to recall participants for an exam. Standard care consists of instructions for self-monitoring visual changes with subsequent self-report to the clinical center. The primary objective of this study is to determine whether home monitoring plus standard care in comparison with standard care alone, results in earlier detection of incident CNV with better present visual acuity. The primary outcome is the decline in visual acuity at CNV diagnosis from baseline. Detection of CNV prior to substantial vision loss is critical as vision outcome
☆ Financial Support: Financial support to the study was provided by the Notal Vision through a clinical trial agreement with the NEI (CTA-00833) and a service agreement with EMMES Corporation. The AREDS2 study is supported by the intramural program funds and contracts from the National Eye Institute/National Institutes of Health (NEI/NIH), Department of Health and Human Services, Bethesda, MD. Contract No. HHS-N-260-2005-00007-C. ADB Contract No. N01-EY-5-0007. Funds were generously contributed to these contracts by the following NIH institutes: Office of Dietary Supplements (ODS), National Center for Complementary and Alternative Medicine (NCCAM), National Institute on Aging (NIA), National Heart, Lung and Blood Institute (NHLBI), and National Institute of Neurological Disorders and Stroke (NINDS). The study medications and raw materials were provided by Alcon, Bausch and Lomb, DSM, and Pfizer. ⁎ Corresponding author at: NIH, Building 10, CRC, Room 3-2531, 10 Center Drive, MSC-1204, Bethesda, MD 20892-1204, United States. Tel.: +1 301 496 6583; fax: +1 301 496 7295. E-mail address: [email protected] (E.Y. Chew).
http://dx.doi.org/10.1016/j.cct.2014.02.003 1551-7144/Published by Elsevier Inc.
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
295
following anti-angiogenic therapy is dependent on the visual acuity at initiation of treatment. Discussion: HOME Study is the first large scale study to test the use of home tele-monitoring system in the management of AMD patients. Published by Elsevier Inc.
1. Introduction Age-related macular degeneration (AMD) is the leading cause of irreversible blindness in the United States [1,2] and in populations of northern European descent in the developed world [3]. Blindness in advanced AMD occurs from the development of choroidal neovascularization (CNV) and/or geographic atrophy affecting the center of the fovea. While CNV occurs in only 10%–15% of all persons with AMD, it accounts for more than 80% of cases of severe central vision loss associated with AMD [4]. Current therapy for neovascular AMD, intravitreal injections of anti-vascular endothelial growth factor (VEGF) agents, such as bevacizumab, ranibizumab and aflibercept, can potentially stabilize visual acuity in 95% of eyes [5–10]. The main predictor for visual acuity outcome following anti-VEGF treatment is the visual acuity at the time of initiation of CNV therapy [11–13]. Patients with worse visual acuity at initiation of treatment were more likely to gain more letters of vision but their final visual acuity tended to be worse than those who started with good visual acuity and gained fewer letters following treatment [11–13]. Current monitoring of patients at risk for CNV development consists of periodic in-office examinations at varying intervals by ophthalmologists and optometrists with the recommendation for self-monitoring of visual function including either new onset or changes in existing symptoms. This may include the use of an Amsler grid. Such monitoring strategies yielded visual acuity of 20/40 or better at the initial detection of CNV in only 13 to 36% of new onset CNV [11,14]. Increasing time interval between onset of symptoms and initiation of treatment resulted in markedly decreased visual acuities both at presentation of CNV and following anti-VEGF treatment [15]. Detection of the CNV prior to significant vision loss is essential to maximize successful visual outcomes following treatment. This report describes the design of the HOme Monitoring of the Eye (HOME) Study, a Phase 3, unmasked, randomized trial evaluating the role of home monitoring with the ForeseeHome (FH) device plus standard care (referred from here as FH monitoring) compared to standard care alone for eyes at risk of developing CNV. Study subjects were enrolled in 44 retina clinics participating in the Age-Related Eye Disease Study 2 (AREDS2), a randomized trial of nutrient supplements. The ForeseeHome monitoring device utilized macular visual field testing and tele-monitoring. The basis for this test is preferential hyperacuity (or vernier acuity) perimetry which potentially detects the earliest functional abnormalities associated with CNV, prior to patient's awareness of symptoms or visual acuity changes. Such changes transmitted to the monitoring center triggered alerts to the physicians about the potential early development of CNV. Participants were then contacted for exams. This strategy, paired with standard care was compared with standard care alone, which consisted of instructions to regular monocular self-monitoring for visual changes with subsequent self-report to the physician. The hypothesis tested
was that home monitoring would detect the onset of neovascular AMD earlier with better visual acuity (a smaller decrease in visual acuity compared with baseline) at CNV detection compared with the standard of care alone arm. The study was registered at ClinicalTrials.gov with the identification NCT01103505. 2. Research design and methods The HOME Study, an ancillary study of the Age-Related Eye Disease Study 2 (AREDS2), a long-term multicenter randomized controlled clinical trial of oral supplements, was an unmasked, randomized clinical study that enrolled both AREDS2 and non-AREDS2 participants. They were randomly assigned, 1:1, to either the FH monitoring plus standard care or the standard care alone. The Institutional Review Boards for human subject research from the individual clinical sites approved the study protocol and all participants signed informed consents. 2.1. ForeseeHome monitoring device The ForeseeHome (FH) monitoring device (Notal Vision, Ltd. Tel Aviv, Israel), cleared by the FDA in 2009, was developed to detect visual field disturbances associated with neovascular AMD prior to overt visual symptoms. This technology is based on hyperacuity (known also as vernier acuity) – the ability to perceive a very small difference in the relative spatial localization of two or more visual stimuli – that can be locally impaired before visual acuity is affected [16,13]. Early CNV causes subtle separation of the different layers of the retina, resulting in localized distortion of vision (metamorphopsia), often with accompanying loss of sensitivity (scotoma). The FH tests the central area of the visual field of 14° with approximately 500 data points sampled rapidly 3 to 5 times. The stimulus is a dot deviation signal flashed in a pseudo-random predetermined order (Fig. 1). During the course of the testing, the patient is presented with artificial distortions similar to those seen by individuals with neovascular AMD. The patient indicates where he/she perceives distortion along the dotted line using a computer mouse. When a signal is presented to the retina over a CNV lesion, the pathological distortion and the artificial distortion compete for the participant's attention. Preferential looking will subconsciously direct the user to select the more extreme distortion. Varying amplitudes of the artificial distortions serve as control stimuli to quantify the extent of possible pathologic distortions originating from retinal pathology [12]. A grayscale map of the visual field is generated and the intensity of the visual
Fig. 1. An image of a dotted line with a distortion is briefly displayed at a random location. The user is required to point to the specific location of the perceived distortion using the computer mouse.
296
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
field defect is determined. Based on a comparison of this intensity with a normative database and the participant's baseline performance with the FH device, the system determines whether there are visual responses which may be associated with progression of AMD to CNV. Participants are instructed to use the FH device ideally on a daily basis to monitor each eye separately, which takes approximately 3 min per eye. Results of each home-testing session are transmitted via a cellular modem to a data monitoring center. When the monitoring center detects a significant change compared with baseline, the physician is notified to recall the patient for an exam. The sensitivity and specificity of the FH home monitor were demonstrated to be similar to that of a clinical device using preferential hyperacuity perimeter technology [17] that was used to determine visual field defects associated with AMD progression in an office setting in a cross-sectional study. This monitoring device plus standard care, coupled with telemedicine is currently tested for home use prospectively. The tele-monitoring component of the solution, based on cellular communication between the device and the monitoring center, allowed for ongoing monitoring of participants adherence. Whenever the frequency of received tests fell below a pre-determined threshold, the monitoring center called the participant to provide a reminder or actively resolve any technical issue preventing data transfer. That feature is deemed important when elderly AMD participants, at times not technologically savvy, are engaged. Knowing that the monitoring center is watching their performance provides user with timely feedback and re-assurance. 2.2. Standard care In both the standard care (control) arm and the FH monitoring arm, participants received their investigator specific standard care instructions for self-monitoring of symptoms associated with AMD progression. In general, they were instructed to regularly test their visual function of each eye separately. Tools such as an Amsler grid may be used. Study participants were instructed to contact the clinic within 72 h to arrange for an exam when changes in visual function were noted. 3. Study objectives and endpoints The primary objective of this study was to determine whether home monitoring with the preferential hyperacuity perimetry and tele-monitoring solution based on the FH device plus standard care in participants at high risk for progression to CNV results in earlier detection of progression to CNV when compared with standard care alone. Progression to CNV was determined by the investigator, based on clinical examination and ancillary office testing. Ocular images obtained for the documentation of CNV were graded by a central reading center to confirm the diagnosis of CNV onset. The main clinical question was whether this potential early detection using the FH device would result in better visual acuity at diagnosis of CNV, a major predictor of future visual function following anti-VEGF therapy. Secondary vision function objectives included the comparison of the proportion of eyes in each monitoring strategy that at the time of CNV detection maintained visual acuity of 20/40
or better, lost ≤5 letters or ≥15 letters in the visual acuity score, or had visual acuity of 20/200 or worse. Other secondary objectives included the comparison between the two arms for the following outcome measures that are graded by the reading center including the differences in lesion characteristics measured on fluorescein angiography and the optical coherence tomography (OCT) measurements at the time of CNV diagnosis. The performance of the monitoring strategies was analyzed. The sensitivity of the FH and the strategy that included tele-monitoring was evaluated by comparing the two study arms in the analyses of the “first modality to alert”, which included alerts generated by the device, reported symptoms and clinical signs identified by the investigator at either scheduled or unscheduled clinic visits. The specificity of each monitoring strategy was evaluated by calculating the annual false positive rate. The HOME Study had a tertiary safety objective in participants who had confirmed CNV development whether the use of FH device with standard care was associated with better visual acuity outcome following 3 consecutive months of intravitreal anti-VEGF therapy when compared with standard care alone. In addition, self-reported adverse events that precluded monitoring with the FH were collected by the monitoring center for the analysis of interruptions to the use of the device. 3.1. Study population 3.1.1. Inclusion criteria Participants meeting the following criteria were enrolled during a 24-month recruitment period (Aug-2010 to Nov-2012). Participants were men and women, 55 to 90 years old, residents of the USA, who agreed to be followed for at least 2 years. Participants were required to be English speaking with the ability to understand the informed consent, instructions for self-monitoring, and the use of the monitoring device. The ocular inclusion criteria included either large drusen (≥125 μm) in both eyes (participants with potential for two study eyes) or large drusen in one eye (the study eye) and advanced AMD (neovascular AMD or central geographic atrophy) in the fellow eye (non-study eye) as determined by the study ophthalmologist. Visual acuity letter score following a standardized refraction had to be at least 54 (Snellen equivalent of 20/60) in the study eye(s). The ocular media were sufficient to permit quality fundus photographs and other retinal imaging. Participants also agreed to the examinations by the HOME Study ophthalmologist when changes were detected by the FH device or symptoms noted by standard care. For those randomized to the device monitoring arm, they agreed to take their monitoring device with them when they were away from the home for 14 days or more and were willing to provide their personal contact information to the monitoring center. 3.1.2. Exclusion criteria Individuals were excluded if they had evidence of macular or retinal disorders other than AMD in the study eye(s) that might confound evaluation of the outcome measurement, including diabetic retinopathy, unless it was limited to fewer than 10 microaneurysms and/or small retinal hemorrhages; angioid streaks; central serous choroidopathy; high myopia; macular hole or pseudohole; retinal vein occlusion; active uveitis, presumed ocular histoplasmosis syndrome; and other
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
297
clinical center was to schedule a study visit with the participant and the ForeseeHome study ophthalmologist within 72 h. Following a device alert that was determined to be a false alert, the participant was required to establish a new baseline with the device. In the event that that eye could not establish a new baseline, the monitoring center notified the patient to stop monitoring that eye and continue with standard care self-monitoring. These eyes remained in the FH monitoring arm for analysis purposes in the ITT cohort.
sight-threatening retinopathies. Previous retinal or other intraocular surgeries that may complicate the assessment of progression of AMD in the study eye(s) rendered the eye ineligible. Exceptions included cataract surgery that occurred more than 1 month prior to enrollment, and laser or cryotherapy for peripheral retinal tears. Refusal or inability to undergo fluorescein angiography was an exclusion criterion. Individuals routinely receiving ophthalmic examination more frequently than every 4 months were not eligible for enrollment. Individuals with chronic requirements for any systemic or ocular medications known to be toxic to the retina or optic nerve or chronic systemic steroid use of at least 10 mg per day or more were excluded. Potential participants who were not likely to adhere to the study program because of concomitant medical conditions were also excluded. The AREDS2 was the only additional clinical trial that was permitted for simultaneous participation by HOME Study participants.
3.2.2.2. Standard care only arm. The participants randomized to the standard care only arm received instructions for monocular self-checking of vision function, which was to be performed regularly. They were asked to call the clinic upon recognition of any new symptoms. The clinical center was to schedule a study visit with the symptomatic participant and the ForeseeHome study ophthalmologist within 72 h.
3.2. Study visits
3.3. Follow-up visits
3.2.1. Baseline visit/randomization At the baseline visit of the HOME Study, participants completed an informed consent process as documented by a signed informed consent form. Potential participants underwent an in office tutorial with the FH device followed by a qualification test with the device (http://www.foreseehome. com/foreseehome_demo.html). Individuals remained eligible for randomization if this process demonstrated their comprehension of instructions and successful use of the FH device (Fig. 2). Best corrected visual acuity was measured using the AREDS2 electronic visual acuity (EVA) test protocol, [18] and standardized digital color fundus photographs of 3 stereoscopic fields were obtained. All test procedures were conducted on all study eyes by certified personnel using standardized AREDS2 protocols. Participants underwent an eye examination and those that fulfilled all the eligibility criteria were then randomized to the FH device plus standard care or to standard care only. All participants were given investigator-specific instructions for self-monitoring visual function changes monocularly which may include the use of an Amsler grid.
HOME Study participants were instructed to return to the clinical center for a study visit within 72 h if they noted a vision function change, regardless of the randomization assignment. For participants randomly assigned to FH monitoring, changes in home testing that triggered an alert from the monitoring center prompted a study visit within 72 h of clinic notification. At these protocol mandated study visits, participants underwent BCVA, clinical evaluation, and ocular imaging, including color stereoscopic fundus photographs, fluorescein angiography and OCT, for the study eye(s) that had symptoms or device triggered alert. If the study eye(s) progressed to CNV the management of CNV was at the investigator's discretion. However, if anti-VEGF treatment was prescribed, the investigator was encouraged to initiate therapy within 7 (+3) days of CNV diagnosis and to administer intravitreal ranibizumab, bevacizumab, or aflibercept, using a monthly regimen for the first 3 doses. Therapy after the first 3 doses was at the discretion of the investigator. The clinical center collected BCVA 3 months after the initial anti-angiogenic treatment. If CNV was confirmed but anti-VEGF therapy was not performed within the suggested time window then, no further follow-up was required for that study eye for the study purposes. There were two types of visits to the clinical center: “scheduled” and “unscheduled”. A scheduled visit may be part of the AREDS2 study, or one that was part of routine care, as determined by the investigator. An unscheduled visit referred to visits initiated by the participant to evaluate new symptoms or visits triggered by a device alert. A participant who offers a complaint describing new symptoms at a scheduled visit would not be considered as an “alert” unless the participant also contacted the clinical center prior to the scheduled visit to request an unscheduled visit. During any scheduled or unscheduled visit, the study ophthalmologist was responsible for determining whether any symptoms or signs the physician observes on the exam were consistent with a CNV event. If the visit was a result of a device alert, participant calling for an unscheduled symptoms visit or the investigator suspects or diagnoses CNV, then the study protocol required complete testing which includes BCVA, OCT, color fundus photographs, and fluorescein angiography. These images were graded by the reading center to confirm the
3.2.2. Randomized arms 3.2.2.1. Device monitoring arm. In addition to receiving standard care instructions, participants in this arm received a FH device at home with instructions for installation. The set-up, initial use and transmission of test results via a cellular modem from the device were expected within 5 days of receipt of the device. During the first several test sessions, the device was required to establish a baseline value for future comparisons. While participants were encouraged to use the device daily, a minimum of two tests per week was the minimum suggested study requirement. For participants who were contributing only one study eye, the device was pre-programmed and mechanically secured to be used solely with the designated study eye. This avoided erroneous testing of a non-study eye. The central monitoring center alerted the participants if the testing frequency fell below twice a week. If the test results from the device were consistent with a change in function when compared with baseline values, an alert was sent from the monitoring center to the participant's clinical center. The
298
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
Fig. 2. Study visits by randomization in the HOME study. BCVA: best-corrected visual acuity. FH: ForeseeHome. FA: Fluorescein angiography. OCT: Optical coherence tomography. CNV: Choroidal neovascularization.
presence of CNV and grade characteristics of the lesion. At the study visit in which the study investigator could not identify CNV but the reading center identified presence of CNV, a followup visit 4 weeks later was required. The study duration was anticipated to be at least 2 years following closure of enrollment, with proposed end date of May 2014. Exit visits aimed at identifying CNV were scheduled in the final 3 months of the study for all study eyes that had not been documented to progress to CNV.
in at least one study eye were anticipated to have progression to neovascular AMD by 3 years. This will provide at least 80% power to detect a minimum difference between the two monitoring strategies of at least 5 letters in the mean decrease in BCVA at the time of CNV diagnosis in a study eye, assuming a standard deviation of 10 letters and a 0.05 one-sided significance level. The power of the study remains stable if the number of confirmed CNV events is imbalanced between the study arms. Three CNV event cohorts will be analyzed as follows:
3.4. Analyses plan
1) The intent-to-treat (ITT) cohort which includes all participants who had an investigator-confirmed CNV event in either of the treatment arms. The adherence to the use of the FH device did not factor into the analysis. 2) The initial per protocol cohort (PP1), in which the device monitoring arm was restricted to those participants who were using the device at the time of CNV detection
A sample size of 1400 participants (700 per arm) was considered adequate to determine the magnitude of effect for the primary efficacy analysis. It was assumed that approximately 20% of the 1400 participants would be lost to follow up. About 12% of the participants (~50 per arm) with intermediate AMD
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
regardless of adherence to minimal recommended frequency of monitoring. 3) The second per protocol cohort (PP2), which further restricted the device monitoring arm to only those participants who met the minimum use criteria of two tests per week in their study eye(s) for each of the four weeks prior to the event, or met the minimum use criteria for ≥6 weeks out of the 8 weeks in their study eye(s) prior to the event. Primary and secondary analyses will be conducted for those initial and second eye CNV events identified during the course of the study. Efficacy analyses based on the ITT population will be regarded as the primary analyses for statistical inferences regarding efficacy for the trial. The performance of the FH device will be assessed. The study design was based on a longitudinal followup with few competing modalities of diagnosis in each arm: device vs. symptoms vs. scheduled clinic visits in the FH arm and symptoms vs. scheduled clinic visits in the control arm. Once one of the modalities alerted an event that resulted in a diagnosis of incident CNV, there was essentially censoring. That eye is no longer available for further analyses of events. This design, which reflects real life practice, in contrast to cross sectional studies, required implementation of novel approaches for the analyses of performance. The assessment of the sensitivity of the FH and the strategy that included tele-monitoring will be evaluated by comparing the distribution within each of the study arms and between the two study arms of the “first modality to alert”. Further analyses of the “first modality to alert” comparing the ITT, PP1 and PP2 cohorts separately, compare the change of BCVA from baseline subgrouped by each alert modality, and compare the scheduled visits to between-visits (combining device and symptoms triggered events) findings. The specificity of the device and symptoms was evaluated by calculating the annual false positive rate for false alerts triggered by the device or by symptoms and describing the frequency in which participants had no false alert or any number of false alerts. The average frequency of device use throughout the study for participants who continued to use the device, the proportion of participants who fell below the recommended weekly frequency of use, and the proportion of FH device assigned participants who stopped using the device will also be reported. An analysis of the primary study outcomes will be conducted with stratification for AREDS2 vs. non AREDS2 participation. 3.4.0.1. Interim analyses. Interim analyses were planned for the following scenarios: to stop the trial early for efficacy, re-calculate the number of CNV events to increase the power of the study, or to stop the trial early for futility based on a calculation of the conditional power and pre-specified conditional power cut points. Two interim analyses were planned at approximately 50% and 75% of the total planned number of CNV events. Guidelines were followed for each of these scenarios. For interim monitoring of efficacy, the group sequential method with O'Brien-Fleming alpha spending function was used. 4. Discussion The HOME Study was designed to address whether implementation of the FH device and tele-monitoring strategy plus standard care will result in earlier detection of CNV manifested
299
by minimal impairment in the presenting visual acuity at the time of CNV diagnosis, potentially allowing for better visual function after intravitreal anti-VEGF therapy to manage the CNV. Study enrollment was completed in November 2012. The study design appeared to be feasible despite challenges to conducting such a multi-site randomized study for a tele-monitoring device, resulting in the enrollment of greater number of participants than the anticipated sample size in a large number of clinics, with potential for answering an important clinically relevant research question. Limitation of the study includes decreased generalizability of the results for all persons with intermediate AMD as an anticipated number of participants may not successfully pass the screening for the testing. However, the results may have impact on the management of persons with AMD, a disease of public health importance. This study was terminated early at the recommendation of the data and safety monitoring committee for efficacy. Initial study results have been reported [19] and additional data will be forthcoming. Conflict of interest disclosures All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr. Susan Bressler reported receiving grant and travel support from the EMMES Corporation; serving as a consultant for GlaxoSmithKline; receiving grants or grants pending from Allergan, Bayer Healthcare, Genentech, Lumenis Inc, Notal Vision Ltd, Novartis, Regeneron, Thrombogenics, and Sanofi-Aventis; receiving payment for lectures from providers of continuing medical education materials; and these grants are negotiated and administered by the School of Medicine, which receives the grants through the Office of Research Administration (individual investigators who participate in such sponsored projects are not directly compensated by the sponsor but may receive salary or other support from the institution to support their effort on the projects). Ronald Danis, MD is a consultant to GSK, Thrombogenics, Oraya, Inc., Merck, Inc. He receives travel for meetings from Thrombogenics. He has ownership interest in EyeKor LLC. Dr. Danis also receives research funding from grants and contracts for reading center functions from GSK, Allergan, Advanced Cell Technologies, Perceptive Informatices, Lilly Research Laboratories, Pharmanet, KangHong Pharmaceuticals, Tactical Therapeutics, and through EMMES Corporation, grants from Notal Vision. Dr. Michael Elman reported receiving grant and travel support from the Emmes Corporation and the Jaeb Center; serving on the Speakers Bureau for Genentech-Roche; serving as a consultant for Genentech-Roche and Ohr Pharmaceuticals; serving on the Scientific Advisory Board for Ohr Pharmaceuticals; receiving research grants from Novartis, Merck, Ohr Pharmaceuticals, and X-Covery Vision. He owns warrants in Ohr Pharmaceuticals. Dr. Richard Garfinkel receives support for the study from Notal Vision through EMMES. He is a stock holder for Covalent Medical. Dr. Jeffrey Heier reported receiving consulting and scientific advisory fees from: Allergan, Bayer, Forsight Labs, Genetech, Genzyme/Sanofi, Kanghong, Notal Vision, Ohr Pharmaceuticals, Oraya, Regenron, Sequenon, and Xcovery. He also received research funding from Allergan, Bayer, Genentech, Genzyme/
300
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
Sanofi, GlaxoSmithKline, Lpath, Notal Vision, Ohr Pharmaceuticals, Ophthotech, and Regeneron. Dr Judy E. Kim serves on the Advisory Board of Thrombogenics; and receives unrestricted grant from Research to Prevent Blindness, Inc. and support for the study from Notal Vision through EMMES. Acknowledgment Emily Chew and Traci Clemons had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analyses. The investigators designed and executed the study in collaboration with the sponsors of the study. The analyses were performed independently in the Coordinating Center. The Data and Safety Monitoring Committees evaluated both the study design and the study data. The manuscript was drafted by the investigators with collaboration and input from the sponsors. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.cct.2014.02.003. References [1] Congdon N, O'Colmain B, Klaver CC, Klein R, Munoz B, Friedman DS, et al. Causes and prevalence of visual impairment among adults in the United States. Arch Ophthalmol 2004;122:477–85. [2] Friedman DS, O'Colmain BJ, Muñoz B, Tomany SC, McCarty C, de Jong PT, et al. Prevalence of age-related macular degeneration in the United States. Arch Ophthalmol 2004;122:564–72. [3] Pascolini D, Mariotti SP, Pokharel GP, Pararajasegaram R, Etya'ale D, Negrel AD, et al. 2002 global update of available data on visual impairment: a compilation of population-based prevalence studies. Ophthalmic Epidemiol 2004;11:67–115. [4] Ferris 3rd FL, Fine SL, Hyman L. Age-related macular degeneration and blindness due to neovascular maculopathy. Arch Ophthalmol 1984; 102:1640–2. [5] Brown DM, Kaiser PK, Michels M, Soubrane G, Heier JS, Kim RY, et al. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med 2006;355:1432–44.
[6] Rosenfeld PJ, Brown DM, Heier JS, Boyer DS, Kaiser PK, Chung CY, et al. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med 2006;355(14):1419–31. [7] IVAN Study Investigators, Chakravarthy U, Harding SP, Rogers CA, Downes SM, Lotery AJ, et al. Ranibizumab versus bevacizumab to treat neovascular age-related macular degeneration: one-year findings from the IVAN randomized trial. Ophthalmology 2012;119:1399–411. [8] Heier JS, Boyer D, Nguyen QD, Marcus D, Roth DB, Yancopoulos G, et al. The 1-year results of CLEAR-IT 2, a phase 2 study of vascular endothelial growth factor trap-eye dosed as-needed after 12-week fixed dosing. Ophthalmology Jun 2011;118(6):1098–106. [9] Brown DM, Heier JS, Ciulla T, Benz M, Abraham P, Yancopoulos G, et al. Primary endpoint results of a phase II study of vascular endothelial growth factor trap-eye in wet age-related macular degeneration. Ophthalmology 2011;118:1089–97. [10] CATT Research Group, Martin DF, Maguire MG, Ying GS, Grunwald JE, Fine SL, et al. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med May 19 2011;364(20):1897-908 [Epub 2011 Apr 28]. [11] Ying GS, Huang J, Maguire MG, et al. Baseline predictors for one-year visual outcomes with ranibizumab or bevacizumab for neovascular age-related macular degeneration. Ophthalmology 2013;120:122–9. [12] Kaiser PK, Brown DM, Zhang K, Hudson HL, Holz FG, Shapiro H, et al. Ranibizumab for predominantly classic neovascular age-related macular degeneration: subgroup analysis of first-year ANCHOR results. Am J Ophthalmol 2007;144:850–7. [13] Boyer DS, Antoszyk AN, Awh CC, Bhisitkul RB, Shapiro H, Acharya NR. Subgroup analysis of the MARINA study of ranibizumab in neovascular age-related macular degeneration. Ophthalmology 2007;114:246–52. [14] Fong DS, Custis P, Howes J, Hsu JW. Intravitreal bevacizumab and ranibizumab for age-related macular degeneration. A multi-center, retrospective study. Ophthalmology 2010;117(2):298–302. [15] Rauch R, Weingessel B, Maca SM, Vecsei-Marlovits PV. Time to first treatment — the significance of early treatment of exudative agerelated macular degeneration. Retina 2012;32:1260–4. [16] Westheimer G. Editorial: visual acuity and hyperacuity. Invest Ophthalmol 1975;145(8):570–2. [17] Loewenstein A, Ferencz JR, Lang Y, Yeshurun I, Pollack A, Siegal R, et al. Toward earlier detection of choroidal neovascularization secondary to agerelated macular degeneration: multicenter evaluation of a preferential hyperacuity perimeter designed as a home device. Retina 2010;30:1058–64. [18] Beck RW, Moke PS, Turpin AH, Ferris 3rd FL, San Giovanni JP, Johnson CA, et al. A computerized method of visual acuity testing: an adaptation of the Early Treatment Diabetic Retinopathy Study testing protocol. Am J Ophthalmol 2003;135(2):194–205. [19] The AREDS2-HOME Study Research Group. Randomized trial of a home monitoring system for early detection of choroidal neovascularization. Home Monitoring of the Eye (HOME) Study. Ophthalmology Nov 8 2013. http://dx.doi.org/10.1016/j.ophtha.2013.10.027 [pii: S01616420(13)00952-4, Epub ahead of print].
Contents lists available at ScienceDirect
Contemporary Clinical Trials journal homepage: www.elsevier.com/locate/conclintrial
Randomized trial of the ForeseeHome monitoring device for early detection of neovascular age-related macular degeneration. The HOme Monitoring of the Eye (HOME) study design — HOME Study report number 1☆ Emily Y. Chew a,⁎, Traci E. Clemons b, Susan B. Bressler c, Michael J. Elman d, Ronald P. Danis e, Amitha Domalpally e, Jeffrey S. Heier f, Judy E. Kim g, Richard A. Garfinkel h, Appendix 1 for AREDS2-HOME Study Research Group a
National Eye Institute, Bethesda, MD, USA The EMMES Corporation, Rockville, MD, USA c Department of Ophthalmology, Wilmer Eye Institute, Johns Hopkins Hospital, Baltimore, MD, USA d Elman Retina Group, Baltimore, MD, USA e Department of Ophthalmology and Visual Sciences, University of Wisconsin–Madison School of Medicine and Public Health, USA f Ophthalmic Consultant of Boston, Boston, MA, USA g Department of Ophthalmology, Medical College of Wisconsin, Milwaukee, USA h Retina Group of Washington, Bethesda, MD, USA b
a r t i c l e
i n f o
Article history: Received 24 September 2013 Received in revised form 25 January 2014 Accepted 5 February 2014 Available online 13 February 2014 Keywords: Age-related macular degeneration Controlled clinical trial Home-monitoring Choroidal neovascularization Best-corrected visual acuity
a b s t r a c t Objective: To evaluate the effects of a home-monitoring device with tele-monitoring compared with standard care in detection of progression to choroidal neovascularization (CNV) associated with age-related macular degeneration (AMD), the leading cause of blindness in the US. Patients and methods: Participants, aged 55 to 90 years, at high risk of developing CNV associated with AMD were recruited to the HOme Monitoring of Eye (HOME) Study, an unmasked, multi-center, randomized trial of the ForeseeHome (FH) device plus standard care vs. standard care alone. The FH device utilizes preferential hyperacuity perimetry and tele-monitoring to detect changes in vision function associated with development of CNV, potentially prior to symptom and visual acuity loss. After establishing baseline measurements, subsequent changes on follow-up are detected by the device, causing the monitoring center to alert the clinical center to recall participants for an exam. Standard care consists of instructions for self-monitoring visual changes with subsequent self-report to the clinical center. The primary objective of this study is to determine whether home monitoring plus standard care in comparison with standard care alone, results in earlier detection of incident CNV with better present visual acuity. The primary outcome is the decline in visual acuity at CNV diagnosis from baseline. Detection of CNV prior to substantial vision loss is critical as vision outcome
☆ Financial Support: Financial support to the study was provided by the Notal Vision through a clinical trial agreement with the NEI (CTA-00833) and a service agreement with EMMES Corporation. The AREDS2 study is supported by the intramural program funds and contracts from the National Eye Institute/National Institutes of Health (NEI/NIH), Department of Health and Human Services, Bethesda, MD. Contract No. HHS-N-260-2005-00007-C. ADB Contract No. N01-EY-5-0007. Funds were generously contributed to these contracts by the following NIH institutes: Office of Dietary Supplements (ODS), National Center for Complementary and Alternative Medicine (NCCAM), National Institute on Aging (NIA), National Heart, Lung and Blood Institute (NHLBI), and National Institute of Neurological Disorders and Stroke (NINDS). The study medications and raw materials were provided by Alcon, Bausch and Lomb, DSM, and Pfizer. ⁎ Corresponding author at: NIH, Building 10, CRC, Room 3-2531, 10 Center Drive, MSC-1204, Bethesda, MD 20892-1204, United States. Tel.: +1 301 496 6583; fax: +1 301 496 7295. E-mail address: [email protected] (E.Y. Chew).
http://dx.doi.org/10.1016/j.cct.2014.02.003 1551-7144/Published by Elsevier Inc.
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
295
following anti-angiogenic therapy is dependent on the visual acuity at initiation of treatment. Discussion: HOME Study is the first large scale study to test the use of home tele-monitoring system in the management of AMD patients. Published by Elsevier Inc.
1. Introduction Age-related macular degeneration (AMD) is the leading cause of irreversible blindness in the United States [1,2] and in populations of northern European descent in the developed world [3]. Blindness in advanced AMD occurs from the development of choroidal neovascularization (CNV) and/or geographic atrophy affecting the center of the fovea. While CNV occurs in only 10%–15% of all persons with AMD, it accounts for more than 80% of cases of severe central vision loss associated with AMD [4]. Current therapy for neovascular AMD, intravitreal injections of anti-vascular endothelial growth factor (VEGF) agents, such as bevacizumab, ranibizumab and aflibercept, can potentially stabilize visual acuity in 95% of eyes [5–10]. The main predictor for visual acuity outcome following anti-VEGF treatment is the visual acuity at the time of initiation of CNV therapy [11–13]. Patients with worse visual acuity at initiation of treatment were more likely to gain more letters of vision but their final visual acuity tended to be worse than those who started with good visual acuity and gained fewer letters following treatment [11–13]. Current monitoring of patients at risk for CNV development consists of periodic in-office examinations at varying intervals by ophthalmologists and optometrists with the recommendation for self-monitoring of visual function including either new onset or changes in existing symptoms. This may include the use of an Amsler grid. Such monitoring strategies yielded visual acuity of 20/40 or better at the initial detection of CNV in only 13 to 36% of new onset CNV [11,14]. Increasing time interval between onset of symptoms and initiation of treatment resulted in markedly decreased visual acuities both at presentation of CNV and following anti-VEGF treatment [15]. Detection of the CNV prior to significant vision loss is essential to maximize successful visual outcomes following treatment. This report describes the design of the HOme Monitoring of the Eye (HOME) Study, a Phase 3, unmasked, randomized trial evaluating the role of home monitoring with the ForeseeHome (FH) device plus standard care (referred from here as FH monitoring) compared to standard care alone for eyes at risk of developing CNV. Study subjects were enrolled in 44 retina clinics participating in the Age-Related Eye Disease Study 2 (AREDS2), a randomized trial of nutrient supplements. The ForeseeHome monitoring device utilized macular visual field testing and tele-monitoring. The basis for this test is preferential hyperacuity (or vernier acuity) perimetry which potentially detects the earliest functional abnormalities associated with CNV, prior to patient's awareness of symptoms or visual acuity changes. Such changes transmitted to the monitoring center triggered alerts to the physicians about the potential early development of CNV. Participants were then contacted for exams. This strategy, paired with standard care was compared with standard care alone, which consisted of instructions to regular monocular self-monitoring for visual changes with subsequent self-report to the physician. The hypothesis tested
was that home monitoring would detect the onset of neovascular AMD earlier with better visual acuity (a smaller decrease in visual acuity compared with baseline) at CNV detection compared with the standard of care alone arm. The study was registered at ClinicalTrials.gov with the identification NCT01103505. 2. Research design and methods The HOME Study, an ancillary study of the Age-Related Eye Disease Study 2 (AREDS2), a long-term multicenter randomized controlled clinical trial of oral supplements, was an unmasked, randomized clinical study that enrolled both AREDS2 and non-AREDS2 participants. They were randomly assigned, 1:1, to either the FH monitoring plus standard care or the standard care alone. The Institutional Review Boards for human subject research from the individual clinical sites approved the study protocol and all participants signed informed consents. 2.1. ForeseeHome monitoring device The ForeseeHome (FH) monitoring device (Notal Vision, Ltd. Tel Aviv, Israel), cleared by the FDA in 2009, was developed to detect visual field disturbances associated with neovascular AMD prior to overt visual symptoms. This technology is based on hyperacuity (known also as vernier acuity) – the ability to perceive a very small difference in the relative spatial localization of two or more visual stimuli – that can be locally impaired before visual acuity is affected [16,13]. Early CNV causes subtle separation of the different layers of the retina, resulting in localized distortion of vision (metamorphopsia), often with accompanying loss of sensitivity (scotoma). The FH tests the central area of the visual field of 14° with approximately 500 data points sampled rapidly 3 to 5 times. The stimulus is a dot deviation signal flashed in a pseudo-random predetermined order (Fig. 1). During the course of the testing, the patient is presented with artificial distortions similar to those seen by individuals with neovascular AMD. The patient indicates where he/she perceives distortion along the dotted line using a computer mouse. When a signal is presented to the retina over a CNV lesion, the pathological distortion and the artificial distortion compete for the participant's attention. Preferential looking will subconsciously direct the user to select the more extreme distortion. Varying amplitudes of the artificial distortions serve as control stimuli to quantify the extent of possible pathologic distortions originating from retinal pathology [12]. A grayscale map of the visual field is generated and the intensity of the visual
Fig. 1. An image of a dotted line with a distortion is briefly displayed at a random location. The user is required to point to the specific location of the perceived distortion using the computer mouse.
296
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
field defect is determined. Based on a comparison of this intensity with a normative database and the participant's baseline performance with the FH device, the system determines whether there are visual responses which may be associated with progression of AMD to CNV. Participants are instructed to use the FH device ideally on a daily basis to monitor each eye separately, which takes approximately 3 min per eye. Results of each home-testing session are transmitted via a cellular modem to a data monitoring center. When the monitoring center detects a significant change compared with baseline, the physician is notified to recall the patient for an exam. The sensitivity and specificity of the FH home monitor were demonstrated to be similar to that of a clinical device using preferential hyperacuity perimeter technology [17] that was used to determine visual field defects associated with AMD progression in an office setting in a cross-sectional study. This monitoring device plus standard care, coupled with telemedicine is currently tested for home use prospectively. The tele-monitoring component of the solution, based on cellular communication between the device and the monitoring center, allowed for ongoing monitoring of participants adherence. Whenever the frequency of received tests fell below a pre-determined threshold, the monitoring center called the participant to provide a reminder or actively resolve any technical issue preventing data transfer. That feature is deemed important when elderly AMD participants, at times not technologically savvy, are engaged. Knowing that the monitoring center is watching their performance provides user with timely feedback and re-assurance. 2.2. Standard care In both the standard care (control) arm and the FH monitoring arm, participants received their investigator specific standard care instructions for self-monitoring of symptoms associated with AMD progression. In general, they were instructed to regularly test their visual function of each eye separately. Tools such as an Amsler grid may be used. Study participants were instructed to contact the clinic within 72 h to arrange for an exam when changes in visual function were noted. 3. Study objectives and endpoints The primary objective of this study was to determine whether home monitoring with the preferential hyperacuity perimetry and tele-monitoring solution based on the FH device plus standard care in participants at high risk for progression to CNV results in earlier detection of progression to CNV when compared with standard care alone. Progression to CNV was determined by the investigator, based on clinical examination and ancillary office testing. Ocular images obtained for the documentation of CNV were graded by a central reading center to confirm the diagnosis of CNV onset. The main clinical question was whether this potential early detection using the FH device would result in better visual acuity at diagnosis of CNV, a major predictor of future visual function following anti-VEGF therapy. Secondary vision function objectives included the comparison of the proportion of eyes in each monitoring strategy that at the time of CNV detection maintained visual acuity of 20/40
or better, lost ≤5 letters or ≥15 letters in the visual acuity score, or had visual acuity of 20/200 or worse. Other secondary objectives included the comparison between the two arms for the following outcome measures that are graded by the reading center including the differences in lesion characteristics measured on fluorescein angiography and the optical coherence tomography (OCT) measurements at the time of CNV diagnosis. The performance of the monitoring strategies was analyzed. The sensitivity of the FH and the strategy that included tele-monitoring was evaluated by comparing the two study arms in the analyses of the “first modality to alert”, which included alerts generated by the device, reported symptoms and clinical signs identified by the investigator at either scheduled or unscheduled clinic visits. The specificity of each monitoring strategy was evaluated by calculating the annual false positive rate. The HOME Study had a tertiary safety objective in participants who had confirmed CNV development whether the use of FH device with standard care was associated with better visual acuity outcome following 3 consecutive months of intravitreal anti-VEGF therapy when compared with standard care alone. In addition, self-reported adverse events that precluded monitoring with the FH were collected by the monitoring center for the analysis of interruptions to the use of the device. 3.1. Study population 3.1.1. Inclusion criteria Participants meeting the following criteria were enrolled during a 24-month recruitment period (Aug-2010 to Nov-2012). Participants were men and women, 55 to 90 years old, residents of the USA, who agreed to be followed for at least 2 years. Participants were required to be English speaking with the ability to understand the informed consent, instructions for self-monitoring, and the use of the monitoring device. The ocular inclusion criteria included either large drusen (≥125 μm) in both eyes (participants with potential for two study eyes) or large drusen in one eye (the study eye) and advanced AMD (neovascular AMD or central geographic atrophy) in the fellow eye (non-study eye) as determined by the study ophthalmologist. Visual acuity letter score following a standardized refraction had to be at least 54 (Snellen equivalent of 20/60) in the study eye(s). The ocular media were sufficient to permit quality fundus photographs and other retinal imaging. Participants also agreed to the examinations by the HOME Study ophthalmologist when changes were detected by the FH device or symptoms noted by standard care. For those randomized to the device monitoring arm, they agreed to take their monitoring device with them when they were away from the home for 14 days or more and were willing to provide their personal contact information to the monitoring center. 3.1.2. Exclusion criteria Individuals were excluded if they had evidence of macular or retinal disorders other than AMD in the study eye(s) that might confound evaluation of the outcome measurement, including diabetic retinopathy, unless it was limited to fewer than 10 microaneurysms and/or small retinal hemorrhages; angioid streaks; central serous choroidopathy; high myopia; macular hole or pseudohole; retinal vein occlusion; active uveitis, presumed ocular histoplasmosis syndrome; and other
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
297
clinical center was to schedule a study visit with the participant and the ForeseeHome study ophthalmologist within 72 h. Following a device alert that was determined to be a false alert, the participant was required to establish a new baseline with the device. In the event that that eye could not establish a new baseline, the monitoring center notified the patient to stop monitoring that eye and continue with standard care self-monitoring. These eyes remained in the FH monitoring arm for analysis purposes in the ITT cohort.
sight-threatening retinopathies. Previous retinal or other intraocular surgeries that may complicate the assessment of progression of AMD in the study eye(s) rendered the eye ineligible. Exceptions included cataract surgery that occurred more than 1 month prior to enrollment, and laser or cryotherapy for peripheral retinal tears. Refusal or inability to undergo fluorescein angiography was an exclusion criterion. Individuals routinely receiving ophthalmic examination more frequently than every 4 months were not eligible for enrollment. Individuals with chronic requirements for any systemic or ocular medications known to be toxic to the retina or optic nerve or chronic systemic steroid use of at least 10 mg per day or more were excluded. Potential participants who were not likely to adhere to the study program because of concomitant medical conditions were also excluded. The AREDS2 was the only additional clinical trial that was permitted for simultaneous participation by HOME Study participants.
3.2.2.2. Standard care only arm. The participants randomized to the standard care only arm received instructions for monocular self-checking of vision function, which was to be performed regularly. They were asked to call the clinic upon recognition of any new symptoms. The clinical center was to schedule a study visit with the symptomatic participant and the ForeseeHome study ophthalmologist within 72 h.
3.2. Study visits
3.3. Follow-up visits
3.2.1. Baseline visit/randomization At the baseline visit of the HOME Study, participants completed an informed consent process as documented by a signed informed consent form. Potential participants underwent an in office tutorial with the FH device followed by a qualification test with the device (http://www.foreseehome. com/foreseehome_demo.html). Individuals remained eligible for randomization if this process demonstrated their comprehension of instructions and successful use of the FH device (Fig. 2). Best corrected visual acuity was measured using the AREDS2 electronic visual acuity (EVA) test protocol, [18] and standardized digital color fundus photographs of 3 stereoscopic fields were obtained. All test procedures were conducted on all study eyes by certified personnel using standardized AREDS2 protocols. Participants underwent an eye examination and those that fulfilled all the eligibility criteria were then randomized to the FH device plus standard care or to standard care only. All participants were given investigator-specific instructions for self-monitoring visual function changes monocularly which may include the use of an Amsler grid.
HOME Study participants were instructed to return to the clinical center for a study visit within 72 h if they noted a vision function change, regardless of the randomization assignment. For participants randomly assigned to FH monitoring, changes in home testing that triggered an alert from the monitoring center prompted a study visit within 72 h of clinic notification. At these protocol mandated study visits, participants underwent BCVA, clinical evaluation, and ocular imaging, including color stereoscopic fundus photographs, fluorescein angiography and OCT, for the study eye(s) that had symptoms or device triggered alert. If the study eye(s) progressed to CNV the management of CNV was at the investigator's discretion. However, if anti-VEGF treatment was prescribed, the investigator was encouraged to initiate therapy within 7 (+3) days of CNV diagnosis and to administer intravitreal ranibizumab, bevacizumab, or aflibercept, using a monthly regimen for the first 3 doses. Therapy after the first 3 doses was at the discretion of the investigator. The clinical center collected BCVA 3 months after the initial anti-angiogenic treatment. If CNV was confirmed but anti-VEGF therapy was not performed within the suggested time window then, no further follow-up was required for that study eye for the study purposes. There were two types of visits to the clinical center: “scheduled” and “unscheduled”. A scheduled visit may be part of the AREDS2 study, or one that was part of routine care, as determined by the investigator. An unscheduled visit referred to visits initiated by the participant to evaluate new symptoms or visits triggered by a device alert. A participant who offers a complaint describing new symptoms at a scheduled visit would not be considered as an “alert” unless the participant also contacted the clinical center prior to the scheduled visit to request an unscheduled visit. During any scheduled or unscheduled visit, the study ophthalmologist was responsible for determining whether any symptoms or signs the physician observes on the exam were consistent with a CNV event. If the visit was a result of a device alert, participant calling for an unscheduled symptoms visit or the investigator suspects or diagnoses CNV, then the study protocol required complete testing which includes BCVA, OCT, color fundus photographs, and fluorescein angiography. These images were graded by the reading center to confirm the
3.2.2. Randomized arms 3.2.2.1. Device monitoring arm. In addition to receiving standard care instructions, participants in this arm received a FH device at home with instructions for installation. The set-up, initial use and transmission of test results via a cellular modem from the device were expected within 5 days of receipt of the device. During the first several test sessions, the device was required to establish a baseline value for future comparisons. While participants were encouraged to use the device daily, a minimum of two tests per week was the minimum suggested study requirement. For participants who were contributing only one study eye, the device was pre-programmed and mechanically secured to be used solely with the designated study eye. This avoided erroneous testing of a non-study eye. The central monitoring center alerted the participants if the testing frequency fell below twice a week. If the test results from the device were consistent with a change in function when compared with baseline values, an alert was sent from the monitoring center to the participant's clinical center. The
298
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
Fig. 2. Study visits by randomization in the HOME study. BCVA: best-corrected visual acuity. FH: ForeseeHome. FA: Fluorescein angiography. OCT: Optical coherence tomography. CNV: Choroidal neovascularization.
presence of CNV and grade characteristics of the lesion. At the study visit in which the study investigator could not identify CNV but the reading center identified presence of CNV, a followup visit 4 weeks later was required. The study duration was anticipated to be at least 2 years following closure of enrollment, with proposed end date of May 2014. Exit visits aimed at identifying CNV were scheduled in the final 3 months of the study for all study eyes that had not been documented to progress to CNV.
in at least one study eye were anticipated to have progression to neovascular AMD by 3 years. This will provide at least 80% power to detect a minimum difference between the two monitoring strategies of at least 5 letters in the mean decrease in BCVA at the time of CNV diagnosis in a study eye, assuming a standard deviation of 10 letters and a 0.05 one-sided significance level. The power of the study remains stable if the number of confirmed CNV events is imbalanced between the study arms. Three CNV event cohorts will be analyzed as follows:
3.4. Analyses plan
1) The intent-to-treat (ITT) cohort which includes all participants who had an investigator-confirmed CNV event in either of the treatment arms. The adherence to the use of the FH device did not factor into the analysis. 2) The initial per protocol cohort (PP1), in which the device monitoring arm was restricted to those participants who were using the device at the time of CNV detection
A sample size of 1400 participants (700 per arm) was considered adequate to determine the magnitude of effect for the primary efficacy analysis. It was assumed that approximately 20% of the 1400 participants would be lost to follow up. About 12% of the participants (~50 per arm) with intermediate AMD
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
regardless of adherence to minimal recommended frequency of monitoring. 3) The second per protocol cohort (PP2), which further restricted the device monitoring arm to only those participants who met the minimum use criteria of two tests per week in their study eye(s) for each of the four weeks prior to the event, or met the minimum use criteria for ≥6 weeks out of the 8 weeks in their study eye(s) prior to the event. Primary and secondary analyses will be conducted for those initial and second eye CNV events identified during the course of the study. Efficacy analyses based on the ITT population will be regarded as the primary analyses for statistical inferences regarding efficacy for the trial. The performance of the FH device will be assessed. The study design was based on a longitudinal followup with few competing modalities of diagnosis in each arm: device vs. symptoms vs. scheduled clinic visits in the FH arm and symptoms vs. scheduled clinic visits in the control arm. Once one of the modalities alerted an event that resulted in a diagnosis of incident CNV, there was essentially censoring. That eye is no longer available for further analyses of events. This design, which reflects real life practice, in contrast to cross sectional studies, required implementation of novel approaches for the analyses of performance. The assessment of the sensitivity of the FH and the strategy that included tele-monitoring will be evaluated by comparing the distribution within each of the study arms and between the two study arms of the “first modality to alert”. Further analyses of the “first modality to alert” comparing the ITT, PP1 and PP2 cohorts separately, compare the change of BCVA from baseline subgrouped by each alert modality, and compare the scheduled visits to between-visits (combining device and symptoms triggered events) findings. The specificity of the device and symptoms was evaluated by calculating the annual false positive rate for false alerts triggered by the device or by symptoms and describing the frequency in which participants had no false alert or any number of false alerts. The average frequency of device use throughout the study for participants who continued to use the device, the proportion of participants who fell below the recommended weekly frequency of use, and the proportion of FH device assigned participants who stopped using the device will also be reported. An analysis of the primary study outcomes will be conducted with stratification for AREDS2 vs. non AREDS2 participation. 3.4.0.1. Interim analyses. Interim analyses were planned for the following scenarios: to stop the trial early for efficacy, re-calculate the number of CNV events to increase the power of the study, or to stop the trial early for futility based on a calculation of the conditional power and pre-specified conditional power cut points. Two interim analyses were planned at approximately 50% and 75% of the total planned number of CNV events. Guidelines were followed for each of these scenarios. For interim monitoring of efficacy, the group sequential method with O'Brien-Fleming alpha spending function was used. 4. Discussion The HOME Study was designed to address whether implementation of the FH device and tele-monitoring strategy plus standard care will result in earlier detection of CNV manifested
299
by minimal impairment in the presenting visual acuity at the time of CNV diagnosis, potentially allowing for better visual function after intravitreal anti-VEGF therapy to manage the CNV. Study enrollment was completed in November 2012. The study design appeared to be feasible despite challenges to conducting such a multi-site randomized study for a tele-monitoring device, resulting in the enrollment of greater number of participants than the anticipated sample size in a large number of clinics, with potential for answering an important clinically relevant research question. Limitation of the study includes decreased generalizability of the results for all persons with intermediate AMD as an anticipated number of participants may not successfully pass the screening for the testing. However, the results may have impact on the management of persons with AMD, a disease of public health importance. This study was terminated early at the recommendation of the data and safety monitoring committee for efficacy. Initial study results have been reported [19] and additional data will be forthcoming. Conflict of interest disclosures All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr. Susan Bressler reported receiving grant and travel support from the EMMES Corporation; serving as a consultant for GlaxoSmithKline; receiving grants or grants pending from Allergan, Bayer Healthcare, Genentech, Lumenis Inc, Notal Vision Ltd, Novartis, Regeneron, Thrombogenics, and Sanofi-Aventis; receiving payment for lectures from providers of continuing medical education materials; and these grants are negotiated and administered by the School of Medicine, which receives the grants through the Office of Research Administration (individual investigators who participate in such sponsored projects are not directly compensated by the sponsor but may receive salary or other support from the institution to support their effort on the projects). Ronald Danis, MD is a consultant to GSK, Thrombogenics, Oraya, Inc., Merck, Inc. He receives travel for meetings from Thrombogenics. He has ownership interest in EyeKor LLC. Dr. Danis also receives research funding from grants and contracts for reading center functions from GSK, Allergan, Advanced Cell Technologies, Perceptive Informatices, Lilly Research Laboratories, Pharmanet, KangHong Pharmaceuticals, Tactical Therapeutics, and through EMMES Corporation, grants from Notal Vision. Dr. Michael Elman reported receiving grant and travel support from the Emmes Corporation and the Jaeb Center; serving on the Speakers Bureau for Genentech-Roche; serving as a consultant for Genentech-Roche and Ohr Pharmaceuticals; serving on the Scientific Advisory Board for Ohr Pharmaceuticals; receiving research grants from Novartis, Merck, Ohr Pharmaceuticals, and X-Covery Vision. He owns warrants in Ohr Pharmaceuticals. Dr. Richard Garfinkel receives support for the study from Notal Vision through EMMES. He is a stock holder for Covalent Medical. Dr. Jeffrey Heier reported receiving consulting and scientific advisory fees from: Allergan, Bayer, Forsight Labs, Genetech, Genzyme/Sanofi, Kanghong, Notal Vision, Ohr Pharmaceuticals, Oraya, Regenron, Sequenon, and Xcovery. He also received research funding from Allergan, Bayer, Genentech, Genzyme/
300
E.Y. Chew et al. / Contemporary Clinical Trials 37 (2014) 294–300
Sanofi, GlaxoSmithKline, Lpath, Notal Vision, Ohr Pharmaceuticals, Ophthotech, and Regeneron. Dr Judy E. Kim serves on the Advisory Board of Thrombogenics; and receives unrestricted grant from Research to Prevent Blindness, Inc. and support for the study from Notal Vision through EMMES. Acknowledgment Emily Chew and Traci Clemons had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analyses. The investigators designed and executed the study in collaboration with the sponsors of the study. The analyses were performed independently in the Coordinating Center. The Data and Safety Monitoring Committees evaluated both the study design and the study data. The manuscript was drafted by the investigators with collaboration and input from the sponsors. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.cct.2014.02.003. References [1] Congdon N, O'Colmain B, Klaver CC, Klein R, Munoz B, Friedman DS, et al. Causes and prevalence of visual impairment among adults in the United States. Arch Ophthalmol 2004;122:477–85. [2] Friedman DS, O'Colmain BJ, Muñoz B, Tomany SC, McCarty C, de Jong PT, et al. Prevalence of age-related macular degeneration in the United States. Arch Ophthalmol 2004;122:564–72. [3] Pascolini D, Mariotti SP, Pokharel GP, Pararajasegaram R, Etya'ale D, Negrel AD, et al. 2002 global update of available data on visual impairment: a compilation of population-based prevalence studies. Ophthalmic Epidemiol 2004;11:67–115. [4] Ferris 3rd FL, Fine SL, Hyman L. Age-related macular degeneration and blindness due to neovascular maculopathy. Arch Ophthalmol 1984; 102:1640–2. [5] Brown DM, Kaiser PK, Michels M, Soubrane G, Heier JS, Kim RY, et al. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med 2006;355:1432–44.
[6] Rosenfeld PJ, Brown DM, Heier JS, Boyer DS, Kaiser PK, Chung CY, et al. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med 2006;355(14):1419–31. [7] IVAN Study Investigators, Chakravarthy U, Harding SP, Rogers CA, Downes SM, Lotery AJ, et al. Ranibizumab versus bevacizumab to treat neovascular age-related macular degeneration: one-year findings from the IVAN randomized trial. Ophthalmology 2012;119:1399–411. [8] Heier JS, Boyer D, Nguyen QD, Marcus D, Roth DB, Yancopoulos G, et al. The 1-year results of CLEAR-IT 2, a phase 2 study of vascular endothelial growth factor trap-eye dosed as-needed after 12-week fixed dosing. Ophthalmology Jun 2011;118(6):1098–106. [9] Brown DM, Heier JS, Ciulla T, Benz M, Abraham P, Yancopoulos G, et al. Primary endpoint results of a phase II study of vascular endothelial growth factor trap-eye in wet age-related macular degeneration. Ophthalmology 2011;118:1089–97. [10] CATT Research Group, Martin DF, Maguire MG, Ying GS, Grunwald JE, Fine SL, et al. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med May 19 2011;364(20):1897-908 [Epub 2011 Apr 28]. [11] Ying GS, Huang J, Maguire MG, et al. Baseline predictors for one-year visual outcomes with ranibizumab or bevacizumab for neovascular age-related macular degeneration. Ophthalmology 2013;120:122–9. [12] Kaiser PK, Brown DM, Zhang K, Hudson HL, Holz FG, Shapiro H, et al. Ranibizumab for predominantly classic neovascular age-related macular degeneration: subgroup analysis of first-year ANCHOR results. Am J Ophthalmol 2007;144:850–7. [13] Boyer DS, Antoszyk AN, Awh CC, Bhisitkul RB, Shapiro H, Acharya NR. Subgroup analysis of the MARINA study of ranibizumab in neovascular age-related macular degeneration. Ophthalmology 2007;114:246–52. [14] Fong DS, Custis P, Howes J, Hsu JW. Intravitreal bevacizumab and ranibizumab for age-related macular degeneration. A multi-center, retrospective study. Ophthalmology 2010;117(2):298–302. [15] Rauch R, Weingessel B, Maca SM, Vecsei-Marlovits PV. Time to first treatment — the significance of early treatment of exudative agerelated macular degeneration. Retina 2012;32:1260–4. [16] Westheimer G. Editorial: visual acuity and hyperacuity. Invest Ophthalmol 1975;145(8):570–2. [17] Loewenstein A, Ferencz JR, Lang Y, Yeshurun I, Pollack A, Siegal R, et al. Toward earlier detection of choroidal neovascularization secondary to agerelated macular degeneration: multicenter evaluation of a preferential hyperacuity perimeter designed as a home device. Retina 2010;30:1058–64. [18] Beck RW, Moke PS, Turpin AH, Ferris 3rd FL, San Giovanni JP, Johnson CA, et al. A computerized method of visual acuity testing: an adaptation of the Early Treatment Diabetic Retinopathy Study testing protocol. Am J Ophthalmol 2003;135(2):194–205. [19] The AREDS2-HOME Study Research Group. Randomized trial of a home monitoring system for early detection of choroidal neovascularization. Home Monitoring of the Eye (HOME) Study. Ophthalmology Nov 8 2013. http://dx.doi.org/10.1016/j.ophtha.2013.10.027 [pii: S01616420(13)00952-4, Epub ahead of print].
