Proof of Concept, Randomized, Placebo-Controlled Study of the Effect of Simvastatin on the Course of Age-Related Macular Degeneration

Background HMG Co-A reductase inhibitors are ubiquitous in our community yet their potential role in age-related macular degeneration (AMD) remains to be determined. Methodology/Principal Findings Objectives: To evaluate the effect of simvastatin on AMD progression and the effect modification by polymorphism in apolipoprotein E (ApoE) and complement factor H (CFH) genes. Design: A proof of concept double-masked randomized controlled study. Participants: 114 participants aged 53 to 91 years, with either bilateral intermediate AMD or unilateral non-advanced AMD (with advanced AMD in fellow eye), BCVA≥20/60 in at least one eye, and a normal lipid profile. Intervention: Simvastatin 40 mg/day or placebo, allocated 1∶1. Main outcome measures: Progression of AMD either to advanced AMD or in severity of non-advanced AMD. Results. The cumulative AMD progression rates were 70% in the placebo and 54% in the simvastatin group. Intent to treat multivariable logistic regression analysis, adjusted for age, sex, smoking and baseline AMD severity, showed a significant 2-fold decrease in the risk of progression in the simvastatin group: OR 0.43 (0.18–0.99), p = 0.047. Post-hoc analysis stratified by baseline AMD severity showed no benefit from treatment in those who had advanced AMD in the fellow eye before enrolment: OR 0.97 (0.27–3.52), p = 0.96, after adjusting for age, sex and smoking. However, there was a significant reduction in the risk of progression in the bilateral intermediate AMD group compared to placebo [adjusted OR 0.23 (0.07–0.75), p = 0.015]. The most prominent effect was observed amongst those who had the CC (Y402H) at risk genotype of the CFH gene [OR 0.08 (0.02–0.45), p = 0.004]. No evidence of harm from simvastatin intervention was detected. Conclusion/Significance Simvastatin may slow progression of non-advanced AMD, especially for those with the at risk CFH genotype CC (Y402H). Further exploration of the potential use of statins for AMD, with emphasis on genetic subgroups, is warranted. Trial Registration Australian New Zealand Clinical Trial Registry (ANZCTR) ACTRN1260500032065


INTRODUCTION
Age-related macular degeneration (AMD) is a progressive, late onset disease affecting central vision that is currently the leading cause of irreversible blindness in our community. Early signs of the disease are present in 15% of the population aged 50 years and older and by 90 years of age, more than two-thirds of the population are affected 1 .
The high prevalence of AMD, the anticipated increase in the ageing population and the very limited treatment options available highlight the urgency for further research and the implementation of a preventative strategy to retard the progression of this debilitating disease.
The pathogenesis of AMD remains largely unknown, however it is regarded as a genetic disorder where environmental risk factors impact on the genetic background.
The potential environmental risk factors are numerous however, smoking 2 is currently the only factor consistently associated with AMD. Other cardiovascular risk factors such as hypertension, atherosclerosis, high serum cholesterol and high dietary fat intake have also been linked to AMD although inconsistently [3][4][5][6] .
Recently, several pieces of evidence have further implicated cholesterol in AMD pathogenesis. Firstly, came the report that the cholesterol pathway gene apolipoprotein E (APOE) may play a role in AMD 7 and other degenerative diseases such as atherosclerosis 8 and Alzheimer's disease 9 . Secondly, the association of AMD with cardiovascular disease was reported in several studies.
Apolipoprotein E being a plasma lipoprotein plays a basic role in the degradation of particles rich in cholesterol and triglycerides. The polymorphic nature of the gene affects various diseases in different ways depending on which allele of the gene is present. That is, in cardiovascular disease the 2 allele is protective and the 4 allele a risk factor for disease 10 . In AMD, the 4 allele has been shown to be protective whilst the 2 allele was significantly associated with an increased risk of AMD 11 .
There are many postulated mechanisms by which atherosclerosis could play a part in the development of AMD. A proposed model for the development of AMD is that deficiencies in the choroidal vasculature leads to deterioration in the retinal pigment epithelium resulting from the build up of waste products 12  An overlap between the risk factors for cardiovascular disease and AMD have lead to the proposal that the two conditions may have a common aetiologic pathway. Components of cardiovascular disease such as hypertension and atherosclerosis have been studied in relation to AMD with inconsistent findings.
There have been several studies that have reported a significant association between hypertension and AMD [13][14][15] . Of particular interest are the findings from the AREDS group that reported an increased risk of large soft drusen or neovascular AMD in people with hypertension 15 .
The Rotterdam Study was the first population-based study to report a significant association between carotid atheroscleorsis determined by the presence of atherosclerotic plaques and late AMD 6 . This was supported by the findings of Chaine et al in 1998 who reported a significant association between coronary heart disease and advanced forms of AMD 16 .
Dietary fat intake may also influence the risk of developing age-related macular degeneration by raising the levels of cholesterol in the bloodstream thereby increasing the risk of atherosclerosis or alternatively, there may be an increase in the deposition of fat in Bruch's membrane that adversely affects the flow supply of nutrients and the removal of waste products from the retinal pigment epithelium 17 . In recent years both the Blue Mountains Eye Study 18 and the Beaver Dam Eye Study 19 have shown that higher intake of specific fats associated with higher cholesterol levels were significantly associated with the presence of AMD.
Data collected from our population-based study of eye diseases (Melbourne Visual Impairment Project) was used to examine whether the progression of early AMD was delayed in people taking cholesterol-lowering medications. Our results showed for the first time, that participants on cholesterol-lowering medications were almost four times less likely to experience progression of AMD than participants who did not, although the major limitation was the small number of people who actually took such medications 20 .
Further evidence to support our findings emerged from a survey of 379 men and women residing in Sheffield, England. The survey found that men and women aged 66-75 who took statins had an eleventh the risk of AMD compared to those not taking the drug 21 .
Given these findings, our aim is to assess the effect of a specific class of cholesterollowering medication 'statins' in reducing the progression of AMD in people with early signs of the disease. The rates of progression will be assessed in terms of APOE genotype to see if any correlation can be made.
People participating in this study will have fasting blood cholesterol levels below which are normally recommended to treat.

Endpoint
The endpoint of the study is the development of late stages AMD. If late AMD develops in both eyes, the participant exits the study and no further scheduled examinations will be required. If late AMD occurs in one eye, the fellow eye remains under observation and the follow up continues.

STUDY DESIGN
The study is a randomised, double-masked, controlled study. The study is based at the Royal Victorian Eye and Ear Hospital and Caulfield General Medical Centre. .
Participants are being randomised in a ratio 1:1, to receive either the active 'statin' or the placebo once they have completed both the ophthalmic and cardiovascular requirements outlined in the protocol.
Participants receive the same study treatment for the duration of their involvement in the study.

Sample size
In photo-based detection, the rate of progression of early AMD in people with high-risk features was found to be 5% per year in the AREDS study 22 . In the VIP project, AMD progression in people on cholesterol-lowering medications was 4 times lower than those not on any such medication 20 . To detect a similar reduction in AMD progression, a sample size of 118 subjects in each arm, followed for three years, was needed to reach 80% power at a 5% level of significance. In the more sensitive functional tests, a smaller sample size would be required. Thus, a sample size of 50 patients per treatment group would be sufficient to detect a 22% to 41% change in sensitivity for the different functional methods, or it will be sufficient to detect a 3.4 min difference in rod-cone break over the trial period, with statistical power of 80% .

Inclusion criteria
(1) male or female aged 50 years or older, (2) the ability to assess the macula in at least one eye, (3) VA better than or equal to 6/18 in the study eye(s). Note: the study eye is allowed to have non-central GA and/or non-neovascular PED (5) cholesterol level within the normal limits according to the person's medical history (6) not currently on any cholesterol-lowering medication.

Exclusion criteria:-
(1) Medical and ophthalmic conditions which potentially affects visual functionincluding visually significant cataract (as defined by WILMER's grading method-that is nuclear opacity score of 2.00 or greater, cortical opacity score of greater than 3, any posterior subcapsular cataract), history of diabetes and glaucoma (2) Use of medications which may affect visual function, such as plaquenil, chloroquine, major tranquilizers. Administration: Patients receive their study medication following the completion of their baseline assessment.
Unused tablets and empty jars are requested to be returned to the study centre for assessment of compliance.

Randomisation of study treatment.
Randomisation of study medication has been computer generated with 50% of participants being randomised to receive the placebo and 50% the 'statin'. Staff involved in the examination of participants are masked as to who receives the placebo or the active 'statin'.

Patient recruitment
Participants are being recruited from both public and private ophthalmic clinics.
Letters have been sent to all Victorian ophthalmologists (Appendix 1). The letter has outlined the aims and requirements of our research study and also contained a flyer (Appendix 2) that can be given to those patients who express an interest in participating. The flyer provided a brief outline of the study and what is involved for the participants as well as providing a contact number for the study centre. Also included may be a letter that patients can take to their GP explaining the need for a cholesterol test (Appendix 3).
Potentially suitable subjects are contacted by research staff, who outline the objectives of the study, the requirements for the participant and order a set of baseline blood tests (including fasting lipids, LFT, CRP and hsCRP). Questions are asked regarding cardiovacular disease and risk factors status, in order to interpret the test results. The participant are notified of the blood test result regardless of whether it is normal, including suggestion of further actions (such as seeing his/her GP). If the blood test result is within the acceptable range for recruitment in the study then further details of the study is discussed with the participant, and prerandomization/baseline visits are arranged.

Informed consent.
Upon arrival at the study centre, the purpose, aims and requirements of the study are being explained to the participant by a Study Investigator or Research Assistant. Participants are being given the Patient Information Sheet (Appendix 3) and have an opportunity to ask any questions before they are asked to sign the informed consent form (Appendix 4). This forms points out that the participant has received an explanation regarding the aims of the study that the requirements of the study have been outlined to them, that their participation is voluntary and they consent to the ophthalmic and cardiovascular examinations and to the taking of a blood sample by a trained staff member for cholesterol, liver function testing and for DNA analyses.
(A separate consent form needs to be signed specifically for the DNA analyses.) A separate agreement to our privacy policy has to be signed. 9

Study assessments
The assessments required for the study include:

Questionnaire
(1) Ethnic origins and family history

Allocation of randomised study treatment
Once participants have completed all baseline assessments they receive their either simvastatin 40mg or placebo. For tracking, each participant is allocated a study medication number which is to be recorded in the database as well as a log book. Each medication number will correspond to either active medication or placebo, which is unknown to the participant and the study team. Participants will keep the same medication number for the duration of their participation in the study.

STUDY PROCEDURES
Refraction and visual acuity will be performed for all participants by a trained vision examiner.

Visual acuity testing
Best-corrected visual acuity is measured using a retroilluminated ETDRS chart, using the EDTRS scoring system.
 Visual acuity is assessed with participant's most comfortable presenting correction  (monofocal for distance, bifocals or multifocals) if applicable.
 Participant is seated 4 metres from the ETDRS visual acuity chart.
 Conventionally, the right eye is examined first, with the left eye being occluded.
 The participant is asked to select the line they can read most comfortably and is then encouraged to continue until they cannot read any further with accuracy.
 If fewer than 20 letters are read, the chart is moved to 1 metre.
 If the participant cannot read the top line at 1 metre, vision is checked with a pinhole.
 The distance and chart used are then recorded.
 Scoring of visual acuity is worked out according to the number of letters read correctly for each eye.
 Procedure is repeated for the left eye.

Dilation of pupils
One drop of tropicamide (0.5%) and phenylephrine hydrochloride (10%) in each eye is administered for pupillary dilation.

Ophthalmic examination
Ophthalmic examination is done with a standard slit lamp to assess the retinal changes related to AMD.
It also allows assessment of any other abnormality of the ocular media and retina which may complicate assessment. Retinal photography is performed with a standard fundus camera.

Rod Psychophysics
Rod psychophysics is performed using the dark adaptometer. This utilizes a circular test target of white light, 2 degree size at 2 degree ecentricity. A small red fixation target is used. Initially there is a bleaching period of 5 minutes, after which all light is turned off. The participant is asked to fixate on the red target. The test target is made to appear at approximately 1 minute intervals. The participant is asked to indicate when he/she first sees the light, as soon as it appears, and as quickly as possible. The intensity of the light is then recorded.
The test is to be continued until the rod component has near "flattened out" in the last 2-3 readings, or if the rod-cone break has not been reached after 45 minutes of dark adaptation.
Total duration of test = 0.5 to 1 hour

Cone Psychophysics
Subjects are given practice at all tasks before beginning tests.
Subjects are given rests regularly and when required.
 Visual Acuity measurement: One eye is covered and the subject is asked to read the smallest line on a standard logMAR letter chart with the test eye (the subjects' current spectacles or best correction are worn

General adverse events relating to study assessments
 Blood test will be performed by trained staff, but may cause minor discomfort and possibly a small bruise at the site.
 Cardiovascular measurements are all non-invasive and are not associated with any pain or discomfort.

Adverse events relating to ophthalmic assessments.
 Participants will experience blurred vision for several hours following the administering of the dilating drops; this is often associated with greater sensitivity to sunlight. It would be our recommendation for participants to bring their own sunglasses with them.

Reporting and documentation of adverse events.
At each visit, research staff will question all participants regarding any adverse experiences particularly muscle soreness as an indicator of possible rhabdomyolysis. Any illness, sign, symptom or clinically significant laboratory test abnormality that appear or worsen during the course of the clinical trial will be documented and pursued regardless of the causal relation to the active 'statin' therapy. Participants would also be encouraged to report any problems as they occur to the study centre where all details would be documented and upon the Chief Investigator's discretion further action would be recommended to the participant.
(Adverse Event form located in Appendix 5).
Laboratory blood analyses and repeat liver function test carried out after 4 weeks will allow the monitoring of changes in cholesterol levels as well as liver enzymes.
The safety issue of the study would be evaluated by a Safety Officer appointed to review all the reported incidents of adverse events between the two groups.

Abnormal laboratory results.
Participants with laboratory results from the blood samples that appear outside the 'normal' range will be notified and a suitable course of action will be suggested such as the need to see their GP.

CARDIOVASCULAR ASSESSMENT
 blood pressure measurements.

Digital photography of the retina.
In this study, non-stereoscopic 45° macular photography using a digital Canon CR10-45NM Non-Mydriatic Retinal Camera with a resolution of 3072 x 2048 pixels will be used. Images will be viewed immediately and repeated if unsatisfactory. Images are then transferred onto a portable hard disk at the test site to allow them to be backed up on the central server at CERA each week, where additional image storage space has been purchased specifically for this project. Images will be graded using the "OptoMize Detailed results from grading will be manually entered on a specifically designed grading sheet and then entered into the custom-written MS Access database.
The grading classification is based on the anatomical definition of the macula i.e. that part of the retina centred on the fovea in which the ganglion cell layer is more than one cell in thickness and that has an appropriate diameter of 5.5mm. For descriptive purposes, the inner macula is defined as the area within a circle centred on the foveola of diameter 3000 microns or approximately two disk diameters across. The outer macula is defined as the area between the inner macula (diameter 3000 microns) and a circle of 6000 microns.
The presence or absence of any feature in the macular area (within 6000 microns diameter across the macula centred on the fovea) was documented on a specifically designed grading sheet.
To enable the study to examine the role of vitamin E on early AMD features all drusen types within the grid area, their maximal size, total number, and area covered within each grading circle were documented and subsequently entered into a database. Other than these two additions to the AMD International group's guidelines, the definitions and protocol were strictly adhered to. Reliability sessions will need to be carried out to assess intra-and inter-grader reproducibility. Graders remain masked to both the study treatment and cardiovascular health of all study participants. The reported agreement was based on the ability to correctly identify the presence or absence of all features thus an unweighted kappa statistic with 95% confidence interval was used as a measure of reliability.

APPENDIX 1: Letter to Ophthalmologists
Dear xxx,

Re: ARMSS (A new trial of Cholesterol lowering medication in AMD)
As part of the ongoing research into Age related Macular Degeneration, we are commencing a new study to examine the possible role of cholesterol-lowering medication ('statins') in slowing the progression of AMD. The impetus of this comes from results of 'Melbourne Visual Impairment Project' that has indicated for the first time, that participants on cholesterol-lowering medications were almost four times less likely to experience progression of AMD than participants who did not ( This study is a collaborative project between Centre for Eye Research Australia and the Department of Optometry, The University of Melbourne. The study is a 3-year, randomised trial that would involve participants undergoing a fundus examination and a cardiovascular assessment. 50% of participants will receive the cholesterol-lowering medication and the other 50% will receive a placebo. Participants need to have normal lipid levels and cannot be on lipid lowering medication for this study. We are therefore very keen to recruit potential participants with high risk fundus changes of AMD (see below for definition) and therefore seek your assistance in identifying suitable patients who may be interested in participating in either of our studies. We have included an information sheet for your patients and a letter they can take to their GP to arrange a fasting blood lipid test as this result will determine which study they could enter.
We hope that you will be able to copy these sheets for interested patients. Please contact Nicola Hunt, on 9929 8360 regarding particular patients so a suitable time can be arranged to see them in a study clinic, or write to us with the patients details and we will be very happy to contact them.
Many thanks for taking the time to assist us with our research into AMD. Will the cholesterol lowering medication "statins" slow the progression of AMD?
Age-related macular degeneration (AMD) is the leading cause of legal blindness in the elderly population. You have some age-related changes in your eyes that put you at significant risk of progressing to the late complications of AMD resulting in loss of central vision in one or both eyes (one eye may already have this complication). At present there is no effective way to reduce your risk of this progression except for giving up smoking if you smoke. Otherwise eating a good diet of green leafy vegetables and possibly corn may help but we have no proven treatments to reduce the risk of progression of this disease.
We have some early results that suggest that cholesterol-lowering medication may have a role in reducing the risk of AMD progression. These are only very preliminary results and have proven nothing at this stage. However, we are hopeful that these medications might reduce the amount of fat in the back of the eye (irrespective of your blood cholesterol level) and thereby lower the risk of severe vision threatening complications. It is the accumulation of lipid or fatty deposits behind the retina that is thought to lead to bleeding or degeneration that occurs as a complication of AMD.
As a result of our preliminary study we commenced a 3 years study to test whether statins, a family of cholesterol lowering medication, can effectively slow the progression of early AMD. Over 900 people will be enrolled and half will take the active drug whilst the other half will be on a placebo (no active drug). Both drugs will look identical so neither yourself nor the investigating team will know which group you are in for the 3 years of the study.
You will be asked to do a fasting cholesterol test, and if it is within a certain normal range you would be a suitable candidate for this study. If your cholesterol level is high then you should discuss this with your doctor and you cannot be included in the study as you might be selected to the placebo group which would not be right if your cholesterol was at a level that needed attention. Regardless of your cholesterol level, you will be notified of the result and what action to take.
The medication for the study (Statins) has been used for many years to lower the risks of heart disease, and is generally very well tolerated. People in the study will take the medication for 3 years. Over this time, monitoring would involve examinations every 6 months. Each set of examinations will involve going to either Caulfield General Medical Centre in Caulfield, or Royal Victorian Eye and Ear Hospital in East Melbourne). The thorough examination ensures that we monitor not only the physical appearance in the eyes but also the visual function.
The finer details of the study will be discussed with you should you become eligible, and agree to enter the study.  This study is a 3-year randomised controlled trial assessing the role of Simvastatin as a cholesterollowering medication, in the progression of early AMD. Participants in this study will receive an ophthalmic examination that includes photographs of the macular area. Participants enrolled in this study will be randomly assigned to the active or placebo arm of the trial. To take part in this study the participants must have lipids levels that are within a range that treatment is not recommended, as half will be on placebo.
Your patient has been identified by their ophthalmologist as having signs of AMD that fit our inclusion criteria and has shown interest in taking part in one of our research studies. We need to know what the fasting blood lipid levels are to determine which study they can take part in. Therefore, we would appreciate if you could organise for your patient to have their fasting cholesterol level assessed. (We require: total cholesterol level, HDL and LDL-cholesterol and triglyceride level). Or provide them with a copy of a recent test (with in the last month) so that they can bring it to their initial study visit.
Should you require further information please feel free to contact me on 9929 8393 or 0418618227.
Many thanks for your time and assistance.

Robyn Guymer
Head of Age-related Macular Degeneration Unit CERA. The chief investigator is Dr. Robyn Guymer who is an ophthalmologist at CERA.
AMD is currently an irreversible disease of the eye whereby the area affected is located at the centre of the retina referred to as the macula. As the name implies the disease is degenerative and increases in frequency with age, especially in people over 65 years of age. The early stages of the disease is characterised by the presence of lipid deposits in the retina known as "drusen" that are not uncommon in people aged over 40 years.
The progression of AMD may eventually lead to the loss of central vision that results in everyday activities such as reading, driving, writing and recognition of facial features being impaired.
To date, there is very limited effective treatment available for the many people affected with AMD. Therefore, it is important to find some way of curbing the increasing number of people who are affected. Many studies have looked at risk factors such as antioxidant status, environmental factors and a family history of the disease in an attempt to understand the way the disease develops. However, the only consistent factors associated with the disease have been smoking and a genetic link.
There is clear evidence that AMD is an inherited disease, however the actual genes responsible remain elusive. Therefore, part of this study will involve examining participants to search for genes involved in AMD. One gene of current interest is the APOE gene that is involved in cholesterol metabolism.
Rationale for the study.
The data obtained from a population-based study that was recently completed in our department, showed for the first time that the progression of AMD might be slowed in people who had reported taking cholesterol-lowering medications. As a result of this finding we have embarked upon this current study.

Objective of the study.
This study is designed to assess the affect of cholesterol-lowering medications on the progression of AMD. The study is a randomised trial, which means that 50% of participants will receive the active medication and 50% will receive a placebo. Neither you the participant, nor we the investigators, will be aware of which you will be receiving and this will be the situation until the study is completed. You will also be asked to participate in the cardiovascular component of the study that will assess the changes to your arteries over the first month whilst you are taking the study medication.

Possible side effects of statins.
Statins have been reported to be generally well tolerated in numerous studies however there are some side-effects that have been associated with this medication. These include: muscle pain and weakness, rash, mild and temporary headache and an increase in liver function. These side effects generally occur in only a small number of people taking such medication. For the purpose of this study, we will be monitoring reports of such effects to ensure your health is not jeopardized.
What the study involves for you as a participant.
As a participant in this study we would require you to attend Caulfield General Medical Centre and the Royal Victorian Eye and Ear Hospital (RVEEH).
Your eyes will be examined and questionnaires answered. The visual component will involve the assessment of your vision followed by the administering of drops that will dilate the pupils of your eyes. This will enable the ophthalmologist to examine the back of the eye where the macula is located and to take photographs of your eyes that will later be used to detect signs of AMD. We will also need to take a fasting blood sample from you to determine your cholesterol level and for DNA analyses.
You will undergo psychophysical tests (tests which measure subtly how well you can see). These tests are non-invasive. However, they will once again require the use of eye drops which will dilate your pupils.
Once all these examinations have been completed you will receive the study medication that you are to take for the duration of your participation in this study.
Four weeks after your initial examination, you will be required to undertake a blood test to assess any side effects of the study medication. Three months after the initial visit and annually thereafter we will require you to return to the study centres (Caulfield, RVEEH and VCO) for the duration of the study that will be 3 years. These appointments will be organised by the study staff.
We thank you for your time and assistance in helping us undertake this important research. Please let us know if you have any problems with transport. For more details or any questions please contact Dr Robyn Guymer or Ms Nicola Hunt on (03) 9929 8360.

APPENDIX 8: Letter to participant on commencing medication
The study medication is either simvastatin, at a dose of 40mg, or placebo.

Instruction
Take two tablets in the evening with food. Avoid drinking grapefruit juice within 4 hours of taking the medication as this may interfere with absorption.
Severe side effects are rare, but seek medical advice promptly if you experience muscle pain, tenderness or weakness, dark urine, or yellowing of skin or the whites of your eyes.
Please give the enclosed "information for GP" to your local doctor. Please remind him/her that you are now in this trial, especially when he is about to prescribe any drug (particularly antibiotics), as there are a few but important interactions.
Please call Dr Luba Robman on 92766175, or Dr Robyn Guymer on 99298360 if you have any questions regarding the medication or the trial.