GSK Clinical Study Register

Transcription

1 In February 2013, GlaxoSmithKline (GSK) announced a commitment to further clinical transparency through the public disclosure of GSK Clinical Study Reports (CSRs) on the GSK Clinical Study Register. The following guiding principles have been applied to the disclosure: Information will be excluded in order to protect the privacy of patients and all named persons associated with the study Patient data listings will be completely removed* to protect patient privacy. Anonymized data from each patient may be made available subject to an approved research proposal. For further information please see the Patient Level Data section of the GSK Clinical Study Register. Aggregate data will be included; with any direct reference to individual patients excluded *Complete removal of patient data listings may mean that page numbers are no longer consecutively numbered

2 CONFIDENTIAL 2012N137612_00 GlaxoSmithKline Group of Companies Division: Worldwide Development Retention Category: GRS019 Information Type: Meta-Analysis Results Report Title: Meta-Analysis Results Report for, compound CCI22890: A Meta-Analysis to compare the efficacy of acyclovir 3% ophthalmic ointment to idoxuridine in curing herpes keratitis on Day 7 of treatment Compound Number: CCI22890 Effective Date: 09-APR-2012 Description: Acyclovir 3% ophthalmic ointment has been evaluated for the treatment of herpes simplex keratitis (HSK) in many clinical trials and has demonstrated healing rates superior to idoxuridine ophthalmic ointment [Wilhelmus, 2009]. Idoxuridine is currently approved in the United States for the treatment of herpes simplex keratitis. The purpose of this meta-analysis was to demonstrate that acyclovir 3% ophthalmic ointment is superior to idoxuridine in treating herpetic keratitis as well as dendritic and geographic ulcer sub-types. There were seven studies of patients with herpetic keratitis included in the meta-analysis. From all cases of herpetic keratitis (N=432) included in the meta-analysis, 185 cases were identified as dendritic ulcers and 35 cases were identified as geographic ulcers. There was not sufficient information to classify the remaining cases as any specific ulcer sub-type. In the acyclovir group, the odds of healing at day 7 for patients with herpetic keratitis was 3.95 times of the odds in the idoxuridine group (95% confidence interval (CI): 2.60, 6.00; p < ). On average, the day 7 healing rate was 79% (95%CI: 67%, 87%) for acyclovir and 47% (95%CI: 34%, 61%) for idoxuridine. In the acyclovir group, the odds of healing at day 7 for patients with a dendritic ulcer was 4.22 times of the odds in the idoxuridine group (95% CI: 2.14, 8.32; p < ). On average, the day 7 healing rate was 87% (95% CI: 59%, 97%) for acyclovir and 56% (95% CI: 24%, 83%) for idoxuridine. Copyright 2012 The GlaxoSmithKline group of companies. All rights reserved. Unauthorised copying or use of this information is prohibited. 1

3

4 CONFIDENTIAL 2012N137612_00 Table of Contents TITLE PAGE... ABBREVIATIONS INTRODUCTION OBJECTIVE(S) INVESTIGATIONAL PLAN Design Selection of Population Colin J, Tournoux A, Chastel C, Renard, G. Superficial herpessimplex keratitis. Double-blind comparative trial of acyclovir and idoxuridine. Nouv Presse Med. 1981;10(36): Collum LM, Benedict-Smith A, Hillary IB. (1980). Randomised double-blind trial of acyclovir and idoxuridine in dendritic corneal ulceration. Br J Ophthalmol. 1980;64: Coster DJ, Wilhelmus KR, Michaud R, et al. A comparison of acyclovir and idoxuridine as treatment for ulcerative herpetic keratitis. Br J Ophthalmol. 1980;64: Professors and - Brief Summary of French report of a double blind clinical trial comparing acyclovir and idoxuridine in herpetic keratitis: Kitano S. Efficacy of acyclovir in dendritic keratitis: A double blind comparative study with IDU. Therapeutic Research 1985;2: Klauber A, Ottovay E. acyclovir and idoxuridine treatment of herpes simplex keratitis - A double blind clinical study. Acta Ophthalmologica.1982;60: McCulley JP, Binder PS, Kaufman HE, et al. A double-blind, multicenter clinical trial of acyclovir vs. idoxuridine for treatment of epithelial herpes simplex keratitis. Ophthalmology. 1982;89: Treatments Statistical Analyses RESULTS DISCUSSION CONCLUSIONS REFERENCES... EFFICACY DATA SOURCE FIGURES... Figure 1 Meta Analysis Result with CMH Method (Herpetic)... Figure 2 Funnel Plot (Herpetic)... Figure 3 Meta Analysis Result with CMH Method (Dendritic)... Figure 4 Meta Analysis Result with CMH Method (Geographic)... Figure 5 Meta Analysis Result with CMH Method (Herpetic, with Collum 1980 Excluded)... Page

5 CONFIDENTIAL 2012N137612_00 Figure 6 Meta Analysis Result with CMH Method (Dendritic, with Collum 1980 Excluded)... EFFICACY DATA SOURCE TABLES... Table 1 Summary of Data (Herpetic)... Table 2 Summary of Data (Dendritic)... Table 3 Summary of Data (Geographic)... Table 4 Summary of Meta Analysis Result by Population with Cochran-Mantel- Haenszel Model... Table 5 Summary of Cure Rate Estimation by Population with Non-Linear Mixed Effect Model... Table 6 Summary of Data (Herpetic, with Collum 1980 Excluded)... Table 7 Summary of Data (Dendritic, with Collum 1980 Excluded)... Table 8 Summary of Meta Analysis Result with CMH Model by Population (without Collum 1980)... Table 9 Summary of Cure Rate Estimation by Population with NLME Model (without Collum 1980)... Table 10 Sensitivity Analysis Summary for Homogeneity Test of Meta Analysis with Cochran-Mantel-Haenszel Model

6 CONFIDENTIAL 2012N137612_00 ABBREVIATIONS ACV BDT CI CMH CRF FDA GSK HSK IDU MESH NLME PAD UK acyclovir Breslow-Day-Tarone test Confidence Interval Cochran-Mantel-Haenszel method Case Report Form Food and Drug Administration GlaxoSmithKline herpes simplex keratitis Idoxuridine Medical Subject Headings Non-Linear Mixed Effect model Post Approval Document United Kingdom Trademark Information Trademarks of the GlaxoSmithKline group of companies ZOVIRAX Trademarks not owned by the GlaxoSmithKline group of companies None 5 4

7 CONFIDENTIAL 2012N137612_00 1. INTRODUCTION Acyclovir (ACV) 3% ophthalmic ointment has been evaluated for the treatment of herpes simplex keratitis (HSK) in many clinical trials and has demonstrated healing rates superior to idoxuridine ophthalmic ointment [Wilhelmus, 2009]. Idoxuridine (IDU) is currently approved in the United States for the treatment of herpes simplex keratitis. Herpetic keratitis can involve all parts of the cornea. The main forms of herpetic keratitis lesions are (a) dendritic and geographic ulceration (often referred to collectively as epithelial herpetic keratitis or superficial herpetic keratitis) and (b) stromal keratitis. All cases of herpetic keratitis included in the meta-analysis were either dendritic or geographic ulcer sub-types. The purpose of this meta-analysis was to demonstrate that acyclovir 3% ophthalmic ointment is superior to idoxuridine in healing herpetic keratitis on Day 7 of treatment by combining the results of available double masked, randomized head-to-head trials comparing these two medicines. A secondary objective was to evaluate these data by sub-type of ulcer and show that acyclovir was effective in these sub-types independently. 2. OBJECTIVE(S) The primary objective of this meta-analysis was to compare the number of patients with herpetic keratitis healed by Day 7 of treatment for acyclovir 3% ophthalmic ointment and idoxuridine. Secondary objectives were to compare the efficacy of acyclovir ophthalmic ointment to idoxuridine in healing sub-types of herpetic keratitis by Day 7 of treatment. The null hypothesis was that the efficacy of acyclovir ophthalmic ointment was similar to idoxuridine and the alternative was that it was superior to idoxuridine. 3. INVESTIGATIONAL PLAN 3.1. Design Using the Ovid search engine for the Ovid MEDLINE database through April 2011, combining medical subject headings (MESH) Keratitis, Herpetic/ AND Acyclovir/ resulted in 400 peer reviewed publications. Further limiting these searches by the key words topical OR ointment resulted in 150 articles. Of these, 125 pertained to human subjects. Similarly, when the 400 articles were restricted to MESH Administration, Topical/ OR Ointments/, 80 articles remained. Limiting these 80 papers to human subjects yielded 67 papers. The total number of unique papers between these sets yielded

8 CONFIDENTIAL 2012N137612_00 Attention was then placed on selected review articles including the 2009 Cochrane report [Wilhelmus, 2009] Therapeutic interventions for herpes simplex virus epithelial keratitis ) and the internal GSK 2007 Post Approval Document (PAD) Clinical Overview for ZOVIRAX Ophthalmic ointment. The reference list from these documents was cross referenced with the above set of 132 papers to identify 47 additional articles discussing the clinical use of acyclovir ophthalmic ointment that were not captured in the primary literature search. The total number of peer reviewed publications reporting on the clinical use of acyclovir ophthalmic ointment was 179. An extensive review of internal GSK document archives was performed. GSK was able to locate 13 clinical documents archived from the acyclovir program; an additional 10 were destroyed in a fire. After extensive searching, no source data, such as case report forms (CRFs), were located. In some instances, patient listings are available in documents but cannot be source-verified. Randomized, double-masked studies in patients with herpetic keratitis with head to head comparator arms of ACV ophthalmic ointment and topical IDU were then selected from the 179 publications and from the 13 internal clinical documents identified in the search of the GSK archives. From these, those that had actual or calculable healing rates at Day 7 were used for the primary efficacy analysis. Ultimately, seven study reports (6 publications and 1 internal summary report) were identified for the meta-analysis to support primary efficacy. Four of the published study reports had corresponding internal documentation. In three instances, these were summaries produced from the publications and previously submitted with Burrough s Wellcome UK Submission of acyclovir in 1981(Table 1). In the fourth instance, this was a clinical/statistical report that corresponded to [McCulley, 1982]. When the publication [McCulley, 1982] was compared to the internal report, efficacy information in the publication was consistent and complete, and therefore the publication was used as the source in the meta-analysis report. 7 6

9 CONFIDENTIAL 2012N137612_00 Table 1 Publications/Studies/Study Reports of Controlled Clinical Studies Pertinent to the Claimed Indication Herpetic Keratitis (refer to m ) Colin J, Tournoux A, Chastel C, Renard, G. Superficial herpes-simplex keratitis. Double-blind comparative trial of acyclovir and idoxuridine. Nouv Presse Med. 1981;10 (36): (original in French; English translation: Colin, 1981a). Corresponding GSK archival document: UK Submission Trial 7 summary results- France; A double blind randomised coded comparative trial of acyclovir and IDU in patients with dendritic herpetic corneal ulceration. Collum LM, Benedict-Smith A, Hillary IB. Randomised double-blind trial of acyclovir and idoxuridine in dendritic corneal ulceration. Br J Ophthalmol. 1980;64: Corresponding GSK archival document: UK Submission Trial 3 summary results, Randomised coded double blind trial of acyclovir and IDU in herpes simplex dendritic ulceration of the eye. Coster DJ, Wilhelmus KR, Michaud R, et al. A comparison of acyclovir and idoxuridine as treatment for ulcerative herpetic keratitis. Br J Ophthalmol. 1980;64: Corresponding GSK archival document: UK Submission Trial 2 summary results, Double-blind randomised coded comparative trial of acyclovir ophthalmic ointment vs topical IDU therapy in patients with dendritic or geographic herpetic ulcers Professors and - Brief Summary of French report [GSK Document Number BKCT/82/0002/00] (no corresponding publication of these data has been found) Kitano S. Efficacy of acyclovir in dendritic keratitis: A double blind comparative study with IDU. Therapeutic Research. 1985;2: (original in Japanese; English translation: Kitano, 1985a). Klauber and Ottovay. Acyclovir and idoxuridine treatment of herpes simplex keratitis - A double blind clinical study. Acta Ophthalmologica. 1982;60: McCulley JP, Binder PS, Kaufman HE, et al. A double-blind, multicenter clinical trial of acyclovir vs idoxuridine for treatment of epithelial herpes simplex keratitis. Ophthalmology. 1982;89: Corresponding GSK archival document: Study BW248U-014; THRS/88/0017 Following the completion of this review, the 2010 update to the Cochrane Review [Wilhelmus, 2010] was published. It was analyzed for significant changes or revisions to the information included for acyclovir; none were determined to have an impact on the outcome of the original search or the meta-analysis plan or results, or any conclusion based on information from the 2009 document [Wilhelmus, 2009]. Seven-day healing rates, defined as proportion of study participants (patients) healed at seven days after study entry, were chosen as the primary outcome measure for this analysis. The choice of Day 7 was influenced by the disease evolution over time and its use as a common timeframe in head to head clinical trials both as presented in the published literature and the fact that resolution of herpetic keratitis in seven days is a favorable outcome and a reasonable therapeutic goal [Wilhelmus, 2009]. 8 7

10 3.2. Selection of Population CONFIDENTIAL 2012N137612_00 Table 1 described the 6 publications and 1 internal report that were included in the metaanalysis, and identified the GSK internal documents that correspond to 4 of the publications. A summary of each study can be found in Section 4.1 of the Clinical Overview (refer to m2.5). All 7 publications/reports only enrolled patients with herpetic keratitis who had either dendritic or geographic ulcer sub-types. When interpreting the publications/reports the following two terms were used interchangeably: superficial herpetic keratitis and epithelial herpetic keratitis. Additionally, the meta-analysis did not include patients that were excluded in the final analysis by the publication or report author. The definition of cure and healing in the selected publications are provided in Table 2. Table 2 Definition of Endpoints and Cure/Healing of the Publications/Report Pertinent to the Claimed Indication Article Colin 1981 Collum 1980 Coster (GSK Document Number BKCT/82/0002/00) Kitano 1985 Klauber 1982 McCulley 1982 (GSK Document Number THRS/88/0017) Definition Endpoints: Number of cures during treatment period, mean time to cure. Definition of cure: the absence of epithelial ulceration after instillation of fluorescein using biomicroscopic examination. Endpoints: Days to heal and total number of patients healed. Definition of healing: Ulcers were considered to have healed when there was no fluorescein uptake Endpoints: Treatment failures, number of days it took ulcers to heal. Definition of healing: no epithelial defect demonstrable with Rose Bengal and fluorescein staining Endpoints: cumulative cure rate, average healing time. Definition of healing: Resolution of the ulcer based on fluorescein staining (inferred). Endpoints: comparing treatment efficacy between study groups. Definition of healing: Based on ulcer appearance over time; reported by categories, as follows: Excellent (ulcer disappeared within 7 days), Good (ulcer disappeared within 14 days or reduced by 50% within 7 days), Fair (ulcer reduced by 50% within 14 days), None (ulcers not reduced by 50% within 14 days). Endpoints: Cumulative healing rates. Definition of cure: The keratitis was judged to have been cured, at the time when there were no demonstrable epithelial defects by fluorescein and Rose Bengal and with the disappearance of the stromal affection and injection Endpoints: Corneal epithelial healing is the primary measurement of efficacy. Definition of healing: the observation of the absence of fluorescent staining and/or faint, segmented staining ("ghost figures") in the area of previous corneal ulceration within the therapy period without recrudescence. 9 8

11 CONFIDENTIAL 2012N137612_00 Some of the studies did not provide exact numbers for day 7 healing or cure. As a consequence, the data from articles without exact values were extracted based on tables and figures in the articles. Where Figures were used, a ruler was used and replicated by 2 or more individuals to confirm results. As per the GSK Meta-Analysis Plan, where a patient did not provide data on Day 7 of treatment: If the patient provided data showing the disease status before the completion of day 7, the same disease status was used as his/her disease status for Day 7 of treatment; If the patient did not provide any data before Day 7, but provided data showing disease status after that, failure of cure was used for the patient at Day 7; If the patient was enrolled but did not provide any data on disease status, the patient was excluded from meta-analysis. The details of data extraction for Day 7 healing or cure are described below for each study Colin J, Tournoux A, Chastel C, Renard, G. Superficial herpessimplex keratitis. Double-blind comparative trial of acyclovir and idoxuridine. Nouv Presse Med. 1981;10(36): Data were available directly from the second paragraph of the results section in [Colin, 1981]. Patients were diagnosed with epithelial herpes simplex keratitis (dendritic or dendrito-geographic corneal ulceration). There were insufficient data to sub-type patients for the secondary analysis. Table 3 Colin 1981 day 7 healing counts Colin 1981 Acyclovir Idoxuridine Cured Total Cured Total Herpetic keratitis No data by sub-type Collum LM, Benedict-Smith A, Hillary IB. (1980). Randomised double-blind trial of acyclovir and idoxuridine in dendritic corneal ulceration. Br J Ophthalmol. 1980;64: These data are available directly from Table 2 and Table 3 in [Collum, 1980]. There are multiple sections of the text that describe the lesion type as dendritic ulceration. Thus, this group, in its entirety, was treated as dendritic ulcers in the secondary analysis. As indicated in the article, although 30 patients per arm were enrolled into the trial, one patient in the idoxuridine treated group failed to attend for follow-up and did not provide any data. This patient was excluded from the analysis per the GSK Meta-Analysis Plan and is not included in Table

12 CONFIDENTIAL 2012N137612_00 Table 4 Collum 1980 day 7 healing counts Collum 1980 Acyclovir Idoxuridine Cured Total Cured Total Herpetic keratitis Dendritic ulcers Coster DJ, Wilhelmus KR, Michaud R, et al. A comparison of acyclovir and idoxuridine as treatment for ulcerative herpetic keratitis. Br J Ophthalmol. 1980;64: The total numbers of patients by treatment in each ulcer type are available in Table 1 of [Coster, 1980] and the numbers cured at day 7 had to be extracted from the cumulative time course plots of healing, i.e. Figure 2 and Figure 3 from [Coster, 1980]. The numbers of patients healed at different days are tabulated in Table 5. Although 30 patients were enrolled per arm, One patient treated with acyclovir failed to present regularly for follow-up, though he responded favourably in that his geographic ulcer had healed when he returned 10 days after beginning therapy. As this patient was not included in the analysis in the publication, he was also excluded from the GSK analysis. For herpetic keratitis treated with acyclovir, 27 out of 29 were cured at day 6. Day 7 data is not available for the acyclovir group on the figure. At day 8, 28 out of 29 were cured. Therefore, per the missing data rules, 27 out of 29 was used as the day 7 healing rate for the acyclovir group. Twenty-eight of these patients with herpetic keratitis had dendritic ulcers and one had geographic ulcer. A similar calculation was performed for dendritic ulcers treated with acyclovir and 27 out of 28 were cured at day 7. For geographic ulcer, 0 out of 1 was cured at day 7. For patients with herpetic keratitis treated with idoxuridine, since a total of 21 patients were cured at day 6, and 19 of these were dendritic ulcers, it is concluded that 2 patients with geographic ulcers were cured at day 6. For day 7, 22 dendritic ulcers were reported healed, no additional reports of healed ulcers were provided, and therefore the number of healed geographic ulcers remained 2 for day 7. In summary, for patients with herpetic keratitis treated with idoxuridine the number of herpetic keratitis ulcers cured at day 7 was 24 out of 30. For dendritic ulcers, 22 out of 26 were cured at day 7. For geographic ulcers, 2 out of 4 were cured at day

13 CONFIDENTIAL 2012N137612_00 Table 5 Number of patients healed based on Figure 2 and Figure 3 of [Coster, 1980] Treatment Group Day 6 Day 7 Day 8 Day 9 Day 13 Day 14 Day 25 Total Acyclovir Herpetic keratitis Dendritic ulcers Geographic ulcers Idoxuridine Herpetic keratitis Dendritic ulcers Geographic ulcers Note: Shaded cells were derived from other cells, unshaded cells were determined directly from figure 2 and figure 3 of [Coster, 1980] Table 6 Coster 1980 day 7 healing counts Coster 1980 Acyclovir Idoxuridine Cured Total Cured Total Herpetic keratitis Dendritic ulcers Geographic ulcers Professors and - Brief Summary of French report of a double blind clinical trial comparing acyclovir and idoxuridine in herpetic keratitis: No corresponding publication of these data was found. The study enrolled patients with epithelial herpetic keratitis. Total number of patients is available from Table 1 in [GlaxoSmith K Document Number BKCT/82/0002/00] and the numbers of patients cured at day 7 were extracted based on Figure 1 the cumulative frequency plot in the report. Although the word dendritic was used in the abstract, the enrolment states superficial herpetic keratitis, and geographic ulceration was not an exclusion criterion. Therefore, the disease type in this study appears to be epithelial or superficial keratitis generally, and not dendritic ulcer only. No breakdown is provided based on dendritic ulcer or geographic ulcer sub-type in this report, and hence was excluded from the sub-type analysis. Table day 7 healing counts 1982 Acyclovir Idoxuridine Cured Total Cured Total Herpetic keratitis No data by sub-type 12 11

14 CONFIDENTIAL 2012N137612_ Kitano S. Efficacy of acyclovir in dendritic keratitis: A double blind comparative study with IDU. Therapeutic Research 1985;2:643 7 Data are available directly from Table 3 in [Kitano, 1985] for patients with herpetic keratitis. There was not sufficient information to separate data for dendritic or geographic ulcers, hence was excluded from the sub-type analysis. Table 8 Kitano 1985 day 7 healing counts Kitano 1985 Acyclovir Idoxuridine Cured Total Cured Total Herpetic keratitis No data by sub-type Klauber A, Ottovay E. acyclovir and idoxuridine treatment of herpes simplex keratitis - A double blind clinical study. Acta Ophthalmologica.1982;60: The total number of patients was available in Table 1 of [Klauber, 1982] and the number cured at day 7 was extracted based on Figure 1 and Figure 2, the cumulative frequency plots in the article. Table 9 Klauber 1982 day 7 healing counts Klauber 1982 Acyclovir Idoxuridine Cured Total Cured Total Herpetic keratitis Dendritic ulcers Geographic ulcers McCulley JP, Binder PS, Kaufman HE, et al. A double-blind, multicenter clinical trial of acyclovir vs. idoxuridine for treatment of epithelial herpes simplex keratitis. Ophthalmology. 1982;89: This article was supported by a final medical/statistical report [GlaxoSmith Kline Document Number THRS/88/0017]. The total numbers of patients by treatment for each ulcer sub-type were available in Table 1 of [McCulley, 1982]. Based on Figure 3 in the article, at day 7 the total number of herpetic keratitis patients cured was 19 of 30 for acyclovir and 18 of 34 for the idoxuridine group. The results were also subset by ulcer sub-type in the article. For dendritic ulcers, based on Figure 1 in the article, at day 7 the total number cured was 16 for acyclovir and 17 for idoxuridine

15 CONFIDENTIAL 2012N137612_00 Based on Figure 2 in the article at day 7, 3 out of 4 patients with geographic ulcers were cured in the acyclovir group and 1 out of 8 patients with geographic ulcer was cured in the idoxuridine group. Table 10 McCulley 1982 day 7 healing counts McCulley 1982 Acyclovir Idoxuridine Cured Total Cured Total Herpetic keratitis Dendritic ulcers Geographic ulcers Treatments All trials in the selected publications were double-masked, randomized and used idoxuridine as the active comparator. The details of treatment information from each study are listed in Table

16 CONFIDENTIAL 2012N137612_00 Table 11 Summary of treatment information from publications included in the meta-analysis Study Acyclovir Idoxuridine Drug Dosing schedule Drug Dosing schedule strength strength Colin 1981 Acyclovir ophthalmic ointment 3% Five times a day for 7 days Idoxuridine ophthalmic ointment 0.5% Five times a day for 7 days Collum 1980 Acyclovir ophthalmic ointment 3% Coster 1980 Acyclovir ophthalmic ointment 3% 1982 Acyclovir ophthalmic ointment 3% Kitano 1985 Acyclovir ophthalmic ointment 3% Klauber 1982 Acyclovir ophthalmic ointment 3% McCulley 1982 Acyclovir ophthalmic ointment 3% Five times a day. If no improvement by day four, patient was withdrawn. Otherwise, drug was continued until epithelium was healed. Five times a day until lesion was healed, then three times a day for three days Not reported. Five times a day for 2 weeks, except dose reduced to three times a day after healing observed. Five times a day until significant healing noted, then the frequency of administration was reduced to three times daily and therapy was stopped when the keratitis healed. Five times a day for 14 days Idoxuridine ophthalmic ointment 0.5% Idoxuridine ophthalmic ointment 1% Idoxuridine ophthalmic ointment 0.5% Idoxuridine ophthalmic ointment 0.25% Idoxuridine ophthalmic ointment 0.5% Idoxuridine ophthalmic ointment 0.5% Five times a day. If no improvement by day four, patient was withdrawn. Otherwise, drug was continued until epithelium was healed. Five times a day until lesion was healed, then three times a day for three days Not reported. Five times a day for 2 weeks, except dose reduced to three times a day after healing observed. Five times a day until significant healing noted, then the frequency of administration was reduced to three times daily and therapy was stopped when the keratitis healed. Five times a day for 14 days 3.4. Statistical Analyses The formal meta-analyses were performed in a sequential manner to control the overall type I error rate at 5%: data from patients with herpetic keratitis were analyzed first. Based on a sequential hypothesis testing, the analysis with data from patients with dendritic and geographic ulcers was further performed at the 2.5% level to maintain the overall type I error rate at 5% level

17 CONFIDENTIAL 2012N137612_00 To compare the efficacy of acyclovir to idoxuridine, the meta-analysis using the regulatory agency agreed Cochran-Mantel-Haenszel (CMH) method [Cochran, 1954; Mantel, 1959] controlling for study was performed as the primary analysis (FDA minutes dated 22 Mar 2011 under IND 93,923) using SAS v9. Point estimates and the corresponding 95% confidence intervals (CIs) of the estimated common odds ratio across studies were provided together with the CMH test statistics and p-value for each ulcer type. To estimate the efficacy of acyclovir and of idoxuridine, the non-linear mixed effect (NLME) model (logistic regression model, based on maximum likelihood method) fitting treatment as a fixed effect and study as a random effect, was performed with SAS v9. Point estimates and the corresponding 95% CIs for the log odds of Day 7 healing rate for acyclovir and idoxuridine were derived. The log odds and the corresponding 95% CIs were then back calculated to the scale of odds and healing rate for each treatment. As per the GSK Meta-Analysis Plan, homogeneity was assessed using Breslow-Day- Tarone (BDT) test [Breslow, 1980; Tarone, 1985] based on the odds ratio. The publications deemed as statistical outliers were removed and the analysis was reconducted to check the potential impact of the outlier on the result. To summarize the data available, the summary of Day 7 healing rate for acyclovir and idoxuridine was calculated by summing up the patients healed over selected studies, and dividing by the sum of the total number of patients over the selected studies. The 95% CIs for the healing rates for each treatment group were also calculated using a normal approximation

18 CONFIDENTIAL 2012N137612_00 4. RESULTS The formal meta-analysis results comparing ACV versus IDU are presented in Figure 1 and Table 12. The results of healing rate estimation for ACV and IDU are presented in Table 13. Summary of data used in the meta-analysis can be found in Table 14. Based on the result from the meta-analysis with the CMH method (Table 12), in patients with herpetic keratitis, the odds of healing at Day 7 in the ACV treatment group is 3.95 times (95% CI: 2.60, 6.00; p-value: <0.0001) of the odds in the IDU treatment group. Per the GSK Meta-Analysis Plan, hypothesis testing was performed in a sequential manner to control the overall type I error. Since the result based on herpetic keratitis was statistically significant at the 5% level, similar analyses were performed for the two ulcer sub-types: dendritic ulcer and geographic ulcer and the results for both sub-types were statistically significant at the 2.5% level (for dendritic ulcer, the common odds ratio was 4.22 (95% CI: 2.14, 8.32; p <0.0001); and for geographic ulcer, the common odds ratio was 5.31 (95% CI: 1.09, 25.93; p= ). Results from CMH method were also presented in forest plots (Data Source Figure 1; Data Source Figure 3; Data Source Figure 4). Table 12 Summary of meta-analysis results on healing of herpetic keratitis at Day 7 for acyclovir versus idoxuridine with CMH method Group # of Articles included Total number of Patients ACV IDU Herpetic keratitis Dendritic ulcers Geographic ulcers Sensitivity Analysis (without data from Collum 1980) Herpetic keratitis < Dendritic 1.24 ulcers *d.f.: degrees of freedom Data Source Table 4; Data Source Table 8 CMH Test for comparing ACV over IDU CMH Common Odds Statistics Ratio P value 95% CI < (2.60, 6.00) < (2.14, 8.32) (1.09, 25.93) 2.96 (1.89, 4.66) 1.62 (0.69, 3.81) BDT test for homogeneity of odds ratio χ 2 statistics (d.f.*) p value (6) (3) (2) (5) (2)

19 CONFIDENTIAL 2012N137612_00 Figure 1 Primary Meta Analysis Result with CMH Method for Herpetic Keratitis CMH: Cochran-Mantel-Haenszel method; OR: Odds Ratio; CI Low: Lower bound of 95% confidence interval; CI Up: Upper bound of 95% confidence interval; ACV: acyclovir; IDU: idoxuridine; Confidence limits were truncated and presented as arrows if they were outside the range 0 to 10; (e.g. the upper confidence interval from Colin 1981). The entire confidence interval for the estimate from Collum, 1980 was above 10 (with a lower CI of 12.48) and therefore only the lower limit was shown as an arrow. The point estimate of the Odds Ratio for each study is represented by the squares, where the size of the square is proportional to the precision of the estimate. The kite shaped quadrilateral at the bottom of the graph has left and right endpoints at the lower and upper confidence intervals respectively, and the vertices of the kite shape that are its highest and lowest point vertically align at the point estimate of the Odds Ratio from the meta analysis. Homogeneity was evaluated across all studies included in the meta analysis for herpetic keratitis, as well as across all studies included in the meta analyses for each of the two ulcer sub-types: dendritic ulcer and geographic ulcer. As a result, not study was excluded from the primary analysis. The Breslow-Day-Tarone (BDT) test for the homogeneity of the odds ratios did result in significant p-values for the herpetic keratitis (overall) studies (p=0.0072), as well as the dendritic ulcer sub-type (p=0.0001). The test was not significant for the geographic ulcer subtype (p=0.1357). Collum [Collum, 1980] was deemed a potential outlier based on larger efficacy effect in favour or acyclovir (Figure 1, Data Source Figure 2). For both the analysis of herpetic keratitis and the dendritic ulcer subtype, a jack-knife analysis was done, where each study was removed in turn and the analysis was reperformed (Data Source Table 10). The results for the BDT test for both the herpetic keratitis group and the dendritic ulcer subtype lost significance when Collum was removed ( for herpetic keratitis and for dendritic ulcer) while they remained significant when any of the other studies were removed. This suggests that Collum was the driver behind the statistically significant BDT test for homogeneity. The jack-knifed results for the CMH statistic (Data Source Table 10) remained significant and consistent with the original analysis in all cases except for one. The result from the jack-knife that lost statistical significance was the analysis on dendritic ulcers with Collum removed (common odds ratio: 1.62; 95% CI: 0.69, 3.81; p= ), although the direction was consistent and still in favour of acyclovir

20 CONFIDENTIAL 2012N137612_00 Of note, the herpetic keratitis reanalysis without Collum remained highly significant (common odds ratio: 2.96; 95% CI: 1.89, 4.66; p-value: <0.0001) and was consistent with the analysis with all data for the herpetic keratitis ulcers. This shows the robustness of the conclusion that acyclovir provided improved 7 day healing rates for herpetic keratitis Results from analysis of the CMH method without Collum for the herpetic keratitis and dendritic ulcer subtype were also presented in forest plots (Data Source Figure 5 for herpetic keratitis and Data Source Figure 6 for dendritic ulcer). The healing rates observed from the acyclovir (97%, 29/30) and from the IDU (21%, 6/29) arms (and therefore the difference between them) were reasonable given the efficacy outcomes for dendritic ulcers reported in the literature. Day 7 healing rates for ACV treated dendritic ulcers reached up to 27/28, or 96% [Coster 1980]. Day 7 healing rates for IDU treated dendritic ulcers healing rates were as low as 4/20, 20% [Markham, 1977]. The collum study was a randomised, double-blind trial, which met the inclusion criteria for the meta-analysis plan and provided day 7 healing rates that were consistent with those in the available literature. Therefore it is not excluded from the primary comparison of interest. The estimates of treatment effects of ACV and of IDU from the Non-Linear-Mixed Effect (NLME) model for different ulcer types are presented in Table

21 CONFIDENTIAL 2012N137612_00 Table 13 Results of Day 7 Healing Rate Estimation for ACV and IDU with Non- Linear-Mixed Effect (NLME) Model Group Herpetic keratitis Dendritic ulcers Geographic ulcers # of Articles Total number of Patients point estimate (95% confidence interval) included ACV IDU Scale ACV IDU Odds (2.02, 6.78) (0.51, 1.55) Log Odds (0.70, 1.91) (-0.68, 0.44) Healing Rate (0.67, 0.87) (0.34, 0.61) Odds (1.43, 30.71) (0.32, 4.98) Log Odds (0.36, 3.42) (-1.15, 1.60) Healing Rate (0.59, 0.97) (0.24, 0.83) Odds (0.11, 12.78) (0.02, 2.40) Log Odds (-2.24, 2.55) (-3.88, 0.87) (0.10, 0.93) (0.02, 0.71) Healing Rate Sensitivity Analysis (without data from Collum 1980) Herpetic keratitis Dendritic ulcers Data Source Table 5, Data Source Table 9 Odds Log Odds Healing Rate Odds Log Odds Healing Rate 3.12 (1.55, 6.30) 1.14 (0.44, 1.84) 0.76 (0.61, 0.86) 4.08 (0.43, 38.79) 1.41 (-0.84, 3.66) 0.80 (0.30, 0.97) 1.06 (0.55, 2.05) 0.06 (-0.61, 0.72) 0.51 (0.35, 0.67) 2.50 (0.29, 21.88) 0.92 (-1.25, 3.09) 0.71 (0.22, 0.96) On average, for herpetic keratitis, the day 7 healing rate was 79% (95%CI: 67%, 87%) for acyclovir and 47% (95%CI: 34%, 61%) for idoxuridine. On average, for dendritic ulcers, the day 7 healing rate was 87% (95% CI: 59%, 97%) for acyclovir and 56% (95% CI: 24%, 83%) for idoxuridine. On average, for geographic ulcers, the day 7 healing rate was 54% (95% CI: 10%, 93%) for acyclovir and 18% (95% CI: 2%, 71%) for idoxuridine. Data from the publications described in Section 3.2 was listed and summarized in Table 14 by the following three groups: herpetic keratitis, dendritic ulcers, and geographic ulcers

22 CONFIDENTIAL 2012N137612_00 Table 14 Summary of healing rate for herpetic keratitis, dendritic and geographic ulcers at Day 7 Healed/Total (%) Herpetic keratitis Dendritic ulcers Geographic ulcers Study ACV IDU ACV IDU ACV IDU Colin /25 (76) 11/27 (41) Collum /30 (97) 6/29 (21) 29/30 (97) 6/29 (21) Coster /29 (93) 24/30 (80) 27/28 (96) 22/26 (85) 0/1 (0) 2/4 (50) /28 (75) 12/23 (52) Kitano /54 (74) 26/55 (47) Klauber /18 (67) 6/20 (30) 8/10 (80) 5/10 (50) 4/8 (50) 1/10 (10) McCulley /30 (63) 18/34 (53) 16/26 (62) 17/26 (65) 3/4 (75) 1/8 (12) Summary (95% CI) 167/214 (78) 103/218 (47) 80/94 (85) 50/91 (55) 7/13 (54) 4/22 (18) (72%, 84%) (41%, 54%) (78%, 92%) (45%, 65%) (27%, 81%) (2%, 34%) Data Source Table 1; Data Source Table 2 and Data Source Table 3 5. DISCUSSION The meta-analysis addressed the objective of comparing efficacy of acyclovir 3% ophthalmic ointment to idoxuridine in healing herpetic keratitis by Day 7 of treatment and showed that acyclovir had a statistically higher healing rate than idoxuridine in terms of odds ratio (common odds ratio of acyclovir versus idoxuridine: 3.95; 95% CI: 2.60, 6.00; p < ). Each individual report that met inclusion criteria for the meta-analysis studied one or both ulcer sub-types of herpetic keratitis exclusively. Analysis results also demonstrated the statistically significant superiority of acyclovir over idoxuridine in both ulcer sub-types dendritic ulcers (common odds ratio of acyclovir versus idoxuridine: 4.22; 95% CI: 2.14, 8.32; p <0.0001) and geographic ulcers (common odds ratio of acyclovir versus idoxuridine: 5.31; 95% CI: 1.09, 25.93; p= ). No obvious publication bias was observed based on a funnel plot (Data Source Figure 2). Outlier analysis did not change the inference of the primary analysis for herpetic keratitis. 6. CONCLUSIONS There were seven studies of patients with herpetic keratitis included in the meta-analysis. From all cases of herpetic keratitis (N=432) included in the meta-analysis, 185 cases were identified as dendritic ulcers and 35 cases were identified as geographic ulcers. There is not sufficient information to classify the remaining cases as specific ulcer sub-type. In the acyclovir group, the odds of healing at day 7 for patients with herpetic keratitis was 3.95 times of the odds in the idoxuridine group (95% CI: 2.60, 6.00; p < ). On average the day 7 healing rate was 79% (95%CI: 67%, 87%) for acyclovir and 47% (95%CI: 34%, 61%) for idoxuridine

23 CONFIDENTIAL 2012N137612_00 In the acyclovir group, the odds of healing at day 7 for patients with a dendritic ulcer was 4.22 times of the odds in the idoxuridine group (95% CI: 2.14, 8.32; p < ). On average, the day 7 healing rate was 87% (95% CI: 59%, 97%) for acyclovir and 56% (95% CI: 24%, 83%) for idoxuridine In the acyclovir group, the odds of healing at day 7 for patients with a geographic ulcer was 5.31 times of the odds in the idoxuridine group (95% CI: 1.09, 25.93; p= ). On average, the day 7 healing rate was 54% (95% CI: 10%, 93%) for acyclovir and 18% (95% CI: 2%, 71%) for idoxuridine

24 CONFIDENTIAL 2012N137612_00 7. REFERENCES Breslow NE, Day NE. Statistical methods in cancer research, Volumn I: The analysis of case-control studies. International Agency for Research on Cancer Scientific publications. 1980;I:32. Cochran WG. Some methods for strengthening the common chi-square tests. Biometrics. 1954;10: Colin J, Tournoux A, Chastel C, Renard G. Superficial herpes-simplex keratitis: Doubleblind comparative trial of acyclovir and idoxuridine. Nouv Presse Med. 1981;10: Colin J, Tournoux A, Chastel C, Renard, G. Superficial herpes-simplex keratitis. Doubleblind comparative trial of acyclovir and idoxuridine. Nouv Presse Med. 1981a;10(36): (English translation). Collum LM., Benedict-Smith A., Hillary I.B. Randomised double-blind trial of acyclovir and idoxuridine in dendritic corneal ulceration. Br J Ophthalmol. 1980;64: Coster DJ., Wilhelmus K.R., Michaud R., Jones B.R. A comparison of acyclovir and idoxuridine as treatment for ulcerative herpetic keratitis. Br J Ophthalmol. 1980;64: GSK 2007 Post Approval Document (PAD) Clinical Overview for ZOVIRAX Ophthalmic Ointment GSK Document Number BKCT/82/0002/00. Brief summary of a French report of a double blind clinical trial comparing acyclovir and idoxuridine in herpetic keratitis (Professor and GSK Document Number THRS/88/0017 BW 248U-14. A Double-masked coded clinical evaluation of Zovirax brand acyclovir ophthalmic ointment, 3% versus Stoxil brand idoxuridine Ophthalmic ointment, 0.5% in the treatment of herpes simplex keratitis. GSK Meta-Analysis Plan for, compound CCI22890: A meta-analysis to compare the efficacy of acyclovir 3% ophthalmic ointment to idoxuridine in curing herpes keratitis after 7-days of treatment. Kitano S. Efficacy of acyclovir in dendritic keratitis: A double-blind comparative study with IDU. Ther Res. 1985;2: Kitano S. Efficacy of acyclovir in dentritic keratitis: A double blind comparative study with IDU. Ther Res. 1985a;2: (English Translation). Klauber A, Ottovay E. Acyclovir and idoxuridine treatment of herpes simplex keratitis - A double blind clinical study. Acta Ophthalmol. 1982;60:

25 CONFIDENTIAL 2012N137612_00 Luntz MH, MacCallum FO. Treatment of herpes simplex keratitis with 5-iodo-2 - deoxyuridine. Br J Ophthalmol.1963;47: Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Nat Cancer Inst. 1959;22: Markham RH, Carter C, Scobie MA, Metcalf C, Easty DL. Double-blind clinical trial of adenine arabinoside and idoxuridine in herpetic corneal ulcers. Trans Ophthal Soc UK 1977;97: McCulley JP., Binder P.S., Kaufman H.E., O'Day D.M., Poirier R.H. A double-blind, multicenter clinical trial of acyclovir vs idoxuridine for treatment of epithelial herpes simplex keratitis. Ophthalmology. 1982;89: Tarone RE. On heterogeneity tests based on efficient scores. Biometrika. 1985;72: UK Submission Trial 2 summary results, A doubleblind randomised coded comparative trial of acyclovir ophthalmic ointment vs topical IDU therapy in patients with dendritic or geographic herpetic ulcers. UK Submission Trial 3 summary results, A randomised coded double blind trial of acyclovir and IDU in herpes simplex dendritic ulceration of the eye. UK Submission Trial 7 summary results- France; A double blind randomised coded comparative trial of acyclovir and IDU in patients with dendritic herpetic corneal ulceration. Wilhelmus KR, Coster DJ Jones, BR. Acyclovir and debridement in the treatment of ulcerative herpetic keratitis. Am J Ophthalmol. 1981;91: Wilhelmus KR. Antiviral treatment and other therapeutic interventions for herpes simplex virus epithelial keratitis (Review). Cochrane Database of Systematic Reviews. 2010;12: Wilhelmus KR. Therapeutic interventions for herpes simplex virus epithelial keratitis (Review). Cochrane Database of Systematic Reviews. 2009;1:

26 Protocol: Population: Herpetic Page 1 of 1 Figure 1. Meta Analysis Result with CMH Method (Odds Ratio Scale) 25 Study Colin 1981 Collum 1980 Coster Kitano 1985 Klauber 1982 McCulley 1982 OR Legend: CMH: Cochran Mantel Haenszel method OR: Odds Ratio CI Low: Lower bound of 95% confidence interval CI Up: Upper bound of 95% confidence interval CI Low CI Up ACV 19 / / / / / / / 30 IDU 11 / 27 6 / / / / 55 6 / / < IDU ACV > CONFIDENTIAL 2012N137612_00

27 Protocol: Population: Herpetic Page 1 of 1 Figure 2. Funnel Plot 26 Inverse of Standard Error McCulley 1982 Kitano Colin 1981 Klauber 1982 Coster 1980 Collum Log Odds Ratio The plot shows the point estimate and 95% confidence interval of log odds ratio of ACV versus IDU from each study. The vertical line indicates the log common odds ratio. CONFIDENTIAL 2012N137612_00

28 Protocol: Population: Dendritic Page 1 of 1 Figure 3. Meta Analysis Result with CMH Method (Odds Ratio Scale) 27 Study Collum 1980 Coster 1980 Klauber 1982 McCulley 1982 OR Legend: CMH: Cochran Mantel Haenszel method OR: Odds Ratio CI Low: Lower bound of 95% confidence interval CI Up: Upper bound of 95% confidence interval CI Low CI Up ACV 29 / / 28 8 / / 26 IDU 6 / / 26 5 / / < IDU ACV > CONFIDENTIAL 2012N137612_00

29 Protocol: Population: Geographic Page 1 of 1 Figure 4. Meta Analysis Result with CMH Method (Odds Ratio Scale) 28 Study Coster 1980 Klauber 1982 McCulley 1982 OR NaN Legend: CMH: Cochran Mantel Haenszel method OR: Odds Ratio CI Low: Lower bound of 95% confidence interval CI Up: Upper bound of 95% confidence interval CI Low CI Up NaN ACV 0 / 1 4 / 8 3 / 4 IDU 2 / 4 1 / 10 1 / < IDU ACV > CONFIDENTIAL 2012N137612_00

30 Protocol: Population: Herpetic Page 1 of 1 Figure 5. Meta Analysis Result with CMH Method: with Collum 1980 excluded (Odds Ratio Scale) 29 Study Colin 1981 Coster Kitano 1985 Klauber 1982 McCulley 1982 OR Legend: CMH: Cochran Mantel Haenszel method OR: Odds Ratio CI Low: Lower bound of 95% confidence interval CI Up: Upper bound of 95% confidence interval CI Low CI Up ACV 19 / / / / / / 30 IDU 11 / / / / 55 6 / / < IDU ACV > CONFIDENTIAL 2012N137612_00

31 Protocol: Population: Dendritic Page 1 of 1 Figure 6. Meta Analysis Result with CMH Method: with Collum 1980 excluded (Odds Ratio Scale) 30 Study Coster 1980 Klauber 1982 McCulley 1982 OR Legend: CMH: Cochran Mantel Haenszel method OR: Odds Ratio CI Low: Lower bound of 95% confidence interval CI Up: Upper bound of 95% confidence interval CI Low CI Up ACV 27 / 28 8 / / 26 IDU 22 / 26 5 / / < IDU ACV > CONFIDENTIAL 2012N137612_00

32 Protocol: Page 1 of 1 Population: Herpetic Table 1 Summary of Data Study ACV Cures IDU Cures n/n(%) n/n(%) 31 Colin /25 (76) 11/27 (41) Collum /30 (97) 6/29 (21) Coster /29 (93) 24/30 (80) /28 (75) 12/23 (52) Kitano /54 (74) 26/55 (47) Klauber /18 (67) 6/20 (30) McCulley /30 (63) 18/34 (53) Summary 167/214 (78) 103/218 (47) 95% CI (72%, 84%) (41%, 54%) CONFIDENTIAL 2012N137612_00

33 Protocol: Page 1 of 1 Population: Dendritic Table 2 Summary of Data Study ACV Cures IDU Cures n/n(%) n/n(%) 32 Collum /30 (97) 6/29 (21) Coster /28 (96) 22/26 (85) Klauber /10 (80) 5/10 (50) McCulley /26 (62) 17/26 (65) Summary 80/94 (85) 50/91 (55) 95% CI (78%, 92%) (45%, 65%) CONFIDENTIAL 2012N137612_00

34 Protocol: Page 1 of 1 Population: Geographic Table 3 Summary of Data Study ACV Cures IDU Cures n/n(%) n/n(%) 33 Coster /1 ( 0) 2/4 (50) Klauber /8 (50) 1/10 (10) McCulley /4 (75) 1/8 (12) Summary 7/13 (54) 4/22 (18) 95% CI (27%, 81%) (2%, 34%) CONFIDENTIAL 2012N137612_00

35 Protocol: Page 1 of 1 Table 4 Summary of Meta Analysis Result by Population with Cochran-Mantel-Haenszel Model Total CMH CMH COR COR BDT BDT BDT Population Patients Stat. p value 95% CI Stat. D.F. p value Herpetic < (2.60,6.00) Dendritic < (2.14,8.32) Geographic (1.09,25.93) The result shows the odds ratio of cure for acyclovir versus idoxuridine CONFIDENTIAL 2012N137612_00

36 Protocol: Page 1 of 1 Table 5 Summary of Cure Rate Estimation by Population with Non-linear Mixed Effect Model Population Scale Treatment Estimate 95% CI Herpetic Log Odds Acyclovir 1.31 (0.70,1.91) Idoxuridine (-0.68,0.44) Odds Acyclovir 3.70 (2.02,6.78) Idoxuridine 0.89 (0.51,1.55) Healing Rate Acyclovir 0.79 (0.67,0.87) Idoxuridine 0.47 (0.34,0.61) 35 Dendritic Log Odds Acyclovir 1.89 (0.36,3.42) Idoxuridine 0.23 (-1.15,1.60) Odds Acyclovir 6.63 (1.43,30.71) Idoxuridine 1.26 (0.32,4.98) Healing Rate Acyclovir 0.87 (0.59,0.97) Idoxuridine 0.56 (0.24,0.83) Geographic Log Odds Acyclovir 0.15 (-2.24,2.55) Idoxuridine (-3.88,0.87) Odds Acyclovir 1.17 (0.11,12.78) Idoxuridine 0.22 (0.02,2.40) Healing Rate Acyclovir 0.54 (0.10,0.93) Idoxuridine 0.18 (0.02,0.71) CONFIDENTIAL 2012N137612_00