Aqueous proteins help predict the response of patients with neovascular age-related macular degeneration to anti-VEGF therapy

Background To reduce the treatment burden for patients with neovascular age-related macular degeneration (nvAMD), emerging therapies targeting vascular endothelial growth factor (VEGF) are being designed to extend the interval between treatments, thereby minimizing the number of intraocular injections. However, which patients will benefit from longer-acting agents is not clear. Methods Eyes with nvAMD (n = 122) underwent 3 consecutive monthly injections with currently available anti-VEGF therapies, followed by a treat-and-extend protocol. Patients who remained quiescent 12 weeks from their prior treatment entered a treatment pause and were switched to pro re nata (PRN) treatment (based on vision, clinical exam, and/or imaging studies). Proteomic analysis was performed on aqueous fluid to identify proteins that correlate with patients’ response to treatment. Results At the end of 1 year, 38 of 122 eyes (31%) entered a treatment pause (≥30 weeks). Conversely, 21 of 122 eyes (17%) failed extension and required monthly treatment at the end of year 1. Proteomic analysis of aqueous fluid identified proteins that correlated with patients’ response to treatment, including proteins previously implicated in AMD pathogenesis. Interestingly, apolipoprotein-B100 (ApoB100), a principal component of drusen implicated in the progression of nonneovascular AMD, was increased in treated patients who required less frequent injections. ApoB100 expression was higher in AMD eyes compared with controls but was lower in eyes that develop choroidal neovascularization (CNV), consistent with a protective role. Accordingly, mice overexpressing ApoB100 were partially protected from laser-induced CNV. Funding This work was supported by the National Eye Institute, National Institutes of Health grants R01EY029750, R01EY025705, and R01 EY27961; the Research to Prevent Blindness, Inc.; the Alcon Research Institute; and Johns Hopkins University through the Robert Bond Welch and Branna and Irving Sisenwein professorships in ophthalmology. Conclusion Aqueous biomarkers could help identify patients with nvAMD who may not require or benefit from long-term treatment with anti-VEGF therapy.

independent and masked graders (TPP and ZY); OCTs performed at time points closest to 6 and 12 months (i.e., within 4 weeks of the 6 time point and 6 weeks of the 12 month time point from treatment initiation) were included in this analysis. Each image was classified as having no fluid, subretinal fluid (SRF), intraretinal fluid (IRF), or both. Differences in image grading between the two blinded graders was reconciled and if an agreement could not be reached, a third investigator (AS) cast the tie-breaker vote. The location of fluid for each patient and at each time point was graphically represented in a table with different color cells representing no fluid, SRF, IRF or both.
Patients were grouped based on frequency of treatment at 12 months to compare those who were able to successfully pause treatment with those who were unable to reach treatment pause.

Sample preparation
Samples from control patients, dry (non-neovascular) AMD patients, untreated nvAMD patients, and nvAMD patients (untreated and treated up to three times) from the TEP/M study who required monthly or q12 weeks (or longer) treatment at 1 year were used for proteomics studies.
For each sample, ~10 µg proteins from pooled patient samples in 10µL of 8.0 M urea lysis buffer was reduced with 6 mM dithiothreitol (DTT) for 1h at 37 o C, and then alkylated with 12 mM iodoacetamide (IAA) for 45 minutes in the dark at room temperature. The proteins were then digested to peptides by Lys-C and trypsin. After the digestion, the peptides were acidified with 50% formic acid to pH <3, and then desalted on C18 stage-tip. Desalted peptides were then dried in a Speed Vacuum and then resuspended in 3% acetonitrile (ACN) with 0.1% HCOOH for future liquid chromatography mass spectrometry analysis (LC-MS/MS).

LC-MS/MS analysis using data independent acquisition (DIA)
1 ug peptides of each sample were spiked with iRT peptides and separated by ACN gradient on an in-house packed C18 column (0.75 µm I.D. x 26.5 cm length) on the Easy nLC 1200 UHPLC system (Thermo Scientific). Separated peptides were analyzed through Nanospray Flex™ Ion Source (Thermo Scientific) using Orbitrap Fusion™ Lumos™ Tribrid™ MS instrument (Thermo Scientific) by DIA as described previously (16).

Analysis of DIA data
The DIA data was analyzed using library free strategy. All the 9 DIA runs were performed using Spectronaut™ (version 14.10,Biognosys,Schlieren,Switzerland) with directDIA™ analysis following the BGS factory search setting with uniprot_reviewed_human database (release 20,418 entries). The variable modifications on amino acid were set as oxidation(M) and acetylation (protein N-terminus), and the fixed modification on amino acid was set as carbamidomethylation (C). Cross run normalization was applied to normalize the DIA together with normalized on median.

Gene Ontology
For gene ontology analyses, DAVID was utilized to identify relevant biological processes enriched in the expressed proteins (19,20). FDR =0.05 was chosen for the statistical significance.

Hierarchical clustering
Heatmap of protein expression level was generated using unsupervised hierarchical clustering with command of scipy.cluster.hierarchy.linkage(method="complete", metric="euclidean" from Scipy library (21) in python. The distance matrix used in the clustering was first constructed from z-score of protein expression across 6 groups of patients. Farthest Point algorithm (also known as Complete) was used to calculate Euclidean distance between 6 groups of patients. Visualization of heatmap was generated by Matplotlib library (22) in python.

Laser CNV Model
Mice were anesthetized with a mixture of ketamine (100 mg/kg) and xylazine (5mg/kg).
After pupils were dilated with tropicamide (Alcon Laboratories), lubricating hypromellose eye drops (Alcon Laboratories) were applied to the cornea. The fundus was viewed using a hand-held coverslip as a contact lens and laser photocoagulation was performed using the diode laser   Table 3 Abbreviations: No., number; ETDRS, Early Treatment Diabetic Retinopathy Study letter score. CST, central subfield thickness; q8-12, requiring treatment every 8 -12 weeks; TEP/M, treat and extend pause/monitor. Values displayed as mean ± standard error of the mean. Statistical analysis was performed using Chi-square and Mann-Whitney test. Data in bold are statistically significant.  Table 5. Recovery of vision after re-treatment with anti-VEGF in patients who were weaned from therapy but were later diagnosed with new CNV activity.

Supplemental
Abbreviations: SRH, subretinal hemorrhage; RPE, retinal pigment epithelium; SR fibrosis, subretinal fibrosis. q4-6, every 4-6 weeks; q8-12, every 8 to 12 weeks; TEP/M, treat and extend pause/monitor. Adverse events were recorded over the initial three years of the study for patients with a minimum of 24 months of follow-up up to a maximum of 36 months of follow-up. Statistical analysis was performed using 3-sample test for equality of proportions.

Supplemental
Abbreviations: n, sample size; q4, received treatment every 4 weeks; q6-8, received treatment every 6-8 weeks; q10-12, received treatment every 10-12 weeks; VEGF, vascular endothelial growth factor; ELISA, enzyme-linked immunoassay; and yr, years. Values displayed as mean ± standard error of the mean. *Unable to obtain standard error of the mean with a sample size of 1. Statistical analysis was performed using ANOVA and Fisher's exact test. Data in bold are statistically significant. Supplemental Table 7 Supplemental Abbreviations: n, sample size; q4, received treatment every 4 weeks; q12+, received treatment every 12 or more weeks; nnvAMD, nonneovascular Age-related Macular Degeneration; nvAMD, neovascular Age-related Macular Degeneration; and yr, years. Values displayed as mean ± standard error of the mean. Statistical analysis was performed using ANOVA and Fisher's exact test. Data in bold are statistically significant.  Table 9. Proteins identified in Proteomic Analyses which had sequences that overlap with current anti-VEGF therapies.

Supplemental
Supplemental Table 10. Proteins identified in Proteomics Analyses that were "highly variable". "Highly variable" proteins were those proteins identified in the Proteomics comparison of NVAMD patients that were arbitrarily divided into two group: NVAMD1 and NVAMD2. Proteins were designated as "highly variable" if they were increased or decreased two-fold or more in this comparison.  Eyes of patients receiving injections with a diagnosis of nvAMD (n=207) Eyes of patients with diagnoses other than nvAMD (n=123) Eyes of nvAMD patients receiving anti-VEGF therapy using TEP/M who met inclusion criteria for analysis at 12 months (n=102) Eyes of patients who did not meet inclusion criteria for study (n=105): did not return for follow-up (23), failed to adhere to TEP/M regimen (56), received less than 1 year of treatment (15), not newly diagnosed or newly active AMD (8), other underlying ischemic retinal disease (3) Eyes of nvAMD patients receiving anti-VEGF therapy who met inclusion criteria for analysis at 24 months (n=65) Eyes of patients who did not meet inclusion at 24 months (n = 37): did not return for follow-up (6), received less than 2 years of treatment (20), passed away (5), discontinued due to futility/development of other ischemic retinal disease (3), or left the practice/continued care with another provider (3)

Supplemental Figure 2
Eyes of patients receiving injections for any ocular neovascular disease at a separate tertiary hospitalbased clinical site between 2013 and 2020 (n=58) Eyes of patients receiving injections with a diagnosis of nvAMD (n=32) Eyes of patients with diagnoses other than nvAMD (n=26) Eyes of nvAMD patients receiving anti-VEGF therapy using TEP/M who met inclusion criteria for analysis at 12 months (n=20) Eyes of patients who did not meet inclusion criteria for study (n=12): failed to adhere to regimen or missed appointments (2), less than 1 year of treatment (1), initiated treatment by other provider (1), received intraocular steroid or other anti-VEGF therapy (6), discontinued due to futility (   "weanable" C "Anti-VEGF Curable" q6-12 weeks B "Anti-VEGF Treatable"

May further benefit from New Therapies targeting other vasoactive mediators
Current anti-VEGF therapies may be adequate (may not require longer acting anti-VEGF agents).

May further benefit from New Therapies targeting other vasoactive mediators
May benefit from longer-acting anti-VEGF agents Require New Therapies targeting other vasoactive mediators Supplemental Figure 7