Predictive Markers Useful in the Treatment of Wet Age-Related Macular Degeneration
20220155295 · 2022-05-19
Assignee
Inventors
Cpc classification
G01N2800/56
PHYSICS
G01N33/564
PHYSICS
International classification
G01N33/564
PHYSICS
Abstract
The invention is directed to the use of biomarkers to determine responsiveness of an individual with wet AMD to treatment with a VEGF antagonist, to diagnose and distinguish dry and wet AMD and to determine the risk of conversion of a dry AMD disease to a wet AMD disease.
Claims
1-15. (canceled)
16. A method of determining whether an individual suffers from dry AMD and/or wet AMD, the method comprising: (i) obtaining a sample from the individual (ii) determining the quantity of at least one antibody selected from Group B antibodies directed against OGFR, alpha-synuclein, SELO, SPRY, GAPDH-H2, Annexin-V, THAP, VTI-B, HSP10, ESD, PKC80, ACO2-C2, PBP-I2, CAZ-C3, EIFA1, MAPK3, ENO1-H7, Chromosome 17, Aconitate Hydratase, and GPX4; and (iii) identifying the individual as suffering from dry AMD and/or wet AMD disease if the amount of the at least one antibody selected from Group B antibodies is increased by at least 20% or decreased by at least 20% compared to control values.
17. The method according to claim 16, wherein the at least one antibody is selected from a subgroup B-1 of the Group B antibodies directed against alpha-synuclein, SELO, SPRY, GAPDH-H2, Annexin-V, THAP, VTI-B, and HSP10.
18. A method to determine that an individual with dry AMD is at risk of developing wet AMD, the method comprising: (i) obtaining a sample from the individual with dry AMD; (ii) determining the quantity of at least one antibody selected from autoantibodies which are members of Group A and/or Group B, wherein Group A antibodies are directed against OGFR, MAPK3, PolyRp2, Chromosome 17, EIFA1, GPX4, SRP14, Gamma-synuclein, Jo-1, Pre-Albumin, GPD2, AP1M1, ENO1-H7, PRG2, ATP synthase, GNB1-A3, TUBB3, HSP60, ICA1, SELO, SOD, and ENO2, and wherein Group B antibodies are directed against alpha-synuclein, SELO, SPRY, GAPDH-H2, Annexin-V, THAP, VTI-B, HSP10, ESD, PKC80, ACO2-C2, OGFR, PBP-I2, CAZ-C3, EIFA1, MAPK3, ENO1-H7, Chromosome 17, Aconitate Hydratase, and GPX4; and (iii) identifying the individual as being at risk of developing wet AMD if the amount of the at least one antibody is increased by at least 20% or decreased by at least 20% compared to control values.
19. The method according to claim 18, wherein the at least one antibody is selected from a subgroup A-1 of the Group A antibodies and/or a subgroup B-1 of the Group B antibodies, wherein antibodies of subgroup A-1 are directed against OGFR, MAPK3, SRP14, ENO2, SOD, Pre-Albumin, Jo-1, PolyRp2, Chromosome 17, and EIFA1, and wherein antibodies of subgroup B-1 are directed against alpha-synculein, SELO, SPRY, GAPDH-H2, Annexin-V, THAP, VTI-B, and HSP10.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
[0077] The invention will be better understood and objects other than those set forth above will become apparent when consideration is given to the following detailed description thereof. This description makes reference to the annexed figures, wherein:
[0078]
[0079]
[0080]
[0081]
[0082]
[0083]
DETAILED DESCRIPTION OF THE INVENTION
[0084] Good results can be achieved for many patients suffering from wet AMD when treated with the approved VEGF antagonists (ranibizumab, aflibercept), meaning either a gain of vision or at least stability of the visual acuity. However, there exists still a certain percentage of patients which do not respond to those approved therapies. These patients suffer from further vision loss and are so called non-responders.
[0085] A patient can be considered a responder if treatment with a VEGF antagonist is useful and effective, as regards not only improvement but also stability, meaning a visual acuity (VA) loss of fewer than 15 ETDRS letters (ETDRS stands for Early Treatment Diabetic Retinopathy Study) after three successive follow-up visits (the follow-up visits start after three anti-VEGF injections). A patient can be considered as responder who gains visual acuity and maintains the gain over time. Patients who gain visual acuity but do not maintain the gain, losing visual acuity to not lower than at baseline are also considered responders. Responders are also considered as patients who gain visual acuity but do not maintain the gain, losing visual acuity to less than 15 ETDRS letters lower than at baseline. Patients who do not gain visual acuity but remain stable, with visual acuity loss not lower than at baseline are also considered as responders. Patients who do not gain visual acuity but whose visual acuity loss does not exceed 15 ETDRS letters compared to a baseline are also considered as responders.
[0086] A patient can be considered a non-responder if treatment with a VEGF antagonist gives no overall clinical benefit, assessed on the basis of morphological and functional parameters, and there is immediate or late visual loss. Non-responders are patients who lose more than 15 ETDRS letters in total in three successive follow-up visits (the follow-up visits start after three anti-VEGF injections). Non-responders are also considered as patients who lose more than 30 ETDRS letters compared with baseline and or best visual acuity recorded at baseline. Non-responders are also defined as patients whose best corrected visual acuity (BCVA) had worsened in the log MAR score. Besides the functional definitions, non-responders are also determined by fundus findings including OCT (optical coherence tomography). A non-responder is also defined as a patient in whom exudative fundus findings (pigment epithelial detachment, subretinal fluid, macular oedema, haemorrhage) had increased or had appeared after treatment, or in whom the central retinal thickness (CTR) increased by more than 100 μm within 12 months after the initial anti-VEGF injection.
[0087] It is known that two third of serum immunoglobulins in healthy individuals are natural occurring autoantibodies. The autoantibody patterns of patients suffering from wet and/or suffering from dry AMD and the autoantibody patterns of patients suffering from wet AMD receiving VEGF antagonist therapy were analysed.
[0088] It was surprisingly found that the amount of autoantibodies directed against antibodies of Group A comprising MAPK3, OGFR, PolyRp2, Chromosome 17, EIFA1, GPX4, SRP14, Gamma-synuclein, Jo-1, Pre-Albumin, GPD2, AP1M1, ENO1-H7, PRG2, ATP synthase, GNB1-A3, TUBB3, HSP 60, ICA1 and SELO, SOD, ENO2 is significantly different in the serum of patients suffering from either dry or wet AMD in comparison to healthy patients. The quantity of one or several autoantibodies selected from autoantibodies directed against MAPK3, OGFR, PolyRp2, Chromosome 17, EIFA1, GPX4, SRP14, Gamma-synuclein, Jo-1, Pre-Albumin, GPD2, AP1M1, ENO1-H7, PRG2, ATP synthase, GNB1-A3, TUBB3, HSP 60, ICA1 and SELO, SOD, ENO2 in the patient serum can therefore be used as a means to determine if a patient suffering from dry or wet AMD will respond to a VEGF antagonist therapy, before the VEGF antagonist therapy is initiated.
[0089] It was also found that in wet AMD patients with vision gain after ranibizumab therapy the serum level of antibodies directed against OGFR, Gamma-synuclein, Pre-Albumin and PRG2 were significantly increased, whereas in patients who showed a vision loss after ranibizumab therapy the level of said antibodies was significantly decreased.
[0090] Furthermore, e.g. immunoreactivities against PolyRp2, GNB1-A3, and TUBB3 are down-regulated in patients improving vision during 12 months under lucentis treatment and upregulated in those who do not.
[0091] The detectable amount of autoantibodies in the serum of a patient suffering from wet AMD changes over time. A positive correlation between an optimal dosing regimen and the level of autoantibodies could have been identified. The quantity of one or several autoantibodies selected from the Group A of autoantibodies directed against MAPK3, OGFR, PolyRp2, Chromosome 17, EIFA1, GPX4, SRP14, Gamma-synuclein, Jo-1, Pre-Albumin, GPD2, AP1M1, ENO1-H7, PRG2, ATP synthase, GNB1-A3, TUBB3, HSP 60, ICA1 and SELO, SOD, ENO2 is therefore also a predictive marker when the VEGF antagonist needs to be applied to the patient.
[0092] Surprisingly it was also found, that autoantibody patterns are able to differentiate between patients suffering from wet or dry AMD. In particular, changes in the autoantibody levels of one or several autoantibodies selected from autoantibodies of Group B of autoantibodies directed against alpha-synuclein, SELO, SPRY, GAPDH-H2, Annexin-V, THAP, VTI-B, HSP10, ESD, PKC80, ACO2-C2, OGFR, PBP-I2, CAZ-C3, EIFA1, MAPK3, ENO1-H7, Chromosome17, Aconitate Hydratase, GPX4.
[0093] Furthermore, it was found that changes in the autoantibody levels of one or several autoantibodies selected from autoantibodies of Group A or of Group B as defined above directed have a prognostic value to indicate the conversion of the dry AMD form to the wet AMD form.
[0094] VEGF is a well-characterised signal protein which stimulates angiogenesis. A “VEGF antagonist” refers to a molecule capable of neutralizing, blocking, inhibiting, abrogating, reducing or interfering with VEGF activities including its binding to one or more VEGF receptors. The VEGF antagonist may be administered intravitreally, e.g. through injection, or topically, e.g. in form of eye drops. VEGF antagonists include anti-VEGF antibodies and antigen-binding fragments thereof, receptor molecules and derivatives which bind specifically to VEGF thereby sequestering its binding to one or more receptors, anti-VEGF receptor antibodies and VEGF receptor antagonists such as small molecule inhibitors of the VEGFR tyrosine kinases, and fusions proteins. The invention also provides non-antibody VEGF antagonists.
[0095] a) Antibody VEGF Antagonists
[0096] In one embodiment, the VEGF antagonist is an antibody. In one embodiment, the VEGF antagonist is a mimetic of the VEGF receptor. In one embodiment, the VEGF antagonist is ranibizumab. In one embodiment, the VEGF antagonist is bevacizumab.
[0097] b) Non-Antibody VEGF Antagonists
[0098] In one embodiment, the VEGF antagonist is a non-antibody VEGF antagonist. In one aspect of the invention, the non-antibody VEGF antagonist is an immunoadhesin. One such immuoadhesin is aflibercept (Eylea®), which has recently been approved for human use and is also known as VEGF-trap (Holash et al., PNAS USA, 2002; Riely & Miller, Clinical Cancer Research, 2007). Aflibercept is the preferred non-antibody VEGF antagonist for use with the invention. Aflibercept is a recombinant human soluble VEGF receptor fusion protein consisting of portions of human VEGF receptors 1 and 2 extracellular domains fused to the Fc portion of human IgG1. It is a dimeric glycoprotein with a protein molecular weight of 97 kilodaltons (kDa) and contains glycosylation, constituting an additional 15% of the total molecular mass, resulting in a total molecular weight of 115 kDa. It is conveniently produced as a glycoprotein by expression in recombinant CHO K1 cells. Each monomer can have the following amino acid sequence (SEQ ID NO: 1):
TABLE-US-00001 SDTGRPFVEMYSEIPEIIHMTEGRELVIPCRVTSPNITVTLKKFP LDTLIPDGKRIIWDSRKGFIISNATYKEIGLLTCEATVNGHLYKT NYLTHRQTNTIIDVVLSPSHGIELSVGEKLVLNCTARTELNVGID FNWEYPSSKHQHKKLVNRDLKTQSGSEMKKFLSTLTIDGVTRSDQ GLYTCAASSGLMTKKNSTFVRVHEKDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVH NAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLPPSRDELTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPG
and disulfide bridges can be formed between residues 30-79, 124-185, 246-306 and 352-410 within each monomer, and between residues 211-211 and 214-214 between the monomers.
[0099] Another non-antibody VEGF antagonist immunoadhesin currently in pre-clinical development is a recombinant human soluble VEGF receptor fusion protein similar to VEGF-trap containing extracellular ligand-binding domains 3 and 4 from VEGFR2/KDR, and domain 2 from VEGFR1/n Flt-1; these domains are fused to a human IgG Fc protein fragment (Li et al., Molecular Vision, 2011). This antagonist binds to isoforms VEGF-A, VEGF-B and VEGF-C. The molecule is prepared using two different production processes resulting in different glycosylation patterns on the final proteins. The two glycoforms are referred to as KH902 (conbercept) and KH906. The fusion protein can have the following amino acid sequence (SEQ ID NO:2):
TABLE-US-00002 MVSYWDTGVLLCALLSCLLLTGSSSGGRPFVEMYSEIPEIIHMTE GRELVIPCRVTSPNITVTLKKFPLDTLIPDGKRIIWDSRKGFIIS NATYKEIGLLTCEATVNGHLYKTNYLTHRQTNTIIDVVLSPSHGI ELSVGEKLVLNCTARTELNVGIDFNWEYPSSKHQHKKLVNRDLKT QSGSEMKKFLSTLTIDGVTRSDQGLYTCAASSGLMTKKNSTFVRV HEKPFVAFGSGMESLVEATVGERVRLPAKYLGYPPPEIKWYKNGI PLESNHTIKAGHVLTIMEVSERDTGNYTVILTNPISKEKQSHVVS LVVYVPPGPGDKTHTCPLCPAPELLGGPSVFLFPPKPKDTLMISR TPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTY RVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPRE PQVYTLPPSRDELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENN YKATPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNH YTQKSLSLSPGK
and, like VEGF-trap, can be present as a dimer. This fusion protein and related molecules are further characterized in EP1767546.
[0100] Other non-antibody VEGF antagonists include antibody mimetics (e.g. Affibody® molecules, affilins, affitins, anticalins, avimers, Kunitz domain peptides, and monobodies) with VEGF antagonist activity. This includes recombinant binding proteins comprising an ankyrin repeat domain that binds VEGF-A and prevents it from binding to VEGFR-2. One example for such a molecule is DARPin® MP0112. The ankyrin binding domain may have the following amino acid sequence (SEQ ID NO: 3):
TABLE-US-00003 GSDLGKKLLEAARAGQDDEVRILMANGADVNTADSTGWTPLHLAV PWGHLEIVEVLLKYGADVNAKDFQGWTPLHLAAAIGHQEIVEVLL KNGADVNAQDKFGKTAFDISIDNGNEDLAEILQKAA
[0101] Recombinant binding proteins comprising an ankyrin repeat domain that binds VEGF-A and prevents it from binding to VEGFR-2 are described in more detail in WO2010/060748 and WO2011/135067.
[0102] Further specific antibody mimetics with VEGF antagonist activity are the 40 kD pegylated anticalin PRS-050 and the monobody angiocept (CT-322).
[0103] The non-antibody VEGF antagonist may be modified to further improve its pharmacokinetic properties or bioavailability. For example, a non-antibody VEGF antagonist may be chemically modified (e.g., pegylated) to extend its in vivo half-life. Alternatively or in addition, it may be modified by glycosylation or the addition of further glycosylation sites not present in the protein sequence of the natural protein from which the VEGF antagonist was derived.
[0104] Variants of the above-specified VEGF antagonists that have improved characteristics for the desired application may be produced by the addition or deletion of amino acids. Ordinarily, these amino acid sequence variants will have an amino acid sequence having at least 60% amino acid sequence identity with the amino acid sequences of SEQ ID NO: 1, SEQ ID NO: 2 or SEQ ID NO: 3, preferably at least 80%, more preferably at least 85%, more preferably at least 90%, and most preferably at least 95%, including for example, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, and 100%. Identity or homology with respect to this sequence is defined herein as the percentage of amino acid residues in the candidate sequence that are identical with SEQ ID NO: 1, SEQ ID NO: 2 or SEQ ID NO: 3, after aligning the sequences and introducing gaps, if necessary, to achieve the maximum percent sequence identity, and not considering any conservative substitutions as part of the sequence identity.
[0105] Sequence identity can be determined by standard methods that are commonly used to compare the similarity in position of the amino acids of two polypeptides. Using a computer program such as BLAST or FASTA, two polypeptides are aligned for optimal matching of their respective amino acids (either along the full length of one or both sequences or along a pre-determined portion of one or both sequences). The programs provide a default opening penalty and a default gap penalty, and a scoring matrix such as PAM 250 [a standard scoring matrix; see Dayhoff et al., in Atlas of Protein Sequence and Structure, vol. 5, supp. 3 (1978)] can be used in conjunction with the computer program. For example, the percent identity can then be calculated as: the total number of identical matches multiplied by 100 and then divided by the sum of the length of the longer sequence within the matched span and the number of gaps introduced into the longer sequences in order to align the two sequences.
[0106] The autoantibody pattern, the presence or lack of a certain quantity of an antibody in a sample from an individual having AMD can individually, or in combination, serve as biomarkers to predict responsiveness of that individual to treatment with a VEGF antagonist. The presence of these biomarkers can be determined in a sample from an individual of interest. The sample can be any sample including but is not limited to a fluid sample such as blood, serum, tears, saliva, urine, a cell sample such as buccal cells, aqueous humor or vitreous body of the eye.
[0107] As used herein, “predicting” indicates that the methods described herein provide information to enable a health care provider to determine the likelihood that an individual having wet AMD will respond to VEGF antagonist treatment (anti-VEGF therapy). Following a positive determination of the relevant biomarker(s) in a sample of interest, the individual will be administered a VEGF antagonist.
[0108] To determine the levels of autoantibodies any known method in the art can be used. Methods to quantify the autoantibody contents include but are not limited to standard immunological analytical techniques such as Western blot or immunoblot assays, enzyme-linked immunoabsorbent assays (ELISA), radioimmunoassays, real-time PCR, microarrays, lateral flow, microfluidic assays, bead based assays, nontargeted proteomics involving e.g. massanalyzers such as ion-trap detection, fourier transformion cyclotron resonance, time-of-flight (TOF) mass spectrometry, targeted proteomics involving e.g. the use of selected reaction monitoring (SRM) with triple quadrupole mass spectrometry (TQMS), immunoaffinity mass spectrometry, surface enhanced laser desorption/ionization in time of flight mass spectrometry (SELDI-TOF-MS), matrix assisted laser desorption/ionization mass spectrometry (MALDI) or other antibody chip techniques.
[0109] The invention is not limited by the types of methods used to determine the quantities of autoantibodies.
[0110] Using the methods of the invention an individual is classified as a VEGF antagonist responder if the amounts of one or several of Group A autoantibodies selected from antibodies directed against MAPK3, OGFR, PolyRp2, Chromosome 17, EIFA1, GPX4, SRP14, Gamma-synuclein, Jo-1, Pre-Albumin, GPD2, AP1M1, ENO1-H7, PRG2, ATP synthase, GNB1-A3, TUBB3, HSP 60, ICA1, SELO, SOD, ENO2 are increased or decreased by at least 10%, by 20%, by 30%, by 40%, by 50%, by 100%, by 150%, by 200%, by 250%, preferably by 25%-150%, more preferably by 50-100%, in the sample compared to a control (healthy individual).
[0111] Using the methods of the invention an individual having dry AMD is classified as having a risk that the dry AMD converts into wet AMD if the amounts of one or several autoantibodies selected from antibodies of Group A antibodies comprising antibodies directed against: MAPK3, OGFR, PolyRp2, Chromosome 17, EIFA1, GPX4, SRP14, Gamma-synuclein, Jo-1, Pre-Albumin, GPD2, AP1M1, ENO1-H7, PRG2, ATP synthase, GNB1-A3, TUBB3, HSP 60, ICA1, SELO, SOD, ENO2 or if the amounts of one or several autoantibodies selected from antibodies of Group B antibodies comprising antibodies directed against: alpha-synuclein, SELO, SPRY, GAPDH-H2, Annexin-V, THAP, VTI-B, HSP10, ESD, PKC80, ACO2-C2, OGFR, PBP-I2, CAZ-C3, EIFA1, MAPK3, ENO1-H7, Chromosome17, Aconitate Hydratase, GPX4 detected and quantified in two samples taken in a time period of about 2 weeks, of about 1 month, of about 2 months, of about 6 months, of about 12 months, of at least about 12 months, decrease by at least 20%, by 30%, by 40%, by 50%, by 60%, by 75%, preferably by at least 30%, more preferably by at least 50%, in the sample compared to a control (healthy individual).
[0112] The methods described herein can be utilized as a diagnostic assay to identify those subjects having wet AMD who are likely to respond to a VEGF antagonist. The methods of the invention can be used to determine whether a subject should be administered a VEGF antagonist to reduce the severity of wet AMD. The methods described herein can also be utilized as a prognostic assay to identify those subjects having dry AMD who are at risk to develop wet AMD and who would benefit from receiving a VEGF antagonist. Prognostic assays can be used for predictive purposes or prophylactic purposes to treat an individual who is at risk to develop wet AMD.
[0113] The invention also encompasses kits for determining the level of one or several autoantibodies directed against MAPK3, OGFR, PolyRp2, Chromosome 17, EIFA1, GPX4, SRP14, Gamma-synuclein, Jo-1, Pre-Albumin, GPD2, AP1M1, ENO1-H7, PRG2, ATP synthase, GNB1-A3, TUBB3, HSP 60, ICA1 and SELO, SOD, ENO2.
MODES FOR CARRYING OUT THE INVENTION
[0114] I. Comparison of Immunoreactivities in Dry and Wet AMD
[0115] The autoantibody patterns against retinal antigens in sera of patients with “wet” AMD were analyzed and compared to healthy control subjects and patients with “dry” AMD by mass spectrometry (MS) approach. It is known that two third of serum immunoglobulins in healthy individuals are natural occurring autoantibodies, so that complex profiles exist even in healthy people and disease specific changes of circulating autoantibodies are known from several other diseases, e.g. glaucoma or Sicca syndrome. After successful de novo screening of immunoreactivities using MS-based approach and high density Protagen antigen microarrays, a customized antigen microarray containing 61 antigens was built (Table 2). Each microarray contained each antigen as triplicate. Patients were included in the study based on the protocol following all inclusion- and exclusion criteria.
TABLE-US-00004 TABLE 1 Breakdown Table of Descriptive Statistics. CTRL = healthy volunteers, 17 = dry AMD, 18 = wet AMD starting new anti-VEGF treatment, 19 = wet AMD continuous anti-VEGF treatment Gender Age - Number of Age - Group (m/f) Means subjects Std. Dev. CTRL f 71.8 9 9.4 CTRL m 74.5 11 11.0 17 f 80.9 14 8.5 17 m 73.2 6 7.2 18 f 81.0 120 7.9 18 m 81.4 60 5.1 19 f 82.8 64 6.9 19 m 78.4 45 5.1 All 80.4 329 7.5 Groups
[0116] Analysis of immunoreactivities of IgG against these 61 antigens was performed in 329 samples. For all samples, complex patterns of immunoreactivities could be found.
TABLE-US-00005 TABLE 2 List of antigens on customized microarray UniProt MW Abbreviation ID kDa Protein name UniProt Protein name in Study P62937 18.0 Peptidyl-prolyl cis-trans Cyclophilin A human Cyclophilin B isomerase A (Cyclophilin A) P61604 10.9 10 kDa heat shock Chaperonin 10, HSP 10 protein, mitochondrial Recombinant, (Hsp10) Human P00441 15.9 Superoxide dismutase Superoxide Dismutase SOD [Cu—Zn] from bovine erythrocytes P02686 33.1 Myelin basic protein Myelin Basic Protein MBP (MBP); Isoform 1 from bovine brain P04792 22.8 Heat shock protein Hsp27 Protein - Low HSP 27 beta-1 (Heat shock 27 Endotoxin kDa protein; Hsp27) P08107 70.1 Heat shock 70 kDa Heat Shock Protein 70 HSP 70 protein 1A/1B from bovine brain (Hsp70.1/Hsp70.2) P02751 262.6 Fibronectin; Isoform 1 Fibronectin from human Fibronektin plasma P01009 46.7 Alpha-1-antitrypsin α1-Antitrypsin from Alpha-1- human plasma Antitrypsin P08758 35.9 Annexin A5 Annexin V from human Annexin V placenta Q14694 87.1 Ubiquitin carboxyl-terminal Ubiquitin human Ubiquitin hydrolase 10 (USP10) P49773 13.8 Histidine triad nucleotide- Protein Kinase C PKC Inhibitor binding protein 1 Inhibitor, Myristoylated (Protein kinase C inhibitor 1) P02766 15.9 Transthyretin Prealbumin from human PreAlbumin (Prealbumin) plasma O76070 13.3 Gamma-synuclein γ-Synuclein human Gamma- synuclein P14136 49.9 Glial fibrillary acidic Anti-Glial Fibrillary GFAP protein (GFAP) Acidic Protein P27797 48.1 Calreticulin Calreticulin from bovine Calretikulin liver P02549 280.0 Spectrin alpha chain, Spectrin from human Spektrin erythrocyte erythrocytes P12081 57.4 Histidine-tRNA ligase, JO-1 human Jo-1 cytoplasmic (JO-1) P10809 61.1 60 kDa heat shock HSP60 (human), HSP 60 protein, mitochondrial (recombinant) (Hsp60) P53674 28.0 Beta-crystallin B1 βL-Crystallin from Beta-L- bovine eye lens Chrystalin P09211 23.4 Glutathione Glutathione S-Transferase GST S-transferase P from bovine liver P68133 42.1 Actin, alpha skeletal Actin from bovine muscle Actin muscle P15104 42.1 Glutamine synthetase Glutamine synthetase GLUL Q99798 83.4 Aconitase 2, aconitase 2, ACO2 mitochondrial mitochondrial E5RFU4 18.3 Dihydropyrimidinase- Dihydropyrimidinase- DBYSL2 like2 like 2 P09936 24.8 Ubiquitin carboxyl-terminal Ubiquitin carboxyl-terminal VCHC1 hydrolase isozyme L1 (UCHL1) hydrolase isozyme L1 P30086 21.1 Phosphatidylethanolamine- Phosphatidylethanola- PBP binding protein 1 mine-binding protein 1 P00918 29.2 Carbonic anhydrase 2 Carbonic Anhydrase II CAZ P12277 42.6 Creatine kinase B-type Creatine kinase B CKB P62873 37.4 Guanine nucleotide- Guanine nucleotide-binding GNB1 binding protein protein G(I)/G(S)/G(T) subunit G(1)/G(S)/G(T) subunit beta-1 (GNB1) beta 1 P06733 47.2 Alpha-enolase Alpha-Enolase ENO1 P04406 36.1 Glyceraldehyde-3- Glyceraldeyde (3-)phosphate GAPDH phosphate dehydrogenase dehydrogenase (GAPDH) P60842 46.2 Eukaryotic initiation Homo sapiens eukaryotic EIFA1 factor 4A-I translation initiation factor 4A isoform 1 (EIF4A1) mRNA A8K318 59.2 Protein kinase C protein kinase C PKC80 substrate 80K-H substrate 80K-H isoform 2 [Homo sapiens] Q68Y55 34.9 Poly(RC) binding poly(rC) binding protein PolyRp2 protein 2 2 isoform g [Homo sapiens] P49761 58.6 CDC-like kinase 3 Homo sapiens CDC-like CLK3 (CLK3), transcript kinase 3 (CLK3); variant phclk3, mRNA transcript variant phclk3; mRNA Q9P2Z0 28.4 THAP domain-containing THAP domain containing 10 THAP protein 10 [Homo sapiens] Q9BXS5 48.6 AP-1 complex subunit Homo sapiens adaptor- AP1M1 mu-1 (AP1M1) related protein complex 1; mu 1 subunit (AP1M1); mRNA P63330 35.6 Serine/threonine- protein phosphatase type pp2A protein phosphatase 2A 2A catalytic subunit catalytic subunit alpha alpha isoform isoform [Mus musculus] Q9NZT2 73.3 Opioid growth factor Homo sapiens opioid OGFR receptor growth factor receptor (OGFR). mRNA Homo sapiens plasticity- PRG2 related gene 2 (PRG2) mRNA P43235 37.0 Cathepsin K cathepsin K preproprotein Catepsin [Homo sapiens] Q53G92 50.4 Tubulin beta-3 chain Homo sapiens tubulin TUBB3 beta 3 (TUBB3) mRNA P37108 14.6 Signal recognition Homo sapiens signal SRP14 particle 14 kDa recognition particle protein 14 kDa (homologous Alu RNA binding protein) (SRP14) mRNA Q7Z6Z7 481.9 E3 ubiquitin-protein HUWE1 protein [Homo HUWE1 ligase HUWE1; sapiens] Isoform 1 Homo sapiens Chromo- chromosome X genomic contig, somX reference assembly Q96S16 36.9 JmjC domain-containing jumonji domain containing jumonji protein 8 (Jumonji 8 [Homo sapiens] domain-containing protein 8) Q96N21 55.1 Uncharacterized protein Homo sapiens Chromosome C17orf56 chromosome 17 open reading 17 frame 56 (C17orf56). mRNA Q96HG3 54.6 Islet cell autoantigen 1, Homo sapiens islet cell ICA1 69 kDa autoantigen 1. 69 kDa (ICA1). transcript variant 2. mRNA P10768 31.5 S-formylglutathione Homo sapiens esterase ESD hydrolase (Esterase D) D/formylglutathione hydrolase (ESD) P25325 33.2 3-mercaptopyruvate mercaptopyruvate MSI 2 sulfurtransferase sulfurtransferase isoform 2 [Homo sapiens] P27361 43.1 Mitogen-activated Homo sapiens mitogen- MAPK3 protein kinase 3; activated protein kinase 3 Isoform 1 (MAPK3); transcript variant 1; mRNA P43304 80.9 Glycerol-3-phosphate Homo sapiens glycerol- GPD2 dehydrogenase, 3-phosphate d ehydrogenase mitochondrial (GPD2); 2 (mitochondrial) Isoform 1 (GPD2); mRNA P36969 22.2 Phospholipid Homo sapiens glutathione GPX4 hydroperoxide peroxidase 4 (phospholipid glutathione hydroperoxidase) peroxidase, (GPX4); transcript mitochondrial variant 1. mRNA B7Z4U7 65.1 Sec1 family domain vesicle transport-related VTI-B containing 1, isoform protein isoform b CRA_b [Homo sapiens] Q9BVL4 73.5 Selenoprotein O Homo sapiens SELO selenoprotein O (SELO) mRNA Q6PJ21 39.4 SPRY domain-contain- SPRY domain-containing SPRX ing SOCS box protein 3 SOCS box protein SSB-3 P35611 81.0 Alpha-adducin adducin 1 (alpha) Adduccin isoform c [Homo sapiens] Q99798 85.4 Aconitate hydratase, Aconitate Hydratase 2 Aconitate mitochondrial (mitochondrial) Hydratase P06576 56.6 ATP synthase subunit ATP synthase ATP Synthase beta, mitochondrial P40926 35.5 Malate dehydrogenase, Malat dehydrogenase Malat mitochondrial Dehydrogenase P37840 14.5 Alpha-synuclein alpha-synuclein Alpha Synuclein P10636 78.9 Microtubule-associated tau TAU protein tau P05067 86.9 Amyloid beta A4 protein beta-amyloid Beta-Amyloid (Alzheimer disease amyloid protein) Q05923 34.4 Dual specificity protein DUSP2 dual specificity DUSP2 phosphatase 2 phosphatase 2 [Homo sapiens] Q14166 74.4 Tubulin-tyrosine ligase- Homo sapiens tubulin TTLL2 like protein 12 tyrosine ligase-like family member 12 (TTLL12) mRNA P09104 47.3 Gamma-enolase Gamma-enolase ENO2 P00441 15.9 Superoxide dismutase Superoxide dismutase SOD [Cu—Zn]
[0117] The immunoreactivities were analyzed in patients suffering from dry AMD, wet AMD, and compared to healthy controls. Table 3 shows the results of ANOVA analysis and their corresponding p-values for the most-significant antigens.
[0118] The immunoreactivities of dry and wet AMD patients are highly significantly different from each other and from controls. Although as part of natural autoimmunity also in healthy subjects complex antibody patterns against the tested antigens could be shown, the patterns in both AMD groups are changed. One of the most prominently changed reactivity is against alpha-synuclein. Alpha-synuclein is known from other neurodegenerative diseases. Others are heat shock proteins (e.g. HSP 10) and also Annexin V, which plays a major role in apoptotic processes. Some others are antithetic regulated in wet and dry-AMD such as VTI-B, PBP-12, and OGFR. Pathway comparison analysis revealed protein functions particularly found in immunological diseases, especially aconitase 2, which plays a role in citric acid cycle, further enolase 1, Annexin V, mitogen-activated protein kinase C (MAPK3) and Alpha-Synuclein. Mutations in Alpha-synuclein e.g. are associated with Parkinson's disease, Alzheimer's disease, and several other neurodegenerative illnesses. Annexin 5 is a phospholipase A2 and protein kinase C inhibitory protein with calcium channel activity and a potential role in cellular signal transduction, inflammation, growth and differentiation. Apart from the immunological markers, bio functions playing a major role in inflammation could be observed, like e.g. Alpha-synuclein, aconitase 2, enolase 1 and GADPH (glyceraldehydephosphate dehydrogenase), which acts in apoptotic processes and is also known to play a role in Alzheimer's disease. Annexin V is proposed to have anti-apoptotic and anti-inflammatory functions, comparison of immunoreactivities in our study revealed higher reactivity in the dry AMD-group. It could clearly be demonstrated that there are huge differences in the immunoreactivities in both wet and dry AMD groups.
TABLE-US-00006 TABLE 3 ANOVA (Analysis of Variance) of the IgG immunoreactivities in patients with wet AMD, dry AMD, and compared to controls. The table reveals the most significant antigens and according p-values. CTR CTR AMDD AMDD AMDW AMDW ANOVA - L- AV L- SE RY- AV RY- SE ET- AV ET- SE P Alpha 6428 487 10667 1974 7221 251 0.003 Synuclein SELO 31155 1755 27412 1733 26330 397 0.01 SPRY 24905 1445 21287 808 22305 265 0.028 GAPDH - H2 11571 821 13277 1137 14926 360 0.031 Annexin V 14977 1821 18675 1422 15258 321 0.034 THAP 14422 1575 10767 715 11889 284 0.044 VTI-B 22121 1816 19333 1493 23773 512 0.065 HSP 10 16099 1313 20613 1716 19235 385 0.071 ESD 29416 1393 24942 995 26008 424 0.082 PKC80 23592 1798 21236 1396 20575 335 0.082 ACO2 - C2 19238 1478 17569 1462 16185 376 0.089 OGFR 18774 2383 21865 3017 17521 516 0.115 PBP - I2 21243 1455 24279 1422 24944 467 0.119 CAZ - C3 5373 595 7534 1274 6402 196 0.148 EIFA1 26612 2773 21695 2045 21915 612 0.15 MAPK3 28505 1571 25186 1110 27123 315 0.15 ENO1 - H7 19297 1419 25493 2798 22685 594 0.15 Chromosome 17 20439 1587 18881 1513 21993 455 0.16 Aconitate 20584 1720 18805 1622 17738 391 0.163 Hydratase GPX4 19207 1159 17929 1277 17190 283 0.177 (AV = average; SE = standard error; p = p-value)
[0119] II. Longitudinal Analysis of IgG Immunoreactivities in Wet AMD
[0120] To answer the question if the injection of Lucentis which is basically a foreign protein to the body, is able to provoke a significant immune reaction, a longitudinal analysis of IgG immunoreactivities was performed.
TABLE-US-00007 TABLE 4 Longitudinal analysis over 12 months of IgG immunoreactivities in wet AMD patients, who started with lucentis treatment. Wilks' Lamda: values in the range of 0 (perfect discrimination) to 1 (no discrimination), Partial Lambda: The Wilks' Lambda associated with the unique contribution of the respective variable to the discriminatory power of the model. Wilks' - Partial - Lambda Lambda p-value TAU-n 0.793690 0.946549 0.143135 ACO2 - 0.802293 0.936398 0.084057 C2-n MAPK3-n 0.810685 0.926705 0.049457 ATP synthase-n 0.788763 0.952462 0.192688 GPD2-n 0.784479 0.957663 0.248013 AP1M1-n 0.777703 0.966006 0.364065
[0121] III. Analysis of IgG Immunoreactivities in Wet AMD: Influence of Success of Treatment
[0122] The immunoreactivities in patients suffering from wet AMD depending on success of treatment were compared. Based on visual acuity (log mar) the patients were divided into those showing vision loss during the 12 months treatment by lucentis (−1), no change in vision (0), or improvement of vision (1). Highly significant changes could be observed between those immunoreactivities from patients showing an improvement of vision during lucentis treatment and those who do not.
TABLE-US-00008 TABLE 5 ANOVA of Immunoreactivities in wet AMD patients depending on treatment. The patients were divided into those showing vision loss (−1), no change in vision (0), or improvement of vision (1). 0-AV 0- SE 1-AV 1- SE -1-AV -1- SE ANOVA - P Cyclopilin B 33179 949 32345 1463 33452 1221 0.813 HSP 10 19235 579 19821 837 18591 704 0.496 HSP70 7866 341 7942 501 8207 492 0.837 Fibronektin 27083 472 26845 512 26608 656 0.816 Alpha-1- 13139 552 12626 725 12866 675 0.853 Antitrypsin Annexin V 15507 487 15107 620 15243 633 0.879 Ubiquitin 17679 597 17507 504 16529 616 0.332 PKC Inhibitor 12519 587 12385 857 13294 669 0.615 PreAlbumin 15562 873 17430 962 12944 756 0.003 Gamma-synuclein 22228 618 23983 746 20546 600 0.003 GFAP 35195 783 34749 1139 35638 1126 0.839 Calretikulin 32672 877 30756 640 33606 945 0.094 Spektrin 24946 619 24708 670 23095 730 0.107 Jo-1 27542 590 31206 1049 30400 892 0.003 HSP 60 23144 988 27559 1267 23609 1274 0.022 Beta-L-Chrystalin 11728 478 10959 666 11006 605 0.538 GST 16706 684 15214 830 15230 811 0.259 Actin 23996 958 26141 1356 24624 1187 0.421 GLUL - A2 19825 607 20624 546 20164 614 0.66 ACO2 - C2 15714 601 16579 781 16185 675 0.665 DBYSL2 - E2 19515 748 21426 1050 19609 732 0.228 VCHC1 - G2 11973 608 10395 722 12151 678 0.164 PBP - I2 24591 697 26074 1009 23293 718 0.069 CAZ - C3 6615 341 5919 300 6646 378 0.306 CKB - K1 4835 167 4906 267 5269 294 0.375 GNB1 - A3 7772 329 6949 320 8550 456 0.021 ENO1 - H7 19856 753 22785 935 23954 1210 0.005 GAPDH - H2 15272 614 15041 719 14734 597 0.826 EIFA1 25391 1158 19280 745 20476 1000 0 PKC80 20766 484 21545 741 20276 588 0.359 PolyRp2 15683 385 14937 350 18082 549 0 CLK3 28064 571 27181 646 27717 639 0.612 THAP 11395 386 12273 567 12608 586 0.178 AP1M1 10985 346 10519 408 12470 482 0.004 pp2A 24322 591 23963 598 22493 655 0.085 OGFR 16924 804 22120 1178 15111 691 0 PRG2 26929 1146 30959 1337 25169 1230 0.008 Catepsin 24577 707 23876 584 24513 778 0.774 TUBB3 12899 382 12244 415 14315 693 0.021 SRP14 25498 1073 20329 692 21662 1301 0.002 HUWE1 25755 585 23901 645 25528 716 0.119 Chromosom X 11364 483 12210 612 11548 497 0.522 jumonji 18742 344 18098 491 18518 442 0.559 Chromosome 17 20052 584 20932 667 24848 1044 0 ICA1 16622 618 17327 725 19514 1099 0.033 ESD 25863 580 25324 700 26872 978 0.38 MSI 2 26033 684 25447 733 24186 586 0.13 MAPK3 28919 577 23712 375 27368 470 0 GPD2 20415 526 20057 731 23181 852 0.003 GPX4 16304 277 16926 435 18737 728 0.001 VTI-B 23423 828 24339 1089 23596 882 0.777 SELO 27500 632 25618 749 25330 735 0.046 SPRX 22907 463 21735 452 21944 457 0.15 Adduccin 12874 419 13792 797 13348 558 0.522 Aconitate 16576 509 18093 817 17913 746 0.193 Hydratase ATP synthase 27814 440 25813 469 27848 601 0.011 Malat 27349 844 25088 847 26175 984 0.216 Dehydrogenase Alpha Synuclein 7553 398 8008 694 6464 231 0.056 TAU 12493 374 13288 576 13714 663 0.219 Beta-Amyloid 8235 161 8344 246 8042 198 0.576 DUSP2 12211 725 13610 856 12014 721 0.33 TTLL2 16754 1387 19730 1426 16204 1103 0.18 SOD 1361 728 287 892 1460 659 ENO2 9538 13415 9267 16430 27101 12134 (AV = average, SE = standard error, p = p-value)
TABLE-US-00009 TABLE 6 ANOVA of immunoreactivities in wet AMD patients depending on treatment. The patients were divided into those showing vision loss (−1), no change in vision (0), or improvement of vision (1). 0-AV 0- SE 1-AV 1- SE -1-AV -1- SE ANOVA - P MAPK3 28919 577 23712 375 27368 470 6 OGFR 16924 804 22120 1178 15111 691 0 PolyRp2 15683 385 14937 350 18082 549 0 Chromosome 20052 584 20932 667 24848 1044 0 17 EIFA1 25391 1158 19280 745 20476 1000 0 GPX4 16304 277 16926 435 18737 728 0.001 SRP14 25498 1073 20329 692 21662 1301 0.002 Gamma- 22228 618 23983 746 20546 600 0.003 synuclein Jo-1 27542 590 31206 1049 30400 892 0.003 PreAlbumin 15562 873 17430 962 12944 756 0.003 GPD2 20415 526 20057 731 23181 852 0.003 AP1M1 10985 346 10519 408 12470 482 0.004 ENO1 - H7 19856 753 22785 935 23954 1210 0.005 PRG2 26929 1146 30959 1337 25169 1230 0.008 ATP 27814 440 25813 469 27848 601 0.011 synthase GNB1 - A3 7772 329 6949 320 8550 456 0.021 TUBB3 12899 382 12244 415 14315 693 0.021 HSP 60 23144 988 27559 1267 23609 1274 0.022 ICA1 16622 618 17327 725 19514 1099 0.033 SELO 27500 632 25618 749 25330 735 0.046 (AV = average, SE = standard error, p = p-value)
[0123] Additionally, a general regression model (GRM) was performed to analyze if there is a longitudinal effect on the antibody patterns depending on success of treatment. Several antigens could be detected which show significant longitudinal effects.
[0124] Thus, the immunoreactivities could have predictive value to determine the treatment effect of lucentis in AMD patients. Involved are antigens such as MAPK3, Gamma-synuclein, different heat shock proteins and many others. Several of the immunoreactivities are regulated in opposed directions. E.g. immunoreactivities against OGFR, Gamma-synuclein, pre-albumin, and PRG2 are up-regulated in patients improving vision during 12 months under lucentis therapy and down-regulated in those who do not. Furthermore, e.g. immunoreactivities against PolyRp2, GNB1-A3, and TUBB3 are down-regulated in patients improving vision during 12 months under lucentis treatment and up-regulated in those who do not. Furthermore, e.g. some heat shock proteins show huge difference in IgG immunoreactivities against them.
[0125] Pathway analysis showed that nine proteins were associated with cellular growth and proliferation, namely ATP synthase, eukaryotic translation initiation factor 4A1 (EIFA1), enolase 1, glutathione peroxidase 4 (GPX4), HSP60, MAPK3, Opioid growth factor receptor (OGFR), Gamma-Synuclein and beta Tubulin (TUBB3). Immunoreactivities against OGFR were up-regulated in patients improving in vision under lucentis therapy, whereas a down-regulation was observed in case of vision loss.
[0126] IV. Predictive Value of Immunoreactivities on Success of Treatment
[0127] Based on visual acuity (log mar) the patients were divided into those showing vision loss during the 12 months treatment by lucentis (−1), no change in vision (0), or improvement of vision (1). Highly significant changes could be observed between those immunoreactivities from patients showing an improvement of vision during lucentis treatment and those who do not. If the immunoreactivities might have predictive value on lucentis treatment, data mining models could be useful to distinguish between those patients just based on their antibody patterns.
[0128] C&RT models (general classification/regression tree models) were performed. Furthermore, an artificial neural network was trained to recognize those antibody patterns which are correlated with a successful treatment with lucentis. After training, those patterns could be recognized with a sensitivity and specificity of about 95%. Nevertheless, it is important to consider that the number of patients in this pilot study is small and further validation studies in larger cohorts are needed to address the predictive value in detail. The most discrimination power was attributed to e.g. OGFR, SPR14, pre-albumin etc. which also were important immunoreactivities in the preceding analyses.
[0129] V. Predictive Value of Immunoreactivities on Course of Vision Loss
[0130] The results of the data mining procedures (C&RT and artificial neural networks) were used as input for this analysis. A survival analysis (Kaplan-Meier analysis, Cumulative proportion surviving) was performed to assess how strongly the classification of the data mining procedures influence the survival time (time to vision loss) in this 12 months study.
[0131] The Kaplan-Meier plot demonstrates the huge impact this a priori classification based on their antibody patterns has on the outcome of lucentis treatment over time.
[0132] Whereas those patients classified as good lucentis responder show nearly no vision loss in survival analysis, the other group (bad lucentis responder) reveal strong vision loss.
[0133] VI. Influence of Other Clinical Parameter on Immunoreactivities
[0134] A component of variance analysis was performed to assess the effect of the parameters vision change, retinal hemorrhage, macular edema, and choroidal neovascularization (CNV) on immunoreactivities. There are no very clear singular effects of these clinical parameters on the immunoreactivity levels for each antigen. For all of the antigens, the effect of change in vision, retinal hemorrhage, macular edema, or CNV are very small and not significant. However, the interaction between those (e.g. 1*2*3*4) can have a larger impact on the antibody patterns.
[0135] Mass Spectrometric Analysis
[0136] The analysis of antigen-antibody profiles can be done in a reliable and sensitive manner using a recently developed proteomics technology: protein G Dynabeads combined with a ProteinChip system based on SELDI-TOF (=surface enhanced laser desorption/ionization-time of flight) mass spectrometry (MS) or Maldi-TOF or ESI-MSMS. The magnetic beads are designed to capture immunoglobulins via a cell wall component, binding a wide range of IgG antibodies during incubation with various body fluids, such as sera. During a subsequent incubation with homogenized antigens it is possible to capture relevant antigens by secondary binding to the antibodies. After elution antigens can be analyzed by SELDI-TOF MS using ProteinChips with different, separating chip surfaces e.g. cationic and anionic exchangers, hydrophobic surfaces and metal-ion affinity-chromatographic surfaces. Resulting mass spectra can be statistically analyzed and compared to gain significantly higher or lower antigen-antibody-reactivity peaks according to the study groups. The identification of potential biomarkers will be done using highly sensitive MALDI-TOF/TOF (=matrix assisted laser desorption/ionization-time of flight) MS.
[0137] Protagen Arrays
[0138] For the analysis of antibody patterns and identification of potential autoantibody biomarker candidates we chose a highly sensitive antigen microarray, which is a promising approach in this field of interest. This method has already been successfully used for the discovery of autoantibodies targeting prostate cancer specific biomarkers and to screen sera of patients with, e.g. different pathological subtypes of multiple sclerosis or autoimmune hepatitis for autoreactive antibodies. To screen autoantibody reactivities in study sera we used an advanced high density microarray approach. Sera of patients before treatment with lucentis (n=10) were compared with sera of the same patients after treatment with lucentis (n=10). Two pools of ten sera each were created for each group, which were incubated on nitrocellulose-coated slides with 3800 immobilized randomly selected human proteins from the UNIclone® library (UNIchip®, Protagen, Dortmund, Germany) as described below. Incubation and washing steps were performed at 4° C. n on an orbital shaker (Micromix 5, DPC, Los Angeles, Calif., USA). Slides were covered with one-pad FAST-frame hybridization chambers (Whatman, Maidstone, UK) and blocked with PBS containing 0.5% BSA for one hour. Afterwards slides were washed three times ten minutes each time with PBS containing 0.5% Tween 20 (PBS-T). Patients' sera were diluted 1:375 in PBS and incubated on the Protagen-Slides overnight. After three washing steps with PBS-T, each time for ten minutes, slides were treated with fluorescence labeled secondary antibody (1:500 diluted in PBS, goat anti-human IgG, Jackson ImmunoResearch Laboratories, West Grove, USA) for one hour in the dark. After three final washing steps, two with PBS-T and one with HPLC-grade ultra pure Water (ten minutes each time) slides were dried under vacuum. By using a high sensitive laser microarray scanner 16-bit TIFF (Tagged Information File Format) were generated. Spot intensities were quantified with ImaGene Software (ImaGene 5.5, Biodiscovery, CA, USA). After data normalization to internal standards with algorithm provided by Protagen, group differences were calculated and compared. For visualization of the resultant antigen-antibody complexes, slides were treated with a secondary fluorescence labeled antibody (Dylight 650) followed by confocal laser scanning. After data normalization spot intensities were compared and group differences were analyzed.
[0139] Analysis
[0140] Blood samples will be centrifuged at 1000 g for ten minutes and the supernatant will be stored at −80° C. for subsequent analysis. Magnetic protein G beads (Dynal, Oslo, Norway) will be incubated with the patient's sera. After several washings steps the patient's antibodies will be covalently bound to the beads using ethanolamine. The bead-antibody complexes will be incubated with homogenized retinal antigens. The antigens bound to the patient's autoantibodies will be eluted, concentrated, and analyzed by SELDI time-of-flight (TOF) MS ProteinChips with two different chromatographic surfaces (CM10 cation exchange and H50 reversed phase). The samples will be measured with a SELDI-TOF MS ProteinChip system (Biorad, Hercules) on a PBS-IIc ProteinChip Reader. Raw data will be transferred to CiphergenExpress 2.1 database software (Biorad, Hercules) for workup and analysis. A recently developed Proteomics Software Project (PSP) will statistically evaluate the spectra using different statistical approaches (trained neuronal networks, tree algorithms and multivariate statistics) to guarantee a high specificity and sensitivity of antibody patterns for the observed study groups. The PSP will additionally search for highly significant biomarkers directing a Statistical based analysis using above mentioned algorithms. The identification of biomarkers will be done by MALDI-TOF/TOF MS analysis. We aim to generate at least eight highly specific biomarkers (significance level α=0.05 and power (1-ß)=90%) for “wet” AMD.
[0141] Statistical calculations of sample sizes were conducted in close cooperation with the Institute of Medical Biometry, Epidemiology and Informatics (IMBEI) at the University in Mainz and are also based on experiences from previous studies: the calculated number of cases (25) is sufficient to detect an effect on the serum antibody profiles, given a significance level α=0.05 and power (1-ß)=90%. The statistical analysis will demonstrate if the antibody composition against retinal antigens within sera changes. A comparison to the control group will show if the modifications are beneficial, i.e. the serum compositions become more similar to the serum of healthy subjects, or not. A subsequent biomarker identification using MALDI-TOF/TOF MS (Bruker) may reveal valuable hints on the systemic effects.
[0142] After electrophoretic separation, proteins will be tryptically digested, crystallized on matrix, and analyzed on a MALDI target. The obtained peptide mass fingerprint data, will be exported into BioTools and used for an internal Mascot database search (SwissProt and NCBI), leading to protein identifications.
[0143] Antigen Microarrays
[0144] In this study, we used highly purified proteins, purchased at Sigma-Aldrich (Germany) and BioMol (Hamburg, Germany), as antigens. The antigen selection is based on previous autoantigen identifications in glaucoma patients by our group and survey of literature related to identifications of autoantigens in autoimmune diseases. Antigens were diluted to 1 μg/μl with PBS buffer optionally containing 1.5% Trehalose for optimal printing conditions. The spotting of antigens was performed with both a non-contact printing technology (sciFLEXARRAYER S3, Scienion, Berlin, Germany), based on piezo dispensing, and the commonly used pin based contact printing technique (OmniGrid100, Digilab Genomic Solutions, Ann Arbor, USA). Results were comparatively evaluated for spot morphology and spot to spot variability. For printing of the whole set of study microarrays the piezo based spotting technique was used. Each antigen was spotted in triplicate onto nitrocellulose-slides (Oncyte, nitrocellulose 16 multi-pad slides, Grace Bio-Labs, Bend, USA). As a positive and negative control we used mouse anti human IgG/A/M or human IgG (10 μg/μl) and spotting buffer. The spotting process was performed at RT and a humidity of 30%. Approximately 1 nl of each antigen-dilution was applied onto the nitrocellulose surface by spotting three times 330 μl on exactly the same position. The accurateness of the spotting volume and the correct positioning of the droplets were monitored prior and after the spotting process of each antigen using the sciDrop-VOLUME and autodrop-detection software (Scienion, Berlin, Germany).
[0145] Incubation and washing steps were performed at 4° C. on an orbital shaker (Titramax 100, Heidolph, Schwabach, Germany). Slides were covered with 16-pad FAST frame hybridization chambers (Whatmann, Maidstone, UK) and blocked with PBS containing 4% BSA or the Super G blocking buffer (Grace Biolabs) for one hour. Afterwards slides were washed three times with PBS containing 0.5% Tween (PBS-T). Patient sera were diluted 1:250 in PBS and aqueous humor in a ratio of 1:10 in PBS. 100-120 μl of these dilutions were randomly incubated on prepared antigen-slides overnight. After several washing steps with PBS-T, slides were incubated with a fluorescent Cy-5 labeled secondary antibody (1:500 diluted in PBS-T, goat anti-human IgG, Jackson ImmunoResearch Laboratories, West Grove, USA) for one hour in the dark. Two washing steps with PBS-T were followed by two final washing steps with HPLC-grade water. All microarrays were air dried before scanning, using a microarray scanner (Affymetrix 428™ Array Scanner, High Wycombe, UK). Generated 16-bit TIFF images (Tagged Information File Format) of slides were analyzed using the Spotfinder 3.1.1 software (TM4, Dana-Faber Cancer Institute, Boston, USA) or ImaGene5 software. Background substraction was performed according to the formula: spot intensity=mean intensity SP−((sumbkg−sumtop5 bkg)/(number of pixelSP−number of pixelstop5 bkg)) where SP represents any spot, bkg the corresponding background and top5 bkg the top five percent of background pixel. The coefficient of variance (CV) was calculated as follows: CV=SDSP3/meanSPX . . . SPn, where SDSP3 represents the standard deviation across three replicate spots of one antigen of one sample, and meanSPX . . . SPn the mean of all spot intensities.
Example
[0146] An artificial neural network was performed prior to the analysis in a study cohort as described above. The autoantibody reactivities were analysed as described above.
[0147] These intensity values of antibody reactivities have been normalized and a calculation of the percentage difference of intensity values to reference values was calculated.
[0148] Based on the algorithm from the artificial neural network, an individual scoring was performed for a single patient, which has not been included prior to this analysis in the training (calculation) of the artificial neural network.
[0149] The different autoimmune reactivates were analysed for patient #22928 (from the study) in 1 μl of sera and used as input to the data mining algorithm trained prior to the study.
[0150] For this individual patient, the confidence levels was calculated as 0.0087 (−1: vision loss); 0,156 (0: no vision change) and 0,835 (1: vision gain). Thus, based on autoreactivity, the patient will respond to the anti-VEGF treatment with highest probability.
[0151] While there are shown and described presently preferred embodiments of the invention, it is to be distinctly understood that the invention is not limited thereto but may be otherwise variously embodied and practiced within the scope of the following claims.