Niculescu

BLOOD BIOMARKERS FOR PSYCHOSIS

CROSS REFERENCE TO RELATED APPLICATIONS

[0001 ] This application claims priority to U S. provisional application Ser. No. 60/917,784, filed May

14, 2007, the disclosure of which is hereby incorporated by reference in its entirety.

[0002] Part of the work during the development of this invention was made with government support from the National Institutes of Health under grant NIMH ROl MH071912-01 The U S Government has certain rights in the invention

BACKGROUND

[0003] Research into the biological basis of psychotic disorders (such as schizophrenia and schizoaffective disorder) has been primarily focused in human and animal studies mostly independently The two avenues of research have complementary strengths and weaknesses In human genetic studies, for example, in samples of patients with psychotic disorders and their family members, positional cloning methods such as linkage analysis, linkage-disequilibrium mapping, and candidate-gene association analysis are narrowing the search for the chromosomal regions harboring risk genes for the illness and, m some cases, identifying plausible candidate genes and polymorphisms that will require further validation Human postmortem bram gene expression studies have also been employed as a way of trying to identify candidate genes for psychotic disorders disorders. In general, human studies suffer from issues of sensitivity- the signal is often difficult to detect due to the noise generated by the genetic heterogeneity of individuals and the effects of diverse environmental exposures on gene expression and phenotypic penetrance.

[0004] In animal studies, carried out in isogenic strains with controlled environmental exposure, the identification of putative neurobiological substrates of psychotic disorders is typically accomplished by modeling human psychotic disorders disorders through pharmacological or genetic manipulations. Animal model studies suffer from issues of specificity- questions regarding direct relevance to the human disorder modeled. Each independent line of investigation (i.e., human and animal studies) is contributing to the incremental gains in knowledge of psychotic disorders disorders etiology witnessed m the last decade

[0005] However, a lack of integration between these two lines of investigations hinders scientific understanding and slows the pace of discovery. Psychiatric phenotypes, as currently defined, are primarily the result of clinical consensus criteria rather than empirical determination. The present disclosure provides methods and compositions that empirically determine disease states for diagnosis and treatment.

[0006] Objective biomarkers of illness and treatment response would make a significant difference in the ability to diagnose and treat patients with psychotic disorders, eliminating subjectivity and reliance

Niculescu

of patient's self-report of symptoms Blood gene expression profiling has emerged as a particularly interesting area of research in the search for peripheral biomarkers Most of the studies to date have focused on human lymphocytes gene expression profiling, comparison between illness groups and normal controls. They suffer from one of both of the following limitations: 1) the sample size used is often small. Given the genetic heterogeneity in human samples and the effects of illness state and environmental history, including medications and drugs, on gene expression, it may not be reliable to extract bona fide findings. 2) Use of lymphoblastoid cell lines — passaged lymphoblastoid cell lines provide a self -renewable source of material, and are purported to avoid the effects of environmental exposure of cells from fresh blood. Fresh blood, however, with phenotypic state information gathered at time of harvesting, may be more informative than immortalized lymphocytes, and may avoid some of the caveats of Epstem-Barr virus (EBV) immortalization and cell culture passaging.

[0007] The current state of the understanding of the genetic and neurobiological basis for psychotic disorders in general and of peripheral molecular biomarkers of the illness in particular, is still inadequate Almost all of the fundamental genetic, environmental, and biological elements needed to delineate the etiology and pathophysiology of psychotic disorders are yet to be completely identified, understood and validated One of the rate-limiting steps has been the lack of concerted integration across disciplines and methodologies The use of a multidisciplinary, integrative research framework as in the present disclosure provided herein, should lead to a reduction in the historically high rate of inferential errors committed in studies of complex diseases like psychotic disorders

[0008] Identification and validation of peripheral biomarkers for psychotic disorders have proven arduous, despite recent large-scale efforts Human genomic studies are susceptible to the issue of being underpowered, due to genetic heterogeneity, the effect of variable environmental exposure on gene expression, and difficulty of accrual of large samples. Animal model gene expression studies, in a genetically homogeneous and experimentally tractable setting, can avoid artifacts and provide sensitivity of detection Subsequent comparisons of the animal datasets with human genetic and genomic datasets can ensure cross-vahdatory power and specificity.

[0009] Convergent functional genomics (CFG), is an approach that translationally cross-matches animal model gene expression data with human genetic linkage data and human tissue data (blood, postmortem brain), as a Bayesian strategy of cross validating findings and identifying candidate genes, pathways and mechanisms for neuropsychiatric disorders. Predictive biomarkers for psychosis are desired for clinical diagnosis and treatment purposes. The present disclosure provides several biomarkers that are predictive of psychotic disorders in clinical settings.

[0001 0] No objective clinical laboratory blood tests for psychosis is available. The current reliance on patient self -report of symptom severity and on the clinicians' impression alone are rate limiting steps in

Niculescu

effective treatment, and in new drug development. Blood biomarkers for psychosis state provide useful tools for diagnosis and therapy.

SUMMARY

[0001 1 ] Methods and compositions to clinically diagnose psychotic disorders using a panel of biomarkers are disclosed. A panel of biomarkers may include 1 to about 100 or more biomarkers. The panel of biomarkers includes one or more biomarkers for psychosis Blood is a suitable sample for measuring the levels or presence of one or more of the biomarkers provided herein

[0001 2] In an aspect, psychotic symptoms measured in a quantitative fashion at time of blood draw in human subjects focus on all or nothing phenomena (genes turned on and off in low symptom states vs. high symptom states). Some of the biomarkers have cross-matched animal and human data, using a convergent functional genomics approach and from blood datasets from animal models

[0001 3] Prioritized list of high probability blood biomarkers, provided herein, for psychotic disorders using cross-matching of animal and human data, provide a unique predictive power of the biomarkers, which have been experimentally tested.

[00014] Integration of human and animal model data, as a way of reducing the false-positives inherent in each approach and helping identify true biomarker molecules were adopted. Whole-genome gene expression differences were measured in fresh blood samples from patients with schizophrenia and related disorders that had no symptoms of hallucinations or delusions vs those that had high symptoms at the time of the blood draw, and separately, changes in gene expression in the brain and blood of a mouse pharmacogenomic model Human blood gene expression data was integrated with animal model gene expression data, human genetic linkage/association data, and human postmortem data, an approach called Convergent Functional Genomics, as a Bayesian strategy for cross-validating and prioritizing findings.

[0001 5] Candidate biomarker genes for hallucinations, include four genes decreased in expression in high hallucinations states (Rhobtb3, Aldhlll, Mpp3, FnI), and two genes increased in high hallucinations states (Arhgef9, S100a6). Five of these genes have evidence of differential expression in human postmortem brains from schizophrenia patients A predictive score developed based on a panel of 10 top candidate biomarkers (5 for no hallucinations, 5 for high hallucinations) shows sensitivity and specificity for high hallucinations and no hallucinations states, in two independent cohorts.

[0001 6] Candidate biomarker genes for delusions include eight genes decreased m expression in high delusions states (Drd2, ApoE, Nabl, Idhl, Scampi, Ncoa2, Aldhlll, Gpmόb), and eight genes increased in high delusions states (Nrgl, Egrl, Dctnl, Nmtl, PlIp, Pvalb, Nmtl, Pctkl) Fourteen of these genes have evidence of differential expression in human postmortem brains from schizophrenia patients. A predictive score developed based on a panel of 10 top candidate biomarkers (5 for no delusions, 5 for high delusions) shows sensitivity and specificity for high delusions and no delusions

Niculescu

states. Blood biomarkers offer an unexpectedly informative window into brain functioning and psychotic diseases states. [0001 7] A method of diagnosing psychosis in an individual, the method includes:

(a) determining the expression of a plurality of biomarkers for delusion or hallucination in a sample from the individual, the plurality of biomarkers selected from the group of biomarkers listed in Table 5 A, Table 5B, Table 6A, and Table 6B; and

(b) diagnosing the presence or absence of psychosis in the individual based on the expression of the plurality of biomarkers.

[0001 8] A plurality of biomarkers include a subset of about 10 biomarkers for delusions designated as

Drd2, ApoE, Scampi, Idhl, Nabl, Nrgl, Egrl, Dctnl, PlIp, and Pvalb or a subset of about 10 biomarkers for hallucinations designated as Rhobtb3, Aldhlll, Mpp3, FnI, Sppl, Arhgef9, S100a6, Adamts5, Pdapl, and Plxndl.

[0001 9] A suitable sample is blood The level of the biomarker can also be determined in a tissue biopsy sample of the individual. The level of the biomarker is determined by a method selected from the group that mlcudes analyzing the expression level of RNA transcripts, analyzing the level of protein, and analyzing the level of peptides or fragments thereof. Suitable analytical techniques include microarray gene expression analysis, polymerase chain reaction (PCR), real-time PCR, quantitative PCR, lmmunohistochemistry, enzyme-linked immunosorbent assays (ELISA), and antibody arrays The level of the plurality of biomarkers is performed by an analysis for the presence or absence of the biomarkers

[00020] A method of predicting the likelihood of a successful treatment for psychosis in a patient includes-

(a) determining the expression level of at least 10 biomarkers for delusion and 10 biomarkers for hallucination, wherein the biomarkers comprise a subset of biomarkers designated as Drd2, ApoE, Scampi, Idhl, Nabl, Nrgl, Egrl, Dctnl, PlIp, and Pvalb for delusion and Rhobtb3, Aldhlll, Mpp3, FnI, Sppl, Arhgef9, S100a6, Adamts5, Pdapl, and Plxndl are present for hallucination, and

(b) predicting the likelihood of successful treatment for psychosis by determining whether the sample from the patient expresses biomarkers for delusion or hallucination

[00021 ] A method of treating a patient suspected of suffering psychosis includes:

(a) diagnosing whether the patient suffers from psychosis by determining the expression level of one or more of the biomarkers listed in Tables 5 A, 5B, 6A, 6B in a sample obtained from the patient;

(b) selecting a treatment for psychosis based on the determination whether the patient suffers from delusion or hallucination, and

(c) administering to the patient a therapeutic agent capable of treating psychosis.

Niculescu

[00022] A treatment plan may include a personalized plan for the patient. A diagnostic microarray for psychosis includes a plurality of nucleic acid molecules representing genes selected from the group of genes listed in Tables 5A-5B and 6A-6B. The diagnostic microarray may consist essentially of biomarkers listed in Table 3A-3B.

[00023] A diagnostic microarray may consist essentially of biomarkers designated as Drd2, ApoE,

Scampi, Idhl, Nabl, Nrgl, Egrl, Dctnl, PlIp, and Pvalb for delusion and Rhobtb3, Aldhlll, Mpp3, FnI, Sppl, Arhgef9, S100a6, Adamts5, Pdapl, and Plxndl for hallucination

[00024] A diagnostic antibody array includes a plurality of antibodies that recognize one or more epitopes corresponding to the protein products of the biomarkers designated as Drd2, ApoE, Scampi, Idhl, Nabl, Nrgl, Egrl, Dctnl, PlIp, and Pvalb for delusion and Rhobtb3, Aldhlll, Mpp3, FnI, Sppl, Arhgef9, S100a6, Adamts5, Pdapl, and Plxndl for hallucination. The diagnostic antibody array may detect the protein levels of the biomarkers from a blood sample.

[00025] A kit for diagnosing psychosis includes a component selected from the group of (i) oligonucleotides for amplification of one or more genes listed in Tables 5 A-5B and 6A-6B (ii) immunohistochemical agents capable of identifying the protein products of one or more biomarkers listed in Tables 5 A-5B and 6A-6B (iii) the microarray of disclosed herein, and (iv) a biomarker expression index representing the genes listed in Tables 5A-5B and 6A-6B for correlation.

BRIEF DESCRIPTION OF THE DRAWINGS

[00026] FIG. 1 shows Prioritization (A) and Conceptualization (B) of results: A.) Convergent

Functional Genomics (CFG) approach for candidate biomarker prioritization Scoring of independent lines of evidence (maximum score= 9 points); B.) Conceptualization of Blood Candidate Biomarker Genes: I — Genes for the illness, whose expression is modulated by medications and by interactions with the environment (stress, drugs, al.); II — Genes for the illness, whose expression is modulated by medications; Ilia — Genes whose expression is modulated by medications and by interactions with the environment (stress, drugs, al ); IHb — Genes for the illness, whose expression is by interactions with the environment (stress, drugs, al ); IVa — Genes whose expression is modulated by medications.

[00027] FIG. 2A illustrates some of the candidate biomarker genes for delusions (Pl). Both Human

Postmortem Bram and Human Blood significance was used. Both Mouse bram and Mouse Blood (italicized), Co-directional m bram-blood *; Convergence with Human Genetic Linkage to Schizophrenia; the high delusions score are not circled; circled genes — associated with low delusions score ; 2B illustrates some of the candidate biomarker genes for hallucinations (P3). Both Human Postmortem Bram and Human Blood; Both Mouse brain and Mouse Blood (italicized); Co-directional in brain-blood *; Convergence with Human Genetic Linkage to Schizophrenia, high hallucinations score (non-circled), associated with low hallucinations score (circled)

Niculescu

[00028] FIG. 3 shows comparison of BioM-10 Hallucinations Prediction Score and actual hallucinations scores in the primary cohort of psychosis subjects (A) (n=31) and secondary psychosis cohort (B) (n=14). For hallucinations scores: blue-no hallucinations red-high hallucinations, white- intermediate hallucinations Hallucinations scores are based on PANSS scale administered at time of blood draw. For biomarkers: A (blue)- called Absent by MAS5 analysis P (red)- called Present by MAS5 analysis. M (yellow)- called Marginally Present by MAS5 analysis A is scored as 0, M as 0 5 and P as 1. BioM Hallucinations Prediction Score is based on the ratio of the sum of the scores for high mood biomarkers and sum of scores for low mood biomarkers, multiplied by 100. A cutoff score of 100 and above was used for high delusions, inf- infinity-denominator is 0. ND- not determined

[00029] FIG. 4 shows comparison of BioM-10 Delusions Prediction Score and actual delusions scores in the primary cohort of psychosis subjects (A) (n=31) and secondary psychosis cohort (B) (n=14). For delusions scores blue-no delusions, red-high delusions, white- intermediate delusions Delusions scores are based on PANSS scale administered at time of blood draw. For biomarkers: A (blue)- called Absent by MAS5 analysis P (red)- called Present by MAS5 analysis M (yellow)- called Marginally Present by MAS5 analysis. A is scored as 0, M as 0.5 and P as 1. BioM Delusions Prediction Score is based on the ratio of the sum of the scores for high mood biomarkers and sum of scores for low mood biomarkers, multiplied by 100 A cutoff score of 100 and above was used for high delusions inf- infinity- denominator is 0

DETAILED DESCRIPTION

[00030] In an aspect, the biomarkers disclosed herein are (i) derived from fresh blood, not immortalized cell lines; (ii) capable of providing quantitative psychosis information obtained at the time of the blood draw; (iii) were derived from comparisons of extremes of low delusion/high delusion and low/high hallucination in patients, as opposed to patients vs normal controls (where the differences could be due to a lot of other environmental factors, medication (side) effects vs. no medications; (iv) based upon a smaller sample size and yet robust in their predictive power; (v) scored based on an all or nothing (Absent/Present) call for gene expression changes, not incremental changes in expression — statistically more robust and avoids false positives; (vi) based on integration of multiple independent lines of evidence that permits extraction of signal from noise (large lists of genes), and prioritization of top candidates; and (vii) used to form the basis of prediction score algorithm based

[00031 ] Integration of animal model and human data were used as a way of reducing the false-positives inherent in each approach and helping identify true biomarker molecules. Gene expression differences were measured in fresh blood samples from patients with psychotic disorders (delusions/hallucinations) at the time of the blood draw. Separately, changes in gene expression were measured in the brains and bloods of a mouse pharmacogenomic models. Human blood gene expression data was integrated with

Niculescu

animal model gene expression data, human genetic linkage/association data, and human postmortem data for cross-validating and prioritizing findings.

[00032] Gene expression changes in specific brain regions and blood from a pharmacogenomic animal model developed in the group were used as cross-validators to help with the identification of potential human blood biomarkers The pharmacogenomic mouse model of relevance to psychosis consists of treatments with an agonist of the illness/psychosis -mimicking drug (phencyclidine, PCP) and an antagonist of the illness/ psychosis-treating drug (clozapine). The pharmacogenomic approach is a tool for tagging genes that may have pathophysiological relevance. As an added advantage, some of these genes may be involved m potential medication effects present in human blood data (FIG 2).

[00033] Human whole blood gene expression studies were initially carried out in a primary cohort of psychosis subjects Whole blood was used as a way of minimizing potential artifacts related to sample handling and separation of individual cell types, and also as a way of having a streamlined approach that lends itself well to scalability and future large scale studies in the field. Genes that were differentially expressed in no symptoms vs. high symptoms subjects were compared with 1) the results of the animal model brain and blood data, as well as 2) published human genetic linkage/association data, and 3) human postmortem brain data, as a way of cross-validating the findings, prioritizing them, and coming up with a short list of high probability candidate biomarker genes (FIGS IA and 2)

[00034] Afocused approach was used for looking separately at two discrete quantitative phenotypic items (phenes), the Hallucinations item and the Delusions item from the PANSS This approach avoids the issue of corrections for multiple comparisons that would arise if we were to look in a discovery fashion at multiple phenes in a comprehensive phenotypic battery (PhenoChippmg) changed in relationship with all genes on a GeneChip microarray

[00035] A panel of top candidate biomarker genes for hallucinations, respectively delusions state identified by the approach herein was then used to generate a prediction score for state (no symptoms vs high symptoms) This prediction score was compared to the actual PANSS testing scores from psychosis subjects in the primary cohort (FIGS. 3 A and 4A). The panels of biomarkers were examined and prediction scores in a separate cohort of psychotic disorders patients (FIG 4C).

[00036] Sample size for human subjects (n=31 for the primary cohort, n= 14 for the secondary cohort) is relatively small, but comparable to the size of cohorts for human postmortem brain gene expression studies. Live donor blood samples were studied instead of postmortem donor brains, with the advantage of better phenotypic characterization, more quantitative state information, and less technical variability. This approach also permits repeated intra-subject measures when the subject is in different psychosis states.

[00037] The experimental approach used in an embodiment for detecting gene expression changes relies on a standard methodology, Affymetrix GeneChip oligonucleotide microarrays. The analyses

Niculescu

have been designed to minimize the likelihood of having false positives, even at the expense of potentially having false negatives, due to the high cost in time and resources of pursuing false leads. For the animal model work, using isogenic mouse strain affords us an ideal control baseline of salme injected animals for the drug-injected animals Three independent de novo biological experiments were performed, at different times, with different batches of mice This overall design is geared to factor out both biological and technical variability. It is to be noted that the concordance between reproducible microarray experiments using the latest generations of oligonucleotide microarrays and other methodologies such as quantitative PCR, with their own attendant technical limitations, is estimated to be over 90%. For the human blood samples differential gene expression analyses, which are the results of single biological experiments, it has to be noted that the approach used a very restrictive and technically robust, all or nothing induction of gene expression (change from Absent Call to Present Call). It is possible that not all biomarker genes for psychosis will show this complete induction related to state, but rather some will show modulation in gene expression levels, and are thus missed by an initial filtering. A classic differential expression analysis to identify additional possible candidate biomarkers (see Tables 6A and 6 B).

[00038] Moreover, given the genetic heterogeneity and variable environmental exposure, it is possible, that not all subjects will show changes in all the biomarker genes Hence have two stringency thresholds were used- changes in 75% of subjects, and in 60% of subjects with no symptoms vs highsymptoms Moreover, an approach described herein is predicated on the existence of multiple cross- validators for each gene that is called a candidate biomarker (FIG IA)- 1) is it changed in human blood, 2) is it changed in animal model bram, 3) is it changed in animal model blood, 4) is it changed in postmortem human brain, and 5) does in map to a human genetic linkage locus. All these lines of evidence are the result of independent experiments. The virtues of this networked Bayesian approach are that, if one or another of the nodes (lines of evidence) becomes questionable/non-functional upon further evidence in the field, the network is resilient and maintains functionality. As more evidence emerges in the field for some of these genes, they will move up in the priontization scoring. Using such an approach, a small number of genes were identified as likely candidate biomarkers, out of the over 40,000 transcripts (about half of which are detected as Present in each subject) measured by the microarrays that were used.

[00039] A validation of is the fact that the primary cohort-derived biomarker panels showed explanatory sensitivity and specificity, of a comparable nature, in the primary cohort. They also showed some predictive sensitivity and specificity in the second (replication) cohort, more so for hallucinations than for delusions. Thus, the approach of using two individual phenes reflecting internal subjective experiences (hallucinations, delusions) which are the hallmark of psychosis (as opposed to more complex and disease specific state/trait clinical instruments), and looking at extremes of state combined

Niculescu

with robust differential expression based on A/P calls, and Convergent Functional Genomics pnoritization, seems to be able to identify state biomarkers for psychosis. Nevertheless, a comparison with existing clinical rating scales, EEG and functional neuroimaging, as well as analysis of biomarker data using such instruments are also suitable for a way of delineating state vs. trait issues, diagnostic boundaries or lack thereof, and informing the design of clinical trials that may incorporate clinical and biomarker measures of response to treatment.

[00040] Human blood gene expression changes may be influenced by the presence or absence of both medications and drugs of abuse That medications and drugs of abuse may have effects on mood state and gene expression is in fact being partially modeled in the mouse pharmacogenomic model, with clozapine and PCP treatments respectively It is the association of blood biomarkers with psychosis state a primary goal of this study, regardless of the proximal causes, which could be diverse (see FIG IB), Candidate biomarkers at a protein level, in larger cohorts of both genders, in different age groups, and in theragnostic settings- measuring responses to specific treatments/medications are also analyzed.

[00041 ] Top candidate biomarker genes for hallucinations include four genes decreased in expression in high hallucinations states (Rhobtb3, Aldhlll, Mpp3, FnI), and two genes increased in high hallucinations states (Arhgef9, S100a6). Five of these genes have evidence of differential expression in human postmortem brains from schizophrenia patients

[00042] Top candidate biomarker genes for delusions include eight genes decreased in expression in high delusions states (Drd2, ApoE, Nabl, Idhl, Scampi, Ncoa2, Aldhlll, Gpm6b), and eight genes increased in high delusions states (Nrgl, Egrl, Dctnl, Nmtl, PlIp, Pvalb, Nmtl, Pctkl) Fourteen of these genes have evidence of differential expression in human postmortem brains from schizophrenia patients

[00043] It is intriguing that genes which have a well-established role in brain functioning should show changes in blood in relationship to psychiatric symptoms state (FIG. 2, Tables 3A and 3B), and moreover that the direction of change should be concordant with that reported in human postmortem brain studies. It is possible that trait promoter sequence mutations or epigenetic modifications influence expression in both tissues (brain and blood), and that state dependent transcription factor changes that modulate the expression of these genes may be contributory as well

[00044] There are to date no clinical laboratory blood tests for psychotic disorders. A translational convergent approach to help identify and prioritize blood biomarkers for psychosis state is disclosed Data demonstrate that blood biomarkers have the potential to offer an unexpectedly informative window into brain functioning and disease state Panels of such biomarkers serve as a basis for objective clinical laboratory tests, a longstanding Holy Grail for psychiatry. Biomarker-based tests help with early intervention and prevention efforts, as well as monitoring response to various treatments. In conjunction with other clinical information, such tests play an important part in personalizing treatment

Niculescu

to increase effectiveness and avoid adverse reactions- personalized medicine m psychiatry. Moreover, they have scientific use m combination with imaging studies (imaging genomics), and is useful to pharmaceutical companies engaged in new neuropsychiatric drag development efforts, at both a preclinical and clinical (Phase I, II and III) stages of the process.

[00045] In an embodiment, the 5 top scoring candidate biomarkers for high delusions and the 5 top scoring candidate biomarkers for low delusions, and doing the same for hallucinations, a panel of 10 biomarkers for delusions, and a panel of 10 biomarkers for hallucinations have been designed for diagnostic and predictive purposes. However, a panel may have more or less number of genes specified in this embodiment.

[00046] To test the predictive value of the panels (to be called the BioM-10 Delusions panel and

BioM-10 Hallucinations panel), a cohort of 30 psychotic disorder datasets was analyzed, containing the datasets from which the candidate biomarker data was derived, as well as additional datasets of subjects with psychosis scores in the intermediate range (self -reported psychosis scores of 2 and 3) (Table 2). A prediction score for each subject was derived, based on the presence or absence of the 10 biomarker of the panel in their blood GeneChip data. Each of the 10 biomarkers gets a score of 1 if it is detected as Present (P) in the blood form that subject, 0 5 if it is detected as Marginally Present (M), and 0 if it is called Absent (A) The ratio of the sum of the high psychosis biomarker scores divided by the sum of the low psychosis biomarker scores is multiplied by 100, and provides a prediction score If the ratio of high biomarker genes to low psychosis biomarker genes is 1 , i e the two sets of genes are equally represented, the psychosis prediction score is 1x100=100 The higher this score, the higher the predicted likelihood that the subject will have a high psychosis symptoms score The predictive score was compared with actual psychosis scores in the cohort of samples with a diagnosis of psychosis (n=30)

[00047] For example, suitable candidate biomarker genes include for example,

[00048] A prediction score above 100 had a 100% sensitivity and an 47 1 % specificity for predicting a high delusions state. A prediction score below 100 had a 62.5% sensitivity and 84.2% specificity for predicting low delusions state (FIG. 3).

[00049] A prediction score above 100 had a 100% sensitivity and an 64 7 % specificity for predicting a high hallucinations state. A prediction score below 100 had a 85.7% sensitivity and 88.9% specificity for predicting a low hallucinations state (FIG. 4).

[00050] The MIT/Broad Institute Connectivity Mapl3 was interrogated with a signature query composed of the genes in BioM -10 Delusions and BioM-10 Hallucinations panels of top biomarkers for low and high psychosis (FIG. 5). It was determined which drugs in the Connectivity Map database have similar effects on gene expression as the effects of high psychosis (delusions, respectively hallucinations) on gene expression, and which drugs have the opposite effect to high psychosis As such, as part of the signature query, separately for delusions and hallucinations, the 5 biomarkers for

Niculescu

high psychosis were considered as genes "Increased " by high psychosis, the 5 biomarkers for low psychosis were genes "Decreased " by high psychosis

[00051 ] The interrogation revealed that deferoxamine had the most similar effects to high delusions, and sulindac the most similar effects to low delusions. For hallucinations, fluphenazine had the most similar effects to high hallucinations, and wortmaninn had the most similar effects to low hallucinations.

[00052] In an embodiment, a comprehensive analysis of: (i) fresh human blood gene expression data tied to illness state (quantitative measures of symptoms), (n) cross-validation of blood gene expression profiling m conjunction with bram gene expression studies in animal models presenting key features of psychotic disorders, and (in) integration of the results in the context of the available human genetic linkage/association and postmortem brain findings in the field is provided.

[00053] A panel of of 289 biomarker genes for Delusions, and 138 biomarker genes for hallucinations were identified, as illustrated in an example described herein, is a suitable subset that is useful in diagnosing psychotic disorders. Larger subsets that includes for example, 300, 350, 400, 450, or 500 markers are also suitable. Smaller subsets that include high-value markers including about 2, 5, 10, 15, 20, 25, 50, 75, and 100 are also suitable. A variable quantitative scoring scheme can be designed using any standard algorithm, such as a variable selection or a subset feature selection algorithms can be used Both statistical and machine learning algorithms are suitable in devising a frame work to identify, rank, and analyze association between marker data and phenotypic data (e g , psychotic disorders)

[00054] A panel of 36 biomarkers, as illustrated in an example described herein, is a suitable subset that is useful in diagnosing a mood disorder Larger subsets that includes for example, 150, 200, 250, 300, 350, 400, 450, 500, 600 or about 700 markers are also suitable Smaller subsets that include high-value markers including about 2, 5, 10, 15, 20, 25, 50, 75, and 100 are also suitable A variable quantitative scoring scheme can be designed using any standard algorithm, such as a variable selection or a subset feature selection algorithms can be used. Both statistical and machine learning algorithms are suitable in devising a frame work to identify, rank, and analyze association between marker data and phenotypic data (e.g , mood disorders).

[00055] In an embodiment, a prediction score for each subject is derived based on the presence or absence of e.g , 10 biomarkers of the panel in their blood. Each of the 10 biomarkers gets a score of 1 if it is detected as "present" (P) in the blood form that subject, 0.5 if it is detected as "marginally present" (M), and 0 if it is called "absent" (A) The ratio of the sum of the high mood biomarker scores divided by the sum of the low mood biomarker scores is multiplied by 100, and provides a prediction score If the ratio of high biomarker genes to low mood biomarker genes is 1, i.e. the two sets of genes are equally represented, the mood prediction score is 1x100=100 The higher this score, the higher the predicted likelihood that the subject will have high mood. The predictive score was compared with actual self-reported mood scores in a primary cohort of subjects with a diagnosis of bipolar mood

Niculescu

disorder. A prediction score of 100 and above had a 84.6 % sensitivity and a 68 8 % specificity for predicting high mood. A prediction score below 100 had a 76.9 % sensitivity and 81.3 % specificity for predicting low mood. The term "present" indicates that a particular biomarker is expressed to a detectable level, as determined by the technique used. For example, in an experiment involving a microarray or gene chip obtained from a commercial vendor Affymetrix (Santa Clara, CA), the embedded software rendered a "present" call for that biomarker. The term "present" refers to a detectable presence of the transcript or its translated protem/peptide and not necessarily reflects a relative comparion to for example, a sample from a normal subject. In other words, the mere presence or absence of a biomarker is assigned a value, e.g , 1 and a prediction score is calculated as described herein. The term "marginally present refers to border line expression level that may be less intense than the "present" but statistically different from being marked as "absent" (above background noise), as determined by the methodology used

[00056] In an embodiment, a prediction score based on differential expression (instead of "present",

"absent") is used. For example, if a subject has a plurality of markers for high or low mood are differentially expressed, a prediction based on the differential expression of markers is determined. Differential expression of about 1 2 fold or 1.3 or 1 5 or 2 or 3 or 4 or 5 -fold or higher for either increased or decreased levels can be used Any standard statistical tool such as ANOVA is suitable for analysis of differential expression and association with high or low mood diagnosis or prediction

[00057] A prediction based on the analysis of either high or low mood markers alone (instead of a ratio of high versus low mood markers) may also be practiced If a plurality of high mood markers (e g , about 6 out of 10 tested) are differentially expressed to a higher level compared to the low mood markers (e g , 4 out of 10 tested), then a prediction or diagnosis of high mood status can be made by analyzing the expression levels of the high mood markers alone without factoring the expression levels of the low mood markers as a ratio

[00058] In an embodiment, a detection algorithm uses probe pair intensities to generate a detection p- value and assign a Present, Marginal, or Absent call. Each probe pair in a probe set is considered as having a potential vote in determining whether the measured transcript is detected (Present) or not detected (Absent). The vote is described by a value called the Discrimination score [R]. The score is calculated for each probe pair and is compared to a predefined threshold Tau. Probe pairs with scores higher than Tau vote for the presence of the transcript. Probe pairs with scores lower than Tau vote for the absence of the transcript. The voting result is summarized as a p-value The greater the number of discrimination scores calculated for a given probe set that are above Tau, the smaller the p-value and the more likely the given transcript is truly Present in the sample The p-value associated with this test reflects the confidence of the Detection call

Niculescu

[00059] Regarding detection p-value, a two-step procedure determines the Detection p-value for a given probe set. The Discrimination score [R] is calculated for each probe pair and the discrimination scores are tested against the user-definable threshold Tau. The detection Algorithm assesses probe pair saturation, calculates a Detection p-value, and assigns a Present, Marginal, or Absent call In an embodiment, the default thresholds of the Affymetrix MAS 5 software were used.

[00060] In spiking experiments by the manufacturer to establish default thresholds (adding of known quantities of test transcripts to a mixture, to measure the sensitivity of the Affymetrix MAS 5 detection algorithm) 80% of spiked transcripts are called Present at a concentration of 1 5 pM. This concentration corresponds to approximately one transcript in 100,000 or 3 5 copies per cell. The false positive rate of making a Present call was roughly 10%, as noted by 90% of the transcripts being called Absent when not spiked into the sample (0 pM concentration).

[00061 ] The term "predictive" or the term "prognostic" does not imply 100% predictive ability. The use of these terms indicates that subjects with certain characteristics are more likely to experience a clinically positive outcome than subjects who do not have such characteristics. For example, characteristics that determine the outcome include one or more of the biomarkers for psychosis disclosed herein Certain conditions are identified herein as associated with an increased likelihood of a clinically positive outcome, e g , biomarkers for delusions and the absence of such conditions or markers will be associated with a reduced likelihood of a clinically positive outcome

[00062] The phrase "clinically positive outcome" refers to biological or biochemical or physical or physiological responses to treatments or therapeutic agents that are generally prescribed for that condition compared to a condition would occur in the absence of any treatment A "clinically positive outcome" does not necessarily indicate a cure, but could indicate a lessening of symptoms experienced by a subject

[00063] The terms "marker" and "biomarker" are synonymous and as used herein, refer to the presence or absence or the levels of nucleic acid sequences or proteins or polypeptides or fragments thereof to be used for associating or correlating a phenotypic state. A biomarker includes any indicia of the level of expression of an indicated marker gene The indicia can be direct or indirect and measure over- or under-expression of the gene given the physiologic parameters and in comparison to an internal control, normal tissue or another phenotype. Nucleic acids or proteins or polypeptides or portions thereof used as markers are contemplated to include any fragments thereof, in particular, fragments that can specifically hybridize with their intended targets under stringent conditions and immunologically detectable fragments. One or more markers may be related Marker may also refer to a gene or DNA sequence having a known location on a chromosome and associated with a particular gene or trait. Genetic markers associated with certain diseases or for pre-disposing disease states can be detected in the blood and used to determine whether an individual is at risk for developing a disease Levels of gene

Niculescu

expression and protein levels are quantifiable and the variation in quantification or the mere presence or absence of the expression may also serve as markers. Using proteins/peptides as biomarkers can include any method known in the art including, without limitation, measuring amount, activity, modifications such as glycosylation, phosphorylation, ADP-nbosylation, ubiquitination, etc , imunohistochemistry (IHC).

[00064] As used herein, "array" or "microarray" refers to an array of distinct polynulceotides, oligonucleotides, polypeptides, or oligopeptides synthesized on a substrate, such as paper, nylon, or other type of membrane, filter, chip, glass slide, or any other suitable solid support. Arrays also include a plurality of antibodies immobilized on a support for detecting specific protein products. There are several microarrays that are commercially available. A microarray may include one or more biomarkers disclosed herein A panel of about 20 biomarkers as nucleic acid fragments can be included in an array. The nucleic acid fragments may include oligonucleotides or amplified partial or complete nucleotide sequences of the biomarkers The term "consisting essentially of generally refers to a collection of markers that substantially affects the determination of the disorder and may include other components such as controls. For example, top biomarkers from Tables 3A-B may be considered as a subset of markers that determine the most association.

[00065] In an embodiment, the microarray is prepared and used according to the methods described in

U S Pat No 5,837,832, Chee et al ; PCT application WO95/11995, Chee et al , Lockhart et al , 1996 Nat Biotech , 14- 1675-80; and Schena et al , 1996 Proc Natl Acad Sci 93-10614-619, all of which are herein incorporated by reference to the extent they relate to methods of making a microarray Arrays can also be produced by the methods described in Brown et al , U S Pat No 5,807,522 Arrays and microarrays may be referred to as "DNA chips" or "protein chips "

[00066] A variety of clustering methods are available for microarray-based gene expression analysis

See for example, Shamir & Sharan (2002) Algorithmic approaches to clustering gene expression data In Current Topics In Computational Molecular Biology (Edited by. Jiang T, Xu Y, Smith T). 2002, 269- 300, Tamames et al , (2002) Bioinformatics methods for the analysis of expression arrays, data clustering and information extraction, J Bwtechnol, 98:269-283.

[00067] "Therapeutic agent" means any agent or compound useful in the treatment, prevention or inhibition of psychosis or a psychosis -related disorder.

[00068] The term "condition" refers to any disease, disorder or any biological or physiological effect that produces unwanted biological effects in a subject.

[00069] The term "subject" refers to an animal, or to one or more cells derived from an animal. The animal may be a mammal including humans. Cells may be in any form, including but not limited to cells retained in tissue, cell clusters, immortalized cells, transfected or transformed cells, and cells derived from an animal that have been physically or phenotypically altered.

Niculescu

[00070] Any body fluid of an animal can be used in the methods of the invention Suitable body fluids include a blood sample (e.g., whole blood, serum or plasma), urine, saliva, cerebrospinal fluid, tears, semen, and vaginal secretions. Also, lavages, tissue homogenates and cell lysates can be utilized.

[00071 ] Many different methods can be used to determine the levels of markers. For example, protein arrays, protein chips, cDNA microarrays or RNA microarrays are suitable More specifically, one of ordinary skill m the art will appreciate that in one example, a microarray may comprise the nucleic acid sequences representing genes listed in Table 1. For example, functionality, expression and activity levels may be determined by immunohistochemistry, a staining method based on lmmunoenzymatic reactions uses monoclonal or polyclonal antibodies to detect cells or specific proteins. Typically, immunohistochemistry protocols include detection systems that make the presence of markers visible (to either the human eye or an automated scanning system), for qualitative or quantitative analyses. Mass-spectrometry, chromatography, real-time PCR, quantitative PCR, probe hybridization, or any other analytical method to determine expression levels or protein levels of the markers are suitable. Such analysis can be quantitative and may also be performed in a high-through put fashion. Cellular imaging systems are commercially available that combine conventional light microscopes with digital image processing systems to perform quantitative analysis on cells and tissues, including immunostained samples (See e g the CAS-200 System (Becton, Dickinson & Co )) Some other examples of methods that can be used to determine the levels of markers include immunohistochemistry, automated systems, quantitative IHC, semi-quantitative IHC and manual methods Other analytical systems include western blotting, immunoprecipitation, fluorescence m situ hybridization (FISH), and enzyme immunoassays

[00072] The term "diagnosis", as used in this specification refers to evaluating the type of disease or condition from a set of marker values and/or patient symptoms where the subject is suspected of having a disorder. This is in contrast to disease predisposition, which relates to predicting the occurrence of disease before it occurs, and the term "prognosis", which is predicting disease progression in the future based on the marker levels/patterns.

[00073] The term "correlating," as used in this specification refers to a process by which one or more biomarkers are associated to a particular disease state, e.g., mood disorder In general, identifying such correlation or association involves conducting analyses that establish a statistically significant association- and/or a statistically significant correlation between the presence (or a particular level) of a marker or a combination of markers and the phenotypic trait in the subject. An analysis that identifies a statistical association (e.g., a significant association) between the marker or combination of markers and the phenotype establishes a correlation between the presence of the marker or combination of markers in a subject and the particular phenotype being analyzed.

Niculescu

[00074] This relationship or association can be determined by comparing biomarker levels in a subject to levels obtained from a control population, e.g., positive control— diseased (with symptoms) population and negative control-disease-free (symptom-free) population. The biomarkers disclosed herein provide a statistically significant correlation to diagnosis at varying levels of probability. Subsets of markers, for example a panel of about 20 markers, each at a certain level range which are a simple threshold, are said to be correlative or associative with one of the disease states. Such a panel of correlated markers can be then be used for disease detection, diagnosis, prognosis and/or treatment outcome Preferred methods of correlating markers is by performing marker selection by any appropriate scoring method or by using a standard feature selection algorithm and classification by known mapping functions A suitable probability level is a 5% chance, a 10% chance, a 20% chance, a 25% chance, a 30% chance, a 40% chance, a 50% chance, a 60% chance, a 70% chance, a 75% chance, a 80% chance, a 90% chance, a 95% chance, and a 100% chance Each of these values of probability is plus or minus 2% or less. A suitable threshold level for markers of the present invention is about 25 pg/mL, about 50 pg/mL, about 75 pg/mL, about 100 pg/mL, about 150 pg/mL, about 200 pg/mL, about 400 pg/mL, about 500 pg/mL, about 750 pg/mL, about 1000 pg/mL, and about 2500 pg/mL.

[00075] Prognosis methods disclosed herein that improve the outcome of a disease by reducing the increased disposition for an adverse outcome associated with the diagnosis Such methods may also be used to screen pharmacological compounds for agents capable of improving the patient's prognosis, e g , test agents for mood disorders

[00076] The analysis of a plurality of markers, for example, a panel of about 20 or 10 markers may be carried out separately or simultaneously with one test sample Several markers may be combined into one test for efficient processing of a multiple of samples In addition, one skilled in the art would recognize the value of testing multiple samples (for example, at successive time points) from the same individual. Such testing of serial samples may allow the identification of changes in marker levels over time, within a period of interest, or in response to a certain treatment

[00077] In another embodiment, a kit for the analysis of markers includes for example, devises and reagents for the analysis of at least one test sample and instructions for performing the assay. Optionally, the kits may contain one or more means for using information obtained from marker assays performed for a marker panel to diagnose psychosis. Probes for markers, marker antibodies or antigens may be incorporated into diagnostic assay kits depending upon which markers are being measured. A plurality of probes may be placed in to separate containers, or alternatively, a chip may contain immobilized probes. In an embodiment, another container may include a composition that includes an antigen or antibody preparation Both antibody and antigen preparations may preferably be provided in a suitable titrated form, with antigen concentrations and/or antibody titers given for easy reference in quantitative applications.

Niculescu

[00078] The kits may also include a detection reagent or label for the detection of specific reaction between the probes provided m the array or the antibody in the preparation for immunodetection. Suitable detection reagents are well known m the art as exemplified by fluorescent, radioactive, enzymatic or otherwise chromogenic ligands, which are typically employed in association with the nucleic acid, antigen and/or antibody, or in association with a secondary antibody having specificity for first antibody. Thus, the reaction is detected or quantified by means of detecting or quantifying the label. Immunodetection reagents and processes suitable for application in connection with the novel methods of the present invention are generally well known in the art.

[00079] The reagents may also include ancillary agents such as buffering agents and protein stabilizing agents, e g., polysaccharides and the like. The diagnostic kit may further include where necessary agents for reducing background interference in a test, agents for increasing signal, software and algorithms for combining and interpolating marker values to produce a prediction of clinical outcome of interest, apparatus for conducting a test, calibration curves and charts, standardization curves and charts, and the like.

[00080] In some embodiments, the methods of correlating biomarkers with treatment regimens can be carried out using a computer database. Computer-assisted methods of identifying a proposed treatment for mood disorders are suitable The method involves the steps of (a) storing a database of biological data for a plurality of patients, the biological data that is being stored including for each of said plurality of patients (i) a treatment type, (ii) at least one marker associated with a mood disorder and (iii) at least one disease progression measure for the mood disorder from which treatment efficacy can be determined; and then (b) querying the database to determine the dependence on the marker of the effectiveness of a treatment type in treating the mood disorder, to thereby identify a proposed treatment as an effective treatment for a subject carrying the marker correlated with the mood disorder.

[00081 ] In an embodiment, treatment information for a patient is entered into the database (through any suitable means such as a window or text interface), marker information for that patient is entered into the database, and disease progression information is entered into the database These steps are then repeated until the desired number of patients has been entered into the database. The database can then be queried to determine whether a particular treatment is effective for patients carrying a particular marker, not effective for patients carrying a particular marker, and the like. Such querying can be carried out prospectively or retrospectively on the database by any suitable means, but is generally done by statistical analysis in accordance with known techniques, as described herein.

Niculescu

EXAMPLES

[00082] The following examples are to be considered as exemplary and not restrictive or limiting in character and that all changes and modifications that come within the spirit of the disclosure are desired to be protected.

EXAMPLE 1 Experimental framework for identification of biomarkers used in diagnosis of psychotic disorders

[00083] Gene expression changes in specific brain regions and blood from a pharmacogenomic animal model developed in the group were used as cross-validators to help with the identification of potential human blood biomarkers Pharmacogenomic mouse model of relevance to bipolar disorder consists of treatments with an agonist of the illness/psychosis -mimicking drug (PCP) and an antagonist of the illness/ bipolar disorder-treating drug (clozapine) 4. The pharmacogenomic approach is a tool for tagging genes that may have pathophysiological relevance.

[00084] Human blood gene expression studies were carried out in a cohort of psychotic disorders subjects. Genes that were differentially expressed in low psychosis vs. high psychosis subjects were compared with: 1) the results of animal model brain and blood data, as well as 2) human genetic linkage/association data, and 3) human postmortem brain data, as a way of cross-validating the findings, prioritizing them, and identifying a short list of high probability candidate biomarker genes (FIG IA and FIG 3)

[00085] PANSS- Pl score for Delusions, and P3 score for Hallucinations were used This approach avoids the issue of corrections for multiple comparisons that would arise if discovery at multiple phenes was considered in a comprehensive phenotypic battery changed in relationship with all genes on a GeneChip microarray Larger sample cohorts would be needed for the latter approach

[00086] In an aspect, the sample size for human subjects (n= 30 for the psychotic disorders cohort) is relatively small, but comparable to the size of cohorts for human postmortem brain gene expression studies. Live donor blood samples were studied instead of postmortem donor brains, with the advantage of better phenotypic characterization, more quantitative state information, and less technical variability.

[00087] Some of the datasets were derived from subjects that were sampled repeatedly, at three months intervals (Table 2). A total of 21 unique subjects were used and 2 of them were sampled three times, and 5 of them were sampled twice. However, this reduction m genetic background diversity may be advantageous in terms of analysis of state related markers, which is an important objective. In mouse models, the isogenic strain background is viewed as advantageous in terms of reducing noise and providing power to pharmacogenomic analyses.

[00088] In an aspect, the experimental approach for detecting gene expression changes relies on a chip methodology, Affymetrix GeneChip oligonucleotide microarrays. It is possible that some of the gene expression changes detected from a single biological experiment, with a one-time assay with this

Niculescu

technology, are biological or technical artifacts. The analyses are designed to minimize the likelihood of having false positives, even at the expense of potentially having false negatives, due to the high cost in time and resources of pursuing false leads. For the animal model work, using isogenic mouse strain affords us an ideal control baseline of saline injected animals for the drug-injected animals. Three independent de novo biological experiments were performed at different times, with different batches of mice. This overall design is geared to factor out both biological and technical variability. It is to be noted that the concordance between reproducible microarray experiments using the latest generations of oligonucleotide microarrays and other methodologies such as quantitative PCR, with their own attendant technical limitations, is estimated to be over 90%. For the human blood samples gene expression analyses, a very restrictive approach was used — all or nothing induction of gene expression (change from Absent Call to Present Call). Moreover, given the genetic heterogeneity and variable environmental exposure, it is possible, indeed likely, that not all subjects will show changes in all the biomarker genes. Therefore, two stringency thresholds were used changes in 75% of subjects, and in 60% of subjects with low psychosis vs. high psychosis. Moreover, the approach, as described above, is predicated on the existence of multiple cross-validators for each gene that is called a candidate biomarker (FIG. IB): 1) is it changed in human blood? 2) is it changed in animal model bram^ 3) is it changed in animal model blood ¹ ? 4) is it changed in postmortem human brain"? and 5) does in map to a human genetic linkage locus'? All these lines of evidence are the result of independent experiments

[00089] Human blood gene expression changes may be influenced by the presence or absence of both medications and drugs of abuse That medications and drugs of abuse may have effects on psychosis state and gene expression is being partially modeled in the mouse pharmacogenomic model, with clozapine and PCP treatments respectively.

[00090] A panel of top candidate biomarker genes for psychosis state identified by the methods disclosed herein was then used to generate a prediction score for psychosis state (low psychosis symptoms vs high psychosis symptoms). This prediction score was compared to the actual psychosis scores from psychotic disorder subjects (FIG. 4A and B) Methods disclosed herein narrow the over 40,000 genes and ESTs (transcript variants) present on the Affymetix Human Genome U133 Plus 2.0 GeneChip, about half of which are detected as Present in each blood sample, to a panel of 10 high probability biomarker genes, which shows surprisingly robust predictive power.

[00091 ] In an aspect, a panel of biomarkers for delusions include a gene associated with low delusions scores (MOBP) and three genes associated with high delusions scores (NRGl, GPM6B, and TPM2). In an aspect, a panel of biomarkers for hallucinations, three genes are associated with high hallucinations scores (TNIK, HSD 17Bl 2 and TPM2) These genes were selected as having a line of evidence CFG score of higher than 5 (Table 4 and 5, and Figure 2A and 2B) That means, in addition to the human blood data, these genes have at least two other independent lines of evidence implicating

Niculescu

them in psychotic disorders. All these genes have evidence of differential expression in human postmortem brains from schizophrenia patients. NRGl (neuroregulin 1) has been implicated in the pathogenesis of schizophrenia by multiple genetic and neurobiological studies.

[00092] It is intriguing that genes which have a well-established role in brain functioning should show changes in blood in relationship to psychiatric symptoms state (FIG. 2, Table 4-5 and Table 6), and moreover that the direction of change should be concordant with that reported m human postmortem brain studies. It is possible that trait promoter sequence mutations or epigenetic modifications influence expression in both tissues (bram and blood), and that state dependent transcription factor changes that modulate the expression of these genes may be contributory as well.

[00093] Data provided herein suggest that genes involved in bram infrastructure changes (myelin, growth factors) are prominent players in psychotic disorders, and are reflected in the blood profile Myelin abnormalities have emerged as a common if perhaps non-specific denominator across a variety of neuropsychiatric disorders. Data regarding cytoskeleton regulating genes (TPM2 and TNIK) changes may provide evidence for a novel and previously underappreciated mechanism for schizophrenia pathophysiology. HSD 17B l 2 (17- beta-hydroxysteroid dehydrogenase) is an enzyme involved in estrogen formation. It is increased in the human blood data in high delusions states, as well as increased in human postmortem brain from schizophrenics and in the animal model bram data Weather these changes are etiopathogenic, compensatory mechanisms, side-effects of medications or results of illness -induced lifestyle changes (FIG IB) is an intriguing area

[00094] The fact that most of the top genes identified are associated with high psycosis states as opposed to low psychosis states (FIG 2 and Table 4-5) may suggest that co-morbid stress- more prevalent in high psychosis than in low/no psychosis- is a factor in the richness of blood gene expression results, as part of a neuro-endocrine-immunological axis. The higher sensitivity than specificity of the test for high psychosis state may reflect this preponderance of candidate biomarker genes for high psychosis state identified relative to candidate biomarker for low psychosis state. The test shows lower sensitivity but higher specificity for low psychosis state.

[00095] Of note, some of the other top candidate genes identified have no previous evidence for involvement in psychosis other than them being mapped to schizophrenia genetic linkage loci (Table 4- 5), and thus constitute novel candidate genes forschizophrenia. They are useful for whole -genome association studies of schizophrenia It is possible that the composition of top biomarker panels for psychosis will be refined or changed for different sub-populations. That being said, it is likely that a large number of the biomarkers that would be of use in different panels and permutations are already present in the complete list of top candidate biomarkers (n=289 top candidate biomarker genes for Delusions, and n= 138 top candidate biomarker genes for Hallucinations). (Tables 10-11)

Niculescu

[00096] The interrogation of the MIT/Broad Institute Connectivity Map 13 with a signature query composed of the genes in the BioM -10 Delusions and BioM-10 Hallucinations panels of top biomarkers revealed that deferoxamine had the most similar effects to high delusions, and sulindac the most similar effects to low delusions. For hallucinations, fluphenazine had the most similar effects to high hallucinations, and wortmaninn had the most similar effects to low hallucinations (FIGS. 4 A and 4B).

[00097] Deferoxamine is a medication used clinically to treat iron overload states. Oligodendrocyte progenitors are highly susceptible to oxidative stress due to their limited content of antioxidants and high iron levels. Iron plays a central role in the toxicity of dopamine to oligodendrocyte progenitors. Dopamine induces accumulation of superoxide, membrane damage and loss in cell viability. The iron chelator deferoxamine reduces superoxide accumulation. Desferrioxamine administration in mice caused a reduction in severity of physical dependence to alcohol Deferoxamine also increases the production of neurons from neural stem/progenitor cells, and showed neuroprotective properties in ischemia states. These observations indicate that deferoxamine activates cellular mechanisms and programs of gene expression that have cell survival and protective effects. Sulindac, a non-steroidal inflammatory drug, has been shown to inhibit liver tryptophan 2,3-dioxygenase activity, a rate-limiting enzyme in tryptophan catabolism, and consequently alter brain neurotransmitter levels, resulting in an increase in serotonin levels and decrease in dopamine levels in rats Taken together, these observations indicate that high delusions are associated with a program of gene expression reflective of a neurotrophic, high dopamine state

[00098] Fluphenazine is a typical (first-generation) anti-psychotic Wortmannm is a phosphoinositide-

3' kinase (PI3K) inhibitor The PI3K pathway is thought to be hypoactive in schizophrenia, suggesting that wortmannin has a schziophrenogenic effect Results demonstrate that the gene expression patterns seen with hallucinations may be reflective of a medication effect in those severely psychotic patients.

[00099] This connectivity map analysis with the BioM -10 psychosis panels genes provides an interesting external biological cross-validation for the internal consistency of the biomarker approach, as well as illustrates the utility of the Connectivity Map for non-hypothesis driven identification of novel drug treatments and interventions. [000100] More profoundly, these results, taken together with candidate biomarker genes results and biological roles categories (3A and 3B), are consistent with a developmental model for genes involved in psychosis.

[000101 ] There are to date no clinical laboratory blood tests for psychosis A translational convergent approach to help identify blood biomarkers of psychosis symptoms (delusions, hallucinations) state is proposed herein. Blood biomarkers have the potential to offer an unexpectedly informative window into

Niculescu

bram functioning and disease state. Panels of such biomarkers serve as a basis for objective clinical laboratory tests.

[000102] Any number of biomarkers can be used as a panel for diagnosis. The panel may contain equal number of biomarkers for delusions and hallucinations. The panel may be tested as a microarray or as any form of diagnostic analysis.

[000103] Thus, the biomarkers identified herein provide quantitative tools for predicting disease states/conditions in subjects suspected of having a psychotic disorder or in any individual for psychiatric evaluation.

[000104] Human subjects: Data from from two cohorts of patients are presented herein. One cohort included 31 different subjects with psychotic disorders (schizophrenia, schizoaffective disorder and substance induced psychosis), from which the primary biomarker data was derived, from testing done at their first visit (vl). A second (replication) cohort consists of 14 subjects from the first cohort, tested 3 moths (v2) or 6 months (v3) later. The diagnosis is established by a structured clinical interview- Diagnostic Interview for Genetic Studies (DIGS), which has details on the course of illness and phenomenology, and is the scale used by the Genetics Initiative Consortia for both Bipolar Disorder and Schizophrenia.

[000105] Subjects included men and women over 18 years of age Subjects were recruited from the patient population at the Indianapolis VA Medical Center, the Indiana University School of Medicine, as well as various facilities that serve people with mental illnesses in Indiana A demographic breakdown is shown in Table 1 Initial studies were focused primarily on an age-matched male population, due to the demographics of the catchment area (primarily male in a VA Medical Center), and to minimize any potential gender-related state effects on gene expression, which would have decreased the discriminative power of the analysis given a relatively small sample size. The subjects were recruited largely through referrals from care providers, the use of brochures left in plain sight in public places and mental health clinics, and through word of mouth Subjects were excluded if they had significant medical or neurological illness or had evidence of active substance abuse or dependence. All subjects understood and signed informed consent forms detailing the research goals, procedure, caveats and safeguards. Subjects completed diagnostic assessments (DIGS), and then a psychosis rating scale (Positive and Negative Symptom Scale- PANSS) at the time of blood draw 10 cc of whole blood were collected in two RNA- stabilizing PAXgene tubes, labeled with an annonymized ID number, and stored at -80C in a locked freezer (Revco) until the time of future processing..

[000106] Human blood gene expression experiments and analysis: RNA extraction 2.5-5 ml of whole blood was collected into each PaxGene tube by routine venipuncture PaxGene tubes contain proprietary reagents for the stabilization of RNA. The cells from whole blood will be concentrated by centrifugation, the pellet washed, resuspended and incubated in buffers containing Proteinase K for

Niculescu

protein digestion. A second centrifugation step is done to remove residual cell debris. After the addition of ethanol for an optimal binding condition the lysate is applied to a silica-gel membrane/column. The RNA bound to the membrane as the column is centrifuged and contaminants are removed m three wash steps. The RNA is then eluted using DEPC -treated water.

[000107] Globin reduction: To remove globin mRNA, total RNA from whole blood is mixed with a biotinylated Capture Oligo Mix that is specific for human globin mRNA. The mixture is then incubated for 15 mm to allow the biotinylated oligonucleotides to hybridize with the globin mRNA. Streptavidm Magnetic Beads are then added, and the mixture is incubated for 30 mm. During this incubation, streptavidm binds the biotinylated oligonucleotides, thereby capturing the globin mRNA on the magnetic beads. The Streptavidm Magnetic Beads are then pulled to the side of the tube with a magnet, and the RNA, depleted of the globin mRNA, is transferred to a fresh tube. The treated RNA is further purified using a rapid magnetic bead-based purification method. This consists of adding an RNA Binding Bead suspension to the samples, and using magnetic capture to wash and elute the GLOBINclear RNA.

[000108] Sample Labeling: Sample labeling is performed using the Ambion MessageAmp II-

BiotmEnhanced aRNA amplification kit. The procedure is briefly outlined below and involves the following steps:

[000109] 1 Reverse Transcription to Synthesize First Strand cDNA is primed with the T7 Oligo(dT)

Primer to synthesize cDNA containing a T7 promoter sequence

[0001 1 0] 2 Second Strand cDNA Synthesis converts the single-stranded cDNA into a double-stranded

DNA (dsDNA) template for transcription The reaction employs DNA Polymerase and RNase H to simultaneously degrade the RNA and synthesize second strand cDNA

[0001 1 1 ] 3. cDNA Purification removes RNA, primers, enzymes, and salts that would inhibit in vitro transcription.

[0001 1 2] 4. In Vitro Transcription to Synthesize aRNA with Biotm-NTP Mix generates multiple copies of biotin-modified aRNA from the double- stranded cDNA templates, this is the amplification step.

[0001 1 3] 5. aRNA Purification removes unincorporated NTPs, salts, enzymes, and inorganic phosphate to improve the stability of the biotm-modified aRNA.

[0001 14] Microarrays: Biotin labeled aRNA are hybridized to Affymetrix HG-U133 Plus 2.0 GeneChips according to manufacturer's protocols http://www.affymetrix.com/support/technical/manual/expressio n_manual.affx. All GAPDH 375' ratios should be less than 2.0 and backgrounds under 50. Arrays are stained using standard Affymetrix protocols for antibody signal amplification and scanned on an Affymetrix GeneArray 2500 scanner with a target intensity set at 250. Present/ Absent calls are determined using GCOS software with thresholds set at default values.

Niculescu

[0001 1 5] Analysis: The subject's psychosis scores at time of blood collection, specifically the scores for hallucinations ( from 1- no symptoms to 7- extreme symptoms) and the scores for delusions (1 to 7), obtained from a PANSS scale were used (Table 1). Only at all or nothing gene expression differences were considered that are identified by Absent (A) vs Present (P) Calls in the Affymetrix MAS software. Genes were classified whose expression is detected as Absent in the Low Psychosis subjects ( score of 1 on a scale of 1 to 7) and detected as Present in the High Psychosis subjects ( score of 4 or above, on a scale of 1 to 7), as being candidate biomarker genes for psychosis (specifically for delusions or hallucinations). Conversely, genes whose expression are detected as Present in the Low Psychosis subjects and Absent in the High Psychosis subjects are being classified as candidate biomarker genes for low psychosis

[0001 1 6] Two thresholds were used for analysis of gene expression differences between low psychosis and high psychosis (Table 3). First, a high threshold was used, with at least 75% of subjects in the cohort showing a change in expression from Absent to Present between low and high psychosis (reflecting an at least 3 fold psychosis state related enrichment of the genes thus filtered). Alow threshold was also used, with at least 60% of subjects in the cohort showing a change in expression from Absent to Present between low and high psychosis (reflecting an at least 1 5 fold psychosis state related enrichment of the genes thus filtered)

[0001 1 7] Animal model data' Schizophrenia pharmacogenomic model includes phencyclidine (PCP) and clozapine treatments in mice

[0001 1 8] All experiments were performed with male C57/BL6 mice, 8 to 12 weeks of age, obtained from

Jackson Laboratories (Bar Harbor, ME), and acclimated for at least two weeks in the animal facility prior to any experimental manipulation Mice were treated by intraperitoneal injection with either smgle-dose saline PCP (7.5 mg/kg), clozapine (2.5 mg/kg), or a combination of PCP and clozapine (7 5 mg/kg and 2.5 mg/kg). Three independent de nσvo biological experiments were performed at different times Each experiment consisted of three mice per treatment condition, for a total of 9 mice per condition across the three experiments.

[0001 1 9] Mouse Blood collection: Twenty-four hours after drug administration, following the 24 hour time-point behavioral test, the mice were decapitated to harvest blood The headless mouse body was put over a glass funnel coated with heparin and approximately 1 ml of blood/mouse was collected into a PAXgene blood RNA collection tubes, BD diagnostic (VWR .com) The Paxgene blood vials were stored in -4 ⁰ C overnight, and then at -80 ⁰ C until future processing for RNA extraction.

[000120] RNA extraction and microarray work: Standard techniques were used to obtain total RNA

(22 gauge syringe homogenization in RLT buffer) and to purify the RNA (RNeasy mini kit, Qiagen, Valencia, CA) from micro-dissected mouse brain regions. For the human and whole mouse blood RNA extraction, PAXgene blood RNA extraction kit (PreAnalytiX, a QIAGEN/ BD company) was used,

Niculescu

followed by GLOBINclear™-Human or GLOBINclear™-Mouse/Rat (Ambion/ Applied Biosystems Inc , Austin, TX) to remove the globin mRNA. All the methods and procedures were carried out as per manufacturer's instructions. The quality of the total RNA was confirmed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). The quantity and quality of total RNA was also independently assessed by 260 nm UV absorption and by 260/280 ratios, respectively (Nanodrop spectrophotometer) Starting material of total RNA labeling reactions was kept consistent within each independent microarray experiment.

[000121 ] For the all the mouse analysis, blood or brain tissues regions from 3 mice were pooled for each experimental condition, and equal amounts of total RNA extracted from tissue samples or blood was used for labeling and microarray assays. Mouse Genome 430 2.0 arrays (Affymetrix, Santa Clara, CA) were used. The GeneChip Mouse Genome 430 2.0 Array contains over 45,000 probe sets that analyze the expression level of over 39,000 transcripts and variants from over 34,000 well-characterized mouse genes. For the human work, we used Affymetrix Human Genome U 133 Plus 2.0 GeneChip with over 40,000 genes and ESTs. Standard Affymetrix protocols were used to reverse transcribe the messenger RNA and generate biotinlylate cRNA. The amount of cRNA used to prepare the hybridization cocktail was kept constant mtra-expenment. Samples were hybridized at 45°C for 17 hours under constant rotation Arrays were washed and stained using the Affymetrix Fluidics Station 400 and scanned using the Affymetrix Model 3000 Scanner controlled by GCOS software All sample labeling, hybridization, staining and scanning procedures were carried out as per manufacturer' s recommendations

[000122] All arrays were scaled to a target intensity of 1000 using Affymetrix MASv 5 0 array analysis software Quality control measures including 375' ratios for GAPDH and beta-actin, scaling factors, background, and Q values were within acceptable limits

[000123] Microarray data analysis: Data analysis was performed using Affymetrix Microarray Suite

5.0 software (MAS v5.0) Default settings were used to define transcripts as present (P), marginal (M), or absent (A). A comparison analysis was performed for each drug treatment, using its corresponding saline treatment as the baseline. "Signal," "Detection," "Signal Log Ratio," "Change," and "Change p- value," were obtained from this analysis Only transcripts that were called Present in at least one of the two samples (saline or drug) intra-experiment, and that were reproducibly changed in the same direction in at least two out of three independent experiments, were analyzed further.

[000124] Cross-validation and integration: Convergent Functional Genomics: Gene identification The identities of transcripts were established using NetAFFX (Affymetrix, Santa Clara, CA), and confirmed by cross-checking the target mRNA sequences that had been used for probe design in the Mouse Genome 430 2.0 Array GeneChip® or the Affymetrix Human Genome U133 Plus 2.0 GeneChip® with the GenBank database. Where possible, identities of ESTs were established by BLAST searches of the nucleotide database A National Center for Biotechnology Information (NCBI) (Bethesda, MD) BLAST

Niculescu

analysis of the accession number of each probe-set was done to identify each gene name BLAST analysis identified the closest known gene existing in the database (the highest known gene at the top of the BLAST list of homologues) which then could be used to search the GeneCards database (Weizmann Institute, Rehovot, Israel). Probe-sets that did not have a known gene were labeled "EST" and their accession numbers kept as identifiers.

[000125] Human Postmortem Brain Convergence: Information about candidate genes was obtained using GeneCards, the Online Mendelian Inheritance of Man database, as well as database searches using PubMed and various combinations of keywords (gene name, schizophrenia, schizoaffective, psychosis, human, brain, postmortem, blood, lymphocytes). Postmortem convergence was deemed to occur for a gene if there were published reports of human postmortem data showing changes in expression of that gene in brains from patients with psychotic disorders (schizophrenia, schizoaffective disorder).

[000126] Human Genetic Data Convergence: To designate convergence for a particular gene, the gene had to have published positive reports from candidate gene association studies, or map within lOcM of a microsatellite marker for which at least one published study showed evidence for genetic linkage to psychotic disorders (schizophrenia, schizoaffective disorder). The University of Southampton's sequence-based integrated map of the human genome (The Genetic Epidemiological Group, Human Genetics Division, University of Southampton( http://cedar genetics soton ac uk/public html) was used to obtain cM locations for both genes and markers The sex-averaged cM value was calculated and used to determine convergence to a particular marker For markers that were not present in the Southampton database, the Marshfield database (Center for Medical Genetics, Marshfield, WI, USA) was used with the NCBI Map Viewer web-site to evaluate linkage convergence

[000127] Ingenuity analysis: Ingenuity Pathway Analysis 3 1 (Ingenuity Systems, Redwood City, CA) was used to analyze the biological roles categories of the top candidate genes resulting from the CFG analysis, as well as employed to identify genes in the datasets that are the target of existing drugs .

[000128] Convergent Functional Genomics (CFG) Analysis Scoring (FIG. 2A): Genes were given the maximum score of 2 points if changed in the human blood samples with high threshold analysis, and only 1 point if changed with low threshold They received 1 point for each external cross-validating line of evidence (human postmortem brain data, human genetic data, animal model brain data, and animal model blood data). Genes received additional bonus points if changed in human brain and blood, as follows 2 points if changed in the same direction, 1 point if changed in opposite direction. Genes also received additional bonus points if changed in brain and blood of the animal model, as follows: 1 point if changed in the same direction in the brain and blood, and 0.5 points if changed in opposite direction. Thus the total maximum CFG score that a candidate biomarker gene can have is 9 (2 + 4 +2 +Y). Human own live subject human blood data was weighted more heavily (if it made the high threshold cut) than literature-derived human postmortem brain data, human genetic data, or the animal model data The

Niculescu

human blood-brain concordance data was weighted more heavily than the animal model blood-brain concordance. Other ways of weighing the scores of line of evidence may give slightly different results m terms of priontization, if not in terms of the list of genes per se

EXAMPLE 2 Hallucinations Biomarkers

[000129] Using the approach for analyzing human blood gene expression data, out of over 40,000 genes and ESTs on the Affymetnx Human Genome U133 Plus 2.0 GeneChip, by using the high threshold (HT), 5 novel candidate biomarker genes were identified (Tables 3A and 5A), of which 1 had at least one line of prior independent evidence for potential involvement in mood disorders (i.e CFG score of 3 or above). In addition to the high threshold genes, by using the low threshold, a larger list totaling 206 genes (Tables 3A and 5A), of which an additional 12 had at least two lines of prior independent evidence for potential involvement m psychotic disorders (i.e. CFG score of 3 or above) were identified. Of interest, one of the low threshold candidate biomarker genes (Phldal9) is reported to be changed in expression in the same direction, in lymphoblastoid cell lines (LCLs) from schizophrenia subjects.

[000130] Making a combined list of all the high value candidate biomarker genes identified as described above- consisting of all the high threshold genes and of the low threshold genes with at least one other external lines of evidence, and the low threshold genes with prior LCL evidence, a list of 50 top candidate biomarker genes for hallucinations, prioritized based on CFG score is identified (Table 3A)

[000131 ] Picking up the 5 top scoring candidate biomarkers for no hallucinations (Rhobtb3, Aldhlll,

Mpp3, FnI, Sppl) and the 5 top scoring candidate biomarkers for high hallucinations (Arhgef9, S100a6, Adamts5, Pdapl, Plxndl), a panel of 10 biomarkers for hallucinations is established that may have diagnostic and predictive value.

[000132] To test the predictive value of this panel (designated as the BioM-10 Hallucinations panel), a cohort of 31 psychotic disorders subjects was tested, containing the 23 subjects (12 no hallucinations, 11 high hallucinations) from which the candidate biomarker data was derived, as well as 8 additional subjects with hallucinations symptoms in the intermediate range (PANSS Hallucinations scores of 2 or 3) A prediction score for each subject was derived, based on the presence or absence of the 10 biomarker of the panel in their blood GeneChip data Each of the 10 biomarkers gets a score of 1 if it is detected as Present (P) in the blood form that subject, 0.5 if it is detected as Marginally Present (M), and 0 if it is called Absent (A) The ratio of the sum of the high hallucinations biomarker scores divided by the sum of the no hallucinations biomarker scores is multiplied by 100, and provides a prediction score. If the ratio of high hallucinations biomarker genes to no hallucinations biomarker genes is 1, i.e. the two sets of genes are equally represented, the prediction score is 1x100=100 The higher this score, the higher the predicted likelihood that the subject will have high hallucinations. The predictive score with

Niculescu

actual PANSS Hallucination scores was compared in the primary cohort of subjects with a diagnosis of psychotic disorders (n=31). A prediction score of 100 and above had a 80.0 % sensitivity and a 65 0 % specificity for predicting high hallucinations. A prediction score below 100 had a 91.7 % sensitivity and 77.8% specificity for predicting no hallucinations (FIG. 3 A and Table 4A).

[000133] Additionally, human blood gene expression analysis was conducted in a second cohort, subsequently collected, consisting of 14 subjects. The subjects in the secondary psychosis cohort had a distribution of no (n=6), intermediate (n=4) and high (n=4) hallucinations scores The second psychosis cohort was used as a replication cohort, to verify the predictive power of the mood state biomarker panel identified by analysis of data from the primary psychosis cohort.

[000134] In the second psychosis cohort (n=14), a prediction score of 100 and above had a 75.0 % sensitivity and a 55.6 % specificity for predicting highhallucinations. A prediction score below 100 had a 66 7 % sensitivity and 71 4 % specificity for predicting no hallucinations (FIG. 3B and Table 4A).

EXAMPLE 3 Delusions Biomarkers

[000135] Using an approach for analyzing human blood gene expression data, out of over 40,000 genes and ESTs on the Affymetrix Human Genome U133 Plus 2.0 GeneChip, by using the high threshold (HT), about 25 novel candidate biomarker genes (Tables 3B and 5B) were identified, of which 13 had at least one line of prior independent evidence for potential involvement in mood disorders (i e CFG score of 3 or above) In addition to the high threshold genes, by using the low threshold, a larger list totaling about 395 genes (Tables 3 A and 5A) were identified, of which an additional 36 had at least two lines of prior independent evidence for potential involvement in psychotic disorders (i e CFG score of 3 or above) Of interest, two of the high threshold candidate biomarker genes (Egrl and Tob2) and two of the low threshold candidate biomarker genes (Nrgl and Gpmόb) are reported to be changed in expression in the same direction, in lymphoblastoid cell lines (LCLs) from schizophrenia subjects.

[000136] Making a combined list of all the high value candidate biomarker genes identified as described above- consisting of all the high threshold genes and of the low threshold genes with at least one other external lines of evidence, including the low threshold genes with prior LCL evidence, a list of 99 top candidate biomarker genes for delusions were identified, prioritized based on CFG score (Table 3B)

[000137] Picking up the 5 top scoring candidate biomarkers for no delusions (Drd2, ApoE, Scampi,

Idhl, Nab I)) and the 5 top scoring candidate biomarkers for high delusions (Nrgl, Egrl, Dctnl, PlIp, Pv alb), a panel of 10 biomarkers for delusions is established that may have diagnostic and predictive value

[000138] To test the predictive value of a panel (designated as BioM-10 Delusions panel), a cohort of

31 psychotic disorders subjects, containing the 22 subjects (9 no delusions, 13 high delusions) from

Niculescu

which the candidate biomarker data was derived, as well as 9 additional subjects with delusions symptoms m the intermediate range (PANSS Delusions scores of 2 or 3) was analyzed. A prediction score for each subject was derived, based on the presence or absence of the 10 biomarkers of the panel m their blood GeneChip data. Each of the 10 biomarkers gets a score of 1 if it is detected as Present (P) m the blood form that subject, 0.5 if it is detected as Marginally Present (M), and 0 if it is called Absent (A). The ratio of the sum of the high delusions biomarker scores divided by the sum of the no delusions biomarker scores is multiplied by 100, and provides a prediction score. If the ratio of high delusions biomarker genes to no delusions biomarker genes is 1, i e. the two sets of genes are equally represented, the prediction score is 1x100=100. The higher this score, the higher the predicted likelihood that the subject will have high delusions. The predictive score was compared with actual PANSS Delusions scores in the primary cohort of subjects with a diagnosis of psychotic disorders (n=31). A prediction score of 100 and above had a 100 % sensitivity and a 55.6 % specificity for predicting high delusions. A prediction score below 100 had a 88 9% sensitivity and 90.9% specificity for predicting no delusions (FIG. 4A and Table 4B).

[000139] Additionally, human blood gene expression analysis was conducted in a second cohort, subsequently collected, consisting of 14 subjects. The subjects in the secondary psychosis cohort had a distribution of no (n=6), intermediate (n=2) and high (n=6) delusions scores The second psychosis cohort was used as a replication cohort, to verify the predictive power of the mood state biomarker panel identified by analysis of data from the primary psychosis cohort

[000140] In the second psychosis cohort (n=14), a prediction score of 100 and above had only a

50 0% sensitivity and a 37 5 % specificity for predicting high delusions A prediction score below 100 had a 33 3 % sensitivity and 50 0 % specificity for predicting no delusions (Figure 4B and Table 4B)

EXAMPLE 4 Clinical Applications

[000141 ] A sample, such as, 5-10 ml of blood is obtained from a patient suspected of having a psychotic disorder. RNA is isolated from the blood using standard protocols, for example with PAXgene blood RNA extraction kit (PreAnalytiX, a QIAGEN/BD company), followed by GLOBINclear™-Human or GLOBINclear™-Mouse/Rat (Ambion/ Applied Biosystems Inc., Austin, TX) to remove the globin mRNA.. Isolated RNA is labeled using any suitable detectable label if necessary for the gene expression analysis.

[000142] The labeled RNA is then quantified for the presence of one or more of the biomarkers disclosed herein. For example, gene expression analysis is performed using a panel of about 10 biomarkers (e.g., BioM 10 panel) for delusions and hallucinations (20 markers total) by any standard technique, for example microarray analysis or quantitative PCR or an equivalent thereof. The gene expression levels

Niculescu

are analyzed and the fold changes (either increased, decreased, no change or absent or present) are determined and a score is established

[000143] Applications of biomarkers for psychosis: There are no reliable clinical laboratory blood tests for psychosis. Given the complex nature of psychosis, the current reliance on patient self -report of symptoms and the clinician's impression on interview of patient is a rate limiting step in delivering the best possible care with existing treatment modalities, as well as in developing new and improved treatment approaches, including new medications.

[000144] Biomarkers disclosed herein are used in the form of panels of biomarkers, as exemplified by a

BioM-10 hallucination/delusion panel, for clinical laboratory tests for psychosis. Such tests can be: 1) at an mRNA level, quantitation of gene expression through polymerase chain reaction, 2) at a protein level, quantitation of protein levels through immunological approaches such as enzyme-linked immunosorbent assays (ELISA).

[000145] In conjunction with other clinical information, biomarker testing of blood and other fluids (CSF, urme) may play an important part of personalizing treatment to increase effectiveness and avoid adverse reactions- personalized medicine in psychiatry.

[000146] Biomarker-based tests for psychosis help: 1) Diagnosis, early intervention and prevention efforts; 2) Prognosis and monitoring response to various treatments; 3) New neuropsychiatry drug development efforts by pharmaceutical companies, at both a pre-clinical and clinical (Phase I, II and III) stages of the process, 4) Identifying vulnerability to psychosis for people in high stress occupations

[000147] Example 4A' Diagnosis, early intervention and prevention efforts A patient with no previous history of psychosis presents to a primary care doctor or internist complaining of non-specific symptoms Such symptoms are reported in conditions such as stress after a job loss, bereavement, mononucleosis, fibromyalgia, and postpartum in the general population, as well as Gulf War syndrome in veterans A panel of psychosis biomarkers can substantiate that the patient is showing objective changes in the blood consistent with a psychosis state. This will direct treatment towards a particular psychotic state.

[000148] Example 4B Clinical diagnosis of a young patient. A young patient (child, adolescent, young adult) with no previous history of psychosis, but coming from a family where one or more blood relatives suffer from psychosis may be monitored with regular lab tests by their primary care doctor/pediatrician using panels of psychosis biomarkers,. These tests may detect early on a change towards delusion or towards hallucination. This indicates and substantiates the need for initiation of a particular mode of treatment. This early intervention may be helpful to prevent full-blown illness and hospitalizations, with their attendant negative medical and social consequences. The decision to start medications in children and adolescents is particularly difficult without objective proof, due to the potential side-effects of medications in that age group (agitation, weigh-gain, sexual side-effects)

Niculescu

[000149] Example 4C Monitoring psychosis biomarkers over an extended period. Many patients with psychosis may present initially with a depressive episode to their primary care doctor or psychiatrist Monitoring psychosis biomarkers over time may also help to differentiate different forms of illness, e.g., depression vs. bipolar disorder (manic-depression), hallucination and others. This distinction is helpful because the first-line treatments for various psychiatric disorders are different. By seeing a change in biomarker profile towards a particular disease state before full blown illness and clinical symptoms, an appropriate addition or change to a medication can be implemented, preventing clinical decompensation, suffering and socio-economic loss (employment, relationships)

[000150] Example 4D: Prognosis and monitoring response to various treatments It takes up to 6-8 weeks to see if a medication truly works. By doing a baseline biomarker panel test, and then a repeat test early one in treatment (after 1 week, for example), there would be an early objective indication if a medication is starting to work or not, and if a switch to another medication is indicated. This would save time and avoid needles suffering for patients, with the attendant socio-economic losses.

[000151 ] Example 4E: Detecting loss of efficacy of an existing treatment. When a patient has been stable for a while on a medication for psychosis, regular biomarker testing may detect early loss of efficacy of the medication or recurrence of the illness, which would indicate the dose needs to be increased, medication changed, or another medication added, to prevent full blown clinical symptoms

[000152] Example 4F: Determining adequacy of treatment plan Objective monitoring with blood biomarker panels of the effect of less reliable or evidence-based interventions: psychotherapy, lifestyle changes, diet and exercise programs for improving mental health This will show whether the particular intervention works, is sufficient, or medications may need to be added to the regimen

EXAMPLE 5 New neuropsychiatric drug development

[000153] Early-stage pre-clinical work and clinical trials of new neuropsychiatric medications for treating psychosis may benefit from biomarker monitoring to help make a decision early on whether the compound is working This will speed up the drug-development process and avoid unnecessary costs. Depending on the expression profile of the biomarkers, the results of clinical trials may be obtained earlier than usual.

[000154] In later-stage large clinical trials, a new compound being tested may show an overall statistically non-significant positive effect, despite working well in a sub-group of people in the study. Biomarker testing may provide an objective signature of the genetic and biological make-up of the responders, which can inform recruitment for subsequent validatory clinical trials with higher likelihood of success, as well as inform which patients should be getting the medication, once it is FDA approved and on the market.

Niculescu

Table 1. Demographics: (a) individual (b) aggregate

[000155] Diagnosis established by DIGS comprehensive structured clinical interview. SZ-schizophrenia,

SZA-schizoaffective disorder. SubPD-substance induced psychosis. Psychosis score at time of blood draw, on a scale 1 (no symptoms) to 7 (severe symptoms).

(a) Individual demographic data

Niculescu

Table 2. High threshold and low threshold analysis in primary psychosis cohort. Genes are considered candidate biomarkers for high psychosis if they are called by the Affymetrix MAS5 software as Absent (A) in the blood of no psychosis subjects and detected as Present (P) in the blood of high psychosis subjects. Conversely, genes are considered candidate biomarkers for no psychosis if they are detected as Present (P) in no psychosis subjects and Absent (A) in high psychosis subjects.

9/12 No Hallucinations vs 9/11

High Threshold Candidate Biomarker Genes (changed in greater than or equal to High Hallucinations 75% subjects; i e. at least 3-fold enrichment) A/P and P/A analysis

8/12 No Hallucinations vs 7/11

Low Threshold Candidate Biomarker Genes (changed in greater than or equal to High Hallucinations 60% subjects; i e. at least 1.5-fold enrichment) A/P and P/A analysis

7/9 No Delusions vs 10/13 High

High Threshold Candidate Biomarker Genes (changed in greater than or equal to

Delusions 75% subjects; i e. at least 3-fold enrichment) A/P and P/A analysis

6/9 No Delusions vs 8/13 High

Low Threshold Candidate Biomarker Genes (changed in greater than or equal to

Delusions 60% subjects; i e. at least 1.5-fold enrichment) A/P and P/A analysis

Table 3A. Top candidate biomarker genes for hallucinations (n=50) prioritized by CFG score for multiple independent lines of evidence.

Top candidate biomarker genes for hallucinations For human blood data I -increased in high hallucinations state, D -decreased in high hallucinations state / increased in no hallucinations state For postmortem brain data Up- increased ; Down -decreased in expression; PCP -phencyclidine, CLZ- clozapine; (HT) High threshold Highlighted with an asterisk - BioM 10 markers

Niculescu

Niculescu

Niculescu

Niculescu

Table 3B. Top candidate biomarker genes for delusions (n=99) prioritized by CFG score for multiple independent lines of evidence.

Top candidate biomarker genes for hallucinations For human blood data I -increased in high delusions state; D -decreased in high delusions state / increased in no delusions state For postmortem brain data Up- increased , Down -decreased in expression, PCP - phencyclidine, CLZ- clozapine; (HT) High threshold Asterisk- BioM 10 markers

Niculescu

Niculescu

Niculescu

Niculescu

Niculescu

Table 4. BioM-10 Psychosis panels sensitivity and specificity for predicting psychosis state, (a) Hallucinations (B) Delusions

A. (Hallucinations).

Primary Psychosis Sensitivity Specificity Cohort

Hallucinations 80.0% 65.0%

No Hallucinations 91.7 % 77.8

Secondary Psychosis Sensitivity Specificity

Cohort

Hallucinations 75.0% 55.6 %

No Hallucinations 66.7% 71.4 %

Niculescu

B. (Delusions). Primary Psychosis

Sensitivity Specificity

Cohort

Delusions 100 0% 55 6%

No Delusions 88.9% 90.9%

Secondary Psychosis Sensitivity Specificity

Cohort

Delusions 50.0 % 37.5 %

No Delusions 33.3 % 50.0 %

Table 5 A: Complete list of candidate biomarker genes for hallucinations (n=211) identified using A/P analysis and CFG scoring

Niculescu

Niculescu

Niculescu

Niculescu

Niculescu

Niculescu

Niculescu

Table 5 B: Complete list of candidate biomarker genes for delusions (n=420) identified using A/P analysis and CFG scoring

Niculescu

Niculescu

Niculescu

Niculescu Atty. Docket No. 29920- 204952

Niculescu

Niculescu

Niculescu

Niculescu

Niculescu

Niculescu

Niculescu

Niculescu

Table 6 A. Additional candidate biomarker genes for hallucinations (n=15) identified by differential gene expression analysis (USing p<0.005). I -increased in high hallucinations state (biomarker for high hallucinations) ; D -decreased in high hallucinations state / increased in no hallucinations state (biomarker for no hallucinations)

Niculescu

Table 6 B. Additional candidate biomarker genes for delusions (n=132) identified by differential gene expression analysis (USing p<0.005). I -increased in high delusions states(biomarker for high delusions) ; D -decreased in high delusions state / increased in no delusions state (biomarker for no delusions)

Niculescu

Niculescu

Niculescu

Niculescu

Niculescu Atty. Docket No. 29920- 204952

Niculescu

Niculescu

Niculescu

Niculescu