Method for Monitoring Glucocorticoid-Mediated Pharmacodynamic Responses in the Body

Abstract

The present invention generally relates to methods for monitoring glucocorticoid-mediated pharmacodynamic responses in vivo. More specifically, the present invention relates to methods for using changes in gene signatures as pharmacodynamic markers of glucocorticoid exposure.

Description

[0001] Cross - Reference to Related Applications

[0002] This application claims the benefit of U.S. Provisional Application Serial No. 62 / 551,839, filed Aug. 30, 2017, the disclosure of which is incorporated herein by reference.

[0003] Throughout the application, various publications are referenced. The disclosures of these publications are hereby incorporated by reference in their entirety into this application to more fully describe the state of the art to which this invention pertains. Field of the Invention

[0004] The present invention generally relates to methods for monitoring glucocorticoid - mediated pharmacodynamic responses following in - vivo administration. Background of the Invention

[0006] Glucocorticoids (GCs) are potent anti - inflammatory drugs that are widely used to treat many human diseases, including rheumatoid arthritis (RA), inflammatory bowel disease, psoriasis, asthma, and systemic lupus erythematosus (SLE) (Buttgereit F. Bull NYU Hosp Jt Dis 2012; 70 Suppl 1: S26–9). However, their utility is limited by the toxicity of these drugs, which includes diabetes, osteoporosis, muscle wasting, fat redistribution, and hypothalamic - pituitary - adrenal (HPA) axis suppression (Desmet SJ et al., J Clin Invest 2017; 127: 1136–45). As the dose increases and the duration of use extends, the risk of adverse side effects increases (Bijlsma JWJ et al., Rheumatology 2016; 55 Suppl 2: ii3–5; Ruiz - Arruza I et al., Rheumatology 2014; 53: 1470–6). Despite the potential side effects, glucocorticoids remain a key standard - of - care treatment.

[0007] Glucocorticoids mediate their biological effects by interacting with the nuclear hormone receptor glucocorticoid receptor alpha (GR). Glucocorticoid receptor alpha is a ligand-activated transcription factor that induces transcription by binding to glucocorticoid response elements as a homodimer (Weikum ER et al., Nat Rev Mol Cell Biol 2017;18:159–74). Many GR-activated genes have anti-inflammatory activity (Colotta F et al., Science 1993;261:472–5; Abraham SM et al., JExp Med 2006;203:1883–9; Beaulieu E et al., Nat Rev Rheum 2011;7:340–8). However, transactivated genes are also associated with side effects (Cain DW et al., Nat Rev Immunol 2017;17:233–47). GR has also been shown to inhibit the activity of several pro-inflammatory transcription factors including NF-κB, AP-1, IR3F, CREB, NFAT, STAT, T-Bet and Gata-3, independent of DNA binding in a process called transrepression (Greulich F et al., Steroids 2016;114:7–15). To widen the therapeutic window, several synthetic glucocorticoids have been developed that have reduced transactivation but intact transrepression activity (Strehl C et al., Exp Opin Invest Drugs 2017;26:187–95).

[0008] In addition to the risk of producing damaging effects, long-term use of glucocorticoids is also associated with tissue-specific resistance (Rodriguez JM et al., Steroids 2016;115:182–92). Several resistance mechanisms have been described, including downregulation of GR expression and upregulation of dominant-negative isoforms of the receptor (Dendoncker K et al., Cytokine Growth FactorRev 2017;35:85–96). GR polymorphisms that regulate agonist sensitivity have also been documented (Straub RH et al., Rheumatology 2016;55Suppl 2:ii6–14). Given the heterogeneity of the clinical response to glucocorticoids, companion biomarkers with glucocorticoid biological activity would be highly valuable. Summary of the Invention

[0010] The present inventors developed a gene signature based on genes modulated by treating peripheral blood mononuclear cells (PBMCs) from normal healthy volunteer (NHV) donors with prednisolone. The sensitivity of the signature was confirmed by analyzing whole blood gene expression in healthy participants after administration of prednisolone or a partial GR agonist. In healthy subjects given prednisolone, the expression of the signature was higher than in those receiving the partial agonist, consistent with the transactivation potential of the compound. The expression of the signature in whole blood from patients with systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) was associated with known glucocorticoid-mediated pharmacodynamic effects, including higher levels of peripheral blood neutrophils and lower levels of peripheral blood lymphocytes. The expression of the signature in these cohorts was also consistent with the reported uses and doses of prednisolone.

[0011] The present invention includes a method for determining a human's response to a glucocorticoid, the method comprising administering to a human a glucocorticoid of interest, drawing blood from the human to whom the glucocorticoid of interest has been administered, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and comparing the post-administration gene signature score to a control gene signature score, wherein an increase in the gene signature scores of FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1, DUSP1 indicates a response to the glucocorticoid.

[0012] The present invention includes a method for determining a human's response to a glucocorticoid, the method comprising administering to a human a glucocorticoid of interest, drawing blood from the human to whom the glucocorticoid of interest has been administered, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and comparing the post-administration gene signature score to a control gene signature score, wherein an increase in the gene signature scores of FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1, DUSP1, PHC2, TLR2, TSC22D3, SLA, CRISPLD2, MAN2A2, FAR2, CEBPD, SPTLC2, HSPA6 indicates a response to the glucocorticoid.

[0013] The present invention includes a method for determining a human's response to a glucocorticoid, the method comprising administering to a human a glucocorticoid of interest, drawing blood from the human to whom the glucocorticoid of interest has been administered, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and comparing the post-administration gene signature score to a control gene signature score, wherein an increase in the gene signature scores of FKBP5, ECHDC3, IL1R2 indicates a response to the glucocorticoid.

[0014] The present invention includes a method for determining a human's response to a glucocorticoid, the method comprising administering a glucocorticoid of interest to a human, drawing blood from the human to whom the glucocorticoid of interest has been administered, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and comparing the post - administration gene signature score with a control gene signature score, wherein an increase in the gene signature scores of FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3 indicates a response to the glucocorticoid.

[0015] In one embodiment of the present invention, the glucocorticoid of interest is cortisone, dexamethasone, hydrocortisone, methylprednisolone, prednisolone, prednisone, triamcinolone, betamethasone, budesonide, fluticasone, and / or a synthetic glucocorticoid.

[0016] In one embodiment of the present invention, the control gene signature score is derived from blood collected from the same human prior to glucocorticoid administration and / or from blood collected from normal healthy controls not administered the glucocorticoid.

[0017] In one embodiment of the present invention, the human blood sample is collected 4 hours after administration from the human to whom the glucocorticoid of interest has been administered.

[0018] In one embodiment of the present invention, the gene signature score of a human who responds to the glucocorticoid of interest is at least 1.5 times that of the control.

[0019] In one embodiment of the present invention, the gene signature score of a human who responds to the glucocorticoid of interest is at least 2 times that of the control.

[0020] The present invention includes a method for treating a human diagnosed with SLE or RA, the method comprising 1) determining a human's response to a glucocorticoid by: administering a glucocorticoid of interest to a human, drawing blood from the human to whom the glucocorticoid of interest has been administered, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and 2) comparing the post - administration gene signature score with a control gene signature score, wherein an increase in the gene signature scores of FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1, DUSP1 indicates that the human will respond to the glucocorticoid of interest, and 3) administering the glucocorticoid to the human.

[0021] The present invention includes a method for treating a person diagnosed with SLE or RA, the method comprising 1) determining the response of the person to a glucocorticoid by administering the glucocorticoid of interest to the person, drawing blood from the person to whom the glucocorticoid of interest has been administered, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and 2) comparing the post - administration gene signature score with a control gene signature score, wherein an increase in the gene signature scores of FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1, DUSP1, PHC2, TLR2, TSC22D3, SLA, CRISPLD2, MAN2A2, FAR2, CEBPD, SPTLC2, HSPA6 indicates that the person will respond to the glucocorticoid of interest, and 3) administering the glucocorticoid to the person.

[0022] The present invention includes a method for treating a person diagnosed with SLE or RA, the method comprising 1) determining the response of the person to a glucocorticoid by administering the glucocorticoid of interest to the person, drawing blood from the person to whom the glucocorticoid of interest has been administered, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and 2) comparing the post - administration gene signature score with a control gene signature score, wherein an increase in the gene signature scores of FKBP5, ECHDC3, IL1R2 indicates that the person will respond to the glucocorticoid of interest, and 3) administering the glucocorticoid to the person.

[0023] The present invention includes a method for treating a person diagnosed with SLE or RA, the method comprising 1) determining the response of the person to a glucocorticoid by administering the glucocorticoid of interest to the person, drawing blood from the person to whom the glucocorticoid of interest has been administered, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and 2) comparing the post - administration gene signature score with a control gene signature score, wherein an increase in the gene signature scores of FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3 indicates that the person will respond to the glucocorticoid of interest, and 3) administering the glucocorticoid to the person. Brief Description of the Drawings

[0025] Figure 1A-1DShows genes regulated by glucocorticoids (GC). Peripheral blood mononuclear cells (PBMC) from normal healthy volunteers (NHV) were cultured in vitro for 6 hours with 1 μM prednisolone or DMSO vehicle alone. Affymetrix profiling was used to analyze gene expression of RNA. Analysis of genes regulated by prednisolone compared to vehicle is shown. The axes represent the FDR-adjusted log10P value versus fold change to the control. Genes upregulated (1A) and downregulated (1B) >2-fold by prednisolone compared to vehicle control with FDR-adjusted P value ≤ 0.05 are shown. ssGSEA scores of upregulated (1C) and downregulated (1D) genes in whole blood samples stimulated with increasing concentrations of prednisolone in vitro. **P = 0.005, ***P < 0.001.

[0026] Figure 2A-2E Shows validation of the glucocorticoid gene signature using partial GR agonists. Mammalian two-hybrid assays of the recruitment of prednisolone, GR modulators BMS-791826 and BMS-776532 to PGC1 (2A) and TIF2 (2B). Data represent the mean of triplicates, normalized to the activity induced by 200 nM dexamethasone. One representative experiment out of 2 is shown. By chromatin immunoprecipitation followed by qPCR analysis, the recruitment of 1 μM prednisolone, 1 μM GR modulator BMS-791826, and 2 μM BMS-776532 of GR (2C) and TIF2 (2D) to the promoters of ANGPTL4, ALOX5AP, and LEPREL1 was analyzed. Values represent the mean and standard deviation of triplicate reactions. Binding values were normalized to input values. ChIP = chromatin immunoprecipitation. P values were calculated by T test. *P < 0.01 versus prednisolone, **P < 0.01 versus prednisolone, ***P < 0.001 versus prednisolone, ns = not significant. (2E) GC gene signature scores of whole blood samples cultured in vitro with DMSO vehicle, 5 μM prednisolone, 5 μM GR modulator BMS-791826, or 10 μM BMS-776532. ***P < 0.001

[0027] Figure 3A-3B Shows in vivo validation of the GC gene signature. (3A) Oral doses of 150 or 300 mg of GR modulator BMS-791826, 10 mg prednisolone, or placebo were administered to NHV. Blood was collected before and 4 hours after administration. The GC gene signature of the whole blood expression profile was analyzed. *P = 0.027; ***P < 0.001. (3B) 5, 10, or 30 mg of prednisolone or placebo (i.e., 0 mg) was administered to NHV. Blood was drawn before and at different times after administration (2, 4, 8, 48, 144, and 216 hours). The GC gene signature of the whole blood expression profile was analyzed.

[0028] Figure 4A-4B Shows the relationship between the GC gene signature and GC use in the SLE and RA cohorts. (4A) Whole blood was collected from NHV as well as RA and SLE patients. RNA was isolated and used to probe an Affymetrix HG-219 array. GC gene signature scores were binned according to current GC users (true) versus patients receiving other standard of care treatments (false). Patients with no treatment information were listed under "NA" (not available). **P = 0.003, ***P < 0.001. (4B) GC gene signature scores from baseline samples of an abatacept SLE phase II trial. Patients were categorized according to GC dose (low, medium, or high). ns = not significant; **P = 0.001.

[0029] Figure 5A-5C Shows the GC gene signature correlation. Percentages of peripheral blood CD4+ T cells, CD8+ T cells, and CD19+ B cells in SLE (5A) and RA (5B) patients were plotted relative to the GC gene signature score for each patient. WBC = white blood cell. (5C) Peripheral blood neutrophil counts from the abatacept SLE trial baseline samples were plotted relative to the GC gene signature score for each patient. Correlations were analyzed by Spearman rank test.

[0030] Figure 6A-6B Shows validation of the 8-gene GC signature. (6A) GC gene signature scores using the abbreviated list of 8 genes for participants from a prednisone healthy male cohort 2 administered placebo, 150 or 300 mg GR modulator BMS-791826, or 10 mg prednisolone. *P = 0.015; ***P < 0.001. (6B) GC gene signature scores using the 8-gene list relative to peripheral blood neutrophil counts for baseline participants from the abatacept SLE study described above. Correlations were calculated using Spearman rank test. DETAILED DESCRIPTION OF THE INVENTION

[0032] The administration of glucocorticoids is carried out as described in the prescribing information for the various drugs. Generally, the initial dose depends on the severity of the particular disease entity being treated. In cases of lower severity, a lower dose is generally sufficient, but higher initial doses may be required in selected patients. The initial dose is usually maintained or adjusted until a satisfactory response is achieved. If there is a lack of a satisfactory clinical response after a reasonable period of time, the glucocorticoid is discontinued and the patient is arranged to receive other appropriate treatment. After a good response is achieved, the initial drug dose is reduced in smaller decrements at appropriate time intervals until the lowest dose that maintains an adequate clinical response is reached, thereby determining the appropriate maintenance dose. Obviously, being able to determine how a patient responds to glucocorticoids will shorten the time required to optimize the patient's maintenance dose.

[0033] The present invention includes a method for determining a human's response to a glucocorticoid by: administering the glucocorticoid of interest to a human, drawing blood from the human to whom the glucocorticoid of interest has been administered, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and comparing the post - administration gene signature with a control gene signature, wherein up - regulation of selected genes indicates a response to the glucocorticoid.

[0034] The inventors of the present invention found that after oral administration of glucocorticoids, the expression of the glucocorticoid gene signature in peripheral blood leukocytes of normal healthy volunteers (NHV) increased significantly. The expression of the signature increased in a dose - dependent manner, peaked at 4 hours after administration, and returned to baseline 48 hours after dosing. Lower expression was detected in NHV treated with a partial glucocorticoid receptor agonist, which is consistent with the reduced trans - activation potential of the compound. In patient cohorts with systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA), the expression of the glucocorticoid signature was negatively correlated with the percentage of peripheral blood lymphocytes and positively correlated with the peripheral blood neutrophil count, which is consistent with the known glucocorticoid receptor biology. Identification of genes regulated by glucocorticoids

[0035] To monitor glucocorticoid - dependent responses in peripheral blood, genes modulated by prednisolone in human PBMCs were identified. PBMCs from 10 independent normal healthy volunteer donors were treated with DMSO control or 1 μM prednisolone for 6 hours. Genes with a >2 - fold change and an FDR - corrected P - value <0.05 were identified as up - regulated or down - regulated genes.

[0036] Upregulated genes include ECHDC3, ACSL1, P2RX5, TPST1, TBC1D8, APBA2, SESN1, RNASE1, ABLIM3, RNASE6, BLM, KIF13B, DNMBP, SAP30, SFMBT1, TMEM2, HSPH1, METTL7A, HSPA1B, SLC25A37, VIPR1, FAR2, HSPA6, PHC2, PELI1, POLR1E, SPON2, GFOD1, SPRY1, NDFIP1, MAN2A2, ISG20, RAB31; FKBP5, IL1R2, ZBTB16, IRS2, IRAK3, DUSP1, SLCO4A1, TSC33D3, CD163, SLC1A3, ALOX15B, CCND3, RHOB, VSIG4, FLT3, CRISPLD2, ADORA3, RGS1, AMPH, CPM, ANG, CD93, SPTLC2, SERPINE1, ESTS2, TLR2, PER1, DNAJB1, PTK2B, CEBPD, SLA( Figure 1A )。

[0037] Downregulated genes include KMO, CCR5, CXCL8, FPR3, PLA2G7, PEA15, TRAF1, CSF2RB, TRDC, OLR1, KIAA0226L, FCGR2B, ATF5, CX3CR1, MYOF, SLAMF7, CD9, IL1RN( Figure 1B )。

[0038] Many of these genes are known glucocorticoid-regulated genes (Chinenov Y et al., BMC Genomics 2014;15:656). However, the upregulated genes ECHDC3, ACSL1, P2RX5, TPST1, TBC1D8, APBA2, SESN1, RNASE1, ABLIM3, RNASE6, BLM, KIF13B, DNMBP, SAP30, SFMBT1, TMEM2, HSPH1, METTL7A, HSPA1B, SLC25A37, VIPR1, FAR2, HSPA6, PHC2, PELI1, POLR1E, SPON2, GFOD1, SPRY1, NDFIP1, MAN2A2, ISG20, RAB31 have not been previously linked to glucocorticoid regulation. Some of the upregulated genes have been previously associated with anti-inflammatory activity, including DUSP1 (Abraham SM et al., J Exp Med 2006;203:1883–9), TSC22D3 (Beaulieu E et al., Nat Rev Rheum 2011;7:340–8), IRAK3 (Miyata M et al., Nat Commun 2015;6:606) and CD163 (Schaer DJ et al., Immunogenetics 2001;53:170–7), while several of the downregulated genes encode chemokines, chemokine receptors and other pro-inflammatory mediator proteins. Network analysis of the regulated genes showed enrichment of immune response pathways. A composite score (or gene signature) was generated using the single-sample gene set enrichment analysis (ssGSEA) algorithm to enrich these genes in the transcriptome of individual samples (Barbie DA et al., Nature 2009;462:108–12). Whole blood was stimulated in vitro with different concentrations of prednisolone for 5 hours, and the expression levels of upregulated and downregulated genes were calculated. The ssGSEA scores of the upregulated genes increased in a dose-dependent manner ( Figure 1C ). Similarly, the expression of the downregulated genes decreased in a dose-dependent manner ( Figure 1D ). The upregulated gene module had a larger dynamic range, and thus this gene module was used for all other analyses.

[0039] To provide further mechanistic evidence that this gene module accurately reflects GR activity, the activities of a subset of glucocorticoid receptor (GR) agonists were analyzed. The in vitro and in vivo activities of two selective GR modulators, BMS-776532 and BMS-791826, have been previously described (Weinstein DS et al., J Med Chem 2011;54:7318–33). Both compounds bind to GR and inhibit AP-1- and nuclear factor-κB-dependent reporter genes, but show significantly weaker induction of GR-dependent reporter genes compared to prednisolone. BMS-791826 is more potent than BMS-776532 in both trans-repression and trans-activation assays. GR modulates transcription by recruiting co-regulators, including TIF2 (Khan SH et al., Biol.Chem.2012;287:44546-44560) and PGC1α (Knutti D et al., Mol.Cell.Biol.2000;20:2411-2422). The trans-activation potential of these compounds was characterized using a mammalian two-hybrid system as well as chromatin immunoprecipitation. Compared to prednisolone, BMS-791826 and BMS-776532 recruited significantly less PGC1α and TIF2 to GR, peaking at 30%–75% of the prednisolone recruitment levels ( Figure 2A , 2B ). Compared to BMS-776532, BMS-791826 recruited more TIF2 (50% vs 30%), but a similar amount of PGC1α. In chromatin immunoprecipitation assays, compared to prednisolone, both compounds recruited significantly lower amounts of GR ( Figure 2C ) as well as TIF2 ( Figure 2D ) to the promoters of three target genes, verifying the reduced trans-activation potential of these compounds. Whole blood from two independent normal healthy volunteer donors was stimulated in vitro with these compounds and prednisolone for 5 h, followed by RNA isolation and Affymetrix analysis. The glucocorticoid gene signature scores of these samples correlated well with the trans-activation potential of these compounds: prednisolone > BMS-791826 > BMS-776532 ( Figure 2E ).

[0040] The present invention includes a method for determining a human's response to glucocorticoids, the method comprising stimulating whole blood collected from a human in need thereof with a glucocorticoid of interest, isolating RNA from the stimulated blood, analyzing the gene expression profile of the isolated RNA, and comparing the post-stimulation gene signature score and a control gene signature score, wherein the gene signature comprises the genes ECHDC3, ACSL1, P2RX5, TPST1, TBC1D8, APBA2, SESN1, RNASE1, ABLIM3, RNASE6, BLM, KIF13B, DNMBP, SAP30, SFMBT1, TMEM2, HSPH1, METTL7A, HSPA1B, SLC25A37, VIPR1, FAR2, HSPA6, PHC2, PELI1, POLR1E, SPON2, GFOD1, SPRY1, NDFIP1, MAN2A2, ISG20, RAB31, and an increase in the gene signature score indicates a response to glucocorticoids.

[0041] The present invention includes a method for determining a human's response to glucocorticoids, the method comprising stimulating whole blood collected from a human in need thereof with one or more glucocorticoids selected from the group consisting of cortisone, dexamethasone, hydrocortisone, methylprednisolone, prednisolone, prednisone, triamcinolone, betamethasone, budesonide, fluticasone, and synthetic glucocorticoids, isolating RNA from the stimulated blood, analyzing the gene expression profile of the isolated RNA, and comparing the post-stimulation gene signature score and a control gene signature score, wherein the gene signature comprises the genes ECHDC3, ACSL1, P2RX5, TPST1, TBC1D8, APBA2, SESN1, RNASE1, ABLIM3, RNASE6, BLM, KIF13B, DNMBP, SAP30, SFMBT1, TMEM2, HSPH1, METTL7A, HSPA1B, SLC25A37, VIPR1, FAR2, HSPA6, PHC2, PELI1, POLR1E, SPON2, GFOD1, SPRY1, NDFIP1, MAN2A2, ISG20, RAB31, and an increase in the gene signature score indicates a response to glucocorticoids.

[0042] In Vivo Evaluation of Glucocorticoid Gene Signatures

[0043] Because the glucocorticoid (GC) signature accurately characterizes the in vitro activity of glucocorticoid receptor (GR) agonists, the in vivo behavior of the GC gene signature was tested. Placebo, 10 mg of prednisolone, or 150 or 300 mg of the GR modulator BMS-791826 was administered to normal healthy volunteers (NHV). Blood was drawn before and 4 hours after dosing, and RNA was analyzed by Affymetrix gene expression profiling. Relative to pre-dose and placebo, the GC signature score was significantly increased at the 4-hour time point in participants given prednisolone ( Figure 3A ). The signature score in participants given BMS-791826 was higher than the pre-dose level and higher than that in participants given placebo, but lower than that in the prednisolone group. To study the kinetics of the GC gene signature response, whole blood RNA profiles of NHV administered different doses of prednisolone were analyzed. The GC gene signature score increased in a dose-dependent manner and peaked at 4 hours after dosing ( Figure 3B ). For all doses except the highest dose of prednisolone, the GC gene signature score had returned to baseline levels by 8 hours after dosing. By 48 hours after dosing, the signature scores in all groups were at baseline levels.

[0044] The composite glucocorticoid gene signature score is a sensitive measure of the in vivo response to glucocorticoid administration.

[0045] The present invention includes a method for determining a human's response to prednisolone, the method comprising administering prednisolone to a human, drawing blood from the human to whom prednisolone has been administered 4 hours after administration, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and comparing the post-administration gene signature score and a control gene signature score, wherein the gene signature comprises the genes ECHDC3, ACSL1, P2RX5, TPST1, TBC1D8, APBA2, SESN1, RNASE1, ABLIM3, RNASE6, BLM, KIF13B, DNMBP, SAP30, SFMBT1, TMEM2, HSPH1, METTL7A, HSPA1B, SLC25A37, VIPR1, FAR2, HSPA6, PHC2, PELI1, POLR1E, SPON2, GFOD1, SPRY1, NDFIP1, MAN2A2, ISG20, RAB31, and wherein an increase in the gene signature score indicates a response to prednisolone.

[0046] Relationship between the glucocorticoid gene signature and reported glucocorticoid use

[0047] To determine whether the glucocorticoid signature could distinguish patients based on treatment status, the expression of the signature was analyzed in cross-sectional cohorts of SLE or RA patients. The signature score was elevated in SLE or RA patients receiving glucocorticoid prescription treatment relative to normal healthy controls or patients receiving other standard-of-care medications ( Figure 4A ). Although the glucocorticoid signature was elevated, there was significant variation in signature scores among patients. The expression of the glucocorticoid gene signature was analyzed in baseline samples from a phase II study of abatacept in SLE (Fleishaker DL et al., BMC Musculoskelet Disord 2016;17:29) ( Figure 4B ). When reported prednisone doses were categorized as high dose (>30 mg), medium dose (10–30 mg), and low dose (<10 mg), the glucocorticoid gene signature scores generally corresponded. However, there remained significant variation in glucocorticoid gene signature scores among patients in all groups. This may reflect steroid resistance in some patients or compliance issues in some patients.

[0048] Correlation of the glucocorticoid gene signature with other pharmacodynamic endpoints

[0049] Glucocorticoids are known to cause redistribution of leukocyte subsets by causing neutrophil demargination from the bone marrow or sequestering lymphocyte populations in lymphoid organs (Merayo-Chalico J et al., Hum Immunol 2016;77:921–6; Spies CM et al., Arthritis Res Ther 2014;16 Suppl 2:S3). To determine whether the glucocorticoid signature was associated with these pharmacodynamic endpoints, CD4+ T cells, CD8+ T cells, and CD19+ B cells in peripheral blood from SLE and RA patients were tested. The expression of the glucocorticoid signature was negatively correlated with the percentages of these subsets in the peripheral blood of SLE patients ( Figure 5A ) and RA patients ( Figure 5B ). In the aforementioned abatacept SLE trial, the glucocorticoid signature score was positively correlated with neutrophil count ( Figure 5C ).

[0050] Thus, the expression of the glucocorticoid gene signature in both SLE and RA patients is associated with the known biological characteristics of glucocorticoids.

[0051] Selection of the glucocorticoid gene signature

[0052] Further refinement of the 64 members of the gene signature would be helpful for clinical practice. By comparing patients receiving prednisolone and placebo in the Prednisolone Healthy Male Cohort 2 trial, the list of 64 upregulated genes was refined to upregulated genes induced more than 1.5-fold with an FDR-adjusted P value < 0.05. This list was further filtered by detectability of expression in the Abatacept SLE trial. Of the initial 64 genes, 18 (FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1, DUSP1, PHC2, TLR2, TSC22D3, SLA, CRISPLD2, MAN2A2, FAR2, CEBPD, SPTLC2, HSPA6) met these criteria. Then, the top 3 genes (FKBP5, ECHDC3, and IL1R2), top 6 genes (FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3), and top 8 genes (FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1, DUSP1) from the list were used to calculate the ssGSEA score.

[0053]

[0054] Analysis of the Prednisolone Healthy Male Cohort 2 study of partial GR agonists with this refined signature fully characterized the behavior of the 64-gene signature ( Figure 6A ). Similar to the signature generated with the 64 upregulated genes, this 8-gene signature accurately reflected the transactivation potential of these compounds after in vivo administration of the partial agonist and prednisolone. This 8-gene signature was also positively correlated with the peripheral blood neutrophil count in the Abatacept SLE trial, with a similar P value to that generated with the list of 64 genes ( Figure 6B ). We conclude that a quantitative polymerase chain reaction (qPCR) assay targeting these 8 genes would be a sensitive biomarker of glucocorticoid pharmacodynamic activity and could be implemented by simple whole blood collection.

[0055] The invention includes a method for determining a human's response to prednisolone, the method comprising administering prednisolone to a human, drawing blood from the human administered with prednisolone 4 hours after administration, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and comparing the post-administration gene signature with a control gene signature, wherein the gene signature comprises the genes FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1, DUSP1, and wherein an increase in the gene signature score indicates a response to prednisolone.

[0056] The present invention includes a method for determining a human's response to prednisolone, the method comprising administering prednisolone to a human, drawing blood from the human to whom prednisolone has been administered 4 hours after administration, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and comparing the post - administration gene signature with a control gene signature, wherein the gene signature includes the genes FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1, DUSP1, PHC2, TLR2, TSC22D3, SLA, CRISPLD2, MAN2A2, FAR2, CEBPD, SPTLC2, HSPA6, and an increase in the gene signature score indicates a response to prednisolone.

[0057] The present invention includes a method for determining a human's response to prednisolone, the method comprising administering prednisolone to a human, drawing blood from the human to whom prednisolone has been administered 4 hours after administration, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and comparing the post - administration gene signature with a control gene signature, wherein the gene signature includes the genes FKBP5, ECHDC3, IL1R2, and an increase in the gene signature indicates a response to prednisolone.

[0058] The present invention includes a method for determining a human's response to prednisolone, the method comprising administering prednisolone to a human, drawing blood from the human to whom prednisolone has been administered 4 hours after administration, isolating RNA from the collected blood, analyzing the gene expression profile of the isolated RNA, and comparing the post - administration gene signature with a control gene signature, wherein the gene signature includes the genes FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, and an increase in the gene signature score indicates a response to prednisolone.

[0059] Conclusion

[0060] Glucocorticoids remain the mainstay of treatment for many autoimmune and inflammatory diseases due to their potent anti - inflammatory activity. However, long - term use is associated with an increased risk of toxic effects. Given this risk and the significant inter - patient variability in the clinical response to glucocorticoids, there is a need for sensitive, objective, pharmacodynamic biomarkers that can assist in correct dose selection.

[0061] The glucocorticoid gene signature of the present invention was developed based on in vitro expression profiling experiments using PBMCs derived from NHV. Focusing on genes induced by glucocorticoid treatment rather than downregulated genes due to a greater dynamic range across all donors. The ssGSEA algorithm was used to generate a composite score applicable to individual samples or patients. Based on several observations, this algorithm appears to sensitively detect glucocorticoid-dependent transcriptional responses. From in vitro whole blood profiling studies and in vivo studies using samples obtained after administration of oral full GR agonists and partial GR agonists, the glucocorticoid signature score accurately reflects the transactivation potential of synthetic partial GR agonists. The signature score also characterizes the in vivo and in vitro dose responses to glucocorticoids.

[0062] When the method of the present invention was applied to samples from a cross-sectional cohort of SLE and RA patients, the glucocorticoid signature scores were higher in patients using glucocorticoids compared to patients receiving other non-glucocorticoid standard-of-care drug therapies. In baseline samples from the abatacept SLE trial, the glucocorticoid signature scores gradually increased with increasing steroid dose.

[0063] The glucocorticoid gene signature of the present invention not only serves as part of clinical practice but also helps to identify potential confounding effects of steroids in clinical trials. In baseline samples from the abatacept SLE trial, the glucocorticoid gene signature scores were generally correlated with the reported steroid doses. However, significant inter-patient variability was observed within each dose group. The method of the present invention can be used to determine whether a patient is resistant to glucocorticoids or not compliant with the study protocol.

[0064] Given the strong anti-inflammatory effects of glucocorticoids, trials typically include requirements to gradually reduce or even discontinue glucocorticoids. The glucocorticoid gene signature of the present invention provides an objective method for assessing compliance with the protocol. Calculation of the 8-gene signature score can be easily performed by qPCR or using other platforms for whole blood collection. In summary, the gene signature of the present invention can be widely used to monitor responses to glucocorticoids in many indications treated with glucocorticoid prescriptions.

[0065] Example 1

[0066] Identification of genes regulated by glucocorticoids

[0067] Lymphocytes were isolated from the blood of 10 independent donors using Ficoll gradient centrifugation. Cells were cultured at 5 million lymphocytes / well in 500 μl of assay medium (RPMI-1640 with GlutaMAX, 10% charcoal-treated fetal bovine serum; Gibco Laboratories, Gaithersburg, MD, USA) in 96-well flat-bottomed, closed plates (Qiagen, Hilden, Germany). Cells were cultured for 6 hours with dimethyl sulfoxide (DMSO) vehicle or 1 μM prednisolone. After 6 hours, the cells were pelleted and resuspended in 1 ml of nucleic acid purification lysis solution (Applied Biosystems, Foster City, CA, USA), which had been diluted 1:2 with phosphate-buffered saline without calcium and magnesium (Invitrogen, Carlsbad, CA, USA). The cells were incubated in the lysis buffer for 10 minutes at room temperature and then stored at –80 °C. RNA was isolated using the Qiagen RNeasy isolation kit according to the manufacturer's instructions.

[0068] For profiling whole blood, whole blood containing anticoagulant citrate phosphate dextrose solution A from 4 normal healthy volunteers was cultured for 5 hours with DMSO vehicle, 200 nM prednisolone, 1 μM prednisolone, 5 μM prednisolone, 5 μM GR modulator BMS-791826, or 10 μM BMS-776532 and then transferred to PAXgene tubes. Total RNA was isolated, then treated with DNase I and purified using the Qiagen RNeasy MinElute Cleanup kit. RNA concentration was determined using NanoDrop (Thermo Fisher Scientific, Waltham, MA, USA), and RNA quality was assessed using the Experion electrophoresis system (Bio-Rad Laboratories, Hercules, CA, USA). All target labeling reagents were purchased from Affymetrix (West Sacramento, CA, USA).

[0069] Double-stranded cDNA was synthesized from 1 μg of total RNA by reverse transcription with an oligo-dT primer containing the T7 RNA polymerase promoter and using a cDNA synthesis system (Invitrogen, Carlsbad, CA, USA) for double-strand conversion. Biotin-labeled cRNA was generated from the cDNA and used to probe the Human Genome HT_HG-U133A array (Affymetrix, Sunnyvale, CA, USA), which consists of 96 individual HG-U133A arrays in a 96-well plate. All cDNA and cRNA target preparation steps were performed on an Affymetrix Caliper GeneChip Array Station. Array hybridization, washing, and scanning were performed according to the manufacturer's recommendations.

[0070] Gene signature development and scoring

[0071] CEL files from the Affymetrix Array Station were processed and normalized using the Robust Multi-Array Average (RMA) algorithm (Gautier I et al., Bioinformatics 2004; 20:307–15), using the “Affy” package in R (version 3.2.1; 16) and Bioconductor (Gentleman RC et al., Genome Biol. 2004, 5(10):R80), where the custom CDF file was from BrainArray (version 18.0.0; Dai M et al., Nucleic Acids Res 2005; 33:e175). Differential gene expression analysis was run in Array Studio (OmicSoft, Cary, NC, USA) using a moderated t-test (Ritchie ME et al., Nucleic Acids Res 2015; 43:e47) to compare gene expression levels in prednisolone-treated samples with control samples. A multiple testing correction method was used to adjust the P-values, which is also known as the false discovery rate (FDR, Benjamini Y et al., J. Royal Statistical Soc., Series B 1995; 57:289). Genes that were upregulated or downregulated by at least 2-fold in all experiments and had an adjusted P-value < 0.05 were reported as the GC gene signature.

[0072] To score individual samples according to the enrichment level of the GC gene signature, we applied the single-sample gene set enrichment analysis (ssGSEA) algorithm (Barbie DA et al., Nature 2009; 462:108–12) to generate a composite score, which was achieved by using the Gene Set Variation Analysis software package in R (version 3.4.0, S et al., BMC Bioinformatics 2013; 14:7). The algorithm was modified to make the enrichment score fall between -1 and 1, which represents the lowest to highest possible rank of the GC gene in the transcriptome.

[0073] Mammalian two-hybrid assay

[0074] Sequences encoding full-length human peroxisome proliferator-activated receptor γ coactivator-1α (PGC1α) or full-length human transcriptional mediator / intermediator 2 (TIF2) were cloned in-frame with the GAL4 DNA-binding domain into the vector pM (Clontech, Mountain View, CA, USA). Full-length human GR was cloned in-frame with the VP16 activation domain into the vector pVP16 (Clontech). Human SK-N-MC neuroblastoma cells (American Type Culture Collection, Manassas, VA, USA) were co-transfected with these plasmids and a GAL4-dependent luciferase reporter gene (pGF-luc; Promega, Madison, WI, USA). Transfectants were stimulated with 200 nM dexamethasone or different concentrations of prednisolone, GR modulators BMS-791826 or BMS-776532. Luciferase activity was measured 48 hours after transfection.

[0075] Chromatin immunoprecipitation

[0076] For chromatin immunoprecipitation, A549 cells were cultured for 1 hour in RPMI with 10% charcoal-treated fetal bovine serum with DMSO, 1 μM prednisolone, 1 μM GR modulator BMS-791826, or 2 μM BMS-776532. Cells were fixed with formaldehyde and then sent to Active Motif (Carlsbad, CA, USA) for analysis of the recruitment of GR and TIF2 to specific promoter sequences using quantitative polymerase chain reaction (qPCR).

[0077] Patient cohort

[0078] Systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) cross-sectional cohort

[0079] During 2014 and 2015, peripheral blood was obtained from 86 SLE patients during routine follow-up at Northwell Health (Great Neck, NY, USA). Patients were receiving standard-of-care treatment for systemic SLE or lupus nephritis, including hydroxychloroquine, mycophenolate mofetil, glucocorticoids, and / or belimumab. Patient characteristics were as follows: age 45 ± 14 years (mean ± SD); 85% female; Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K), 3.7 ± 3.2 (mean ± SD); history of lupus nephritis in 43%; disease duration 15 ± 13 years (mean ± SD).

[0080] For the RA cohort, during 2014 and 2015, blood was obtained from 84 patients during routine follow-up at Brigham and Women’s Hospital, Boston, MA or Northwell Health, Great Neck, New York. Patients were receiving standard-of-care treatment for RA, including methotrexate, hydroxychloroquine, tofacitinib, abatacept, anti-tumor necrosis factor biologics, tocilizumab, glucocorticoids, and / or non-steroidal anti-inflammatory drugs. Patient characteristics were as follows: age 57 ± 14 years (mean ± SD); 77% female; American College of Rheumatology 2010 Rheumatoid Arthritis classification criteria, 7.8 ± 1.6 (mean ± SD); disease duration 17 ± 10 years (mean ± SD). Blood was drawn into PAXgene tubes at each follow-up and heparinized. Blood was shipped overnight and processed upon arrival for fluorescence-activated cell sorting analysis. Blood was also drawn from age- and sex-matched normal healthy volunteers (Bristol-Myers Squibb, Princeton, NJ, USA) into PAXgene tubes. RNA was isolated from PAXgene tube blood and used to probe Affymetrix HG-U219 gene arrays using the protocol described above.

[0081] Abatacept SLE clinical cohort (NCT00119678)

[0082] Baseline PAXgene collections and complete blood counts were obtained from 144 adult SLE patients who met British Isles Lupus Assessment Group (BILAG) criteria A or B. The baseline population consisted of 53% polyarthritis patients, 35% discoid lupus patients, and 12% serositis patients. Overall, 87% of patients received prednisone, 50% received hydroxychloroquine, and 41% received immunosuppressants (methotrexate, azathioprine, or mycophenolate mofetil).

[0083] Prednisolone healthy male cohort 1 (NCT03196557)

[0084] Male normal healthy volunteers were randomly assigned (6 participants / group) to receive 5, 10 or 30 mg of prednisolone per day for 7 days. Two participants received placebo. PAXgene tubes were collected before and 2, 4, 8, 48, 144 and 216 hours after dosing.

[0085] Prednisolone healthy male cohort 2 (NCT03198013)

[0086] Male normal healthy volunteers were randomly assigned to receive placebo polyethylene glycol (PEG)-400 solution (4 participants); a single daily oral dose of the GR modulator BMS-791826 in PEG-400 solution (150 or 300 mg) (6 participants / dose); or a single daily dose of 10 mg prednisolone (4 participants) for 3 consecutive days. PAXgene tubes were collected before and 4 hours after dosing on Day 1.

[0087] Peripheral blood phenotyping

[0088] Heparinized whole blood was stained with a premixed antibody cocktail, followed by lysis and fixation. Antibodies used for the SLE panel included CD3-eF450 (clone OKT3; eBioscience, San Diego, CA, USA), CD4-PE-Cy7 (clone OKT4; BioLegend, San Diego, CA, USA), CD8-APC-H7 (clone SK1; BD Biosciences, San Jose, CA, USA), and CD19-BV421 (clone HIB19; BioLegend).

[0089] Antibodies used for the RA panel included CD19-BV421, CD3-Ax700 (clone OKT3; BioLegend), CD4-Percp-Cy5.5 (clone RPA-T4; eBioscience), and CD8-Bv785 (clone RPA-T8; BioLegend).

Claims

1. Use of a reagent for determining the gene signature scores of FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1 and DUSP1 in the preparation of a kit for implementing a method for determining the response of a human to a glucocorticoid of interest, wherein the human is diagnosed with rheumatoid arthritis, and the method comprises: a) isolating RNA from a blood sample collected from a human after administration of the glucocorticoid of interest, b) performing gene expression profiling on the isolated RNA in step (a), and c) comparing the post - administration gene signature score with a control gene signature score, wherein the control gene score is derived from blood collected from the same human before glucocorticoid administration or from blood collected from a normal healthy control who has not been administered the glucocorticoid, wherein an increase in the gene signature scores of FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1 and DUSP1 indicates a response to the glucocorticoid.

2. Use of a reagent for determining the gene signature scores of FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1 and DUSP1 in the preparation of a kit for implementing a method for determining the response of a human to a glucocorticoid of interest, wherein the human is diagnosed with systemic lupus erythematosus, and the method comprises: a) isolating RNA from a blood sample collected from a human after administration of the glucocorticoid of interest, b) performing gene expression profiling on the isolated RNA in step (a), and c) comparing the post - administration gene signature score with a control gene signature score, wherein the control gene score is derived from blood collected from the same human before glucocorticoid administration or from blood collected from a normal healthy control who has not been administered the glucocorticoid, wherein an increase in the gene signature scores of FKBP5, ECHDC3, IL1R2, ZBTB16, IRS2, IRAK3, ACSL1 and DUSP1 indicates a response to the glucocorticoid.

3. The use according to claim 1 or 2, wherein the glucocorticoid of interest is selected from the group consisting of: cortisone, dexamethasone, hydrocortisone, methylprednisolone, prednisolone, prednisone, triamcinolone, betamethasone, budesonide and fluticasone.

4. The use according to claim 1 or 2, wherein, an increase in the gene signature score to 1.5 - fold compared to the control indicates a response to the glucocorticoid.

5. The use according to claim 1 or 2, wherein, an increase in the gene signature score to 2 - fold compared to the control indicates a response to the glucocorticoid.

6. The use according to claim 1 or 2, wherein the blood sample is collected from the human 4 hours after the glucocorticoid administration.

7. The use according to claim 1 or 2, wherein the glucocorticoid of interest is a synthetic glucocorticoid.

Images