Systems and methods for identifying responders and non-responders to immune checkpoint blockade therapy

1. A method for determining whether or not a subject is likely to respond to a checkpoint blockade therapy, the method comprising:

obtaining RNA expression data for the subject having, suspected of having, or at risk of having cancer;

using the expression data to determine expression levels, in the subject, for at least five genes selected from the set of predictor genes consisting of BRAF, ACVR1B, MPRIP, PRKAG1, STX2, AGPAT3, FYN, CMIP, ROBO4, RAB40C, HAUS8, SNAP23, SNX6, ACVR1B, MPRIP, COPS3, NLRX1, ELAC2, MON1B, ARF3, ARPIN, SPRYD3, FLI1, TIRAP, GSE1, POLR3K, PIGO, MFHAS1, NPIPA1, DPH6, ERLIN2, CES2, LHFP, NAIF1, ALCAM, SYNE1, SPINT1, SMTN, SLCA46A1, SAP25, WISP2, TSTD1, NLRX1, NPIPA1, HIST1H2AC, FUT8, FABP4, ERBB2, TUBA1A, XAGE1E, SERPINF1, RAI14, SIRPA, MT1X, NEK3, TGFB3, USP13, HLA-DRB4, IGF2, and MICAL1; and

determining, using the determined expression levels and a statistical model trained using expression data indicating expression levels for a plurality of genes for a plurality of subjects, that the subject is likely to respond to the checkpoint blockade therapy, wherein the statistical model comprises a logistic regression model,

wherein the checkpoint blockade therapy is a PD1 inhibitor and/or a CTLA4 inhibitor,

wherein the method further comprises administering to the subject a PD1 inhibitor, a CTLA4 inhibitor, or a PD1 inhibitor and a CTLA4 inhibitor.

2. The method of claim 1 , wherein the statistical model comprises a first set of dependent variables, each representing a ratio of a pair of genes, wherein the genes are selected from: BRAF, PRKAG1, STX2, AGPAT3, FYN, CMIP, ROBO4, RAB40C, HAUS8, SNAP23, SNX6, ACVR1B, MPRIP, COPS3, NLRX1, ELAC2, MON1B, ARF3, ARPIN, SPRYD3, FLI1, TIRAP, GSE1, POLR3K, PIGO, MFHAS1, NPIPA1, DPH6, ERLIN2, CES2, LHFP, NAIF1, ALCAM, SYNE1, SPINT1, SMTN, SLCA46A1, SAP25, WISP2, TSTD1, NLRX1, NPIPA1, HIST1H2AC, FUT8, FABP4, ERBB2, TUBA1A, XAGE1E, SERPINF1, RAI14, SIRPA, MT1X, NEK3, TGFB3, USP13, HLA-DRB4, IGF2, and MICAL1.

3. The method of claim 2 , wherein the statistical model comprises a first set of dependent variables, each representing a ratio of a pair of genes, wherein the genes are selected from: BRAF, PRKAG1, STX2, AGPAT3, FYN, CMIP, ROBO4, RAB40C, HAUS8, SNAP23, SNX6, XAGE1E, SERPINF1, RAI14, SIRPA, MT1X, NEK3, TGFB3, USP13, HLA-DRB4, IGF2, and MICAL1.

4. The method of claim 2 , wherein the statistical model comprises a first set of dependent variables, each representing a ratio of a pair of genes, wherein the genes comprise: BRAF, RAI14, PRKAG1, STX2, AGPAT3, FYN, CMIP, ROBO4, RAB40C, HAUS8, SNAP23, and SNX6.

5. The method of claim 2 , wherein the statistical model comprises a first set of dependent variables, each representing a ratio of a pair of genes, wherein the genes consist of: BRAF, RAI14, PRKAG1, STX2, AGPAT3, FYN, CMIP, ROBO4, RAB40C, HAUS8, SNAP23, and SNX6.

6. The method of claim 2 , wherein the first set of dependent variables comprises at least two ratios selected from: BRAF:RAI14, ACVR1B:MPRIP, ACVR1B:COPS3, PRKAG1:STX2, NLRX1:ELAC2, MON1B:STX2, ARF3:MPRIP, ARPIN:MPRIP, SPRYD3:FLI1, TIRAP:MPRIP, GSE1:RAI14, POLR3K:HAUS8, RAB40C:HAUS8, PIGO:MPRIP, MFHAS1:USP13, GSE1:NPIPA1, DPH6:STX2, ERLIN2:RAI14, CES2:LHFP, and NAIF1:HAUS8.

7. The method of claim 2 , wherein the first set of dependent variables comprises at least two ratios selected from: MON1B:STX2, FAM234A:LIN37, DPH6:STX2, BRAF:RAI14, ADCK2:C14ORF80, POLR3K:HAUS8, URB1:TMEM181, GCLC:NEK3, RAB40C:HAUS8, NLRX1:ELAC2, CMIP:ROBO4, NXT2:FBXO5, EIF3H:NEK3, EHHADH:SNX6, DMTN:RASD1, SLC46A1:RBM8A, ACVR1B:GTF2H2, NPEPPS:HDAC2, CXCL16:BASP1, MFHAS1:DYRK3, ACVR1B:USP13, DPH6:C14ORF166, CES2:LHFP, ACVR1B:DCP1B, NAIF1:HDGFRP2, ABCC1:TRIO, GPR107:STX2, ZDHHC7:USP13, PRKAG1:TSEN2, PC:HAUS8, LRBA:CEP192, POM121C:CNPY4, KAT14:SETD5, SLC35A5:SNX6, ATP6V1A:GTDC1, TXNL4B:AKAP8L, SLC36A1:HAUS8, PSAP:SH3BP5, CMIP:OLFML2B, TATDN3:NEK3, TASP1:DDX5, SDC4:FUT8, TMEM254:COPS3, ARF3:MPRIP, SLC46A1:PMF1-BGLAP, ATP6V1A:FBXO30, MFHAS1:C2CD5, ERAP1:SYNE1, F11R:FYN, RCHY1:RNF146, ATP6V1A:PDCD5, ACVR1B:ELAC2, CLN3:HAUS8, NAIF1:HAUS8, PRKAG1:SOCS4, HNRNPH2:USP13, TPD52:MTRF1L, ACVR1B:C14ORF80, IST1:NPIPA1, DPH6:TCEAL1, CSNK2A1:MRGBP, CXCL16:FILIP1L, AGK:USP13, MYO18A:FYN, SIRPA:FLII, C160RF58:FLII, TRIM11:AKAP8L, MFHAS1:DMPK, JMJD8:AKAP8L, DIAPH1:SYNE1, BCKDHA:HAUS8, TMEM254:PSMC5, ACVR1B:HTRA2, MON1B:SYNE1, DCAKD:PMF1-BGLAP, VWA5A:RASD1, TPD52:TRA2A, ZMIZ1:STX2, NUB1:C2CD5, GSE1:RAI14, AGFG1:STX2, NXT2:TRA2A, ACSS1:NPIPA5, FBXW8:USP13, CMIP:TRIO, AGPAT3:FYN, PSMF1:PTOV1, CREG1:TARS, SLC46A1:CEP131, SIRPA:SERPINF1, DNAJA2:HDAC2, ERLIN2:RAI14, FAM234A:ZNF428, CHMP1A:LIN37, FAM110A:TCF7, ACVR1B:COPS3, GSE1:DDX11, CREG1:ARFGAP3, BRPF3:USP13, MFHAS1:USP13, LAMP1:MAPK7, ACSS1:PMF1-BGLAP, SUFU:TRIO, ARF3:DAD1, NLRX1:TRA2A, NLRX1:SLC39A13, CMIP:SH3BP5, PPIF:HAUS8, ANKRD13A:SOCS4, F8:SYNE1, ATP6V1A:USP48, ACVR1B:MPRIP, TMEM141:HAUS8, TIRAP:MPRIP, ZDHHC12:HAUS8, SLC46A1:MED9, MFHAS1:NCAPD2, ERBB2:CENPL, JMJD7:PQBP1, PHKG2:AKAP8L, SLC36A1:SLC26A6, ATP6V1A:SPDL1, DCTN5:CEP89, IPPK:STX2, LAMB3:ADM, ARPIN:MPRIP, SLC46A1:FYN, ACVR1B:LTV1, GDE1:ZNF576, DMTN:GFPT2, LCMT2:RDH11, ACVR1B:CCDC66, ACVR1B:NEK3, SEC24B:MAP3K7, ZNF764:AKAP8L, CHMP1A:HAUS8, PIGO:USP13, ARF3:MAPK7, GSE1:CCDC66, ACVR1B:METTL17, C200RF196:HAUS8, ARF3:CBY1, BRPF3:MPRIP, SLC46A1:PSMC5, CMTR2:MAP3K7, TASP1:CCDC66, BRPF3:PIP5K1A, PIGO:FLII, MYO18A:ACIN1, PSMF1:AKAP8L, FBXW11:LTV1, CXCL16:ADAMTS2, SPRYD3:FLII, DPH6:FRA10AC1, PDXK:HAUS8, ACVR1B:WDR45B, MON1B:SMIM10L1, LAT:PRRX1, WDR24:AKAP8L, EHHADH:CCDC174, ACVR1B:CEP89, MFHAS1:ODF2, ALDH6A1:GOPC, GSE1:NBPF14, EHHADH:MEX3C, BRPF3:STX2, EHHADH:GTF2H2, PIGO:MPRIP, PRKAG1:STX2, EHHADH:SMIM10L1, SLC36A1:SH3PXD2A, ADCK2:STX2, SORD:PDCD5, ACVR1B:LLGL1, LAMTOR3:ZNF644, PIM1:MICAL2, CREG1:NPIPA1, GSE1:NPIPA1, PRR13:HAUS8, WDR55:RAD1, CMIP:NID2, DIAPH1:TARS, SNAP23:SNX6, GSE1:TSPYL2, C2ORF68:NPIPA1, MFHAS1:TRIO, DPH6:CGRRF1, KIF13B:STX2, PTK2B:TCF7, ATP2A2:STX2, ANKS1A:USP13, JRK:NEK3, LRBA:DDX5, IDH2:HAUS8, CCNF:HAUS8, CMIP:CHN1, STAU2:STX2, ACSS1:LHFP, GSR:STX2, IGF2R:FYN, CXCL16:ACVRL1.

8. The method of claim 2 , wherein the first set of dependent variables consists of the ratios: BRAF:RAI14, PRKAG1:STX2, AGPAT3:FYN, CMIP:ROBO4, RAB40C:HAUS8, SNAP23:SNX6.

9. The method of claim 2 , wherein the first set of dependent variables comprises at least three ratios, at least four ratios, at least five ratios, or at least six ratios.

10. The method of claim 1 , wherein the expression data to determine expression levels, in the subject, for at least eight genes from the group consisting of BRAF, ACVR1B, MPRIP, PRKAG1, STX2, AGPAT3, FYN, CMIP, ROBO4, RAB40C, HAUS8, SNAP23, SNX6, ACVR1B, MPRIP, COPS3, NLRX1, ELAC2, MON1B, ARF3, ARPIN, SPRYD3, FLI1, TIRAP, GSE1, POLR3K, PIGO, MFHAS1, NPIPA1, DPH6, ERLIN2, CES2, LHFP, NAIF1, ALCAM, SYNE1, SPINT1, SMTN, SLCA46A1, SAP25, WISP2, TSTD1, NLRX1, NPIPA1, HIST1H2AC, FUT8, FABP4, ERBB2, TUBA1A, XAGE1E, SERPINF1, RAI14, SIRPA, MT1X, NEK3, TGFB3, USP13, HLA-DRB4, IGF2, and MICAL1.

11. The method of claim 1 , wherein the expression data to determine expression levels, in the subject, for at least eleven genes from the group consisting of BRAF, ACVR1B, MPRIP, PRKAG1, STX2, AGPAT3, FYN, CMIP, ROBO4, RAB40C, HAUS8, SNAP23, SNX6, ACVR1B, MPRIP, COPS3, NLRX1, ELAC2, MON1B, ARF3, ARPIN, SPRYD3, FLI1, TIRAP, GSE1, POLR3K, PIGO, MFHAS1, NPIPA1, DPH6, ERLIN2, CES2, LHFP, NAIF1, ALCAM, SYNE1, SPINT1, SMTN, SLCA46A1, SAP25, WISP2, TSTD1, NLRX1, NPIPA1, HIST1H2AC, FUT8, FABP4, ERBB2, TUBA1A, XAGE1E, SERPINF1, RAI14, SIRPA, MT1X, NEK3, TGFB3, USP13, HLA-DRB4, IGF2, and MICAL1.

12. The method of claim 1 , further comprising training the logistic regression model, the logistic regression model having a plurality of regression variables, each of the plurality of regression variables representing a ratio of a pair of genes for respective pairs of the predictor set of genes.

13. The method of claim 12 , wherein the logistic regression model comprises a respective plurality of weights for the plurality of regression variables, wherein estimating the parameters of the statistical model comprises:

estimating the plurality of weights using the expression data for the plurality of subjects and information indicating which of the plurality of subjects responded to the checkpoint blockade therapy and/or which of the plurality of subjects did not respond to the checkpoint blockade therapy.

14. The method of claim 12 , wherein training the statistical model comprises iteratively adding regression variables for respective genes to the statistical model, at least in part by:

identifying a candidate gene in the subset of genes;

augmenting a current statistical model with a regression variable for the candidate gene to obtain an augmented statistical model;

evaluating performance of the augmented statistical model; and

determining to add the regression variable for the candidate gene to the current statistical model based on results of evaluating the performance.

15. The method of claim 1 , wherein the subject has melanoma.

16. The method of claim 1 , wherein the method further comprises providing, to a user, an indication of whether or not the subject is likely to respond to the checkpoint blockade therapy.