Patent Application
20150299795
Several types of human malignancies arise from cells that belong to the hematologic system, including peripheral blood cells, bone marrow, lymph nodes and lymphatic and blood vasculature. Identification of mutations associated with these diseases is helpful for developing diagnostics and treatments. Several candidate gene and GWAS studies of human B cell Non-Hodgkin Lymphoma (NHL), a type of hematologic cancer, have identified germ-line predisposing risk factors in human NHL patients (Conde et al, 2010, Wang et al, 2010, Smedby et al, 2011, Conde et al, 2011). These studies reported conflicting findings of association of polymorphisms in the major histocompatibility complex (MHC) to some subtypes of NHL, highlighting the challenges due to the heterogenic nature of subtypes of NHL and of the human population. Human angiosarcoma, another type of hematologic cancer, is very rare and accounts for 1˜3% of adult sarcomas. Angiosarcoma is a very aggressive cancer due to its ability to facilitate excessive angiogenesis. The rarity of this disease in humans is a limiting factor when undertaking a feasible genetic study of angiosarcoma.
Dogs also suffer from several spontaneously occurring cancers of the hematologic system, including lymphoma (LSA) and hemangiosarcoma (HSA, a cancer of blood vessel endothelial cells). These canine cancers are clinically and histologically similar to human Non-Hodgkin Lymphoma (NHL) and angiosarcoma, respectively. The domesticated dog is an ideal model species to study genetics of human diseases and non-human animal diseases, as each breed has been created and maintained by strict selective breeding, thereby causing the alleles underlying desirable traits and alleles predisposing the dog to specific diseases to become common within certain breeds. Golden retrievers, one of the most popular family breeds in the U.S., have a high lifetime risk of cancer, with over 60% of golden retrievers dying from some type of cancer. Two of the most common cancers in golden retrievers are LSA and HSA, with a lifetime risk of 13% and 25%, respectively.
The invention provides methods for identifying subjects that are at elevated risk of developing certain types of cancers. Subjects are identified based on the presence of one or more germ-line and/or somatic markers shown to be associated with the presence of cancer, in accordance with the invention.
In one aspect, the invention provides a method comprising analyzing genomic DNA from a canine subject for the presence of a risk allele identified by BICF2G63035726 or BICF2G630183630, and identifying a canine subject having a chromosome 5 risk allele identified by BICF2G63035726 or BICF2G630183630 as a subject (a) at elevated risk of developing a hematological cancer or (b) having a hematological cancer that is as yet undiagnosed (e.g., morphologically undetected).
In some embodiments, the genomic DNA is obtained from white blood cells of the subject. In some embodiments, the genomic DNA is analyzed using a single nucleotide polymorphism (SNP) array. In some embodiments, the genomic DNA is analyzed using a bead array.
In another aspect, the invention provides a method comprising analyzing genomic DNA from a canine subject for the presence of a mutation in a locus selected from the group consisting of C11orf7, ANGPTL5, KIAA1377, TRPC6, NTN1, NTN3, STX8, WDR16, USP43, DHRS7C, GLP2R, BIRC3, CD68, MYBBP1A, CHD3, CHRNB1, RANGRF, ZBTB4, and a locus comprising SEQ ID NO:1, and identifying a canine subject having a mutation in a locus selected from the group consisting of C11orf7, ANGPTL5, KIAA1377, TRPC6, NTN1, NTN3, STX8, WDR16, USP43, DHRS7C, GLP2R, BIRC3, CD68, MYBBP1A, CHD3, CHRNB1, RANGRF, ZBTB4, and a locus comprising SEQ ID NO:1 as a subject (a) at elevated risk of developing a hematological cancer or (b) having a hematological cancer that is as yet undiagnosed (e.g., morphologically undetected).
In some embodiments, the genomic DNA is obtained from white blood cells of the subject. In some embodiments, the mutation is in a regulatory region of the locus. In some embodiments, the mutation is in a regulatory region of a locus selected from the group consisting of ANGPTL5, BIRC3, CD68, MYBBP1A, CHD3, CHRNB1, RANGRF, ZBTB4, and a locus comprising SEQ ID NO:1. In some embodiments, the mutation is in a coding region of the locus. In some embodiments, the mutation is in a coding region of a locus selected from the group consisting of ANGPTL5, KIAA1377 and TRPC6. In some embodiments, the mutation is in a coding region of TRPC6.
In another aspect, the invention provides a method comprising analyzing, in a sample from a canine subject, an expression level of a locus selected from the group consisting of ANGPTL5, BIRC3, CD68, MYBBP1A, CHD3, CHRNB1, RANGRF, ZBTB4, and a locus comprising SEQ ID NO:1, and identifying a canine subject having an altered expression level of a locus selected from the group consisting of ANGPTL5, BIRC3, CD68, MYBBP1A, CHD3, CHRNB1, RANGRF, ZBTB4, and a locus comprising SEQ ID NO:1 as compared to a control, as a subject (a) at elevated risk of developing a hematological cancer or (b) having a hematological cancer that is as yet undiagnosed (e.g., morphologically undetected).
In yet another aspect, the invention provides a method comprising analyzing, in a sample from a canine subject, an expression level of a locus selected from the group consisting of TRPC6, KIAA1377, PIK3R6, ANGPTL5, HS3ST3B1, and BIRC3, and identifying a canine subject having an altered expression level of a locus selected from the group consisting of TRPC6, KIAA1377, PIK3R6, ANGPTL5, HS3ST3B1, and BIRC3 as compared to a control, as a subject (a) at elevated risk of developing a hematological cancer or (b) having an undiagnosed hematological cancer. In some embodiments, the locus is TRPC6. In some embodiments, the altered expression level is a decreased expression level of TRPC6, KIAA1377, PIK3R6, ANGPTL5 and/or BIRC3 compared to control, and/or an increased expression level of HS3ST3B1 compared to control. In some embodiments, the locus is TRPC6.
In some embodiments, the sample is a white blood cell sample from a canine subject. In some embodiments, the sample is a tumor sample from a canine subject. In some embodiments, the control is a level of expression in a sample from a canine subject having lymphoma and negative for risk allele identified by BICF2G63035726 and risk allele identified by BICF2G630183630.
In some embodiments, the altered expression level is (a) a decreased expression level of ZBTB4, BIRC3 and/or ANGPTL5 compared to control, and/or (b) an increased expression level of CD68, CHD3, CHRNB1, MYBBP1A and/or RANGRF compared to control.
In some embodiments, the altered expression level is analyzed using an oligonucleotide array or RNA sequencing.
In another aspect, the invention provides a method comprising analyzing genomic DNA in a sample from a canine subject for presence of a mutation in a locus selected from the group consisting of TRAF3, FBXW7, DOK6, RARS, JPH3, LRRN3, MLL2, OGT, POU3F4, SETD2, CACNA1G, DSCAML1, MLL, ADD2, ARID1A, ARNT2, CAPN12, EED, ENSCAFG00000002808, ENSCAFG00000005301, ENSCAFG00000017000, ENSCAFG00000024393, ENSCAFG00000025839, ENSCAFG00000027866, L3MBTL2, LOC483566, MAPKBP1, NCAPH2, PPP6C, Q597P9_CANFA, SGIP1, XM—533169.2, XM—533289.2, XM—541386.2, XM—843895.1, and XM—844292.1, and identifying a canine subject having a mutation in a locus selected from the group consisting of TRAF3, FBXW7, DOK6, RARS, JPH3, LRRN3, MLL2, OGT, POU3F4, SETD2, CACNA1G, DSCAML1, MLL, ADD2, ARID1A, ARNT2, CAPN12, EED, ENSCAFG00000002808, ENSCAFG00000005301, ENSCAFG00000017000, ENSCAFG00000024393, ENSCAFG00000025839, ENSCAFG00000027866, L3MBTL2, LOC483566, MAPKBP1, NCAPH2, PPP6C, Q597P9_CANFA, SGIP1, XM—533169.2, XM—533289.2, XM—541386.2, XM—843895.1, and XM—844292.1, as a subject (a) at elevated risk of developing a hematological cancer or (b) having a hematological cancer that is as yet undiagnosed (e.g., morphologically undetected).
In some embodiments, the genomic DNA comprises a risk allele identified by BICF2G63035726 or BICF2G630183630.
In some embodiments, the genomic DNA comprises a mutation in a locus selected from the group consisting of C11orf7, ANGPTL5, KIAA1377, TRPC6, NTN1, NTN3, STX8, WDR16, USP43, DHRS7C, GLP2R, BIRC3, CD68, MYBBP1A, CHD3, CHRNB1, RANGRF, ZBTB4, and a locus comprising SEQ ID NO:1. In some embodiments, the sample comprises (a) a decreased expression level of ZBTB4, BIRC2 and/or ANGPTL5 compared to control, and/or (b) an increased expression level of CD68, CHD3, CHRNB1, MYBBP1A and/or RANGRF compared to control.
In some embodiments, the genomic DNA is obtained from white blood cells of the subject. In some embodiments, the mutation is in a coding region of the locus. In some embodiments, the mutation (a) is a frame shift mutation, (b) is a premature stop mutation, or (c) results an amino acid substitution. In some embodiments, the hematological cancer is a lymphoma or a hemangiosarcoma. In some embodiments, the lymphoma is a B cell lymphoma.
In another aspect, the invention provides a method comprising analyzing genomic DNA in a sample from a subject for presence of a mutation in a locus selected from the group consisting of ADD2, ARID1A, ARNT2, CAPN12, EED, ENSCAFG00000002808, ENSCAFG00000005301, ENSCAFG00000017000, ENSCAFG00000024393, ENSCAFG00000025839, ENSCAFG00000027866, L3MBTL2, LOC483566, MAPKBP1, NCAPH2, PPP6C, Q597P9_CANFA, SGIP1, XM—533169.2, XM—533289.2, XM—541386.2, XM—843895.1, and XM—844292.1, or an orthologue of such a locus, and identifying a subject having a mutation in a locus selected from the group consisting of ADD2, ARID1A, ARNT2, CAPN12, EED, ENSCAFG00000002808, ENSCAFG00000005301, ENSCAFG00000017000, ENSCAFG00000024393, ENSCAFG00000025839, ENSCAFG00000027866, L3MBTL2, LOC483566, MAPKBP1, NCAPH2, PPP6C, Q597P9_CANFA, SGIP1, XM—533169.2, XM—533289.2, XM—541386.2, XM—843895.1, and XM—844292.1, or an orthologue of such a locus, as a subject (a) at elevated risk of developing a cancer or (b) having a cancer that is as yet undiagnosed (e.g., morphologically undetected).
In some embodiments, the subject is a human subject. In some embodiments, the subject is a canine subject. In some embodiments, the cancer is a hematological cancer. In some embodiments, the cancer is a lymphoma or a hemangiosarcoma. In some embodiments, the cancer is a B cell lymphoma. In some embodiments, the cancer is a hemangiosarcoma. In some embodiments, the cancer is angiosarcoma.
In another aspect, the invention provides isolated nucleic acid molecules. In some embodiments, the isolated nucleic acid molecule comprises SEQ ID NO: 2.
The invention is based in part on the discovery of germ-line and somatic markers associated with particular cancers in canine subjects. The two canine cancer types studied were B-cell lymphoma (referred to herein as LSA) and hemangiosarcoma (referred to herein as HSA). These cancers were chosen for analysis at least in part because they are clinically and histologically similar to human B cell NHL and angiosarcoma. These cancers are also relatively common in canine subjects. For example, golden retrievers in the U.S. have a lifetime risk for developing LSA or HSA of 13% and 25% respectively.
The discovery of germ-line markers was made by genotyping “normal” canine subjects and those having these types of cancers, and identifying markers (or alleles) that associated (or tracked) with either or both of the cancers. Surprisingly, this revealed a non-random association between certain alleles on chromosome 5 of canine genomic DNA and the presence of both of the cancers studied. Remarkably, two regions on chromosome 5 were found to contribute as much as 50% of the total risk associated with both cancers studied. Genes previously mapped to the regions of these alleles were sequenced and/or had their expression levels measured. A number of these genes were found to be differentially expressed in tumors from subjects carrying the risk alleles as compared to tumors from subjects that did not carry the risk alleles. Risk alleles are also referred to herein as risk-associated alleles. The differential expression pattern may be indicative of the downstream mediators of the increased cancer risk associated with the alleles on chromosome 5. In addition, a number of these genes were found to be mutated in their coding regions in tumors from subjects carrying the risk alleles.
Similarly, the discovery of somatic markers associated with particular cancers was made by genomic sequencing of tumor cells and matched normal cells from canine subjects, and then identifying differences between the genomic sequences. A variety of somatic mutations were discovered in tumor cells that were not present in normal cells. In some instances, the observed somatic mutations affected gene products (e.g. frameshift mutations).
The invention therefore provides diagnostic and prognostic methods that involve detecting one or more of the germ-line and somatic markers in canine subjects in order to (a) identify subjects at elevated risk of developing a hematological cancer such as LSA or HAS or (b) identify subjects having a hematological cancer that is as yet undiagnosed (e.g., because it is morphologically undetectable at that time). Accordingly, the methods can be used for prognostic purposes and for early detection. Identifying canine subjects at an elevated risk of developing such cancers is useful in a number of applications. For example, canine subjects identified as at elevated risk may be excluded from a breeding program and/or conversely canine subjects that do not carry the risk alleles may be included in breeding program. As another example, canine subjects identified as at elevated risk may be monitored for the appearance of certain cancers and/or may be treated prophylactically (i.e., prior to the development of the tumor) or therapeutically (including prior to a detectable tumor). Canine subjects carrying one or more of the risk markers may also be used to further study the progression of these cancer types and optionally the efficacy of various treatments.
In addition, in view of the clinical and histological similarity between canine LSA and HSA with human NHL and angiosarcoma, the markers identified by the invention may also be markers and/or mediators of disease progression in these human cancers as well. Accordingly, the invention provides diagnostic and prognostic methods for use in canines, human subjects, as well as others.
The invention refers to the germ-line and somatic markers described herein as risk-associated markers to convey that the presence of these various markers has been shown to be associated with the occurrence of certain cancer types in accordance with the invention. The germ-line markers may also be referred to herein as risk-associated alleles. The somatic markers may also be referred to herein as risk-associated mutations. These various marker types will be discussed in greater detail herein.
The germ-line and somatic markers of the invention can be used to identify subjects at elevated risk of developing a cancer such as a hematological cancer. An elevated risk means a lifetime risk of developing such a cancer that is higher than the risk of developing the same cancer in a population that is unselected for the presence or absence of the marker (i.e., the general population) or a population that does not carry the risk-associated marker.
The germ-line markers associated with HSA and LSA in canine subjects were identified through genome-wide association studies (GWAS) of 148 HSA cases, 43 B cell LSA cases, and 190 healthy older control golden retrievers. The analysis was performed using single nucleotide polymorphism (SNP) arrays customized for canine genomic DNA analysis. Such arrays are commercially available from suppliers such as Affymetrix and Illumina (Illumina 170K canine HD array). Such arrays can be used to analyze genomes for polymorphisms (or alleles) in a population. Each polymorphism will have an expected frequency based on the general population. These GWAS studies identify polymorphisms in a particular subject that are present at a disproportionate frequency (otherwise represented by a “P value” that differs from the expected P value for the polymorphism in the general population). The data set so obtained was also controlled for population stratification given the known high levels of encrypted relatedness and complex family structures in canine populations such as golden retrievers. The data analysis algorithm is shown in
This analysis revealed the presence of one or more regions within chromosome 5 that were disproportionately represented in the subjects having LSA and HSA compared to the control “healthy” subjects, as shown in
Further significant SNPs are listed in Table 2. The use or detection of any SNPS listed herein is contemplated.
The position (i.e., the chromosome coordinates) and SNP ID for each SNP in Table 2 are based on the CanFam 2.0 genome assembly (see, e.g., Lindblad-Toh K, Wade C M, Mikkelsen T S, Karlsson E K, Jaffe D B, Kamal M, Clamp M, Chang J L, Kulbokas E J 3rd, Zody M C, et al.: Genome sequence, comparative analysis and haplotype structure of the domestic dog. Nature 2005, 438:803-819). The first base pair in each chromosome is labeled 0 and the position of the SNP is then the number of base pairs from the first base pair (for example, the SNP on chromosome 5 at position 36,417,176 is located 36,417,176 base pairs from the first base pair of chromosome 5).
A more in-depth analysis revealed the presence of two linkage disequilibrium (LD) regions on chromosome 5 that were independently disproportionately represented in the subjects having LSA and HSA compared to the control subjects. The first region spanned an area on chromosome 5 from about 32.5 Mb to about 33.1 Mb. This region was identified according to the SNP BICF2G63035726. It is also identified as position 32,901,346 bp, CamFam2.0. This region may also be identified using one or more of the SNPs in Table 2 located within the boundaries of the first region. The second region spanned an area on chromosome 5 from about 36.6 Mb to about 37.3 Mb. This region was identified according to the SNP BICF2G630183630. It is also identified as position 36,848,237 bp, CamFam2.0. This region may also be identified using one or more of the SNPs in Table 2 located within the boundaries of the second region. Details relating to these two chromosome 5 LD regions are shown in Table 4 in the Examples section. Schematics of these chromosome 5 regions are provided in
The chromosome 5 risk-associated regions comprise a number of loci that may be the downstream mediators of the elevated cancer risk phenotype.
The locus comprising the nucleotide sequence of SEQ ID NO:1 is a novel locus. The sequence is provided in Table 8. Its coordinates, on CamFam2.0 genome, are chr5:32732962-32766974. The underlined and bolded sequences correspond to a novel transcript made by the locus.
In accordance with the invention, all of these loci were sequenced in order to identify particular mutations that may be associated with elevated cancer risk. These sequencing studies identified a number of loci that are mutated in tumors carrying one or both germ-line risk alleles. Exemplary mutations found within the coding sequence include those in the following loci: KIAA1377, ANGPTL5 and TRP6. Details relating to these mutations are provided in Table 5 in the Examples section. As indicated in the Table, germ-line mutations were detected in these loci but somatic mutations (as described below) were not.
The invention contemplates methods that sequence these chromosome 5 specific markers and identify subjects having mutations in these markers. The presence of such mutations is associated with an elevated risk of developing cancer or the presence of an otherwise undetectable cancer, according to the invention. The invention further contemplates that mutations in these markers may exist in their regulatory and/or coding regions. As a result, sequencing analysis may be performed on mRNA transcripts or cDNA counterparts (for coding region mutations) or on genomic DNA (for regulatory region mutations). As used herein, regulatory regions are those nucleotide sequences (and regions) that control the temporal and/or spatial expression of a gene but typically do not contribute to the amino acid sequence of their gene product. As used herein, coding regions are those nucleotide sequences (and regions) that dictate the amino acid sequence of the encoded gene product. Methods for sequencing such markers are described herein.
An analysis of the expression levels in tumors carrying one or both of the germ-line risk-associated alleles as compared to expression levels in tumors that did not carry the risk-associated allele(s) revealed differential expression of some of the chromosome 5 germ-line markers. Some of the markers, including ANGPTL5, BIRC3, CD68, MYBBP1A, CHD3, CHRNB1, RANGRF, ZBTB4, and a locus comprising SEQ ID NO:1, were differentially expressed in the tumors carrying the germ-line risk-associated allele(s) compared to the tumors that did not contain the germ-line risk-associated allele(s). More specifically, it was further found that certain markers were down-regulated in tumors carrying the germ-line allele(s) while others were up-regulated in tumors carrying the germ-line allele(s) compared to a tumor that does not carry the germ-line allele(s). The markers that are down-regulated include ZBTB4, BIRC3 and ANGPTL5. The markers that are up-regulated include CD68, CHD3, CHRNB1, MYBBP1A and RANGRF. The tumors therefore could be characterized at the molecular level based on the expression profile or one or more of these markers. The expression profile composites from the analysis of LSA and HSA tumors are provided in
Additionally, other markers on chromosome 5, including TRPC6, KIAA1377, PIK3R6, ANGPTL5, HS3ST3B1, and BIRC3, were differentially expressed in the tumors carrying the germ-line risk-associated allele(s) compared to the tumors that did not contain the germ-line risk-associated allele(s). TRPC6, KIAA1377, PIK3R6, ANGPTL5 and BIRC3 were found to be down-regulated in tumors carrying the germ-line allele(s) compared to a tumor that does not carry the germ-line allele(s). HS3ST3B1 was found to be up-regulated in tumors carrying the germ-line allele(s) compared to a tumor that does not carry the germ-line allele(s). In some embodiments, the chromosome 5 marker is TRPC6.
Additionally, other markers on chromosome 5, including XLOC—083025, PLEKHG5, TMPRSS13, TNFRSF18, and TNFRSF4, were differentially expressed in the tumors carrying the germ-line risk-associated allele(s) compared to the tumors that did not contain the germ-line risk-associated allele(s). TMPRSS13, TNFRSF18, and TNFRSF4 were found to be down-regulated in tumors carrying the germ-line allele(s) compared to a tumor that does not carry the germ-line allele(s). XLOC—083025 and PLEKHG5 were found to be up-regulated in tumors carrying the germ-line allele(s) compared to a tumor that does not carry the germ-line allele(s).
Expression data related to these markers are provided in
Accordingly, in view of these findings, the invention contemplates methods for measuring the level of expression of one or more of these markers and then identifying a subject that is at elevated risk of developing cancer or that has an as yet undiagnosed cancer based on an expression level profile similar to that provided herein.
The differential expression of various of the chromosome 5 markers suggests that mutations in these markers may occur in the regulatory region instead of or in addition to the coding region. A marker that appears to have mutations in both regulatory and coding regions is the ANGPTL5 gene.
The invention is further premised in part on the discovery that other non-chromosome 5 genes are differentially expressed in tumors carrying the germ-line risk-associated allele(s) and tumors that do not carry the germ-line risk-associated allele(s). These non-chromosome 5 genes are as follows: ABTB1, AGA, AK1, ANXA1, B4GALT3, BAG3, BAT1, BCAT2, BEX4, BID, BIRC3, BTBD9, CCDC134, CCDC18, CCDC88C, CD1C, CD320, CD68, CDKN1A, CMTM8, COASY, COL7A1, CPT1B, CTSD, DDX41, DENND4B, DGKA, DHRS1, DUSP6, ECM1, EFCAB3, EIF4B, LOC478066, FABP3, FADS1, FBXL6, FBXO11, FBXO33, FBXW7, FNBP4, GALNT6, GBE1, GDPD3, GNGT2, GPR137B, GSTM1, GTF2IRD2, GTF3C3, GUCY1B3, HBD, LOC609402, ICAM4, IRF5, KIF5C, KLHDC1, KLHDC9, LBX2, LOC475952, LOC479273, LOC479683, LOC482085, LOC482088, LOC482361, LOC482532, LOC482790, LOC483843, LOC484249, LOC484784, LOC485196, LOC487557, LOC487994, LOC490377, LOC490693, LOC491116, LOC609521, LOC610353, LOC610841, LOC611771, LOC612387, LOC612917, LOXL3, LZTS2, MED24, MFSD6, MTA3, MYC, MYO19, NAGA, NAPRT1, NEIL1, NQO2, OGT, OSGEPL1, OVGP1, P2RX5, PDLIM7, PER3, PHKA2, PIGV, PIK3R6, PITPNM1, PRKRA, PVRIG, RAB24, RAB25, RABEP2, RASAL1, RBM11, RBM18, RBM35A, RBPJ, REC8, RILPL1, RPA1, SIGLEC12, SLC37A1, SP1, SUOX, TIMM22, TIPARP, TLE4, TMED6, TMEM41B, TNFSF8, TRAF5, TRMT1, TTC39C, TTF1, TUBB2A, UNC93B1, YWHAE, ZFAND2A, ZFC3H1, ZMAT1, ZMYM1, ZNF215, ZNF292, ZNF331, ZNF513, ZNF608, ZNF674, and ZNF711.
The markers that are up-regulated compared to control are as follows: ANXA1, BCAT2, BEX4, BID, BTBD9, CCDC18, CD1C, CD320, CD68, COASY, CTSD, DDX41, EFCAB3, FABP3, FBXL6, FBXO11, FBXO33, FBXW7, FNBP4, GBE1, GTF3C3, GUCY1B3, ICAM4, KLHDC1, LOC484249, LOC485196, LOC487557, LOC487994, LOC491116, LOC609521, LOC610353, LOC610841, LOC611771, LOC612917, LOXL3, MED24, MFSD6, MTA3, MYC, MYO19, NQO2, OGT, OSGEPL1, OVGP1, PDLIM7, PER3, PIGV, PITPNM1, PRKRA, RAB24, RASAL1, RBM35A, RILPL1, SIGLEC12, SLC37A1, SP1, SUOX, TIPARP, TMEM41B, TRMT1, TTF1, UNC93B1, ZFAND2A, ZFC3H1, ZNF215, ZNF331, ZNF608, ZNF674, and ZNF711.
The remaining markers in the list are down-regulated compared to the control.
Further non-chromosome 5 genes are as follows: C1GALT1, FGFR4, SCARA5, GFRA2, CD5L, CXL10, SLC25A48, KRT24, RP11-10N16.3, RPL6, ENSCAFG00000029323, XLOC—011971, ENSCAFG00000013622, XLOC—102336, and HIST1H. C1GALT1, FGFR4, SCARA5, GFRA2, CD5L, CXL10, SLC25A48, KRT24, and RP11-10N16.3 were found to be down-regulated in tumors carrying the germ-line allele(s) compared to a tumor that does not carry the germ-line allele(s). RPL6, ENSCAFG00000029323, XLOC—011971, ENSCAFG00000013622, XLOC—102336, and HIST1H were found to be up-regulated in tumors carrying the germ-line allele(s) compared to a tumor that does not carry the germ-line allele(s).
Further non-chromosome 5 genes are as follows: C1GALT1, EXTL1, B6F250, ENSCAFG00000030890, XLOC—088759, RGS13, KRT24, RP11-10N16.3, TNFAIP3, CD8A, Q95J95, XLOC—094643, SLC25A48, FGFR4, GPC3, NKG7, CXCR3, CD5L, PADI4, CXL10, GFRA2, SLC38A11, FABP4, PTPN22, ENSCAFG00000028509, U6, XLOC—044225, XLOC—100547, CCL22, CCDC168, TNIK, ENSCAFG00000030894, RGS10, HTR4, CNNM1, FBXO11, GRM5, SCARA5, OBSL1, RAB19, GZMK, ENSCAFG00000031494, TRBC2, TNFRSF21, ENSCAFG00000031437, GZMB, XLOC—091705, CHGA, H6BA90, GALNT13, CACNA1D, CD8B, XLOC—024761, EOMES, ZNF662, AFF2, COL6A6, HTRA1, LAD1, ENSCAFG00000029236, SCN2A, XLOC—077615, KIAA1456, CCL19, KIF5C, XLOC—026187, GPR27, ENSCAFG00000028940, ADAMTS2, CCL5, MAPK11, SMOC1, ABCA4, KIAA1598, KLRK1, LAT, FAM190A, ENSCAFG00000029467, PGBD5, TBXA2R, CSF1, MT1, ENSCAFG00000029651, RPL6, CHRM4, CD300A, KEL, RP11-664D7.4, MARCKSL1, TCTEX1D4, PROK2, LBH, NPDC1, CCR6, XLOC—022131, ENSCAFG00000028850, XLOC—068212, HIST1H, DLGAP3, MPO, CD151, XLOC—067564, NETO1, U2, XLOC—011971, GZMA, ENSCAFG00000013622, ENSCAFG00000029323, ENSCAFG00000029323, and XLOC—102336.
The markers that are up-regulated compared to control are as follows:
CSF1, MT1, ENSCAFG00000029651, RPL6, CHRM4, CD300A, KEL, RP11-664D7.4, MARCKSL1, TCTEX1D4, PROK2, LBH, NPDC1, CCR6, XLOC—022131, ENSCAFG00000028850, XLOC—068212, HIST1H, DLGAP3, MPO, CD151, XLOC—067564, NETO1, U2, XLOC—011971, GZMA, ENSCAFG00000013622, ENSCAFG00000029323, ENSCAFG00000029323, and XLOC—102336.
The markers that are down-regulated compared to control are as follows:
C1GALT1, EXTL1, B6F250, ENSCAFG00000030890, XLOC—088759, RGS13, KRT24, RP11-10N16.3, TNFAIP3, CD8A, Q95J95, XLOC—094643, SLC25A48, FGFR4, GPC3, NKG7, CXCR3, CD5L, PADI4, CXL10, GFRA2, SLC38A11, FABP4, PTPN22, ENSCAFG00000028509, U6, XLOC—044225, XLOC—100547, CCL22, CCDC168, TNIK, ENSCAFG00000030894, RGS10, HTR4, CNNM1, FBXO11, GRM5, SCARA5, OBSL1, RAB19, GZMK, ENSCAFG00000031494, TRBC2, TNFRSF21, ENSCAFG00000031437, GZMB, XLOC—091705, CHGA, H6BA90, GALNT13, CACNA1D, CD8B, XLOC—024761, EOMES, ZNF662, AFF2, COL6A6, HTRA1, LAD1, ENSCAFG00000029236, SCN2A, XLOC—077615, KIAA1456, CCL19, KIF5C, XLOC—026187, GPR27, ENSCAFG00000028940, ADAMTS2, CCL5, MAPK11, SMOC1, ABCA4, KIAA1598, KLRK1, LAT, FAM190A, ENSCAFG00000029467, PGBD5, and TBXA2R.
Expression data related to these markers are provided in
The invention therefore contemplates methods for identifying subjects at elevated risk of developing cancer based on aberrant expression levels of one or more of these genes compared to a control.
In some embodiments, the invention contemplates detection and/or use of chromosome 5 genes and non-chromosome 5 genes that are differentially expressed in tumors carrying the germ-line risk-associated allele(s) and tumors that do not carry the germ-line risk-associated allele(s). In some embodiments, the chromosome 5 genes and non-chromosome 5 genes are selected from TRPC6, C1GALT1, RPL6, PIK3R6, ENSCAFG00000029323, XLOC—011971, FGFR4, SCARA5, GFRA2, KIAA1377, ENSCAFG00000013622, CD5L, XLOC—102336, CXL10, SLC25A48, KRT24, ENSCAFG00000029323, RP11-10N16.3, HIST1H, or HS3ST3B1.
TRPC6, C1GALT1, PIK3R6, FGFR4, SCARA5, GFRA2, KIAA1377, CD5L, CXL10, SLC25A48, KRT24, and RP11-10N16.3 were found to be down-regulated in tumors carrying the germ-line allele(s) compared to a tumor that does not carry the germ-line allele(s).
RPL6, ENSCAFG00000029323, XLOC—011971, ENSCAFG00000013622, XLOC—102336, ENSCAFG00000029323, HIST1H, and HS3ST3B1 were found to be up-regulated in tumors carrying the germ-line allele(s) compared to a tumor that does not carry the germ-line allele(s). Expression data related to these markers are provided in
The invention is also based in part on the discovery of various somatic mutations present in tumors carrying the germ-line allele(s) as compared to tumors that do not carry the germ-line allele(s). Somatic mutations were identified by performing a genome-wide sequencing of tumor cells and normal cells from dogs with LSA. The markers demonstrating a mutation are TRAF3, FBXW7, DOK6, RARS, JPH3, LRRN3, MLL2, OGT, POU3F4, SETD2, CACNA1G, DSCAML1, MLL, ADD2, ARID1A, ARNT2, CAPN12, EED, ENSCAFG00000002808, ENSCAFG00000005301, ENSCAFG00000017000, ENSCAFG00000024393, ENSCAFG00000025839, ENSCAFG00000027866, L3MBTL2, LOC483566, MAPKBP1, NCAPH2, PPP6C, Q597P9_CANFA, SGIP1, XM—533169.2, XM—533289.2, XM—541386.2, XM—843895.1, and XM—844292.1.
The invention therefore provides methods for detecting the presence of a mutation in one or more of these genes and identifying a subject at elevated risk of developing cancer or having an as yet undiagnosed cancer based on the presence of such mutation(s).
In some instances, the invention provides methods for detecting the presence of a mutation in one or more of ADD2, ARID1A, ARNT2, CAPN12, EED, ENSCAFG00000002808, ENSCAFG00000005301, ENSCAFG00000017000, ENSCAFG00000024393, ENSCAFG00000025839, ENSCAFG00000027866, L3MBTL2, LOC483566, MAPKBP1, NCAPH2, PPP6C, Q597P9_CANFA, SGIP1, XM—533169.2, XM—533289.2, XM—541386.2, XM—843895.1, and XM—844292.1, and identifying a subject at elevated risk of developing cancer based on the presence of such mutation(s). The subject may be a canine subject or a human subject, although it is not so limited.
Table 3 lists the NCBI database accession numbers for several of these markers in the canine genome and in the human genome. In some instances, a human orthologue of the locus has not yet been identified. In those instances, the invention contemplates that the human orthologue possesses at least 60% homology, or at least 70% homology, or at least 75% homology to the canine sequence and the methods described herein can be based on an analysis of loci in the human genome that share these degrees of homology.
Methods of genetic analysis are known in the art. Examples of genetic analysis methods and commercially available tools are described below.
Affymetrix:
The Affymetrix SNP 6.0 array contains over 1.8 million SNP and copy number probes on a single array. The method utilizes at a simple restriction enzyme digestion of 250 ng of genomic DNA, followed by linker-ligation of a common adaptor sequence to every fragment, a tactic that allows multiple loci to be amplified using a single primer complementary to this adaptor. Standard PCR then amplifies a predictable size range of fragments, which converts the genomic DNA into a sample of reduced complexity as well as increases the concentration of the fragments that reside within this predicted size range. The target is fragmented, labeled with biotin, hybridized to microarrays, stained with streptavidin-phycoerythrin and scanned. To support this method, Affymetrix Fluidics Stations and integrated GS-3000 Scanners can be used.
Illumina Infinium:
Examples of commercially available Infinium array options include the 660W-Quad (>660,000 probes), the 1 MDuo (over 1 million probes), and the custom iSelect (up to 200,000 SNPs selected by user). Samples begin the process with a whole genome amplification step, then 200 ng is transferred to a plate to be denatured and neutralized, and finally plates are incubated overnight to amplify. After amplification the samples are enzymatically fragmented using end-point fragmentation. Precipitation and resuspension clean up the DNA before hybridization onto the chips. The fragmented, resuspended DNA samples are then dispensed onto the appropriate BeadChips and placed in the hybridization oven to incubate overnight. After hybridization the chips are washed and labeled nucleotides are added to extend the primers by one base. The chips are immediately stained and coated for protection before scanning. Scanning is done with one of the two Illumina iScan™ Readers, which use a laser to excite the fluorophore of the single-base extension product on the beads. The scanner records high-resolution images of the light emitted from the fluorophores. All plates and chips are barcoded and tracked with an internally derived laboratory information management system. The data from these images are analyzed to determine SNP genotypes using Illumina's BeadStudio. To support this process, Biomek F/X, three Tecan Freedom Evos, and two Tecan Genesis Workstation 150s can be used to automate all liquid handling steps throughout the sample and chip prep process.
Illumina BeadArray:
The Illumina Bead Lab system is a multiplexed array-based format. Illumina's BeadArray Technology is based on 3-micron silica beads that self-assemble in microwells on either of two substrates: fiber optic bundles or planar silica slides. When randomly assembled on one of these two substrates, the beads have a uniform spacing of ˜5.7 microns. Each bead is covered with hundreds of thousands of copies of a specific oligonucleotide that act as the capture sequences in one of Illumina's assays. BeadArray technology is utilized in Illumina's iScan System.
Sequenom:
During pre-PCR, either of two Packard Multiprobes is used to pool oligonucleotides, and a Tomtec Quadra 384 is used to transfer DNA. A Cartesian nanodispenser is used for small-volume transfer in pre-PCR, and another in post-PCR. Beckman Multimeks, equipped with either a 96-tip head or a 384-tip head, are used for more substantial liquid handling of mixes. Two Sequenom pin-tool are used to dispense nanoliter volumes of analytes onto target chips for detection by mass spectrometry. Sequenom Compact mass spectrometers can be used for genotype detection.
Methods of genome sequencing are known in the art. Examples of genome sequencing methods and commercially available tools are described below.
Illumina Sequencing:
89 GAIIx Sequencers are used for sequencing of samples. Library construction is supported with 6 Agilent Bravo plate-based automation, Stratagene MX3005p qPCR machines, Matrix 2-D barcode scanners on all automation decks and 2 Multimek Automated Pipettors for library normalization.
454 Sequencing:
Roche® 454 FLX-Titanium instruments are used for sequencing of samples. Library construction capacity is supported by Agilent Bravo automation deck, Biomek FX and Janus PCR normalization.
SOLiD Sequencing:
SOLiD v3.0 instruments are used for sequencing of samples. Sequencing set-up is supported by a Stratagene MX3005p qPCR machine and a Beckman SC Quanter for bead counting.
ABI Prism® 3730 XL Sequencing:
ABI Prism® 3730 XL machines are used for sequencing samples. Automated Sequencing reaction set-up is supported by 2 Multimek Automated Pipettors and 2 Deerac Fluidics—Equator systems. PCR is performed on 60 Thermo-Hybaid 384-well systems.
Ion Torrent:
Ion PGM™ or Ion Proton™ machines are used for sequencing samples. Ion library kits (Invitrogen) can be used to prepare samples for sequencing.
Other Technologies:
Examples of other commercially available platforms include Helicos Heliscope Single-Molecule Sequencer, Polonator G.007, and Raindance RDT 1000 Rainstorm.
The invention contemplates that elevated risk of developing certain cancers is associated with an altered expression pattern of one or more genes some but not all of which are located on chromosome 5 at or near the germ-line risk-associated alleles identified by the invention. The invention therefore contemplates methods that involve measuring the mRNA or protein levels for these genes and comparing such levels to control levels, including for example predetermined thresholds.
mRNA Assays
The art is familiar with various methods for analyzing mRNA levels. Examples of mRNA-based assays include but are not limited to oligonucleotide microarray assays, quantitative RT-PCR, Northern analysis, and multiplex bead-based assays.
Expression profiles of cells in a biological sample (e.g., blood or a tumor) can be carried out using an oligonucleotide microarray analysis. As an example, this analysis may be carried out using a commercially available oligonucleotide microarray or a custom designed oligonucleotide microarray comprising oligonucleotides for all or a subset of the germ-line markers described herein. The microarray may comprise any number of the germ-line markers, as the invention contemplates that elevated risk may be determined based on the analysis of single differentially expressed markers or a combination of differentially expressed markers. The markers may be those that are up-regulated in tumors carrying a risk allele (compared to a tumor that does not carry the risk allele), or those that are down-regulated in tumors carrying a risk allele (compared to a tumor that does not carry the risk allele), or a combination of these. The number of markers measured using the microarray therefore may be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or more markers selected from the group consisting of ANGPTL5, BIRC3, CD68, MYBBP1A, CHD3, CHRNB1, RANGRF, ZBTB4, and a locus comprising SEQ ID NO:1, and/or any other markers listed in Tables 12 and/or 13. It is to be understood that such arrays may however also comprise positive and/or negative control markers such as housekeeping genes that can be used to determine if the array has been degraded and/or if the sample has been degraded or contaminated. The art is familiar with the construction of oligonucleotide arrays.
Commercially available gene expression systems include Affymetrix GeneChip microarrays as well as all of Illumina standard expression arrays, including two GeneChip 450 Fluidics Stations and a GeneChip 3000 Scanner, Affymetrix High-Throughput Array (HTA) System composed of a GeneStation liquid handling robot and a GeneChip HT Scanner providing automated sample preparation, hybridization, and scanning for 96-well Affymetrix PEGarrays. These systems can be used in the cases of small or potentially degraded RNA samples. The invention also contemplates analyzing expression levels from fixed samples (as compared to freshly isolated samples). The fixed samples include formalin-fixed and/or paraffin-embedded samples. Such samples may be analyzed using the whole genome Illumina DASL assay. High-throughput gene expression profile analysis can also be achieved using bead-based solutions, such as Luminex systems.
Other mRNA detection and quantitation methods include multiplex detection assays known in the art, e.g., xMAP® bead capture and detection (Luminex Corp., Austin, Tex.).
Another exemplary method is a quantitative RT-PCR assay which may be carried out as follows: mRNA is extracted from cells in a biological sample (e.g., blood or a tumor) using the RNeasy kit (Qiagen). Total mRNA is used for subsequent reverse transcription using the SuperScript III First-Strand Synthesis SuperMix (Invitrogen) or the SuperScript VILO cDNA synthesis kit (Invitrogen). 5 μl of the RT reaction is used for quantitative PCR using SYBR Green PCR Master Mix and gene-specific primers, in triplicate, using an ABI 7300 Real Time PCR System.
mRNA detection binding partners include oligonucleotide or modified oligonucleotide (e.g. locked nucleic acid) probes that hybridize to a target mRNA. Probes may be designed using the sequences or sequence identifiers listed in Table 3 or using sequences associated with the provided Ensembl gene IDs. Methods for designing and producing oligonucleotide probes are well known in the art (see, e.g., U.S. Pat. No. 8,036,835; Rimour et al. GoArrays: highly dynamic and efficient microarray probe design. Bioinformatics (2005) 21 (7): 1094-1103; and Wernersson et al. Probe selection for DNA microarrays using OligoWiz. Nat Protoc. 2007; 2(11):2677-91).
The art is familiar with various methods for measuring protein levels. Protein levels may be measured using protein-based assays such as but not limited to immunoassays, Western blots, Western immunoblotting, multiplex bead-based assays, and assays involving aptamers (such as SOMAmer™ technology) and related affinity agents.
An exemplary immunoassay may be carried out as follows: A biological sample is applied to a substrate having bound to its surface marker-specific binding partners (i.e., immobilized marker-specific binding partners). The marker-specific binding partner (which may be referred to as a “capture ligand” because it functions to capture and immobilize the marker on the substrate) may be an antibody or an antigen-binding antibody fragment such as Fab, F(ab)2, Fv, single chain antibody, Fab and sFab fragment, F(ab′)2, Fd fragments, scFv, and dAb fragments, although it is not so limited. Other binding partners are described herein. Markers present in the biological sample bind to the capture ligands, and the substrate is washed to remove unbound material. The substrate is then exposed to soluble marker-specific binding partners (which may be identical to the binding partners used to immobilize the marker). The soluble marker-specific binding partners are allowed to bind to their respective markers immobilized on the substrate, and then unbound material is washed away. The substrate is then exposed to a detectable binding partner of the soluble marker-specific binding partner. In one embodiment, the soluble marker-specific binding partner is an antibody having some or all of its Fc domain. Its detectable binding partner may be an anti-Fc domain antibody. As will be appreciated by those in the art, if more than one marker is being detected, the assay may be configured so that the soluble marker-specific binding partners are all antibodies of the same isotype. In this way, a single detectable binding partner, such as an antibody specific for the common isotype, may be used to bind to all of the soluble marker-specific binding partners bound to the substrate.
It is to be understood that the substrate may comprise capture ligands for one or more markers, including two or more, three or more, four or more, five or more, etc. up to and including all nine of the markers provided by the invention.
Other examples of protein detection and quantitation methods include multiplexed immunoassays as described for example in U.S. Pat. Nos. 6,939,720 and 8,148,171, and published US Patent Application No. 2008/0255766, and protein microarrays as described for example in published US Patent Application No. 2009/0088329.
Protein detection binding partners include marker-specific binding partners. Marker-specific binding partners may be designed using the sequences or sequence identifiers listed in Table 3 or using sequences associated with the provided Ensembl gene IDs. In some embodiments, binding partners may be antibodies. As used herein, the term “antibody” refers to a protein that includes at least one immunoglobulin variable domain or immunoglobulin variable domain sequence. For example, an antibody can include a heavy (H) chain variable region (abbreviated herein as VH), and a light (L) chain variable region (abbreviated herein as VL). In another example, an antibody includes two heavy (H) chain variable regions and two light (L) chain variable regions. The term “antibody” encompasses antigen-binding fragments of antibodies (e.g., single chain antibodies, Fab and sFab fragments, F(ab′)2, Fd fragments, Fv fragments, scFv, and dAb fragments) as well as complete antibodies. Methods for making antibodies and antigen-binding fragments are well known in the art (see, e.g. Sambrook et al, “Molecular Cloning: A Laboratory Manual” (2nd Ed.), Cold Spring Harbor Laboratory Press (1989); Lewin, “Genes IV”, Oxford University Press, New York, (1990), and Roitt et al., “Immunology” (2nd Ed.), Gower Medical Publishing, London, New York (1989), WO2006/040153, WO2006/122786, and WO2003/002609).
Binding partners also include non-antibody proteins or peptides that bind to or interact with a target marker, e.g., through non-covalent bonding. For example, if the marker is a ligand, a binding partner may be a receptor for that ligand. In another example, if the marker is a receptor, a binding partner may be a ligand for that receptor. In yet another example, a binding partner may be a protein or peptide known to interact with a marker. Methods for producing proteins are well known in the art (see, e.g. Sambrook et al, “Molecular Cloning: A Laboratory Manual” (2nd Ed.), Cold Spring Harbor Laboratory Press (1989) and Lewin, “Genes IV”, Oxford University Press, New York, (1990)) and can be used to produce binding partners such as ligands or receptors.
Binding partners also include aptamers and other related affinity agents. Aptamers include oligonucleic acid or peptide molecules that bind to a specific target. Methods for producing aptamers to a target are known in the art (see, e.g., published US Patent Application No. 2009/0075834, U.S. Pat. Nos. 7,435,542, 7,807,351, and 7,239,742). Other examples of affinity agents include SOMAmer™ (Slow Off-rate Modified Aptamer, SomaLogic, Boulder, Colo.) modified nucleic acid-based protein binding reagents.
Binding partners also include any molecule capable of demonstrating selective binding to any one of the target markers disclosed herein, e.g., peptoids (see, e.g., Reyna J Simon et al., “Peptoids: a modular approach to drug discovery” Proceedings of the National Academy of Sciences USA, (1992), 89(20), 9367-9371; U.S. Pat. No. 5,811,387; and M. Muralidhar Reddy et al., Identification of candidate IgG biomarkers for Alzheimer's disease via combinatorial library screening. Cell 144, 132-142, Jan. 7, 2011).
Detectable binding partners may be directly or indirectly detectable. A directly detectable binding partner may be labeled with a detectable label such as a fluorophore. An indirectly detectable binding partner may be labeled with a moiety that acts upon (e.g., an enzyme or a catalytic domain) or is acted upon (e.g., a substrate) by another moiety in order to generate a detectable signal. These various methods and moieties for detectable labeling are known in the art.
Some of the methods provided herein involve measuring a level of a marker in a biological sample and then comparing that level to a control in order to identify a subject having an elevated risk of developing a cancer such as a hematological cancer. The control may be a control level that is a level of the same marker in a control tissue, control subject, or a population of control subjects.
The control may be (or may be derived from) a normal subject (or normal subjects). Normal subjects, as used herein, refer to subjects that are apparently healthy and show no tumor manifestation. The control population may therefore be a population of normal subjects.
In other instances, the control may be (or may be derived from) a subject (a) having a similar tumor to that of the subject being tested and (b) who is negative for the germ-line risk allele.
It is to be understood that the methods provided herein do not require that a control level be measured every time a subject is tested. Rather, it is contemplated that control levels of markers are obtained and recorded and that any test level is compared to such a predetermined level (or threshold).
The methods provided herein detect and sometimes measure (and thus analyze) levels or particular markers in biological samples. Biological samples, as used herein, refer to samples taken or derived from a subject. These samples may be tissue samples or they may be fluid samples (e.g., bodily fluid). Examples of biological fluid samples are whole blood, plasma, serum, urine, sputum, phlegm, saliva, tears, and other bodily fluids. In some embodiments, the biological sample is a whole blood sample, or a sample of white blood cells from a subject. In some embodiments, the biological sample is a tumor, a fragment of a tumor, or a tumor cell(s). The sample may be taken from the mouth of a subject using a swab or it may be obtained from other mucosal tissue in the subject.
Certain methods of the invention are intended for canine subjects, including for example golden retrievers. Other methods of the invention may be used in a variety of subjects including but not limited to humans and canine subjects.
Methods of computation analysis of genomic and expression data are known in the art. Examples of available computational programs are: Genome Analysis Toolkit (GATK, Broad Institute, Cambridge, Mass.), Expressionist Refiner module (Genedata AG, Basel, Switzerland), GeneChip-Robust Multichip Averaging (CG-RMA) algorithm, PLINK (Purcell et al, 2007), GCTA (Yang et al, 2011), the EIGENSTRAT method (Price et al 2006), EMMAX (Kang et al, 2010).
Other aspects of the invention relate to use of the diagnostic methods in connection with a breeding program. A breeding program is a planned, intentional breeding of a group of animals to reduce detrimental or undesirable traits and/or increase beneficial or desirable traits in offspring of the animals. Thus, a subject identified using the methods described herein as not having a risk marker of the invention may be included in a breeding program to reduce the risk of developing hematological cancer in the offspring of said subject. Alternatively, a subject identified using the methods described herein as having a risk marker of the invention may be excluded from a breeding program. In some embodiments, methods of the invention comprise exclusion of a subject identified as being at elevated risk of developing hematological cancer or having undiagnosed hematological cancer in a breeding program or inclusion of a subject identified as not being at elevated risk of developing hematological cancer or having undiagnosed hematological cancer in a breeding program.
Other aspects of the invention relate to diagnostic or prognostic methods that comprise a treatment step (also referred to as “theranostic” methods due to the inclusion of the treatment step). Any treatment for a hematological cancer, such as LSA or HSA, is contemplated herein. In some embodiments, treatment comprises one or more of surgery, chemotherapy, and radiation. Examples of chemotherapy for treatment of hematological cancers include rituximab, cyclophosphamide, doxorubicin, vincristine, and/or prednisone.
In some embodiments, a subject identified as being at elevated risk of developing hematological cancer or having undiagnosed hematological cancer is treated. In some embodiments, the method comprises selecting a subject for treatment on the basis of the presence of one or more risk markers as described herein. In some embodiments, the method comprises treating a subject with a hematological cancer characterized by the presence of one or more risk markers as defined herein.
Administration of a treatment may be accomplished by any method known in the art (see, e.g., Harrison's Principle of Internal Medicine, McGraw Hill Inc.). Administration may be local or systemic. Administration may be parenteral (e.g., intravenous, subcutaneous, or intradermal) or oral. Compositions for different routes of administration are well known in the art (see, e.g., Remington's Pharmaceutical Sciences by E. W. Martin). Dosage will depend on the subject and the route of administration. Dosage can be determined by the skilled artisan.
According to one aspect of the invention, isolated nucleic acid molecules are provided selected from the group consisting of: (a) nucleic acid molecules which hybridize under stringent conditions to a molecule consisting of a nucleic acid of SEQ ID NO: 1 or SEQ ID NO: 2, (b) deletions, additions and substitutions of (a), (c) nucleic acid molecules that differ from the nucleic acid molecules of (a) or (b) in codon sequence due to the degeneracy of the genetic code, and (d) complements of (a), (b) or (c). In some embodiments, the isolated nucleic acid molecule comprises SEQ ID NO: 1 or SEQ ID NO: 2. In some embodiments, the isolated nucleic acid molecule comprises SEQ ID NO:2.
The invention in another aspect provides an isolated nucleic acid molecule selected from the group consisting of (a) a unique fragment of nucleic acid molecule of SEQ ID NO:1 or SEQ ID NO: 2 (of sufficient length to represent a sequence unique within the canine genome) and (b) complements of (a).
In one embodiment, the sequence of contiguous nucleotides is selected from the group consisting of (1) at least two contiguous nucleotides nonidentical to the sequence group, (2) at least three contiguous nucleotides nonidentical to the sequence group, (3) at least four contiguous nucleotides nonidentical to the sequence group, (4) at least five contiguous nucleotides nonidentical to the sequence group, (5) at least six contiguous nucleotides nonidentical to the sequence group, (6) at least seven contiguous nucleotides nonidentical to the sequence group.
In another embodiment, the fragment has a size selected from the group consisting of at least: 8 nucleotides, 10 nucleotides, 12 nucleotides, 14 nucleotides, 16 nucleotides, 18 nucleotides, 20, nucleotides, 22 nucleotides, 24 nucleotides, 26 nucleotides, 28 nucleotides, 30 nucleotides, 40 nucleotides, 50 nucleotides, 75 nucleotides, 100 nucleotides, 200 nucleotides, 1000 nucleotides and every integer length there between.
According to another aspect, the invention provides expression vectors, and host cells transformed or transfected with such expression vectors, comprising the nucleic acid molecules described above.
Table 3 provides a list of the germ-line and somatic markers associated with elevated risk of tumors in canines. The canine Ensembl gene identifiers are based on the CanFam 2.0 genome assembly (see, e.g., Lindblad-Toh K, Wade C M, Mikkelsen T S, Karlsson E K, Jaffe D B, Kamal M, Clamp M, Chang J L, Kulbokas E J 3rd, Zody M C, et al.: Genome sequence, comparative analysis and haplotype structure of the domestic dog. Nature 2005, 438:803-819). The Ensembl gene ID provided for each gene can be used to determine the nucleotide sequence of the gene, as well as associated transcript and protein sequences, by inputting the Ensemble ID into the Ensemble database (Ensembl release 70).
To search for inherited (i.e. germ-line) risk factors predisposing to B-cell LSA or HSA, a Genome-Wide Association Study (GWAS) was performed on golden retrievers. DNA was extracted from whole blood samples taken from 43 B-cell LSA cases, 148 HSA cases, and 190 healthy controls>10 years of age were genotyped using the Illumina 170K canine HD array (CamFam2.0, Illumina, San Diego, Calif.).
Since the dog population contained high levels of encrypted relatedness and complex family structures, it was necessary to apply a method that could successfully control for the population stratification present in this data set (see
Chromosome 5 was identified with significant association to disease status (
Among the dogs with LSA, 88% of the dogs had at least one risk allele (19 dogs/45%+/+, 18 dogs/43%+/−, and 5 dogs/12%−/−. Among the dogs with HSA, 94% of the dogs have at least one risk allele (78 dogs/53%+/+, 60 dogs/41%−/−, and 10 dogs/7%−/−). Frequency of the risk alleles is shown in
Several genes were found to be located within the two disease-associated regions ch5:32.5 Mb˜33.1 Mb and ch5:36.6 Mb˜37.3 Mb. These genes include C11orf7, ANGPTL5, TRPC6, KIAA1377, NTN1, NTN3, STX8, WDR16, USP43, GLP2R, novel transcript chr:5:32732962-32766974 (SEQ ID NO:1) and DHRS7C (
First, whole genome sequencing was performed on tumor and normal DNA paired samples form 3 B-cell LSA dogs, and the coding exons of genes within the associated regions were examined to find germ-line mutations that associated with the risk haplotypes at the 32.9 and 36.8 Mb loci. Three genes, KIAA1377, ANGPTL5, and TRPC6, were found to have germ-line mutations within the coding sequence associated with the risk haplotypes and are summarized in Table 5.
Secondly, gene expression profiles of tumor samples from nine B-cell LSA dogs were generated using the Affymetrix canine expression array (Affymetrix, Santa Clara, Calif.). The nine dogs were divided into two groups (risk vs. non-risk) defined by the genotypes of the 32.9 Mb or the 36.8 Mb top SNPs. The risk group defined by the 32.9 Mb SNP contained 5 homozygotes for the risk allele (T/T), and non-risk group included 3 heterozygotes (T/C) and 1 homozygote for the non-risk allele (C/C). The risk group for the 36.8 Mb SNP included 4 heterozygotes (T/C) and 1 homozygote for the risk allele (T/T), and non-risk group included 4 homozygotes (C/C) for the non-risk allele. The expression data was analyzed by the methods described below to detect differentially expressed genes between the risk and non-risk groups.
Following hybridization on the array described above, each chip passed quality assurance and control procedures using the Affymetrix quality control algorithms provided in Expressionist Refiner module (Genedata AG, Basel, Switzerland). Probe signal levels were quantile-normalized and summarized using the GeneChip-Robust Multichip Averaging (CG-RMA) algorithm. Normalized files were imported into the Expressionist Analyst module for principal component analysis (PCA), unsupervised clustering, and to assess significant differences in gene expression. There are no precise tests to develop sample size estimates for gene expression profiling, theoretical principles and empirical observations were applied to support the sample size for these experiments a priori. The Power Atlas available online, provided an empirical estimate that the sample sets used for these experiments should have provided >90% power at p=0.05 to identify true positives, although the power to identify true negatives could have been be lower. The correlation coefficient (r2) for expression values of all probes between duplicated samples was >0.95. Probe IDs were mapped to corresponding canine Entrez Gene IDs using Affymetrix NetAffx EntrezGene Annotation. Prior to hierarchical clustering, normalized chip data were median-centered and log 2-transformed. Supervised groups included all of the tumors available for each defined genotype. Two group t-tests were done to determine genes that were differentially expressed between groups.
Data were confirmed by quantitative real time reverse transcriptase-polymerase chain reaction (qRT-PCR) using 11 samples where 4 had been included in the original set of 9 dogs on the array and 7 were independent of the samples from the array dataset. The genes that had a P value less than 0.05 and greater than two-fold difference between the two groups were considered significant. The genes located within 1 Mb from, or in between the two SNPs were considered differentially expressed if the P value was less than 0.05 regardless of fold-change.
This analysis identified eight candidate genes, ZBTB4, BIRC3, CD68, CHD3, CHRNB1, MYBBP1A, RANGRF, and ANGPTL5, located at or proximal to the disease associated regions, as differentially expressed (
These 141 genes are ABTB1, AGA, AK1, ANXA1, B4GALT3, BAG3, BAT1, BCAT2, BEX4, BID, BIRC3, BTBD9, CCDC134, CCDC18, CCDC88C, CD1C, CD320, CD68, CDKN1A, CMTM8, COASY, COL7A1, CPT1B, CTSD, DDX41, DENND4B, DGKA, DHRS1, DUSP6, ECM1, EFCAB3, EIF4B, LOC478066, FABP3, FADS1, FBXL6, FBXO11, FBXO33, FBXW7, FNBP4, GALNT6, GBE1, GDPD3, GNGT2, GPR137B, GSTM1, GTF2IRD2, GTF3C3, GUCY1B3, HBD, LOC609402, ICAM4, IRF5, KIF5C, KLHDC1, KLHDC9, LBX2, LOC475952, LOC479273, LOC479683, LOC482085, LOC482088, LOC482361, LOC482532, LOC482790, LOC483843, LOC484249, LOC484784, LOC485196, LOC487557, LOC487994, LOC490377, LOC490693, LOC491116, LOC609521, LOC610353, LOC610841, LOC611771, LOC612387, LOC612917, LOXL3, LZTS2, MED24, MFSD6, MTA3, MYC, MYO19, NAGA, NAPRT1, NEIL1, NQO2, OGT, OSGEPL1, OVGP1, P2RX5, PDLIM7, PER3, PHKA2, PIGV, PIK3R6, PITPNM1, PRKRA, PVRIG, RAB24, RAB25, RABEP2, RASAL1, RBM11, RBM18, RBM35A, RBPJ, REC8, RILPL1, RPA1, SIGLEC12, SLC37A1, SP1, SUOX, TIMM22, TIPARP, TLE4, TMED6, TMEM41B, TNFSF8, TRAF5, TRMT1, TTC39C, TTF1, TUBB2A, UNC93B1, YWHAE, ZFAND2A, ZFC3H1, ZMAT1, ZMYM1, ZNF215, ZNF292, ZNF331, ZNF513, ZNF608, ZNF674, and ZNF711.
Whole-genome sequencing of lymphomatic tumors as well as a matched normal blood sample from six dogs diagnosed with canine lymphoma was performed with the goal of identifying candidate somatic mutations, including SNPs, insertion/deletion events (indels), copy number variants (CNVs) and structural rearrangements. Four tumors were of B-cell type LSA and two were of T-cell type LSA.
For each sample, approximately 1 billion 101 base-pair paired-end reads were generated using Illumina HiSeq (Illumina, San Diego, Calif.), of which 98% were aligned to the CanFam2.0 reference genome with the Picard pipeline (Broad Institute, Cambridge, Mass.). The resulting depth of coverage per sample was roughly 40×. Genotype calls were made from the aligned reads using the Genome Analysis Toolkit's (GATK, Broad Institute, Cambridge, Mass.) UnifiedGenotyper in multi-sample mode. Hard filters were applied for low quality, strand bias, clustering, and excessive read depth. Accurate genotype calls were possible at 95% of the bases in the genome. Alignments were subsequently cleaned around novel insertion/deletion events in accordance with the GATK Best Practices guide.
Comparison of the SNP calls produced by the UnifiedGenotyper to GWAS data from the same samples showed greater than 99% concordance at common sites in 11 of the 12 samples. One tumor sample did not match its GWAS data, although the matched normal from the same sample did. Both the tumor sample and the matched normal sample were excluded from further analysis.
Somatic SNP variants were called with the MuTect software package v1.0.18339 (Broad Institute, Cambridge, Mass.) using standard practices and the pathology-based estimates of tumor purity provided in the table below. Somatic insertion/deletion variants were called using the GATK's SomaticIndelDetctor in ‘somatic’ mode. Results were filtered using standard practices. Both somatic SNP and indel variants were then annotated with the software package snpEff (“SNP effect predictor”; which is available at the snpeff website through sourceforge) using the CanFam2.61 (ENSEMBL version 61) gene annotation database.
Passing variants were segregated into two bins. Those variants observed in only a tumor sample were declared somatic candidate mutations. Those variants observed in both a tumor sample and its matched normal were declared germ-line variation. A second filter was then applied to the list of both germ-line variation and somatic candidate mutations such that the confidence of the reported genotype call exceeded that of finished sequence. The freely available software snpEff (“SNP effect predictor”; snpeff.sourceforge.net) was used to annotate the variants with snpEff's CanFam2.61 (ENSEMBL version 61) annotation database.
Several genes were identified with somatic mutations associated with LSA. These genes include TRAF3, FBXW7, DOK6, RARS, JPH3, LRRN3, MLL2, OGT, POU3F4, SETD2, CACNA1G, DSCAML1, MLL, ADD2, ARID1A, ARNT2, CAPN12, EED, ENSCAFG00000002808, ENSCAFG00000005301, ENSCAFG00000017000, ENSCAFG00000024393, ENSCAFG00000025839, ENSCAFG00000027866, L3MBTL2, LOC483566, MAPKBP1, NCAPH2, PPP6C, Q597P9_CANFA, SGIP1, XM—533169.2, XM—533289.2, XM—541386.2, XM—843895.1, and XM—844292.1. Several of these genes, TRAF3, FBXW7, DOK6, RARS, JPH3, LRRN3, MLL2, OGT, POU3F4, SETD2, CACNA1G, DSCAML1, MLL, were known previously to occur in human lymphoma or leukemia. These genes were found to have disease-associated somatic mutations in dogs and are summarized in Table 6.
The remaining genes, ADD2, ARID1A, ARNT2, CAPN12, EED, ENSCAFG00000002808, ENSCAFG00000005301, ENSCAFG00000017000, ENSCAFG00000024393, ENSCAFG00000025839, ENSCAFG00000027866, L3MBTL2, LOC483566, MAPKBP1, NCAPH2, PPP6C, Q597P9_CANFA, SGIP1, XM—533169.2, XM—533289.2, XM—541386.2, XM—843895.1, and XM—844292.1, have not been identified in association with human lymphoma or leukemia but were found to have somatic mutations associated with canine LSA (Table 7).
TGTGCATGGTTGGTCCAGATTTGGGGTGTACGTGACTACCACTGTTTGTGTTATACATGATCAGGAGAATGAACAG
GAAGAATATGGAATCCACTTAACACATGTGGTGTCGCCAGGAGGAGCATGATTCTGCCAGCATTGACTAGACTTAT
CTCTGATATGTTTTCAATCTGCAAAACTGAGAAGCTAGTAAATTTGTATCAATTTAGACTCTTTTCTTCAATGAGA
AGTGAGTTGAAG
GTGGGTGGATGGAGACCCTCTGTTTTCTTCTATCCATAAATTGCCAATATAGAAATTTATGCAT
TCCAGCTGCTGGAAATCTAAGAGCTGTCCAGAGGCAAGAAATACTTAAAACTTAGAGTGGTTAAG
GTAAGTAGATG
AATTTTGGAAAGGTGACTGGCTCCAGGGTGATGAACTAGTTGAAGGGAAGCCACACTCCTGTCAGAAGACCAGCTG
AAGTTACTGCAGTGGAGGAAGGTCAGAGCCCGTGGAGGCAG
(SEQ ID NO: 1)
TGTGCATGGTTGGTCCAGATTTGGGGTGTACGTGACTACCACTGTTTGTGTTATACATGATCA
GGAGAATGAACAGGAAGAATATGGAATCCACTTAACACATGTGGTGTCGCCAGGAGGAGCATG
ATTCTGCCAGCATTGACTAGACTTATCTCTGATATGTTTTCAATCTGCAAAACTGAGAAGCTA
GTAAATTTGTATCAATTTAGACTCTTTTCTTCAATGAGAAGTGAGTTGAAGAAGTGGCACTCA
GAAATTACCTTTGAGTTATCCAGCTGCTGGAAATCTAAGAGCTGTCCAGAGGCAAGAAATACT
TAAAACTTAGAGTGGTTAAGGTATGACAAGAACAGATGTGAATTTTGGAAAGGTGACTGGCTC
CAGGGTGATGAACTAGTTGAAGGGAAGCCACACTCCTGTCAGAAGACCAGCTGAAGTTACTGC
AGTGGAGGAAGGTCAGAGCCCGTGGAGGCAG
(SEQ ID NO: 2)
Dogs can serve as excellent model of human complex disease, as many canine diseases including cancer show similar clinical and molecular profiles to their human correlate. Dogs receive modern health care, have recorded family structures, and largely share the human environment. In addition, based on the recent breed creation, purebred dogs have megabase-sized haplotypes and linkage-disequilibrium (LD), allowing genome-wide association studies (GWAS) in dogs to be performed with 10-fold fewer SNPs than in humans (Lindblad-Toh, Wade et al. 2005). Power calculations and proof of principle studies have shown that 100-300 cases and 100-300 controls can suffice to map risk factors contributing a 2-5 fold increased risk (Lindblad-Toh, Wade et al. 2005).
Golden retrievers, one of the most popular family-owned dog breeds in the U.S., have a high prevalence of cancer with over 60% eventually dying from cancer (Glickman 2000). Two of the most common cancer types in golden retrievers are lymphoma and hemangiosarcoma with a lifetime risk of 13% and 20%, respectively (Glickman 2000). Canine lymphoma and hemangiosarcoma are clinically and histologically similar to human Non-Hodgkin Lymphoma (NHL) and visceral angiosarcoma, respectively (Priester 1976; Paoloni and Khanna 2007).
Approximately 50% of lymphoma in the golden retriever is of B-cell origin, within which the most common subtype is diffuse large B-cell lymphoma (DLBCL) (Modiano, Breen et al. 2005). In human adults, DLBCL and follicular lymphoma (FL) are the two most common subtypes of NHL, accounting for 60% of all B-cell NHL in North America (Anderson, Armitage et al. 1998
In contrast, while canine hemangiosarcoma is relatively common, angiosarcoma is rare in humans, accounting for 2-3% of adult sarcomas (Penel, Marreaud et al. 2011). The rarity of this disease in human limits the feasibility of genetic studies. Angiosarcoma is a very aggressive cancer in both species where the angiogenesis caused by the tumor is accompanied by highly invasive and metastatic nature.
As described herein, a GWAS of B-cell lymphoma and hemangiosarcoma in 373 golden retrievers from the U.S. was performed. The study revealed two major loci on chromosome 5 that together explain approximately half of the disease risk in this cohort. In addition, RNA sequence analysis of differential gene expression in B-cell lymphoma identified that the risk alleles of those 2 loci significantly altered expressions of genes that affect T-cell mediated immune responses.
To search for inherited risk factors predisposing to B-cell lymphoma in golden retrievers, GWAS was performed using the canineHD Illumina 170 k SNP array (Vaysse, Ratnakumar et al. 2011) (
An independent GWAS for hemangiosarcoma in golden retrievers identified significant association to 10 loci on six chromosomes (
Because both B-cell lymphoma and hemangiosarcoma showed association to the same region of chromosome 5, the datasets for the two diseases were combined. After quality control, the combined dataset included 187 cases (144 hemangiosarcoma cases and 43 B-cell lymphoma cases) and 186 controls, and 127,188 SNPs for the association analysis. The QQ plot deviated from the null distribution at 1×10−4, identifying 21 SNPs with p-values ranging from 3.5×10−7 to 8.0×10−5 to be significant (
Importantly, the two loci located 4 Mb apart constitute two independent association signals, rather than a single signal due to the long linkage disequilibrium (LD) in dogs. The top SNP in each region show high LD (r2>0.8) with SNPs in the same peak, but low LD (r2<0.2) to the associated SNPs in the other peak (
To define the exact risk haplotypes and their boundaries, an r2-based clumping analysis was performed by PLINK and Haploview (Purcell; Barrett, Fry et al. 2005; Purcell, Neale et al. 2007) methods) identifying risk and protective haplotypes in both loci. In the 32.9 Mb region two associated haplotype blocks were seen; a 19-SNP block (“32.9 Mb block1”) spanning 182 Kb, and a 4-SNP block (“32.9 Mb block2”) spanning 26 Kb (Table 10 and 11). In the 36.8 Mb region, a 12-SNP haplotype block (“36.8 Mb block1”) spanning 266 Kb was identified (Table 10 and 11).
#For the B-cell lymphoma analysis, haplotypes that are not significantly associated in this analysis are also listed for comparison purposes.
The risk haplotype at the 32.9 Mb locus had high frequency (
In contrast, the risk haplotype at the 36.8 Mb locus had a much lower frequency: 27.9% in dogs with B-cell lymphoma (9% homozygous, 37% are heterozygous) and 20.1% in dogs with hemangiosarcoma (3% homozygous, 33% heterozygous) as compared to 8.9% in controls (0% homozygous, 18% are heterozygous) (
To determine the proportion of disease risk explained by the genotypes of these two loci, a restricted maximum likelihood (REML) analysis was performed using GCTA software (Yang, Lee et al. 2011). Together, the two loci accounted for 52.5%±17.8% of the risk for canine B-cell lymphoma and hemangiosarcoma, suggesting that these risk loci are major drivers of disease in the golden retriever breed. The risk contributed by these two loci may be slightly higher for hemangiosarcoma (56%±20%) than for B-cell lymphoma (46%±32%).
To evaluate potential candidate genes within the regions of association, two approaches were taken. First the coding exons of genes within the most associated regions were examined for risk-haplotype-concordant amino acid changing germ-line mutations using ˜40× coverage of Illumina sequence. None of the genes near the 36.8 Mb locus (NTN1, NTN3, STX8, and WDR16) had any amino acid changes. At the 32.9 Mb locus, the three genes, KIAA1377, ANGPTL5 and TRPC6 each had one SNP leading to amino acid substitutions. However, none of these variants were associated with the risk haplotype.
Because no coding changes were evident, it was investigated whether the risk haplotypes were associated with transcriptional changes in tumors. 22 B-cell lymphoma samples were studied, from which RNA-Seq and correlated expression levels for protein-coding genes genome-wide with the risk haplotypes were obtained. For the 32.9 Mb locus, 12 samples that were homozygous for the risk allele were compared to the remaining 10 samples (8 heterozygous, 2 lacking the risk allele). The risk haplotype (both block 1 and 2) in homozygous state significantly altered the expression of TRPC6, the closest gene to block 2 (log FCrisk=−6.70, p-value=2.85×10-16, FDR=5.38×10-12, Table 4,
In addition, the expression levels of three nearby genes KIAA1377, ANGPTL5 and BIRC3 were also reduced significantly or near significantly by the risk haplotype, respectively (KIAA1377: log FCrisk=−2.25, p-value=1.98×10-6, FDR=0.004, ANGPTL5: log FCrisk=−2.60, p-value=1.30×10-4, FDR=0.06, and BIRC3: log FCrisk=−0.76, p-value=1.05×10-3, FDR=0.21, Table 4,
Across the genome 28 additional genes had significantly altered expression linked to the 32.9 Mb risk haplotypes (Table 12 and 13). All genes, except for PIK3R6, are located on other chromosomes, suggestive of trans-regulation or downstream effects triggered by the predisposing mutation(s). Pathway analysis using Ingenuity Pathway Analysis (IPA; Ingenuity Systems) did not cluster significantly in any particular canonical pathways, nor was any common downstream biological functions identified. Instead nine micro RNAs (miRNAs) were identified as the upstream regulator of the genes with the observed expression changes.
2.85 × 10−16
5.38 × 10−12
4.10 × 10−14
3.87 × 10−10
7.63 × 10−12
Expression comparison between the 36.8 Mb locus risk (1 homozygous and 5 heterozygous dogs) and non-risk (16 dogs lacking the risk haplotype) haplotypes identified 89 alternatively expressed genes elsewhere in the genome, with no sign of cis-regulation (Table 13). However, the IPA analysis of the 89 genes showed that the expression changes of many of those genes were linked to decrease in the activation of immune cells. The analysis strongly associated various immune modulators to the observed expression changes, including TCR (poverlap=6.24×10-10), cytokines; IL2 (poverlap=1.53×10-9), IL15 (poverlap=9.41×10-6), which is essential for activation and survival of T-cells and NK cells, and TNF (poverlap=5.83×10-5). Seven miRNAs, four of which have been associated with lymphoma/leukemia were also identified to be the regulators. Significant enrichment of the differentially expressed genes was also observed in four canonical pathways that play role in innate and adaptive immunity, and hematopoiesis. All of these changes are consistent with the role of STX8 in fusing lytic granules with the plasma membrane in CD8 T-cells for cytotoxic release.
2.85 × 10−16
5.38 × 10−12
4.10 × 10−14
3.87 × 10−10
7.63 × 10−12
Taken together the almost complete expression decrease of TRPC6 by the homozygous risk state at 32.9 Mb locus, and the expression changes correlated with at least one risk allele at the 36.8 Mb locus strongly suggest that the T-cell activation is severely compromised in the tumors of golden retrievers with B-cell lymphoma.
GWAS of human DLBCL in thousands of human patients have detected tens of loci which together account for <10% of the genetic risk. For human angiosarcoma no GWAS has been performed due to the rarity of the disease. As described herein, GWAS was performed for canine lymphoma and hemangiosarcoma using less than 400 dogs for both diseases combined, and two loci of strong effect accounting for approximately half of the disease risk were identified. The fact that one of the two risk factors on chromosome 5 (32 Mb) is very common in the U.S. golden retriever population may relate to the use of popular sires and may constitute an example of an allele accumulating either through drift or selective breeding for a nearby locus.
Incorporation of additional cases and controls in the future will likely identify additional risk factors, as several candidate loci fall just above or below the significance threshold. In this context it is noted that the 43 B-cell lymphoma cases alone produced a relatively weaker signal for the canine chromosome 5 locus at 32 Mb, suggesting that for this high frequency risk allele a higher sample number would be optimal as predicted by original power calculations that 100 cases and 100 controls are required for detection of a risk factor conferring a 5-fold increased risk.
In this study, the individual risk factors appear to contribute a risk of 1.5- to 7-fold depending on if a simple allelic risk or a context-dependent OR was calculated. The remarkable finding lies partly in the fact that only two loci contribute as much as 50% of the total risk.
The fact that both hemangiosarcoma and B-cell lymphoma predisposition map to the same loci is intriguing. Both diseases stem from the hematopoietic lineage, where hemangioblasts are proposed to have the ability to generate both lymphocytic stem cells and endothelial cells. Previous studies have also proposed that canine hemangiosarcoma carry hematopoietic stem cell markers consistent with an immature precursor for these cells.
Interestingly, the 32 Mb region contained two associated haplotypes, which could either constitute the same larger haplotype tagging a single mutation or each haplotype could be carrying different mutations. While in theory if there were two different mutations, these could be affecting the two different diseases differently, the multipotent function of the genes affected by the expression changes linked to the risk haplotype suggests that even a single mutation could easily have pleiotropic effects in different tissues.
RNA-Seq data from B-cell lymphoma tumors demonstrated an almost complete reduction of TRPC6 transcript associated with the 32 Mb risk haplotype. Intriguingly, TRPC6 is not normally expressed in B-cells (Roedding, Li et al. 2006), but has been reported to play a major role for T-cell and NK-cell activation (Tseng, Lin et al. 2004; Finney-Hayward, Popa et al. 2010; Carrillo, Hichami et al. 2012). Without wishing to be bound by any theory or mechanism, it is hypothesized that B-cell lymphoma is at least partially promoted by a lack of T-cell and/or NK-cell response in the lymph nodes. This hypothesis is further supported by the effect of the 36 Mb locus, where ˜90 genes involved in T-cell activation were suppressed by the risk allele (
The 32 Mb risk allele also correlated, although less strongly, with a reduced expression in other genes near the 32 Mb locus in B-cell lymphoma tumors. These genes include genes involved in both in B-cell anti-apoptotic signaling in human DLBCL (BIRC3) a potent hematopoietic stem cells growth factor (ANGPTL5) (Zhang, Kaba et al. 2008; Drake, Khoury et al. 2011; Khoury, Drake et al. 2011), a novel centrosomal protein with mitotic regulatory potential (KIAA1377) (Tipton, Wang et al. 2012) and could affect other aspects of tumor formation. The risk haplotype contained a potential lincRNA not previously reported, which hypothetically could be involved in regulating the expression of multiple genes in the region.
All golden retrievers in this study were privately owned pet dogs in the U.S., and participated in this study with owner consent. The diagnosis of B-cell lymphoma or hemangiosarcoma was confirmed by histology including immunohistochemistry and PCR based methods (Supplemental methods). Genomic DNA samples were isolated from whole blood and genotyped for 170,000 SNPs using the Illumina 170K canine HD array (Vaysse, Ratnakumar et al. 2011). To successfully control for the population stratification present in the dataset, analysis approach was taken based on a method described by Price et al. (Price, Zaitlen et al. 2010). The discovery of germ-line was performed by generating 40× whole genome sequencing by Illumina HiSeq of DNA samples from three dogs with that had B-cell lymphoma and had been included in the GWAS. The gene expression levels of twenty-two B-cell lymphoma tumors were profiled by strand-specific RNA-Seq, and analyzed for changes by the germ-line risk alleles at 32.9 Mb and 36.8 Mb loci on chromosome 5. To test if the genes affected by the germ-line risk alleles were enriched in particular biological pathways/biological functions, Ingenuity Pathway Analysis (IPA), proteins encoded in genomic regions associated with immune-mediated were used.
All references recited herein are incorporated by reference herein in their entirety. The definitions and disclosures provided herein govern and supersede all others incorporated by reference. Although the invention herein has been described in connection with preferred embodiments thereof, it will be appreciated by those skilled in the art that additions, modifications, substitutions, and deletions not specifically described may be made without departing from the spirit and scope of the invention as defined in the appended claims. It is therefore intended that the foregoing detailed description be regarded as illustrative rather than limiting, and that it be understood that it is the following claims, including all equivalents, that are intended to define the spirit and scope of this invention.
This application claims the benefit of U.S. Provisional Application No. 61/654,067, filed May 31, 2012, and U.S. Provisional Application No. 61/780,823, filed Mar. 13, 2013. The entire contents of each of these referenced provisional applications are incorporated by reference herein.
This invention was made with government support under RO1CA112211 and U54HG003067 awarded by the National Institutes of Health. The Government has certain rights in the invention.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US2013/043323 | 5/30/2013 | WO | 00 |
| Number | Date | Country | |
|---|---|---|---|
| 61654067 | May 2012 | US | |
| 61780823 | Mar 2013 | US |
