Patent Application
20150118681
The present invention relates to a method for detecting an unfavorable prognostic risk of renal cell carcinoma, the method comprising detecting a DNA methylation level. Moreover, the present invention relates to an oligonucleotide used in the method.
Renal cell carcinoma (RCC) often occurs in the working population at the maturity stage. While there are many case groups who are curable by nephrectomy, there are also apparently case groups who develop a distant metastasis rapidly. The two greatly differ in clinical course. Further, there is known a case for which an immunotherapy, molecularly targeted therapeutic drug, or the like is effective even if a metastasis occurs. Cases who are highly likely to have a recurrence should be subjected to a close follow-up observation to diagnose a recurrence at an early stage, and if an additional after-treatment is performed, there is a possibility that the prognosis can be improved. However, cases are experienced, who belong to histopathologically low grade and the most common histological type, clear cell RCC, and rapidly develop a distant metastasis. It is difficult to predict a prognosis utilizing existing clinicopathological parameters and the like.
It is well known that clear cell RCCs are characterized by inactivation of the VHL tumor-suppressor gene. Moreover, systematic resequencing and exome analysis of RCCs are performed by The Cancer Genome Atlas, The Cancer Genome Project, and other international efforts. Then, such efforts have revealed that renal carcinogenesis involves inactivation of histone-modifying genes, such as SETD2, a histone H3 lysine 36 methyltransferase; JARID1C (KDM5C), a histone H3 lysine 4 demethylase; UTX (KDM6A), a histone H3 lysine 27 demethylase; and PBRM1, a SWI/SNF chromatin remodeling complex (NPLs 1 to 3). Furthermore, non-synonymous mutations of the NF2 gene and truncating mutations of the MLL2 gene in RCC have also been reported (NPL1). However, such gene mutations cannot fully explain the aforementioned difference in RCC clinical course and the like (clinicopathological diversity).
Not only genetic but also epigenetic events are observed during carcinogenesis, and these two events reflect the clinicopathological diversity in various tissues in association with each other. In addition, DNA methylation alternation is believed to be one of major epigenetic changes in human cancers.
In fact, on the basis of the analyses of RCCs by methylation-specific PCR (MSP), COBRA (combined bisulfite restriction enzyme analysis), and bacterial artificial chromosome (BAC) array-based methylated CpG island amplification (BAMCA), the present inventors have demonstrated that a non-cancerous renal cortex tissue obtained from RCC patients is already at the precancerous stage associated with DNA methylation alterations (PLT 1 and NPLs 4 to 7). Further, the inventors have revealed by the genome-wide analysis using BAMCA that the DNA methylation alternation status in a non-cancerous renal cortex tissue at the precancerous stage is inherited by the corresponding RCC in the same patient, and successfully developed a method for predicting a prognosis of an RCC case (PLT 1 and NPL 6).
However, the technique of evaluating a DNA methylation status using BAMCA is complex. In addition, in predicting a prognosis of an RCC case using such BAMCA, the region of chromosomes that can be covered by BAC clones was quite limited at the time of the invention. Hence, a methylated CpG site having a truly high diagnostic ability has not been identified.
Moreover, regarding the DNA methylation in cancers, the existence of a cancer phenotype, CpG island methylator phenotype (CIMP), showing that DNA hypermethylation accumulates on CpG islands in a manner correlated with clinicopathological parameters of cases has been revealed in colorectal cancer, stomach cancer, and the like (NPLs 8 to 11).
Nevertheless, regarding renal cell carcinomas, it has been considered that an association between the CIMP-positive phenotype and renal cell carcinomas has not been revealed yet (NPL 12). In fact, on the basis of a finding that the distribution of the number of methylated CpGs in individual tumors was shown to differ from the expected Poisson distribution, a possibility has suggested that a subset of renal cell carcinomas exhibit CIMP. However, the existence of CIMP-positive renal cell carcinomas in kidneys has not been verified, and no distinct CpG site that could become a hallmark for CIMP has been identified (NPL 13).
From such circumstances, desired are methods capable of indicating, in renal cell carcinomas also, the existence of a phenotype (CIMP) showing that DNA methylation accumulates on CpG islands in a manner strongly correlated with clinicopathological RCC parameters, identifying a CpG site serving as a CIMP marker, and predicting a prognosis of RCC easily with quite high sensitivity and specificity. However, such methods are not put into practical use at present.
[PLT 1] Japanese Unexamined Patent Application Publication No. 2010-63413
[NPL 1] Dalgliesh, G. L. et al., Nature, 2010, vol. 463, pp. 360 to 363
[NPL 2] van Haaften, G. et al., Nat. Genet., 2009, vol. 41, pp. 521 to 523
[NPL 3] Varela, I. et al., Nature, 2011, vol. 469, pp. 539 to 542
[NPL 4] Arai, E. et al., Clin. Cancer Res., 2008, vol. 14, pp. 5531 to 5539
[NPL 5] Arai, E. et al., Int. J. Cancer, 2006, vol. 119, pp. 288 to 296
[NPL 6] Arai, E. et al., Carcinogenesis, 2009, vol. 3 0, pp. 214 to 221
[NPL 7] Arai, E. et al., Pathobiology, 2011, vol. 78, pp. 1 to 9
[NPL 8] Issa, J. P., Nat. Rev. Cancer, 2004, vol. 4, pp. 988 to 993
[NPL 9] Toyota, M. et al., Proc. Natl. Acad. Sci. USA, 1999, vol. 96, pp. 8681 to 8686
[NPL 10] Shen, L. et al., Proc. Natl. Acad. Sci. USA, 2007, vol. 104, pp. 18654 to 18659
[NPL 11] Toyota, M. et al., Cancer Res., 1999, vol. 5 9, pp. 5438 to 5442
[NPL 12] Morris, M. R. et al., Genome Med., 2010, 2 (9): 59
[NPL 13] McRonald, F. E. et al., Mol. Cancer, 2009, 8: 31
An object is to provide a method for determining an unfavorable prognostic risk of renal cell carcinoma easily with quite high sensitivity and specificity.
In order to achieve the above object, the present inventors have performed a methylome analysis using a single CpG resolution Infinium array on 29 normal renal cortex tissue (C) samples, and 107 non-cancerous renal cortex tissue (N) samples and 109 tumor tissue (T) samples obtained from patients with clear cell renal cell carcinomas (clear cell RCCs). The result revealed that the DNA methylation level of the N samples was already altered at 4830 CpG sites in comparison with the C samples. Further, DNA methylation alternations occurred in the N samples, and 801 CpG sites where the alternations were inherited by and strengthened in the T samples were identified. An unsupervised hierarchical clustering analysis was performed based on the DNA methylation levels at the 801 CpG sites. As a result, it was found out that renal cell carcinomas was grouped into Cluster A (n=90) and Cluster B (n=14). Then, it was found out that clinicopathologically aggressive tumors were accumulated in this Cluster B, and also that the cancer-free survival rate (recurrence-free survival rate) and overall survival rate of patients belonging to this Cluster B were significantly lower than those of patients belonging to Cluster A. Specifically, it was revealed that renal cell carcinomas belonging to Cluster B were characterized by accumulation of DNA hypermethylation on CpG islands and were CpG island methylator phenotype (CIMP)-positive cancers.
Further, it was also found out for the first time that DNA hypermethylations at CpG sites of FAM150A, GRM6, ZNF540, ZFP42, ZNF154, RIMS4, PCDHAC1, KHDRBS2, ASCL2, KCNQ1, PRAC, WNT3A, TRH, FAM78A, ZNF671, SLC13A5, and NKX6-2 genes were hallmarks of CIMP in renal cell carcinomas.
Note that none of the CpG sites of the 17 genes identified this time were included in the renal cell carcinoma-associated regions (70 BAC clones) having been identified as being effective in predicting a prognosis of renal cell carcinoma, by examining the presence or absence of the DNA methylation described in PLT 1 and NPL 6.
Moreover, it was also verified that it was possible to detect the hypermethylation status at the CpG sites of these 17 genes by methods other than the analysis using the Infinium array (a pyrosequencing method and a DNA methylation analysis method using a mass spectrometer). These have led to the completion of the present invention. More specifically, the present invention is as follows.
(a) a step of preparing a genomic DNA derived from a kidney tissue of a subject;
(b) a step of detecting a DNA methylation level of at least one CpG site of a gene selected from the gene group consisting of FAM150A, GRM6, ZNF540, ZFP42, ZNF154, RIMS4, PCDHAC1, KHDRBS2, ASCL2, KCNQ1, PRAC, WNT3A, TRH, FAM78A, ZNF671, SLC13A5, and NKX6-2 in the genomic DNA prepared in the step (a); and
(c) a step of determining whether or not the subject is classified into an unfavorable prognosis group according to the DNA methylation level detected in the step (b).
(a) an oligonucleotide that is a pair of primers designed to flank at least one CpG site of a gene selected from the gene group; and
(b) an oligonucleotide that is any one of a primer and a probe capable of hybridizing to a nucleotide comprising at least one CpG site of a gene selected from the gene group.
It is made possible to determine an unfavorable prognostic risk of renal cell carcinoma easily with quite high sensitivity and specificity.
The present invention provides a method for detecting an unfavorable prognostic risk of renal cell carcinoma, the method comprising the following steps (a) to (c):
(a) a step of preparing a genomic DNA derived from a kidney tissue of a subject;
(b) a step of detecting a DNA methylation level of at least one CpG site of a gene selected from the gene group consisting of FAM150A, GRM6, ZNF540, ZFP42, ZNF154, RIMS4, PCDHAC1, KHDRBS2, ASCL2, KCNQ1, PRAC, WNT3A, TRH, FAM78A, ZNF671, SLC13A5, and NKX6-2 in the genomic DNA prepared in the step (a); and
(c) a step of determining whether or not the subject is classified into an unfavorable prognosis group according to the DNA methylation level detected in the step (b).
In the present invention, the term “renal cell carcinoma” refers to a cancer originated from the renal tubular epithelial cells in the kidney. According to the pathological features, the cancer is classified into clear cell type, granular cell type, chromophobe type, spindle type, cyst-associated type, cyst-originating type, cystic type, or papillary type. Moreover, examples of the “subject” according to the present invention include patients who have been treated for renal cell carcinomas by nephrectomy or the like.
An example of the “unfavorable prognostic risk of renal cell carcinoma” according to the present invention includes a low survival rate in a prognosis (after nephrectomy or the like) of a subject. More specifically, the examples include a recurrence-free survival rate (cancer-free survival rate) of 50% or less after 500 days from the surgery as illustrated later in
In the present invention, the term “CpG site” means a site where cytosine (C) is linked to guanine (G) with a phosphodiester bond (p), and the term “DNA methylation” means a state where carbon at position 5 of cytosine is methylated at the CpG site. The term “DNA methylation level” means a ratio of the methylation at a particular CpG site to be detected, and can be expressed, for example, as a ratio of the number of methylated cytosines relative to the number of all cytosines (methylated cytosines and unmethylated cytosines) at a particular CpG site to be detected.
The “preparation of a genomic DNA derived from a kidney tissue” according to the present invention is not particularly limited. A known procedure such as a phenol-chloroform treatment method can be appropriately selected and used for the preparation.
Examples of a kidney tissue from which a genomic DNA is prepared by such a method include an intact kidney tissue sampled in nephrectomy or the like, a kidney tissue frozen after sampled in nephrectomy or the like, and a kidney tissue fixed in formalin and embedded in paraffin after sampled at the time of nephrectomy or the like. Among these kidney tissues, a frozen kidney tissue is desirably used from the viewpoints that degradation of a genomic DNA in the kidney tissue and the like are suppressed until the kidney tissue is subjected to the detection method of the present invention, and that a bisulfite treatment, PCR, and so on can be performed more efficiently in the step of detecting a DNA methylation level described later.
Additionally, as described in Examples later, the present inventors have revealed by an Infinium assay that it is possible to clearly distinguish between renal cell carcinomas of unfavorable prognosis (CIMP-positive renal cell carcinomas) and relatively favorable renal cell carcinomas by detecting DNA methylation levels of 18 CpG sites of 17 genes (FAM150A, GRM6, ZNF540, ZFP42, ZNF154, RIMS4, PCDHAC1, KHDRBS2, ASCL2, KCNQ1, PRAC, WNT3A, TRH, FAM78A, ZNF671, SLC13A5, and NKX6-2). Further, the inventors have revealed a DNA methylation analysis method using amass spectrometer that the hypermethylation status in the renal cell carcinomas of unfavorable prognosis continues in all regions of CpG islands comprising the CpG sites also.
Thus, the “CpG site” according to the present invention means CpG sites located at positions closer to at least one gene in the 17-gene group than to the other genes, and is preferably at least one CpG site within a CpG island located at the position closer to the gene than to the other genes, more preferably at least one CpG site located in promoter regions of the 17-gene group, and particularly preferably at least one CpG site at a position on a reference human genome sequence NCBI database Genome Build 37, the position being indicated by the chromosomal number and the position on the chromosome shown in Tables 1 to 4.
Moreover, in the present invention, typically, FAM150A is a gene encoding a protein specified under RefSeq ID: NP—997296, GRM6 is a gene encoding a protein specified under RefSeq ID: NP—000834, ZNF540 is a gene encoding a protein specified under RefSeq ID: NP—689819, ZFP42 is a gene encoding a protein specified under RefSeq ID: NP—777560, ZNF154 is a gene encoding a protein specified under RefSeq ID: NP—001078853, RIMS4 is a gene encoding a protein specified under RefSeq ID: NP—892015, PCDHAC1 is a gene encoding a protein specified under RefSeq ID: NP—061721, KHDRBS2 is a gene encoding a protein specified under RefSeq ID: NP—689901, ASCL2 is a gene encoding a protein specified under RefSeq ID: NP—005161, KCNQ1 is a gene encoding a protein specified under RefSeq ID: NP—000209, PRAC is a gene encoding a protein specified under RefSeq ID: NP—115767, WNT3A is a gene encoding a protein specified under RefSeq ID: NP—149122, TRH is a gene encoding a protein specified under RefSeq ID: NP—009048, FAM78A is a gene encoding a protein specified under RefSeq ID: NP—203745, ZNF671 is a gene encoding a protein specified under RefSeq ID: NP—079109, SLC13A5 is a gene encoding a protein specified under RefSeq ID: NP—808218, and NKX6-2 a gene encoding a protein specified under RefSeq ID: NP—796374.
In the present invention, the “method for detecting a DNA methylation level” may be any method capable of quantifying a DNA methylation level at a particular CpG site. A known method can be appropriately selected for the detection. Examples of such a known method include first to seventh methods described below.
The first method is a method based on the following principle. First, the genomic DNA is treated with bisulfite. Note that this bisulfite treatment converts unmethylated cytosine residues to uracil, but does not convert methylated cytosine residues (see Clark S J et al.,
Nucleic Acids Res, 1994, vol. 22, pp. 2990 to 7). Then, using the bisulfite-treated genomic DNA as a template, the full genome is amplified, enzymatically fragmented (normally fragmented into approximately 300 to 600 bp), and dissociated into single strands.
Moreover, in the first method, a probe is prepared which is capable of hybridizing to the genomic DNA converted by the bisulfite treatment, the base at the 3′ end of the probe being a base complementary to cytosine of the CpG site. Specifically, in a case where the CpG site is methylated, the base at the 3′ end of the probe is guanine; meanwhile, in a case where the CpG site is not methylated, the base at the 3′ end of the probe is adenine.
Then, two types of such probes differing from each other only in the base at the 3′ end complementary to the
CpG site are hybridized to the fragmented genomic DNA, and a single-base extension reaction is carried out in the presence of a fluorescence-labeled base. As a result, in the case where the CpG site is methylated, the fluorescence-labeled base is incorporated into the probe having guanine as the base at the 3′ end (probe for detecting methylation). On the other hand, in the case where the CpG site is not methylated, the fluorescence-labeled base is incorporated into the probe having adenine as the base at the 3′ end (probe for detecting unmethylation). Hence, the DNA methylation level can be calculated from an intensity of fluorescence emitted by the probe for detecting methylation and/or the probe for detecting unmethylation.
Further, as another embodiment of the first method, instead of the above-described probe for detecting methylation and probe for detecting unmethylation, a probe may be used which is capable of hybridizing to the genomic DNA converted by the bisulfite treatment, the base at the 3′ end of the probe being a base complementary to guanine of the CpG site. Then, the probe is hybridized to the fragmented genomic DNA, and a single-base extension reaction is carried out in the presence of guanine labeled with a fluorescent substance and/or adenine labeled with a fluorescent dye different from the fluorescent substance. As a result, in the case where the CpG site is methylated, the fluorescence-labeled guanine is incorporated into the probe. On the other hand, in the case where the CpG site is not methylated, the fluorescence-labeled adenine is incorporated into the probe. Hence, the DNA methylation level can be calculated from an intensity of fluorescence emitted by each fluorescent substance incorporated in the probe.
An example of the first method includes a bead array method (for example, Infinium(registered trademark) assay).
Furthermore, in the first method, the CpG site as the target of the DNA methylation level detection is preferably at least one CpG site located at a position on the reference human genome sequence NCBI database Genome Build 37, the position being selected from the group consisting of position 53,478,454 on chromosome 8, position 178,422,244 on chromosome 5, position 38,042,472 on chromosome 19, position 188,916,867 on chromosome 4, position 58,220,662 on chromosome 19, position 43,438,865 on chromosome 20, position 140,306,458 on chromosome 5, position 62,995,963 on chromosome 6, position 2,292,004 on chromosome 11, position 2,466,409 on chromosome 11, position 46,799,640 on chromosome 17, position 58,220,494 on chromosome 19, position 228,194,448 on chromosome 1, position 129,693,613 on chromosome 3, position 134,152,531 on chromosome 9, position 58,238,928 on chromosome 19, position 6,617,030 on chromosome 17, and position 134,599,860 on chromosome 10. Additionally, in the first method according to the present invention, it is preferable to detect the DNA methylation level at at least one site among the 18 CpG sites. Nevertheless, from the viewpoint that the sensitivity or specificity in detecting an unfavorable prognostic risk can be further improved, the target of the DNA methylation level detection is more preferably multiple CpG sites (for example, 2 sites, 5 sites, 10 sites, 15 sites), and the target of the DNA methylation level detection is particularly preferably all of the 18 CpG sites.
The second method is a method based on the following principle. First, the genomic DNA is treated with bisulfite. Then, using the bisulfite-treated genomic DNA as a template, a DNA comprising at least one of the CpG sites is amplified with a primer to which a T7 promoter is added. Subsequently, the resultant is transcribed into RNA, and a base-specific cleavage reaction is carried out with an RNAse. Thereafter, the cleavage reaction product is subjected to amass measurement with amass spectrometer.
After that, the mass of the methylated cytosine residues (the mass of cytosine) and the mass of the unmethylated cytosine residues (the mass of uracil), which are obtained by the mass measurement, are compared with each other to calculate the DNA methylation level at the CpG site.
An example of the second method includes a DNA methylation analysis method using amass spectrometer (for example, MassARRAY(registered trademark), see Jurinke C et al., Mutat Res, 2005, vol. 573, pp. 83 to 95).
Additionally, in the second method, the CpG site as the target of the DNA methylation level detection is preferably at least one CpG site contained in base sequences of SEQ ID NOs: 1 to 16. From the viewpoint that the sensitivity or specificity in detecting an unfavorable prognostic risk can be further improved, the CpG site is more preferably at least one CpG site among a CpG site group shown in Tables 5 to 8 below and having an area under the ROC curve (AUC) to be described later larger than 0.90, and further preferably at least one CpG site among a CpG site group having an AUC larger than 0.95 shown in Tables 5 to 8 below. The target of the DNA methylation level detection is particularly preferably all among the CpG site group having an AUC larger than 0.95.
Note that “chromosomal number” and “position on chromosome” shown in Tables 5 to 8 indicate a position on the reference human genome sequence NCBI database Genome Build 37. “Target gene name_primer set name_CpG site” indicates the order of CpG sites in PCR products amplified using primer sets shown in Tables 17 and 18 in a DNA methylation analysis using a mass spectrometer to be described later (Example 5). As to “AUC value”, “cutoff value”, “specificity”, “sensitivity”, and “1-specificity”, see Example 5 described later.
The third method is a method based on the following principle. First, the genomic DNA is treated with bisulfite. Note that this bisulfite treatment converts unmethylated cytosine residues to uracil, but uracil is expressed as thymine in the following extension reaction (sequence reaction). Then, using the bisulfite-treated genomic DNA as a template, a DNA comprising at least one of the CpG sites is amplified. Subsequently, the amplified DNAs are dissociated into single strands. Thereafter, only one of the dissociated single stranded DNAs is separated. After that, the extension reaction is performed on each base from one near the base at the CpG site, pyrophosphoric acid generated during this is caused to enzymatically emit light, and the intensity of the luminescence is measured. The intensity of luminescence from the methylated cytosine residue (luminescence intensity of cytosine) and the intensity of luminescence from the unmethylated cytosine residue (luminescence intensity of thymine) thus obtained are compared with each other to calculate the DNA methylation level (%) at the CpG site, for example, according to the following formula. DNA methylation level (%)=luminescence intensity of cytosinex100/(luminescence intensity of cytosine+luminescence intensity of thymine).
Examples of the third method include a pyrosequencing method (registered trademark, Pyrosequencing) (see Anal. Biochem. (2000) 10: 103-110) and the like.
The fourth method is a method based on the following principle. First, the genomic DNA is treated with bisulfite. Next, in a reaction system containing an intercalator which emits fluorescence when inserted between DNA double strands, a nucleotide comprising at least one of the CpG sites is amplified using the bisulfite-treated genomic DNA as a template. Then, the temperature of the reaction system is changed to detect a variation in the intensity of fluorescence emitted by the intercalator. A melting curve of the nucleotide comprising at least one of the CpG sites is compared with a melting curve of an amplification product obtained by using methylated/unmethylated control specimens as templates to then calculate the DNA methylation level at the CpG site.
An example of the fourth method includes a methylation-sensitive high resolution melting analysis (MS-HRM, see Wojdacz T K et al., Nat Protoc., 2008, vol. 3, pp. 1903 to 8).
The fifth method is a method based on the following principle. First, the genomic DNA is treated with bisulfite. Next, prepared are a primer set capable of amplification in the case where the CpG site is methylated, and a primer set capable of amplification in the case where the CpG site is not methylated. Then, using the bisulfite-treated genomic DNA as a template and these primer set, a nucleotide comprising at least one of the CpG sites is amplified. Subsequently, amounts of the obtained amplification products, that is, the amount of the amplification product specific to the methylated CpG site and the amount of the amplification product specific to the unmethylated CpG site, are compared with each other to calculate the DNA methylation level at the CpG site.
Further, as another embodiment of the fifth method, first, the genomic DNA is treated with bisulfite. Next, an oligonucleotide probe is prepared which has a nucleotide capable of hybridizing in the case where the CpG site is methylated, and which is labeled with a reporter fluorescent dye and a quencher fluorescent dye. In addition, an oligonucleotide probe is prepared which has a nucleotide capable of hybridizing in the case where the CpG site is not methylated, and which is labeled with a quencher fluorescent dye and a reporter fluorescent dye different from the aforementioned reporter fluorescent dye. Then, the oligonucleotide probes are hybridized to the bisulfite-treated genomic DNA. Further, using as a template the genomic DNA with the oligonucleotide probes hybridized thereto, a nucleotide comprising the CpG site is amplified. Subsequently, fluorescences emitted by the reporter fluorescent dyes through degradation of the oligonucleotide probes associated with the amplification are detected. The intensity of the fluorescence emitted by the reporter fluorescent dye specific to the methylated cytosine CpG site and the intensity of the fluorescence emitted by the reporter fluorescent dye specific to the unmethylated cytosine CpG site thus detected are compared with each other to calculate the DNA methylation level at the CpG site.
Examples of the fifth method include methylation-specific quantitative PCR (methylation-specific polymerase chain reaction (MS-PCR) using real-time quantitative PCR) such as MethyLight assay using TaqMan probe (registered trademark).
The sixth method is a method based on the following principle. First, the genomic DNA is treated with bisulfite. Next, using as a template a nucleotide comprising the bisulfite-converted CpG site, a sequencing reaction is performed directly. Then, the fluorescence intensities of the determined base sequence, that is, the fluorescence intensity from the methylated cytosine residue (fluorescence intensity of cytosine) and the fluorescence intensity from of the unmethylated cytosine residue (fluorescence intensity of thymine) are compared with each other to calculate the DNA methylation level at the CpG site.
Further, as another embodiment of the sixth method, first, the genomic DNA is treated with bisulfite. Then, a nucleotide comprising the bisulfite-converted CpG site is cloned by a PCR reaction or the like. Subsequently, the base sequence of each of multiple cloned products thus obtained is determined. The number of cloned products having a base sequence specific to the methylated cytosine CpG site and the number of cloned products having a base sequence specific to the unmethylated cytosine CpG site are compared with each other to thereby calculate the DNA methylation level at the CpG site.
Examples of the sixth method include bisulfite direct sequencing and bisulfite cloning sequencing (see Kristensen L S et al., Clin Chem, 2009, vol. 55, pp. 1471 to 83).
The seventh method is a method based on the following principle. First, the genomic DNA is treated with bisulfite. Then, using as a template a nucleotide comprising the bisulfite-converted CpG site, a region comprising the CpG site is amplified by PCR. Subsequently, the amplified DNA fragments are treated with a restriction enzyme capable of recognizing sites differing in sequence from each other in the cases where the CpG site is and is not methylated. Thereafter, band intensities of restriction enzyme fragments from the methylated CpG site and restriction enzyme fragments from the unmethylated CpG site, which are fractionated by electrophoresis, are quantitatively analyzed, so that the DNA methylation level at the CpG site can be calculated.
An example of the seventh method includes COBRA (combined bisulfite restriction enzyme analysis).
Although the methods that can be suitably used as the “method for detecting a DNA methylation level” of the present invention have been described above, the present invention is not limited thereto. Moreover, as described above, the genomic DNA prepared from a subject is further treated with bisulfite in detecting the DNA methylation level. Thus, the method for detecting an unfavorable prognostic risk of renal cell carcinoma of the present invention may be a method, wherein the step (b) is a step of treating the genomic DNA prepared in the step (a) with bisulfite and detecting a DNA methylation level of the CpG site.
Those skilled in the art can set an indicator for determining whether or not the subject is classified into an unfavorable prognosis group according to the DNA methylation level detected in the step (b) in the present invention, as appropriate in accordance with the method for detecting a DNA methylation level. For example, as described in Examples later, a receiver operating characteristic (ROC) analysis is performed on each CpG site to obtain the sensitivity (positive rate) and specificity. Further, a DNA methylation level at which the sum of the sensitivity and the specificity is the maximum can be set as the indicator (cutoff value, diagnostic threshold). If a detected DNA methylation level is higher than the cutoff value, the subject can be classified into the unfavorable prognosis group.
Moreover, in the present invention, from the viewpoint that the sensitivity or specificity in detecting an unfavorable prognostic risk of renal cell carcinoma can be further improved, not only a DNA methylation level but also the number of CpG sites exhibiting a value higher than the cutoff value may be used as an indicator for determining whether or not the subject is classified into the unfavorable prognosis group. For example, as described in Examples later, if the number of sites satisfying the cutoff value is 15 or more among 23 CpG units according to the present invention, the subject may be classified into the unfavorable prognosis group (see
In this manner, the present invention makes it possible to judge an unfavorable prognostic risk of renal cell carcinoma after nephrectomy, which cannot be detected by the existing classification criteria of histological observation and the like. Although nephrectomy is the first choice as a method for treating renal cell carcinoma, if metastasis/recurrence can be discovered at an early stage, an immunotherapy, molecularly-targeted therapeutic drug, or the like can be expected to be effective against the metastasis/recurrence.
Thus, the present invention can also provide a method for treating renal cell carcinoma, the method comprising: a step of administering a molecularly targeted therapeutic drug to the subject classified into the unfavorable prognosis group by the method of the present invention and/or a step of conducting an immunotherapy of the subject.
Further, in the present invention, patients classified into the unfavorable prognosis group among a large number of renal cell carcinoma cases subjected to nephrectomy are subjected to more intensive metastasis/recurrence screening. In this event, it is expected that discovering at an early stage can improve the clinical outcome; on the other hand, for patients not classified into the unfavorable prognosis group, the load of the metastasis/recurrence screening can be reduced.
The present invention provides an oligonucleotide according to any one of the following (a) and (b), which have a length of at least 12 bases, for use in the method for detecting an unfavorable prognostic risk of renal cell carcinoma:
(a) an oligonucleotide that is a pair of primers designed to flank at least one site selected from the CpG site group; and
(b) an oligonucleotide that is any one of a primer and a probe capable of hybridizing to a nucleotide comprising at least one site selected from the CpG site group.
Examples of the pair of primers according to (a) designed to flank at least one site selected from the CpG site group include primers (polymerase chain reaction (PCR) primers (forward primer and reverse primer)) capable of amplifying a DNA comprising at least one site selected from the bisulfite-converted CpG site group. The primers are primers capable of hybridizing to each bisulfite-converted nucleotide on both sides of at least one site selected from the CpG site group.
In addition, an example of the primer according to (b) capable of hybridizing to the nucleotide comprising at least one site selected from the CpG site group includes a primer (sequencing primer) capable of performing an extension reaction on each base from one near the base at the bisulfite-converted CpG site. Further, an example of the probe according to (b) capable of hybridizing to the nucleotide comprising at least one site selected from the CpG site group includes a probe (so-called TaqMan probe) capable of hybridizing to the nucleotide comprising the bisulfite-converted CpG site.
Furthermore, the oligonucleotide of the present invention has a length of at least 12 bases, but preferably at least 15 bases, more preferably at least 20 bases.
The oligonucleotide capable of hybridizing to the particular nucleotide has a base sequence complementary to the particular nucleotide, but the base sequent does not have to be completely complementary as long as the oligonucleotide hybridizes. Those skilled in the art can design the sequences of these oligonucleotides as appropriate on the basis of the base sequence comprising the CpG site either bisulfite-converted or not converted, by a known procedure, for example, as described in Examples later, using MassARRAY primer design software EpiDesigner (http://www.epidesigner.com, manufactured by SEQUENOM, Inc.), pyrosequencing assay design software ver. 1.0 (manufactured by QIAGEN N.V.), or the like. Additionally, the phrase “comprising the CpG site” according to the present invention and similar phrases may mean not only containing all of the CpG site, that is, both of cytosine and guanine, but also containing a part thereof (cytosine, guanine, or uracil or thymine after unmethylated cytosine is converted with bisulfite).
The oligonucleotide of the present invention is preferably a primer selected from the group consisting of base sequences of SEQ ID NOs: 17 to 48 in a DNA methylation analysis method using a mass spectrometer as described in Examples later (see Tables 17 and 18). In addition, in pyrosequencing as described in Examples later, the oligonucleotide of the present invention is preferably a primer selected from the group consisting of base sequences of SEQ ID NOs: 49 to 57 (see Table 9).
Furthermore, the present invention can also provide a kit for use in the method for detecting an unfavorable prognostic risk of renal cell carcinoma, the kit comprising the oligonucleotide.
In a preparation of the oligonucleotide, the oligonucleotide may be fixed if necessary. For example, in the case of detection by an Infinium assay, a probe fixed to beads can be used. Moreover, the oligonucleotide may be labeled if necessary. For example, a biotin-labeled primer may be used in the case of detection by a pyrosequencing method, and a probe labeled with a reporter fluorescent dye and a quencher fluorescent dye may be used in the case of detection by a TaqMan probe method.
The kit of the present invention can comprise a preparation other than the preparation of the oligonucleotide. Such a preparation includes reagents required for bisulfate conversion (for example, a solution of sodium bisulfite and the like), reagents required for PCR reaction (for example, deoxyribonucleotides, thermostable DNA polymerases, and the like), reagents required for Infinium assay (for example, nucleotides labeled with a fluorescent substance), reagents required for MassARRAY (for example, RNAses for base-specific cleavage reaction), reagents required for pyrosequencing (for example, ATP-sulfurylase, adenosine-5′-phosphosulfate, luciferases, and luciferins for detection of pyrophosphoric acid; streptavidin for separation of single stranded DNAs; and the like), reagents required for MS-HRM (for example, intercalators which emit fluorescence when inserted between DNA double strands, and the like). Moreover, the examples include reagents required for detection of the labels (for example, substrates and enzymes, positive controls and negative controls, buffer solutions used for dilution or washing of samples (genomic DNA derived from kidney tissues of subjects, and the like), or the like). The kit may further comprise an instruction thereof.
Hereinafter, the present invention will be more specifically described on the basis of Examples. However, the present invention is not limited to the following Examples. Note that the samples and methods used in Examples are as follows.
<Patients and Tissue Samples>
From materials surgically resected from 110 patients with primary clear cell renal cell carcinomas, 109 tumor tissue (T) samples and corresponding 107 non-cancerous renal cortex tissue (N) samples were obtained. The N samples showed no remarkable histological changes.
Note that these patients did not receive preoperative treatment but underwent nephrectomy at the National Cancer Center Hospital, Tokyo, Japan. The patients included 79 men and 31 women with a mean age of 62.8±10.3 (mean±standard deviation, 36 to 85 years old).
Moreover, histological diagnosis was made on the samples in accordance with the WHO classification (see Eble, J. N. et al., “Renal cell carcinoma. WHO classification of tumours. Pathology and genetics. Tumours of the urinary system and male genital organs”, 2004, IARC Press, Lyon, pp. 10 to 43,
Further, the histological grade of all the tumors was evaluated in accordance with the criteria described in “Fuhrman, S. A. et al., Am. J. Surg. Pathol., 1982, vol. 6, pp. 655 to 663” and classified according to the TNM classification in “Sabin, L. H. et al., International Union Against Cancer (UICC), TNM Classification Of Malignant Tumors, 6th edition, 2002, Wiley-Liss, New York, pp. 193 to 195”.
In addition, the criteria for macroscopic configuration of renal cell carcinoma followed the criteria established for hepatocellular carcinoma (HCC) (see NPLs 4 to 6). Note that type 3 (contiguous multinodular type) HCCs show poorer histological differentiation and a higher incidence of intrahepatic metastasis than type 1 (single nodular type) and type 2 (single nodular type with extranodular growth) HCCs (see Kanai, T. et al., Cancer, 1987, vol. 60, pp. 810 to 819).
The presence or absence of vascular involvement was examined microscopically on slides stained with hematoxylin-eosin and elastica van Gieson.
The presence or absence of tumor thrombi in the main trunk of the renal vein was examined macroscopically. Note that renal cell carcinoma is usually enclosed by a fibrous capsule and well demarcated. Moreover, renal cell carcinoma hardly ever contains fibrous stroma between cancer cells. Hence, cancer cells were successfully obtained from the surgical specimens, avoiding contamination with both non-cancerous epithelial cells and stromal cells.
Furthermore, for comparison with the RCC patients, 29 samples of normal renal cortex tissues (C1 to C29) were obtained from materials that had been surgically resected from 29 patients without any primary renal tumor. The patients without any primary renal tumor from whom the samples were obtained included 18 men and 11 women with a mean age of 61.4 ±10.8 (mean ±standard deviation, 31 to 81 years old). Additionally, 22 of these patients were patients who had undergone nephroureterectomy for urothelial carcinomas of the renal pelvis and ureter, while 6 patients had undergone nephrectomy with resection of retroperitoneal sarcoma around the kidney. The remaining one patient had undergone paraaortic lymph node dissection for metastatic germ cell tumor, which resulted in simultaneous nephrectomy because it was difficult to preserve the renal artery.
All the patients included in this study provided written informed consent. In addition, the study was conducted with the approval of the Ethics Committee of the National Cancer Center, Tokyo, Japan.
<Infinium Assay>
High-molecular-weight DNA from fresh frozen tissue samples obtained from the patients was extracted by treatment with phenol-chloroform, followed by dialysis (see Sambrook, J. et al., Molecular Cloning: A Laboratory Manual. Third Edition, Cold Spring Harbor Laboratory Press, NY, pp. 6.14 to 6.15).
Then, 500-ng aliquots of the DNA were subjected to bisulfite conversion using an EZ DNA Methylation-Gold™ kit (manufactured by Zymo Research Corporation).
Subsequently, DNA methylation status at 27578 CpG sites was analyzed at single-CpG resolution using the Infinium HumanMethylation27 Bead Array (manufactured by Illumina, Inc.). This array contains CpG sites located within the proximal promoter regions of the transcription start sites of 14475 genes (consensus coding sequences) registered in the NCBI database. Moreover, on average, two sites were selected per gene, and furthermore, 3 to 20 CpG sites were selected per gene for 200 or more cancer-related and imprinted genes, and employed for the array. In addition, 40 control probes were employed for each array. These control probes included staining, hybridization, extension, and bisulfate conversion controls, as well as negative controls.
Note that an Evo robot (manufactured by Tecan Group Ltd.) was used for automated processing of the bisulfite-converted DNA. Moreover, whole-genome amplification was performed using the Infinium Assay Kit (manufactured by Illumina, Inc.) (see Bibikova, M. et al., Epigenomics, 2009, vol. 1, pp. 177 to 200).
Then, after hybridization between the DNA fragments thus amplified and the probes on the array, the specifically hybridized DNA was fluorescence-labeled by a single-base extension reaction. Subsequently, the DNA was detected using a BeadScan reader (manufactured by Illumina, Inc.) in accordance with the manufacturer's protocol. The obtained data were analyzed using Genome Studio methyl at ion software (manufactured by Illumina, Inc.).
Note that, at each CpG site, the ratio of the fluorescent signal was measured using a relative ratio of a methylated probe to the sum of the methylated and unmethylated probes. Specifically, the so-called β value (range: 0.00 to 1.00) reflects the methylation level at an individual CpG site.
<Statistical Analysis>
In the Infinium assay, the call proportions (P-values for detection of signals above the background <0.01) for 32 probes in all of the tissue samples analyzed were 90% or less. Since such a low call proportion may be attributable to polymorphism at the probe CpG sites, these 32 probes were excluded from the present assay. In addition, all CpG sites on chromosomes X and Y were excluded, to avoid any gender-specific methylation bias. As a result, 26454 CpG sites on the autosomal chromosomes were left as a final analysis target.
Infinium probes showing significant differences in DNA methylation levels between the 29 C samples and 107 N samples were identified by a logistic model.
Probes on which DNA methylation levels showed ordered differences from C to N and then to T samples were identified by the cumulative logit model using the 29 C, 107 N, and 109 T samples.
Differences of DNA methylation status between 104 paired samples of N and corresponding T derived from a single patient were examined by the Wilcoxon signed-rank test.
A false discovery rate (FDR) of q=0.01 was considered significant.
Unsupervised hierarchical clustering (Euclidean distance, Ward method) based on DNA methylation levels (ΔβT-N) was performed inpatients with clear cell renal cell carcinomas.
Correlations between clusters of patients and clinicopathological parameters were examined by Wilcoxon rank sum test and Fisher's exact test.
Survival curves of patients belonging to each cluster were calculated by the Kaplan-Meier method. Then, the differences were compared by the Log-rank test.
The number of Infinium assay probes showing DNA hypermethylation or DNA hypomethylation in each cluster and the average DNA methylation level (ΔβT-N) of each cluster were examined using Wilcoxon rank sum test at a significance level of P<0.05.
The CpG sites discriminating the clusters were identified by Fisher's exact test and random forest analysis (see Breiman, L., Mach. Learn., 2001, vol. 45, pp. 5 to 32).
<DNA Methylation Alternations during Renal Carcinogenesis>
First, representative CpG sites found based on the Infinium assay were verified by performing a pyrosequencing method under conditions shown in Table 9. As a result, as shown in
This confirmed that the data on the present Infinium assay were highly reliable.
Precancerous conditions in the kidney have been rarely discussed. Nevertheless, the present inventors have suggested that non-cancerous tissues are already at precancerous stages from the viewpoint of altered DNA methylation, despite the absence of any remarkable histological changes and the lack of association with chronic inflammation and persistent infection with viruses or other pathogenic microorganisms (PLT 1 and NPLs 4 to 7).
In this regard, the result of the present Infinium assay was analyzed by the logistic model. The result revealed that the DNA methylation levels on 4830 probes were already altered in the N samples compared to those in the C samples (FDR, q=0.01, see (a) in Table 10).
Further, in order to reveal the DNA methylation alternations inherited by renal cell carcinomas themselves, probes on which DNA methylation levels showed ordered differences from C to N and then to T samples were identified by the cumulative logit model. As a result, such ordered differences of DNA methylation level were observed on 11089 probes (FDR, q=0.01, see (b) in Table 10).
Furthermore, in order to reveal the cancer-prone DNA methylation alternations, 104 paired samples of N and T were examined by the Wilcoxon signed-rank test. As a result, significant differences between the N samples and the corresponding renal cell carcinomas were observed on 10870 probes (FDR, q=0.01, see (c) in Table 10).
The above result revealed that although DNA hypomethylation was also observed during progression to established cancer, DNA hypermethylation frequently occurred at the very early stages of renal carcinogenesis.
Moreover, 801 probes were identified which satisfied all of the criteria shown in (a), (b), and (c) in Table 10; in other words, the DNA methylation alterations thereon were already evident at the non-cancerous stages, and also these alterations were inherited by and strengthened in the renal cell carcinomas.
<Epigenetic Clustering of Renal Cell Carcinomas>
The result of the unsupervised hierarchical clustering using the DNA methylation levels (ΔβT-N) on the 801 probes revealed that 104 patients with clear cell renal cell carcinomas were subclustered into Cluster A (n=90) and Cluster B (n=14) (see
Next, the clinicopathological parameters of clear cell renal cell carcinomas belonging to Clusters A and B, and TNM stage were examined. Table 11 shows the obtained result.
1.07 × 10−4 (b)
6.29 × 10−4 (c)
8.33 × 10−6 (c)
5.67 × 10−4 (c)
2.45 × 10−4 (c)
3.38 × 10−3 (c)
1.86 × 10−4 (c)
2.06 × 10−3 (c)
4.86 × 10−6 (c)
3.98 × 10−2 (c)
5.41 × 10−5 (c)
Note that, among “P-values” in Table 11, “P<0.05” are underlined, the numerical values with (b) are of the Wilcoxon rank sum test, and the numerical values with (c) are of the Fisher's exact test. Moreover, regarding the clinicopathological parameter with (d), findings in the predominant area are described if the tumor showed heterogeneity. Regarding the clinicopathological parameter with (e), if the tumor showed heterogeneity, the most aggressive features of the tumor are described.
Further, the survival rates of patients belonging to these Clusters A and B were also examined. The period of the survival rate analysis was 42 to 4024 days (mean: 1821 days).
As apparent from the result shown in Table 11, Cluster B had larger (or higher) values than Cluster A in terms of: the diameter of clear cell renal cell carcinomas, incidence of single nodular type with extranodular growth (type 2) or contiguous multinodular type (type 3) according to the aforementioned macroscopic configuration, frequencies of vascular involvement, renal vein tumor thrombus formation, infiltrating growth, tumor necrosis, and renal pelvis invasion, histological grade, and pathological TNM stage. Note that it is clear as shown in Table 11 that epigenetic clustering of renal cell carcinomas was dependent on neither sex nor age of the patients.
Moreover, as apparent from the results shown in
<DNA Methylation Profiles of Renal Cell Carcinomas>
Next, the proportions of probes showing various degrees of DNA hypermethylation in T samples compared to the corresponding N samples (ΔβT-N>0.1, 0.2, 0.3, 0.4, or 0.5) for all 26454 probes were analyzed. Moreover, the proportions of probes showing various degrees of DNA hypomethylation in N samples compared to the corresponding T samples (ΔβT-N<−0.1, −0.2, −0.3, −0.4, or −0.5) for all 26454 probes were analyzed.
As apparent from the result shown in
Thus, it was revealed that renal cell carcinomas belonging to Cluster B were characterized by accumulation of DNA hypermethylation.
Further, Tables 12 and 13 shows the top 61 probes on which DNA methylation levels differed markedly between Clusters A and B. Note that, in Tables 12 and 13, “target ID” indicates the probe number for the Infinium HumanMethylation27 Bead Array all assigned by Illumina, Inc., and “chromosomal number” and “position on chromosome” indicate a position on the reference human genome sequence NCBI database Genome Build 37 (hereinafter, the same applies to headings in Tables regarding probes). “Y” under “CpG island” indicates that the corresponding probe is located within the CpG island, while “N” indicates that the corresponding probe is not located within the CpG island (the same applies to Tables 14 and 15). Further, “gene region” indicates that the corresponding probe is located in an exon or an intron, or upstream of the transcription start site (TSS). Furthermore, “P-value” indicates a value calculated by the Wilcoxon rank sum test.
Although only 19246 probes, 72.8%, out of the total of 26454 probes were located within CpG islands, 60 probes, 98.4%, out of the 61 probes located within CpG islands showed DNA hypermethylation in renal cell carcinomas belonging to Cluster B (ΔβT-N>0.097, Tables 12 and 13). Note that the remaining one probe among the 61 probes was located within a non-CpG island and showed DNA hypomethylation (ΔβT-N>−0.425±0.096 in Cluster B).
The results described in Examples 2 and 3 revealed that Cluster B was well correlated with the clinicopathological phenotype and characterized by frequent DNA hypermethylation on CpG islands.
Note that such characteristics of renal cell carcinomas belonging to Cluster B are similar to those of CpG island methylator phenotype (CIMP)-positive cancers in other well-studied organs (for example, colon and stomach) (NPLs 8 to 11). In other words, this single-CpG resolution methylome analysis identified, for the first time, CIMP-positive renal cell carcinomas as Cluster B.
<Identification of Hallmark CpG Sites of CIMP-Positive Renal Cell Carcinomas>
Correlations between DNA methylation levels (β values) in renal cell carcinoma tissues (T samples) and those in non-cancerous renal tissues (N samples) from representative patients with renal cell carcinomas belonging to Clusters A and B were examined.
As apparent from the result shown in
Based on this result, in order to discriminate renal cell carcinomas belonging to Cluster B from those belonging to Cluster A, focused on were probes for which the average β value in all N samples was less than 0.2 and the incidence of more than 0.4 ΔβT-N was markedly high in Cluster B than Cluster A (P<1.98×10−6, Fisher's exact test).
Then, among such probes, 16 probes (15 genes: FAM150A, GRM6, ZNF540, ZFP42, ZNF154, RIMS4, PCDHAC1, KHDRBS2, ASCL2, KCNQ1, PRAC, WNT3A, TRH, FAM78A, and ZNF671) showed more than 0.4 ΔβT-N in 6 or more (42.8% or more) renal cell carcinomas among the 14 renal cell carcinomas belonging to Cluster B. On the other hand, the 16 probes showed more than 0.4 ΔβT-N in 2 or fewer (2.2% or less) renal cell carcinomas among the 90 renal cell carcinomas belonging to Cluster A (see Table 14).
Moreover, as apparent from the result shown in
Further, random forest analysis was performed using 869 probes on which DNA methylation levels (ΔβT-N) differed markedly between Clusters A and B (FDR [q=0.01]) (see
Note that 2 genes (FAM150A and GRM6) were shared by the 15 genes and the top 4 genes which were found by the random forest analysis to be able to discriminate Cluster A from Cluster B.
Thus, CpG sites of these 17 genes (FAM150A, GRM6, ZNF540, ZFP42, ZNF154, RIMS4, PCDHAC1, KHDRBS2, ASCL2, KCNQ1, PRAC, WNT3A, TRH, FAM78A, ZNF671, SLC13A5, and NKX6-2) can be considered as hallmarks of CIMP-positive renal cell carcinomas, for example, renal cell carcinomas belonging to Cluster B. In other words, it was revealed that it was possible to detect an unfavorable prognostic risk of patients with renal cell carcinomas by detecting DNA methylation levels at CpG sites of the 17 genes.
In addition, levels of these genes expressed were analyzed by quantitative RT-PCR. The result revealed that DNA hypermethylation reduced the expression of these genes (see Table 16).
2.40 × 10−11
1.32 × 10−10
Thus, it was demonstrated that the DNA methylation alternations occurring at the precancerous stage determined the aggressiveness of renal cell carcinomas and the prognosis of the patients through alterations of gene expression levels.
<Detection of DNA Methylation Level in Renal Cell Carcinomas, using Mass Spectrometer>
The effectiveness of the DNA methylation level detection at the CpG sites of the 17 genes was verified by a MassARRAY method, a different methylated DNA detection method from the Infinium assay.
The MassARRAY method is a method for detecting a difference in molecular weight between methylated DNA fragments and unmethylated DNA fragments using a mass spectrometer after a bisulfate-treated DNA is amplified and transcribed into RNA, which is further base-specifically cleaved with an RNase.
First, MassARRAY primers were designed using EpiDesigner (manufactured by SEQUENOM, Inc., primer design software for MassARRAY) for CpG islands containing the CpG sites that are the probe site of the Infinium array.
Note that the PCR target sequence in MassARRAY is somewhat long: approximately 100 to 500 bp. Accordingly, DNA methylation levels of a large number of CpG sites around the CpG sites that are the probe site of the Infinium array can be evaluated together.
Moreover, in order to exclude the influence of a bias in PCR, a test was run in such a manner as to average combinations of three DNA polymerases with conditions of approximately four annealing temperatures per primer set, so that optimum PCR conditions for favorable quantification were determined.
Then, it was confirmed that the adopted PCR conditions were favorable in terms of the quantification for all the CpG sites contained in the PCR target sequence and to be analyzed. The MassARRAY analysis was performed on 88 specimens of CIMP-negative renal cell carcinomas and 14 specimens of CIMP-positive renal cell carcinomas.
Specifically, first, in the same manner as in the above-described Infinium assay, a genomic DNA was extracted from each sample and converted with bisulfate. Then, the resultant was amplified by PCR, and an in vitro transcription reaction was carried out. Subsequently, the obtained RNA was specifically cleaved at a uracil site with RNAse A, thereby forming fragments differed from one another in length according to the presence or absence of the methylation on the genomic DNA of each sample. Thereafter, the obtained RNA fragments were subjected to MALDI-TOF MAS (manufactured by SEQUENOM, Inc., MassARRAY Analyzer 4) capable of detecting a difference in mass of a single base to conduct the mass analysis. The obtained mass analysis result was aligned with a reference sequence using analysis software (EpiTYPER, manufactured by SEQUENOM, Inc.). The methylation level was calculated from a mass ratio between the RNA fragment derived from the methylated DNA and the RNA fragment derived from the unmethylated DNA.
Tables 17 and 18 and Sequence Listing show the sequences of the primers used in this analysis and the sequences of PCR products amplified using the primer sets.
As apparent from the results shown in
Thus, it was revealed that one CpG site in a region where strong silencing occurred by the hypermethylation status of all the promoter region had been identified in Example 4; in other words, it was revealed that detecting a DNA methylation level of not only the aforementioned 18 CpG sites but also at least one CpG site located on CpG islands of the 17 genes made it possible to detect an unfavorable prognostic risk of renal cell carcinoma.
Further, the DNA methylation levels at 312 CpG sites of 14 genes in the 14 cases already classified into the CIMP-positive group by the above-described Infinium assay and of the 88 CIMP-negative cases were quantified by the MassARRAY method. Then, based on the result, a receiver operating characteristic (ROC) analysis was performed, and “sensitivity (positive rate)”, “specificity”, and “1-specificity (false-positive rate)” were obtained which are used when the CIMP-positive group is distinguished from the CIMP-negative group on the basis of each CpG site alone. Further, a ROC curve was created from the obtained values of these, and an AUC (area under the curve, the area under the ROC curve) was calculated. Moreover, a cutoff value (diagnostic threshold) at which “sensitivity+specificity” was the maximum was set for each CpG site. Tables 19 to 27 show the obtained results of the CpG sites quantitatively analyzed by the MassARRAY analysis. Note that, in Tables 19 to 27, multiple CpG sites which are close to each other, and whose DNA methylation levels are measured together due to the feature of the MassARRAY method, are collectively shown as a single unit. Additionally, in these tables, “target gene name_primer set name_CpG site” indicates the order of CpG sites in PCR products amplified using the primer sets shown in Tables 17 and 18. Note that, in Table 23, SLC13A5—10_CpG—44 and SLC13A5—13_CpG—1 respectively indicate the 44th CpG site and the 1st CpG site in the region amplified by different primer sets, but their positions on the genome (positions on NCBI database Genome Build 37) are at the same CpG site: position 6617077 on chromosome 17.
As apparent from the results shown in Tables 19 to 27, it was found that a large number of CpG sites having a high diagnostic ability existed in each CpG island besides the Infinium-probe CpG sites. Specifically, t he number of CpG sites having an AUC>0.9 was 141 sites, and the number of CpG sites having an AUC >0.95 was 32 sites.
Moreover, in the MassARRAY method, one measurement value is obtained from consecutive CpG sites such as CGCGCG, for example, “FAM150A—14_CpG—13.14.15”, as a whole. Accordingly, the 141 sites having an AUC>0.9 correspond to 90 measurement values (units) based on the AUC calculation. Similarly, the 32 sites having an AUC>0.95 correspond to 23 measurement values (units) in terms of the measurement value based on the AUC calculation.
Furthermore, as apparent from the result shown in
As has been described above, the present invention makes it possible to clearly classify renal cell carcinomas of unfavorable prognosis (CIMP-positive renal cell carcinomas) and relatively favorable renal cell carcinomas by detecting a DNA methylation level at at least one CpG site of the 17 genes (FAM150A, GRM6, ZNF540, ZFP42, ZNF154, RIMS4, PCDHAC1, KHDRBS2, ASCL2, KCNQ1, PRAC, WNT3A, TRH, FAM78A, ZNF671, SLC13A5, and NKX6-2).
Since the difference in the DNA methylation level between the unfavorable prognosis group and the favorable group is large, such a difference can be easily detected by a PCR method and the like (for example, methylation-specific quantitative PCR, COBRA) already widespread in examination rooms in hospitals and other places. Moreover, a genomic DNA for prognosis can be abundantly extracted from specimens resulting from renal cell carcinoma surgeries without involving unnecessary invasion to patients. Thus, the method for detecting an unfavorable prognostic risk of renal cell carcinoma of the present invention is useful in the clinical field as the method directed to improve the clinical outcome.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2013/062650 | 4/30/2013 | WO | 00 |
| Number | Date | Country | |
|---|---|---|---|
| 61646044 | May 2012 | US |
