Progressive kidney cancer automatic scoring and classifying method based on DCCD subtype marker

Abstract

The invention discloses an automatic grading and classifying method for progressive kidney cancer based on DCCD subtype markers, which comprises the following steps: taking a TGFBI gene, a PLOD2 gene and an SLC2A1 gene as marker genes of DCCD subtype renal clear cell carcinoma tissue samples subjected to clear cell differentiation to obtain an immunohistochemical staining image of a marker gene antibody corresponding to each renal clear cell carcinoma tissue sample; obtaining a staining intensity score of each immunohistochemical staining image of each renal clear cell carcinoma tissue sample; performing image segmentation and cell counting on the immunohistochemical staining image corresponding to each renal clear cell carcinoma tissue sample to obtain a positive cell proportion score corresponding to each renal clear cell carcinoma tissue sample; classifying the renal clear cell carcinoma tissue sample as a progressive renal clear cell carcinoma patient or a non-progressive renal clear cell carcinoma based on the DCCD score; according to the invention, rapid and accurate screening of progressive renal clear cell carcinoma patients is realized, the realization mode is simple, and the cost is low.

Description

Technical Field

[0001] The present invention relates to the technical field of biological information processing, and in particular to an automatic scoring and classification method for progressive renal cancer based on DCCD subtype markers. Background Art

[0002] As one of the most common malignancies of the urinary system, kidney cancer (RCC) has a high mortality rate. The most common pathological type of RCC is clear cell renal cell carcinoma (ccRCC), characterized by prominent abnormal lipid accumulation. Most patients are initially diagnosed at an early stage. In clinical practice, the standard treatment for early-stage ccRCC is surgical resection followed by passive observation. However, approximately one-third of surgically treated patients experience disease progression, such as recurrence or metastasis, resulting in a 5-year survival rate of less than 15%. Early-stage RCC patients account for 80% of those initially diagnosed. Although complete surgical resection is the preferred treatment for early-stage localized RCC, approximately one-third of patients experience disease progression, such as recurrence or metastasis, after surgery, resulting in a severely poor prognosis. This suggests that the current TNM staging system, while prioritizing tumor size, overlooks the differentiation characteristics of the tumor cells themselves, leading to disease progression in patients with early, clinically advanced disease during passive surveillance.

[0003] Invention patent CN117238369A has disclosed the detection of the relative expression levels of IM2-related genes and IM4-related genes in tumor tissue samples from renal cancer patients through sequencing, nucleic acid membrane strips, chips, or kit methods. Based on this, combined with the ssGSEA algorithm, the ssGSEA scores of the IM2- and IM4-related gene sets are calculated for all tumor tissue samples, namely IM2_ssGSEA and IM4_ssGSEA, respectively. When IM4_ssGSEA-IM2_ssGSEA>-0.2, the patient from which the sample was derived is defined as a "de-clear cell differentiation patient" and recorded as the IM4 type. When IM4_ssGSEA-IM2_ssGSEA≤-0.2, the patient from which the sample was derived is defined as a "non-clear cell differentiation patient" and recorded as the IM2 type. Among them, patients with the IM4 subtype have very little lipid accumulation but a significantly worse prognosis, while patients with the IM2 subtype have significant lipid accumulation but a good prognosis.

[0004] Although sequencing technology can achieve accurate classification of patients with renal clear cell carcinoma, its high price and poor timeliness limit its promotion and application. Not every IM4 type patient has progressive renal clear cell carcinoma. Therefore, it is necessary to further determine whether IM4 type renal cancer patients have progressive renal clear cell carcinoma or non-progressive renal clear cell carcinoma. Summary of the Invention

[0005] The purpose of the present invention is to provide an automatic scoring and classification method for progressive renal cell carcinoma based on DCCD subtype markers, so as to solve the problem that sequencing technology in the existing technology can achieve accurate classification of patients with renal clear cell carcinoma, but its high price and poor timeliness limit its promotion and application, and there is no technical problem to further determine whether IM4 type renal cell carcinoma patients are progressive renal clear cell carcinoma or non-progressive renal clear cell carcinoma.

[0006] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0007] An automatic scoring and classification method for progressive renal cancer based on DCCD subtypes comprises the following steps:

[0008] Step 100: using the TGFBI gene, the PLOD2 gene, and the SLC2A1 gene as marker genes for renal clear cell carcinoma tissue samples expressing the clear cell differentiation subtype, and performing DAB color development, nuclear staining, dehydration, mounting, and image scanning on each renal clear cell carcinoma tissue sample to obtain an immunohistochemical staining image;

[0009] Step 200: Obtain staining intensity scores of immunohistochemical staining images corresponding to immunohistochemical staining with TGFBI, PLOD2, and SLC2A1 antibodies for each renal clear cell carcinoma tissue sample;

[0010] Step 300: performing image segmentation and cell counting on the immunohistochemical staining image corresponding to each renal clear cell carcinoma tissue sample to obtain a positive cell ratio score corresponding to each renal clear cell carcinoma tissue sample;

[0011] Step 400: Multiply the staining intensity score of each renal clear cell carcinoma tissue sample by the positive cell proportion score to obtain the IRS score of each marker corresponding to each renal clear cell carcinoma tissue sample; take the sum of the IRS scores of each marker corresponding to each renal clear cell carcinoma tissue sample as the DCCD score; and classify the renal clear cell carcinoma tissue sample into an advanced renal clear cell carcinoma patient sample or a non-advanced renal clear cell carcinoma sample based on the DCCD score.

[0012] As a preferred embodiment of the present invention, in step 100, the gene set related to clear cell renal cell carcinoma of the dedifferentiated subtype includes COL13A1, IL2RA, HTRA3, HTRA1, MOCOS, DIO2, IL1R2, F2, F3, PLEKHG4B, IGFN1, SLC7A5, TRIB3, SPOCD1, SPOCK1, TIMP1, RAB3B, RAB42, RAB6B, P4HA3, PSAT1, DNER, CDCP1, SRPX2, SLC12A8, SERPINE1, SERPINF1, SERPINH1, SERPINA5, ITPKA, SEMA3C, SEM A4B, CHST1, WDR86, MAP3K15, AMIGO2, MLPH, APCDD1L, CTHRC1, MMP9, LTBP1, TRNP1, IBSP, LOXL1, LOXL2, COL10A1, ADAMTS2, KLHDC8A, KIF26B, MRC2, CACN B3, KIF18B, DDIT4, L1CAM, MT1X, MT2A, OLFML3, TENM3, MXRA8, MXRA5, SPTBN2, QSOX1, TNNT1, SSC5D, NT5DC2, SLC18A3, ARL4C, TRIM46, RIN1, ECM1, SLC17 A9, WISP2, IRS2, NAV1, NAV3, ROS1, AHNAK2, NEFL, KCND1, KCNQ3, KCNS3, CARD14, CARD11, SPINK13, MYBL2, ELL2, RFLNA, CEBPB, RTL1, FAM20A, ACAN, RAB2 7B, ADD2, C10orf99, LRRC15, LRRC8E, PTHLH, BACE2, RYR2, LYPD1, JAK3, DCBLD2, C6orf141, HMGA2, COL22A1, CCDC91, MUC12, ADGRG2, SAA1, SAA2, WNT5A, IGDCC4, IL20RB, AMZ1, COL24A1, TFPI2, KDELR3, ANLN, PTPRN, PTPRH, IMPDH1, SCG2, AQP9, AC068580.4, ADAMDEC1, NPTX2, PCBP3, PLPPR4, TFAP2A, GREM1 , PYCR1, JPH2, SOX11, STEAP3, BASP1, DPYSL3, BTBD11, SLPI, GRIK4, SLC4A3, PLAUR, FHL2, FHL3, BMPR1B, SLC2A1, C1S, C1R, LGALS1, TBX15, PLIN3, G0S2,SLC35F3, SLITRK2, EPHA10, GPR173, FAP, TGFBI, NFKBIZ, FGB, FGA, FGG, RGS17, FN1, SPHK1, PLOD2, SYBU, SLC38A5, PPP1R1A, KIF2A, PDLIM7, PRIMA1, PLPP2, COL7A1, LIF, CERCAM, COL6A2, COL6A1, COL6A3, FSCN1, MDK, SPON2, FKBP10, ZFHX4, BMP1, PLXDC1, SPSB1, INHBA, COL5A2, GAL NT17, ADAMTS14, MMP11, MMP16, MMP17, GAD1, TNR, HOXD11, HOXD10, SYT12, CACNA1H, PLXNB3, EFNA5, B3GALT5, SORCS2, KNDC1, RUNX2, RUNX1, UCHL1, B4GALNT4, SPINK1, TPM2, NPEPL1, PDPN, MLLT11, ZNF703, KRT19, KRT17, KRT80, TSKU, ZNF469, FAM78B, ZNF365, TMEM132A, SPRED3, LRR C3, CD44, GPC1, CD82, CDH3, APLP1, PKP3, BCAT1, PLAU, ADAM19, ADAM12, PLP2, PODNL1, PLTP, GRIN2D, GFPT2, B3GNT4, SCNN1G, MELTF, LEF1, P PP2R2C, PRAME, CLMP, ATP8A2, IGFBP1, ATP8B3, PSD3, SLC6A17, C1QL1, PREX2, CPA4, CFAP47, LAMB3, LAMB1, PXDN, IGF2BP3, IGF2BP2, CPNE7, A CTN2, ACTN1, ACTG2, RARRES1, THSD4, FAM167A, GXYLT2, GALNT2, ZPLD1, MEX3B, GPRC5A, BAIAP2L1, N4BP3, MICAL2, CD276, SMOC1, C5orf46, H OXB9, NALCN, ALDH1L2, C16orf74, MFAP5, CHRDL2, DBN1, KIAA1644, OLFML2B, ITGB4, ELFN2, ITGB6, ITGA5, C1QTNF1, GRAMD1B, ZBTB7C, TRAM2. ,

[0013] As a preferred embodiment of the present invention, the method for obtaining the immunohistochemical staining image corresponding to each tumor tissue sample is as follows:

[0014] Each of the tumor tissue samples is wax-solidified and serially sectioned;

[0015] Antigen retrieval was performed on the sections, and endogenous peroxidase activity of the sections was blocked;

[0016] All tumor tissue sample sections were incubated with anti-TGFBI primary antibody, anti-PLOD2 primary antibody, and anti-SLC2A1 primary antibody in sequence;

[0017] Each tumor tissue sample after primary antibody incubation was performed using HRP-conjugated polymer secondary antibody;

[0018] After incubation with the secondary antibody, the sections of each tumor tissue sample were subjected to DAB color development, nuclear staining, dehydration, sealing, and scanning to obtain immunohistochemical staining images.

[0019] As a preferred embodiment of the present invention, in step 200, the staining intensity scores of the immunohistochemical staining images corresponding to multiple sections of each tumor tissue sample are obtained using Image J or Fiji software, and the average of the staining intensity scores of the multiple sections of each tumor tissue sample is used as the staining intensity score of the tumor tissue sample;

[0020] The staining intensity score range is set to 0-3, the staining intensity score of no positive staining is set to 0, the staining intensity score of light yellow is set to 1, the staining intensity score of brown is set to 2, and the staining intensity score of tan is set to 3.

[0021] As a preferred embodiment of the present invention, in step 300, the immunohistochemical staining images corresponding to each renal clear cell carcinoma tissue sample are subjected to machine learning by using Image J or Fiji software.

[0022] The positive cells in the immunohistochemical staining images corresponding to each renal clear cell carcinoma tissue sample were calculated respectively, and the positive cell proportion scores corresponding to multiple sections of each renal clear cell carcinoma tissue sample were obtained. The average of the positive cell proportion scores of multiple sections of each renal clear cell carcinoma tissue sample was used as the positive cell proportion score of the tumor tissue sample.

[0023] As a preferred embodiment of the present invention, when the positive cell ratio is 0-5%, the positive cell ratio score is 0;

[0024] When the positive cell ratio is 6%-25%, the positive cell ratio score is 1;

[0025] When the positive cell percentage is 26%-50%, the positive cell percentage score is 2;

[0026] When the positive cell percentage is 51%-75%, the positive cell percentage score is 3;

[0027] When the positive cell ratio was greater than 75%, the positive cell ratio score was 4.

[0028] As a preferred embodiment of the present invention, in step 400, the IRS scores of the images corresponding to the immunohistochemical staining of TGFBI, PLOD2 and SLC2A1 antibodies in each renal clear cell carcinoma tissue sample are calculated;

[0029] Calculate the sum of the IRS scores of all immunohistochemical images corresponding to the TGFBI gene, PLOD2 gene, and SLC2A1 gene staining in each renal clear cell carcinoma tissue sample, i.e., the DCCD score;

[0030] Based on the numerical range of the DCCD score, all clear cell renal cell carcinoma tissue samples were classified as samples of patients with advanced clear cell renal cell carcinoma or samples of non-progressive clear cell renal cell carcinoma.

[0031] As a preferred embodiment of the present invention, renal clear cell carcinoma tissue samples with a DCCD score value range higher than 21 (16, 25) are used as samples of patients with advanced renal clear cell carcinoma;

[0032] The renal clear cell carcinoma tissue samples with a DCCD score of less than 9 (6, 12) were considered as non-progressive renal clear cell carcinoma patient samples.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The present invention performs a series of immunohistochemical marker staining on tumor tissue samples to be identified and classified (divided into non-progressive or progressive), determines a staining intensity score based on the color of the immunohistochemical marker staining image corresponding to the DCCD subtype of each tumor tissue sample, determines a positive cell proportion score based on the positive cell proportion of the immunohistochemical marker staining image corresponding to each renal clear cell carcinoma tumor tissue sample, takes the sum of the staining intensity score and the positive cell proportion score of each marker staining image as the DCCD score, and uses the DCCD score as a standard for screening progressive renal clear cell carcinoma, thereby achieving rapid and accurate screening of progressive renal clear cell carcinoma patients among DCCD subtype patients, and the implementation method is simple and low-cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other implementation drawings based on the provided drawings without inventive effort.

[0036] Figure 1 Schematic diagram of the process of the automatic scoring and classification method for progressive renal cancer according to an embodiment of the present invention;

[0037] Figure 2 This is an immunohistochemical staining image of a tumor tissue sample in the DCCD scoring control group of an embodiment of the present invention;

[0038] Figure 3 Schematic diagram of processing immunohistochemical staining images when calculating staining intensity provided by an embodiment of the present invention

[0039] Figure 4 A schematic diagram of processing an immunohistochemical staining image when calculating positive cells provided by an embodiment of the present invention;

[0040] Figure 5 This is a statistical graph of the IRS scores of the DCCD subgroup immunohistochemical markers according to an embodiment of the present invention; DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0042] like Figure 1 As shown, the present invention provides an automatic scoring and classification method for progressive renal cancer based on DCCD subtype markers. The present invention aims to disclose a method for screening potential progressive renal clear cell carcinoma, achieve early screening of corresponding patients, and provide a theoretical basis for clinical stratified diagnosis and treatment of renal clear cell carcinoma and personalized precision treatment of patients, including the following steps:

[0043] Step 100: using the TGFBI gene, the PLOD2 gene, and the SLC2A1 gene as marker genes for renal clear cell carcinoma tissue samples expressing the de-clear cell differentiation (DCCD) subtype, and performing DAB color development, nuclear staining, dehydration, mounting, and image scanning on each renal clear cell carcinoma tissue sample to obtain an immunohistochemical staining image;

[0044] Step 200: Obtain staining intensity scores of immunohistochemical staining images corresponding to immunohistochemical staining with TGFBI, PLOD2, and SLC2A1 antibodies for each renal clear cell carcinoma tissue sample;

[0045] Step 300: performing image segmentation and cell counting on the immunohistochemical staining image corresponding to each renal clear cell carcinoma tissue sample to obtain a positive cell ratio score corresponding to each renal clear cell carcinoma tissue sample;

[0046] Step 400: Multiply the staining intensity score of each renal clear cell carcinoma tissue sample by the positive cell proportion score to obtain the IRS score of each marker corresponding to each renal clear cell carcinoma tissue sample; take the sum of the IRS scores of each marker corresponding to each renal clear cell carcinoma tissue sample as the DCCD score; and classify the renal clear cell carcinoma tissue sample into an advanced renal clear cell carcinoma patient sample or a non-advanced renal clear cell carcinoma sample based on the DCCD score.

[0047] Invention patent CN117238369A has disclosed the use of sequencing, nucleic acid membrane strips, chips, or kits to detect the relative expression levels of IM2-related genes and IM4-related genes in tumor tissue samples from patients with renal cancer, thereby classifying renal clear cell carcinoma into type IM2 (patients with obvious lipid droplet accumulation, i.e., clear cell differentiation) and type IM4 (patients with very few lipid droplets, i.e., de-clear cell differentiation). Patients with type IM4 are at high risk for early postoperative tumor recurrence and progression.

[0048] That is to say, invention patent CN117238369A is mainly used to identify IM4 type "de-clear cell differentiation patients" among renal cancer patients. However, it is necessary to further identify the progressive ccRCC and non-progressive ccRCC in IM4 type patients and IM2 type patients, so as to screen patients with progressive clear cell renal cell carcinoma early and prevent it in advance. Although IM2 type patients have a significantly better prognosis than IM4, it does not mean that there are no patients with progressive clear cell renal cell carcinoma among IM2 type patients, nor does it mean that there are no non-progressive clear cell renal cell carcinoma among all IM4 type patients. Therefore, it is necessary to conduct secondary identification and classification of renal clear cell carcinoma patients to further screen out patients with progressive clear cell renal cell carcinoma.

[0049] Not every IM4 type patient has advanced clear cell renal cell carcinoma, and not all IM2 type patients are non-progressive. Therefore, in this context, it is necessary to further screen patients with potential advanced clear cell renal cell carcinoma. In this embodiment, TGFBI gene, PLOD2 gene, and SLC2A1 gene are used as DCCD subtype markers to identify patients with advanced clear cell renal cell carcinoma among IM4 type renal cell carcinoma.

[0050] In step 100, the de-clear cell differentiation (DCCD) subtype tumor-associated gene set includes COL13A1, IL2RA, HTRA3, HTRA1, MOCOS, DIO2, IL1R2, F2, F3, PLEKHG4B, IGFN1, SLC7A5, TRIB3, SPOCD1, SPOCK1, TIMP1, RAB3B, RAB42, RAB6B, P4HA3, PSAT1, DNER, CDCP1, SRPX2, SLC12A8, SERPINE1, SERPINF1, SERPINH1, SERPINA5, ITPKA, SEMA3C, SEMA4B, CHST1, W DR86, MAP3K15, AMIGO2, MLPH, APCDD1L, CTHRC1, MMP9, LTBP1, TRNP1, IBSP, LOXL1, LOXL2, COL10A1, ADAMTS2, KLHDC8A, KIF26B, MRC2, CACNB3, KIF18B, DDIT4, L1CAM, MT1X, MT2A, OLFML3, TENM3, MXRA8, MXRA5, SPTBN2, QSOX1, TNNT1, SSC5D, NT5DC2, SLC18A3, ARL4C, TRIM46, RIN1, ECM1, SLC17A9, WISP2, I RS2, NAV1, NAV3, ROS1, AHNAK2, NEFL, KCND1, KCNQ3, KCNS3, CARD14, CARD11, SPINK13, MYBL2, ELL2, RFLNA, CEBPB, RTL1, FAM20A, ACAN, RAB27B, ADD2, C 10orf99, LRRC15, LRRC8E, PTHLH, BACE2, RYR2, LYPD1, JAK3, DCBLD2, C6orf141, HMGA2, COL22A1, CCDC91, MUC12, ADGRG2, SAA1, SAA2, WNT5A, IGDCC4, IL 20RB, AMZ1, COL24A1, TFPI2, KDELR3, ANLN, PTPRN, PTPRH, IMPDH1, SCG2, AQP9, AC068580.4, ADAMDEC1, NPTX2, PCBP3, PLPPR4, TFAP2A, GREM1, PYCR1, J PH2, SOX11, STEAP3, BASP1, DPYSL3, BTBD11, SLPI, GRIK4, SLC4A3, PLAUR, FHL2, FHL3, BMPR1B, SLC2A1, C1S, C1R, LGALS1, TBX15, PLIN3, G0S2, SLC35F3,<h2 style=";text-align:left;direction:ltr">SLITRK2,EPHA10,GPR173,FAP,TGFBI,NFKBIZ,FGB,FGA,FGG,RGS17, FN1,SPHK1,PLOD2,SYBU,SLC38A5,PPP1R1A,KIF2A,PDLIM7,PRIMA1, PLPP2,COL7A1,LIF,CERCAM,COL6A2,COL6A1,COL6A3,FSCN1,MDK,SP ON2,FKBP10,ZFHX4,BMP1,PLXDC1,SPSB1,INHBA,COL5A2,GALNT17,AD AMTS14,MMP11,MMP16,MMP17,GAD1,TNR,HOXD11,HOXD10,SYT12,CAC NA1H,PLXNB3,EFNA5,B3GALT5,SORCS2,KNDC1,RUNX2,RUNX1,UCHL1, B4GALNT4,SPINK1,TPM2,NPEPL1,PDPN,MLLT11,ZNF703,KRT19,KRT1 7,KRT80,TSKU,ZNF469,FAM78B,ZNF365,TMEM132A,SPRED3,LRRC3,CD 44,GPC1,CD82,CDH3,APLP1,PKP3,BCAT1,PLAU,ADAM19,ADAM12,PLP 2,PODNL1,PLTP,GRIN2D,GFPT2,B3GNT4,SCNN1G,MELTF,LEF1,PPP2R 2C,PRAME,CLMP,ATP8A2,IGFBP1,ATP8B3,PSD3,SLC6A17,C1QL1,PRE X2, CPA4, CFAP47, LAMB3, LAMB1, PXDN, IGF2BP3, IGF2BP2, CPNE7, ACT N2,ACTN1,ACTG2,RARRES1,THSD4,FAM167A,GXYLT2,GALNT2,ZPLD1, MEX3B,GPRC5A,BAIAP2L1,N4BP3,MICAL2,CD276,SMOC1,C5orf46,HOX B9,NALCN,ALDH1L2,C16orf74,MFAP5,CHRDL2,DBN1,KIAA1644,OLFM L2B,ITGB4,ELFN2,ITGB6,ITGA5,C1QTNF1,GRAMD1B,ZBTB7C,TRAM2。,<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0051] This embodiment creatively uses TGFBI, PLOD2, and SLC2A1 as DCCD subtype markers to identify advanced clear cell renal cell carcinoma. The tissue sections of the obtained clear cell renal cell carcinoma patient samples are incubated with anti-TGFBI (1:200, ABclonal, A2407) primary antibodies, anti-PLOD2 (1:2000, Abcam, ab314640) primary antibodies, and anti-SLC2A1 (1:500, Abcam, ab115730) primary antibodies. The sample images are scanned to form immunohistochemical staining images. The staining intensity scores and positive cell proportion scores of the immunohistochemical staining images of TGFBI, PLOD2, and SLC2A1 can identify advanced ccRCC patient samples in patients with clear cell renal cell carcinoma.

[0052] Therefore, the method for obtaining the immunohistochemical staining image corresponding to each tumor tissue sample in this embodiment is as follows:

[0053] Each tumor tissue sample was wax-cured and serially sectioned. Specifically, the renal cancer tumor tissue sample was fixed in 4% paraformaldehyde for 24-48 hours, dehydrated using gradient ethanol (50%-100% ethanol, 30 minutes per level), and transparentized twice in xylene (60 minutes each). The sample was then wax-impregnated using 60°C paraffin wax and infiltrated in stages (low melting point wax → high melting point wax) for a total of approximately 6-8 hours. After embedding using a fully automatic embedding machine, the sample was serially sectioned to a thickness of 3-4 μm. After baking and other anti-slicing treatments, immunohistochemical staining was performed.

[0054] The sections were subjected to antigen repair and endogenous peroxidase activity was blocked. Specifically, after dewaxing with xylene and gradient ethanol, EDTA alkaline repair solution (pH 9.0) was used for antigen repair, and 3% H2O2 was used to block endogenous peroxidase activity.

[0055] All tumor tissue sample sections were incubated with anti-TGFBI, anti-PLOD2, and anti-SLC2A1 antibodies, respectively, at 4°C. Anti-TGFBI (1:200, ABclonal, A2407), anti-PLOD2 (1:2000, Abcam, ab314640), and anti-SLC2A1 (1:500, Abcam, ab115730) were used in this study. Other brands of antibodies can be tested and standards established in actual testing.

[0056] Each tumor tissue sample after primary antibody incubation was performed using HRP-conjugated polymer secondary antibody;

[0057] After incubation with the secondary antibody, the sections of each tumor tissue sample were subjected to DAB color development, nuclear staining, dehydration, sealing, and scanning to obtain immunohistochemical staining images.

[0058] In order to identify samples of patients with advanced or non-advanced renal clear cell carcinoma among renal cancer patients, this embodiment selects TGFBI, PLOD2, and SLC2A1 genes from the IM4-related gene set as DCCD subtype markers. By identifying whether TGFBI, PLOD, and SLC2A1 proteins are widely and highly expressed in each tumor tissue sample, it is determined whether the tumor tissue sample is a sample of a patient with advanced clear cell renal carcinoma. This screening method is simpler, faster, and more convenient.

[0059] Specifically, in this embodiment, the specific implementation method for identifying whether TGFBI, PLOD2 and SLC2A1 proteins are widely and highly expressed in each tumor tissue sample is:

[0060] Obtain the immunohistochemical staining image corresponding to each tumor tissue sample, wherein the DCCD subtype (IM4 type) tumor tissue sample is incubated with anti-TGFBI primary antibody, anti-PLOD2 primary antibody, and anti-SLC2A1 primary antibody, and the obtained immunohistochemical staining image is as follows: Figure 2 As shown in the right figure, the color of the immunohistochemical staining image is relatively dark. After incubating the non-DCCD subtype (IM2 type) tumor tissue sample with anti-TGFBI primary antibody, anti-PLOD2 primary antibody, and anti-SLC2A1 primary antibody, the immunohistochemical staining image obtained is as follows Figure 2 As shown in the left picture, the color of the immunohistochemical staining image is relatively light.

[0061] Therefore, by scoring the staining intensity of the TGFBI gene, PLOD2 gene, and SLC2A1 gene, tumor tissue samples of the DCCD subtype (IM4 type) and tumor tissue samples of the non-DCCD subtype (IM2 type) can be distinguished among tumor tissue samples.

[0062] In clinical studies, the overall survival and disease-free survival of patients with DCCD subtype renal cancer were significantly lower than those of non-DCCD patients. Multiple group tests found that the expression levels of TGFBI, PLOD2, and SLC2A1 genes in tumor tissue samples of patients with DCCD subtype renal cancer were significantly higher than those of patients with non-DCCD subtype renal cancer. Therefore, the immunohistochemical staining images of TGFBI, PLOD2, and SLC2A1 in DCCD subtype tumor tissue samples had a wider coloring range and darker coloring, while the immunohistochemical staining images of non-DCCD subtype tumor tissue samples had a smaller coloring range and lighter coloring.

[0063] In the TCGA-KIRC (clear cell renal cell carcinoma) public database, higher TGFBI, PLOD2, and SLC2A1 gene expression scores (based on RNA levels) were associated with significantly worse overall survival and disease-free survival of patients, indicating the importance of evaluating TGFBI, PLOD2, and SLC2A1 gene expression levels for screening DCCD subtype renal cancer.

[0064] When performing immunohistochemical staining of renal tumor samples, the intensity of positive cell staining for the corresponding indicator depends on the antigen content and distribution density, the labeling method, and its sensitivity. The greater the antigen content and the higher the distribution density of the corresponding indicator, the stronger the positive staining result obtained with the corresponding indicator antibody. The staining intensity of the nucleus / cytoplasm / cell membrane (different indicators may have different cellular localizations) ranges from no staining (only the blue nucleus is visible) to light yellow to brownish yellow to tan.

[0065] This embodiment performs a series of immunohistochemical marker staining on tumor tissue samples to be identified and classified (divided into non-progressive or progressive), determines the staining intensity score based on the color of the immunohistochemical marker staining image corresponding to the DCCD subtype of each tumor tissue sample, determines the positive cell proportion score based on the positive cell proportion of the immunohistochemical marker staining image corresponding to each renal clear cell carcinoma tumor tissue sample, takes the sum of the staining intensity score of each marker staining image and the IRS score consisting of the positive cell proportion score as the DCCD score, and uses the DCCD score as the standard for screening progressive renal clear cell carcinoma, thereby achieving rapid and accurate screening of progressive renal clear cell carcinoma patients among DCCD subtype patients, and the implementation method is simple and low-cost.

[0066] In step 200, the staining intensity scores of the immunohistochemical staining images corresponding to multiple sections of each tumor tissue sample are obtained using the IHC Profiler plug-in of Image J or Fiji software, and the average of the staining intensity scores of the multiple sections of each tumor tissue sample is used as the staining intensity score of the tumor tissue sample.

[0067] The staining intensity score range is set to 0-3, the staining intensity score of no positive staining is set to 0, the staining intensity score of light yellow is set to 1, the staining intensity score of brown is set to 2, and the staining intensity score of tan is set to 3.

[0068] like Figure 3As shown in the figure, the IHC Profiler plug-in of Image J software can automatically evaluate the average gray value of positive cells and the percentage of positive area after importing immunohistochemical staining images, and output the results of four staining intensity scores. The basic principle of this plug-in is to first perform color deconvolution on the input immunohistochemical staining image, separate the blue negative area and the yellow (brown) positive area, and determine the staining type (cytoplasm / nuclear staining) according to the cellular localization of the indicator. Then, the gray value of positive cells and the percentage of positive area are calculated respectively, and the staining intensity score results are output, which represent negative (0 points), weak positive (1 point), positive (2 points), moderate positive (3 points), and strong positive (4 points).

[0069] In step 300, machine learning is performed on the immunohistochemical staining images corresponding to each renal clear cell carcinoma tissue sample using the Trainable_Weka_Segmentation plug-in of Image J or Fiji software. The positive cells in the immunohistochemical staining images corresponding to each renal clear cell carcinoma tissue sample are calculated, and the positive cell proportion scores corresponding to multiple sections of each renal clear cell carcinoma tissue sample are obtained. The average of the positive cell proportion scores of the multiple sections of each renal clear cell carcinoma tissue sample is used as the positive cell proportion score of the tumor tissue sample.

[0070] When the positive cell proportion is 0-5%, the positive cell proportion score is 0;

[0071] When the positive cell ratio is 6%-25%, the positive cell ratio score is 1;

[0072] When the positive cell percentage is 26%-50%, the positive cell percentage score is 2;

[0073] When the positive cell percentage is 51%-75%, the positive cell percentage score is 3;

[0074] When the positive cell ratio was greater than 75%, the positive cell ratio score was 4.

[0075] The IRS score table formed by combining the positive cell proportion score and the staining intensity score is shown below.

[0076]

[0077] For the scoring of the positive cell percentage, the stained image can be machine-learned and segmented using the Trainable_Weka_Segmentation plug-in of Image J or Fiji software, and then the cells can be counted. The positive cell percentage can be calculated and scored according to the criteria of 0-5% cells being positive (0 points), 6-25% cells being positive (1 point), 26%-50% cells being positive (2 points), 51%-75% cells being positive (3 points), and >75% cells being positive (4 points). Figure 4 shown.

[0078] In step 400, the IRS scores of the immunohistochemical images corresponding to the TGFBI gene, PLOD2 gene, and SLC2A1 gene staining in each renal clear cell carcinoma tissue sample are calculated, and the sum of the IRS scores of the immunohistochemical images corresponding to the immunohistochemical staining of TGFBI, PLOD2, and SLC2A1 antibodies in each renal clear cell carcinoma tissue sample is calculated, that is, the DCCD score. Figure 5 shown.

[0079] Based on the numerical range of the DCCD score, all clear cell renal cell carcinoma tissue samples were classified as samples of patients with advanced clear cell renal cell carcinoma or samples of non-progressive clear cell renal cell carcinoma.

[0080] RCC tissue samples with a DCCD score higher than 21 (16, 25) were considered as advanced RCC patient samples.

[0081] The renal clear cell carcinoma tissue samples with a DCCD score of less than 9 (6, 12) were considered as non-progressive renal clear cell carcinoma patient samples.

[0082] In summary, patients with a DCCD score higher than 21 (16, 25) can be defined as patients with progressive clear cell renal cell carcinoma (i.e., progressive ccRCC); patients with a DCCD score lower than 9 (6, 12) can be defined as patients with non-progressive clear cell renal cell carcinoma (i.e., non-progressive ccRCC), thus achieving the classification and identification of patients with clear cell renal cell carcinoma and providing a theoretical basis for further precise management.

[0083] Furthermore, the DCCD score had a screening efficacy of AUC = 0.93 for both progressive and non-progressive ccRCC patient samples. AUC (Area Under Curve) refers to the area under the curve, defined as the area under the receiver operating characteristic curve and the coordinate axis, and ranges from 0.5 to 1. The closer the AUC is to 1.0, the higher the authenticity of the detection method; when it is equal to 0.5, the authenticity is the lowest and has no application value. The AUC calculation in this embodiment is implemented in R language, indicating that the DCCD score has high authenticity in screening samples for both progressive and non-progressive ccRCC patient samples.

[0084] Not every IM4 type patient has advanced clear cell renal cell carcinoma, and not every IM2 type patient does not have advanced clear cell renal cell carcinoma. Therefore, in this context, the present embodiment further provides a method for screening out patients with potential advanced clear cell renal cell carcinoma and patients without advanced clear cell renal cell carcinoma from IM4 type patients, and further allows for early screening and prevention of patients with advanced clear cell renal cell carcinoma.

[0085] The above embodiments are merely exemplary embodiments of the present application and are not intended to limit the scope of the present application. The scope of protection of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and scope of protection of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the scope of protection of the present application.

Claims

1. An automatic scoring and classification method for progressive renal cancer based on DCCD subtype markers, characterized in that: The following steps are involved: Step 100: using the TGFBI gene, the PLOD2 gene, and the SLC2A1 gene as marker genes for renal clear cell carcinoma tissue samples expressing the clear cell differentiation subtype, and performing DAB color development, nuclear staining, dehydration, mounting, and image scanning on each renal clear cell carcinoma tissue sample to obtain an immunohistochemical staining image; Step 200: Obtain staining intensity scores of immunohistochemical staining images corresponding to immunohistochemical staining with TGFBI, PLOD2, and SLC2A1 antibodies for each renal clear cell carcinoma tissue sample; Step 300: performing image segmentation and cell counting on the immunohistochemical staining image corresponding to each renal clear cell carcinoma tissue sample to obtain a positive cell ratio score corresponding to each renal clear cell carcinoma tissue sample; Step 400: Multiply the staining intensity score of each renal clear cell carcinoma tissue sample by the positive cell proportion score to obtain the IRS score of each marker corresponding to each renal clear cell carcinoma tissue sample; take the sum of the IRS scores of each marker corresponding to each renal clear cell carcinoma tissue sample as the DCCD score; and classify the renal clear cell carcinoma tissue sample into an advanced renal clear cell carcinoma patient sample or a non-advanced renal clear cell carcinoma sample based on the DCCD score.

2. The automatic scoring and classification method for progressive renal cancer based on DCCD subtype markers according to claim 1, characterized in that: In step 100, the gene set related to clear cell renal cell carcinoma of the dedifferentiated subtype includes COL13A1, IL2RA, HTRA3, HTRA1, MOCOS, DIO2, IL1R2, F2, F3, PLEKHG4B, IGFN1, SLC7A5, TRIB3, SPOCD1, SPOCK1, TIMP1, RAB3B, RAB42, RAB6B, P4HA3, PSAT1, DNER, CDCP1, SRPX2, SLC12A8, SERPINE1, SERPINF1, SERPINH1, SERPINA5, ITPKA, SEMA3C, SEMA4B, CHST1, W DR86, MAP3K15, AMIGO2, MLPH, APCDD1L, CTHRC1, MMP9, LTBP1, TRNP1, IBSP, LOXL1, LOXL2, COL10A1, ADAMTS2, KLHDC8A, KIF26B, MRC2, CACNB3, KIF18B, DDIT4, L1CAM, MT1X, MT2A, OLFML3, TENM3, MXRA8, MXRA5, SPTBN2, QSOX1, TNNT1, SSC5D, NT5DC2, SLC18A3, ARL4C, TRIM46, RIN1, ECM1, SLC17A9, WISP2, I RS2, NAV1, NAV3, ROS1, AHNAK2, NEFL, KCND1, KCNQ3, KCNS3, CARD14, CARD11, SPINK13, MYBL2, ELL2, RFLNA, CEBPB, RTL1, FAM20A, ACAN, RAB27B, ADD2, C 10orf99, LRRC15, LRRC8E, PTHLH, BACE2, RYR2, LYPD1, JAK3, DCBLD2, C6orf141, HMGA2, COL22A1, CCDC91, MUC12, ADGRG2, SAA1, SAA2, WNT5A, IGDCC4, IL 20RB, AMZ1, COL24A1, TFPI2, KDELR3, ANLN, PTPRN, PTPRH, IMPDH1, SCG2, AQP9, AC068580.4, ADAMDEC1, NPTX2, PCBP3, PLPPR4, TFAP2A, GREM1, PYCR1, J PH2, SOX11, STEAP3, BASP1, DPYSL3, BTBD11, SLPI, GRIK4, SLC4A3, PLAUR, FHL2, FHL3, BMPR1B, SLC2A1, C1S, C1R, LGALS1, TBX15, PLIN3, G0S2, SLC35F3,<h2 style=";text-align:left;direction:ltr">SLITRK2,EPHA10,GPR173,FAP,TGFBI,NFKBIZ,FGB,FGA,FGG,RGS17, FN1,SPHK1,PLOD2,SYBU,SLC38A5,PPP1R1A,KIF2A,PDLIM7,PRIMA1, PLPP2,COL7A1,LIF,CERCAM,COL6A2,COL6A1,COL6A3,FSCN1,MDK,SP ON2,FKBP10,ZFHX4,BMP1,PLXDC1,SPSB1,INHBA,COL5A2,GALNT17,AD AMTS14,MMP11,MMP16,MMP17,GAD1,TNR,HOXD11,HOXD10,SYT12,CAC NA1H,PLXNB3,EFNA5,B3GALT5,SORCS2,KNDC1,RUNX2,RUNX1,UCHL1, B4GALNT4,SPINK1,TPM2,NPEPL1,PDPN,MLLT11,ZNF703,KRT19,KRT1 7,KRT80,TSKU,ZNF469,FAM78B,ZNF365,TMEM132A,SPRED3,LRRC3,CD 44,GPC1,CD82,CDH3,APLP1,PKP3,BCAT1,PLAU,ADAM19,ADAM12,PLP 2,PODNL1,PLTP,GRIN2D,GFPT2,B3GNT4,SCNN1G,MELTF,LEF1,PPP2R 2C,PRAME,CLMP,ATP8A2,IGFBP1,ATP8B3,PSD3,SLC6A17,C1QL1,PRE X2, CPA4, CFAP47, LAMB3, LAMB1, PXDN, IGF2BP3, IGF2BP2, CPNE7, ACT N2,ACTN1,ACTG2,RARRES1,THSD4,FAM167A,GXYLT2,GALNT2,ZPLD1, MEX3B,GPRC5A,BAIAP2L1,N4BP3,MICAL2,CD276,SMOC1,C5orf46,HOX B9,NALCN,ALDH1L2,C16orf74,MFAP5,CHRDL2,DBN1,KIAA1644,OLFM L2B,ITGB4,ELFN2,ITGB6,ITGA5,C1QTNF1,GRAMD1B,ZBTB7C,TRAM2。, 3. The automatic scoring and classification method for progressive renal cancer based on DCCD subtype markers according to claim 2, characterized in that: In step 100, the method for obtaining the immunohistochemical staining image corresponding to each tumor tissue sample is as follows: Each of the tumor tissue samples is wax-solidified and serially sectioned; Antigen retrieval was performed on the sections, and endogenous peroxidase activity of the sections was blocked; All tumor tissue sample sections were incubated with anti-TGFBI primary antibody, anti-PLOD2 primary antibody, and anti-SLC2A1 primary antibody in sequence; Each tumor tissue sample after primary antibody incubation was performed using HRP-conjugated polymer secondary antibody; After incubation with the secondary antibody, the sections of each tumor tissue sample were subjected to DAB color development, nuclear staining, dehydration, sealing, and scanning to obtain immunohistochemical staining images.

4. The automatic scoring and classification method for progressive renal cancer based on DCCD subtype markers according to claim 1, characterized in that: In step 200, the staining intensity scores of the immunohistochemical staining images corresponding to multiple sections of each tumor tissue sample are obtained using Image J or Fiji software, and the average of the staining intensity scores of the multiple sections of each tumor tissue sample is used as the staining intensity score of the tumor tissue sample; The staining intensity score range is set to 0-3, the staining intensity score of no positive staining is set to 0, the staining intensity score of light yellow is set to 1, the staining intensity score of brown is set to 2, and the staining intensity score of tan is set to 3.

5. The automatic scoring and classification method for progressive renal cancer based on DCCD subtype markers according to claim 4, characterized in that: In step 300, the immunohistochemical staining images corresponding to each renal clear cell carcinoma tissue sample are subjected to machine learning using Image J or Fiji software. The positive cells in the immunohistochemical staining images corresponding to each renal clear cell carcinoma tissue sample were calculated respectively, and the positive cell proportion scores corresponding to multiple sections of each renal clear cell carcinoma tissue sample were obtained. The average of the positive cell proportion scores of multiple sections of each renal clear cell carcinoma tissue sample was used as the positive cell proportion score of the tumor tissue sample.

6. The automatic scoring and classification method for advanced renal cancer based on DCCD subtype markers according to claim 5, characterized in that: When the positive cell proportion is 0-5%, the positive cell proportion score is 0; When the positive cell ratio is 6%-25%, the positive cell ratio score is 1; When the positive cell percentage is 26%-50%, the positive cell percentage score is 2; When the positive cell percentage is 51%-75%, the positive cell percentage score is 3; When the positive cell ratio was greater than 75%, the positive cell ratio score was 4.

7. The automatic scoring and classification method for progressive renal cancer based on DCCD subtype markers according to claim 6, characterized in that: In the step 400, the IRS scores of the immunohistochemical images corresponding to the immunohistochemical staining with TGFBI, PLOD2 and SLC2A1 antibodies in each renal clear cell carcinoma tissue sample are calculated respectively; Calculate the sum of the IRS scores of all immunohistochemical images corresponding to TGFBI, PLOD2, and SLC2A1 antibody immunohistochemical staining in each renal clear cell carcinoma tissue sample, i.e., the DCCD score; Based on the numerical range of the DCCD score, renal clear cell carcinoma tissue samples were classified as advanced renal clear cell carcinoma patient samples or non-advanced renal clear cell carcinoma samples.

8. The automatic scoring and classification method for progressive renal cancer based on DCCD subtype markers according to claim 7, characterized in that: The renal clear cell carcinoma tissue samples with a DCCD score range higher than 21 (16, 25) were regarded as samples of patients with advanced renal clear cell carcinoma; The renal clear cell carcinoma tissue samples with a DCCD score of less than 9 (6, 12) were considered as non-progressive renal clear cell carcinoma patient samples.

Images