CN111793689A - Molecular marker related to rectal cancer and application thereof - Google Patents

Abstract

The invention relates to a molecular marker related to rectal cancer and application thereof, belonging to the technical field of medical molecular biology. Two key designs are carried out in the invention, namely, the full-exon large panel of 425 cancer-related genes is applied to carry out ctDNA detection instead of the small panel of a few hot spot genes, so that the correlation between single gene mutation and nCRT curative effect is obtained, and the correlation between gene variation of a plurality of signal paths and the nCRT curative effect is also obtained. Secondly, a plurality of monitoring time points (4 time points) are set before the operation, 425 gene large panels are combined, and the elimination of the mutation in the ctDNA and the dynamic change of the acquired mutation are monitored. Demonstrates the value of ctDNA dynamic monitoring in nCRT efficacy prediction and presents new insights into patient selection for W & W strategies. Demonstrating the role of ctDNA detection in early prediction of the prognosis of LARC patients.

Description

Molecular marker related to rectal cancer and application thereof

Technical Field

The invention relates to a molecular marker related to rectal cancer and application thereof, belonging to the technical field of medical molecular biology.

Background

Colorectal cancer (CRC) is the third most common cancer and fourth most cancer-related cause of death worldwide. Rectal cancer accounts for approximately 30% of CRC, and Locally Advanced Rectal Cancer (LARC) accounts for 50% of all rectal cancers. Currently, neoadjuvant chemoradiotherapy (nCRT) and total rectal resection (TME) in combination with adjuvant chemotherapy are the standard treatment for LARC recommended by the National Comprehensive Cancer Network (NCCN) guidelines. Complete remission of pathology (pCR) can be achieved in approximately 10-35% of nrt patients at the time of surgery. pCR patients have better prognosis, less local recurrence and distant metastasis, and high 5-year overall survival rate. However, traditional radical surgery can lead to serious complications and long term negative effects on the intestinal, urinary and sexual function. Thus, a non-surgical treatment method known as "watch and wait" (W & W) has been widely used for patients who experience complete clinical remission (cCR) after nrct treatment. Currently, cCR evaluation criteria include endoscopic, MRI and digital examination with no residual tumor found. It is clinically desirable to judge pCR by cCR, but unfortunately there is only a 30% to 50% match between the two. In addition, a recent study showed that 5-year overall survival and disease-free survival (DFS) for patients receiving W & W treatment was lower than for patients with TME that demonstrated pCR. In this case, pCR prediction outside of clinical evaluation would help to better select patients using the W & W method.

Circulating tumor dna (ctDNA) has been used as a predictor of the efficacy of chemotherapy for metastatic colorectal cancer, but little research has been done into the role of ctDNA in predicting the pathological response of the nrct in LARC patients. Furthermore, ctDNA has shown its value in the detection of early relapse after adjuvant chemotherapy in colon cancer patients, but the potential role of ctDNA in risk stratification and treatment strategy guidance after ncr treatment in LARC patients has not been fully evaluated.

Disclosure of Invention

The invention proves the feasibility and effectiveness of dynamic monitoring of ctDNA in the curative effect evaluation and prognosis prediction of the novel auxiliary radiotherapy and chemotherapy (nCRT) of the Locally Advanced Rectal Cancer (LARC). We have made two key designs to our study, the first being the use of full-exon large panels of 425 cancer-associated genes instead of small panels of few hot spot genes for ctDNA detection, which resulted in not only the correlation of single gene mutations with nrct efficacy, but also the correlation of gene mutations for multiple signaling pathways with nrct efficacy. Secondly, a plurality of monitoring time points (4 time points) are set before the operation, 425 gene large panels are combined, and the elimination of the mutation in the ctDNA and the dynamic change of the acquired mutation are monitored. Demonstrates the value of ctDNA dynamic monitoring in nCRT efficacy prediction and presents new insights into patient selection for W & W strategies. We also demonstrated the potential role of ctDNA detection in early prediction of the prognosis of LARC patients.

In a first aspect of the present invention, there is provided:

use of a mutation detection reagent for POLD1 and/or FAT1 gene in preparing a reagent for predicting the curative effect of a locally advanced rectal cancer patient after treatment.

In one embodiment, the treatment is nrct treatment.

In one embodiment, the prediction of therapeutic effect is a distinction between complete pathological remission (pCR) or incomplete pathological remission (non-pCR).

In one embodiment, the mutation detection reagent is used for detecting the frequency of gene mutation in plasma at time 1; time 1 is referred to as nrct treatment.

In one embodiment, the nrct treatment can be with capecitabine in combination with irinotecan.

In a second aspect of the present invention, there is provided:

use of a mutation detection reagent for Adhesion Junction (AJ), Histone Methyltransferase (HM), DNA Damage Repair (DR) and/or Osteoclast Differentiation (OD) signaling pathway-related genes for the preparation of a reagent for predicting the post-treatment therapeutic effect of a locally advanced rectal cancer patient.

In one embodiment, the treatment is nrct treatment.

In one embodiment, the prediction of therapeutic effect is a distinction between complete pathological remission (pCR) or incomplete pathological remission (non-pCR).

In one embodiment, the mutation detection reagent is used for detecting the frequency of gene mutation in plasma at time 1; time 1 is referred to as nrct treatment.

In one embodiment, the nrct treatment can be with capecitabine in combination with irinotecan.

In a third aspect of the present invention, there is provided:

use of a reagent for detecting mutations in ctDNA in plasma for the preparation of a prognostic agent for the post-treatment effect in patients with locally advanced rectal cancer.

In one embodiment, the prediction of therapeutic effect is a prediction of Tumor Regression Grade (TRG).

In one embodiment, the treatment is nrct treatment.

In one embodiment, the test agent is for the detection of the frequency of gene mutations or mutations at the highest variant allele frequency in plasma at time 1, time2, time 3, time 4; the time 1 refers to: before nCRT treatment; the time2 refers to: at time 15 of nrct radiotherapy; the time 3 refers to: nrct radiation therapy at time 25; the time 4 refers to: TME preoperative 0-1 d; the time 5 refers to: TME 5-12d after operation.

In one embodiment, the application further comprises: the zero clearing rate of the ctDNA is judged by any one of the following two judgment methods:

(1) determining whether the mutation with the highest Variant Allele Frequency (VAF) in the plasma ctDNA detected at time 1 is cleared at time2, 3 or 4;

(2) it was determined whether all detected mutations in plasma ctDNA detected at time 1 were cleared at time2, 3, or 4.

In a fourth aspect of the present invention, there is provided:

use of a reagent for detecting mutations in TP53, APC or KRAS genes for the preparation of a prognostic reagent for disease-free survival (DFS) following treatment in patients with locally advanced rectal cancer.

In one embodiment, the detection reagent is for detecting a gene mutation in plasma at time 1; time 1 is referred to as nrct treatment.

In one embodiment, the disease-free survival is disease-free survival following combination therapy of neoadjuvant chemoradiotherapy (nCRT) and total rectal resection (TME) in combination with adjuvant chemotherapy.

In a fifth aspect of the present invention, there is provided:

use of a reagent for detecting all detected mutations of plasma ctDNA for the preparation of a Disease Free Survival (DFS) prognostic reagent for locally advanced rectal cancer.

In one embodiment, the reagents for detecting all detected mutations in plasma ctDNA are used for detection at time points 1, 2, 3, 4, and 5, respectively; the application also comprises: determining whether all detected mutations in plasma ctDNA at time 1 are cleared at times 2, 3, 4, or 5; the time 1 refers to: before nCRT treatment; the time2 refers to: at time 15 of nrct radiotherapy; the time 3 refers to: nrct radiation therapy at time 25; the time 4 refers to: TME preoperative 0-1 d; the time 5 refers to: TME 5-12d after operation.

In a sixth aspect of the present invention, there is provided:

use of a reagent for detecting the highest VAF mutation in plasma ctDNA detected at time 1 for the preparation of a disease-free survival (DFS) prognostic reagent for localized late colorectal cancer.

In one embodiment, the application comprises: obtaining plasma ctDNA samples at time 1, 2 and detecting mutations, determining whether the highest VAF mutation in baseline plasma ctDNA is cleared at time 2; the time 1 refers to: before nCRT treatment; the time2 refers to: at time 15 of nrct radiotherapy; the time 3 refers to: nrct radiation therapy at time 25; the time 4 refers to: TME preoperative 0-1 d; the time 5 refers to: TME 5-12d after operation.

In a seventh aspect of the present invention, there is provided:

a pairAn apparatus for predicting the effectiveness of a treatment for a patient with locally advanced rectal cancer, comprising:

the plasma ctDNA extraction module is used for extracting ctDNA of patient plasma;

the sequencing module is used for sequencing the ctDNA obtained by the plasma ctDNA extraction module;

and the gene mutation analysis module is used for counting the gene mutation in the sample obtained by the plasma ctDNA extraction module.

In one embodiment, further comprising: the zero clearing rate judging module is used for judging the zero clearing rate of the ctDNA and comprises any one of the following two judging methods: (1) determining whether the mutation with the highest Variant Allele Frequency (VAF) in the plasma ctDNA detected at time 1 is cleared at time2, 3 or 4; (2) judging whether all detected mutations in the plasma ctDNA detected at the time 1 are cleared at the time2, 3 or 4; the time 1 refers to: before nCRT treatment; the time2 refers to: at time 15 of nrct radiotherapy; the time 3 refers to: nrct radiation therapy at time 25; the time 4 refers to: TME preoperative 0-1 d; the time 5 refers to: TME 5-12d after operation.

In an eighth aspect of the present invention, there is provided:

a computer readable medium is describedTo pairA computer program for a method of predicting the efficacy of a treatment for a patient with locally advanced rectal cancer, said program comprising the steps of:

obtaining a gene mutation in plasma ctDNA of the patient;

patients were judged to be either in complete pathological remission (pCR) or incomplete pathological remission (non-pCR) after nrct treatment based on gene mutations.

In one embodiment, said obtaining a genetic mutation in plasma ctDNA of a patient refers to obtaining a genetic mutation in cfDNA in plasma of a patient at time 1, time2, time 3, time 4, or time 5.

In one embodiment, further comprising: and (3) judging the zero clearing rate of the ctDNA, and predicting the Tumor Regression Grade (TRG) after the patient is treated or predicting the disease-free survival (DFS) according to the judgment result of the zero clearing rate of the ctDNA.

Drawings

FIG. 1: and (3) analyzing the correlation between the detection rate of the baseline plasma ctDNA mutation and the curative effect and prognosis of nCRT. (A) Comparison of rates of detection of baseline plasma ctDNA mutations in pCR and non-pCR patients. (B) Comparison of baseline plasma ctDNA mutation detection rates for different TRG fractionated patients. (C) Correlation analysis of baseline plasma ctDNA detection with disease-free survival (DFS).

FIG. 2: correlation analysis of baseline molecular characteristics with nrct efficacy (TRG grading). (A) The condition of high frequency somatic genetic variation in baseline plasma of LARC patients. (B) Different TRGs rank the proportion of patients with mutations in APC, TP53, POLD1 and FAT 1. The Y-axis represents the proportion of patients carrying a gene mutation to the total number of patients in the corresponding TRG group. (C) Different TRGs grade the proportion of gene mutations of 4 KEGG signaling pathways in the patient (HD: homologous recombination; HM: histone methyltransferase; OD: osteoclast differentiation; AJ: adhesive ligation). The ratio is defined similarly to (B).

FIG. 3: correlation analysis of baseline molecular characteristics with nCRT efficacy (pCR/non-pCR).

FIG. 4: correlation analysis between ctDNA dynamics and nCRT efficacy (TRG grading). (A) Time point 2/3/4 toggles the clear state. Time2_ highest _ non _ clean: the highest VAF mutation was not cleared at time point 2 baseline. Time234_ highest _ non _ clean: of the 2/3/4 time points, the baseline highest VAF mutation was not cleared, i.e., the baseline highest VAF mutation was detected at least at one time point. Time2_ all _ non _ clean: at least one baseline detected mutation at time point 2 is not cleared (not limited to the highest VAF mutation). Time234_ all _ non _ clean: at least one of the time points 2/3/4 is a mutation detected. "highest" refers to the mutation with the highest frequency of variant alleles; "all" refers to all mutations detected at baseline. (B) Distribution of acquired mutations in different TRG grades and pCR/non-pCR grouped patients. (C) The proportion of patients with the above characteristics in different TRG groups.

FIG. 5: baseline molecular mutation characteristics and correlation analysis of pre-operative ctDNA dynamics with disease-free survival (DFS). (A-D) Kaplan-Meier disease-free survival curve analysis was performed based on pCR status (pCR or non-pCR), time point 2-baseline maximum VAF mutation clear status, detection status of TP53 mutation in baseline plasma (TP53_ wt: no TP53 mutation detected; TP53_ mut: TP53 mutation detected), and lymph node metastasis found in pathological examination after operation (lymph node 0: lymph node metastasis negative; lymph node 1: lymph node metastasis positive), respectively.

FIG. 6: baseline molecular mutation characteristics and correlation analysis of pre-operative ctDNA dynamics with disease-free survival (DFS). (E-G) Kaplan-Meier disease-free survival curves stratified by time point 2 at baseline with the highest VAF mutation clear status based on pCR status, detected status of TP53 mutation in baseline plasma, and lymph node metastasis status.

FIG. 7: correlation analysis between APC and KRAS mutations and disease-free survival (DFS) in baseline plasma.

FIG. 8: correlation analysis of pre/post-operative pathological features with disease-free survival (DFS). "═ 0" indicates negative; "═ 1" indicates positive. cMRF: nrct treats anterior rectal fascia. cEVMI: nrct treats anterior wall external vessel invasion.

FIG. 9: correlation analysis between all baseline detected mutation zeroes and disease-free survival (DFS) at dynamic monitoring points. "All clear" means that All mutations were cleared at baseline. "═ 0" means clear without mutation. "═ 1" means mutation clearing.

Detailed Description

To demonstrate the feasibility and effectiveness of ctDNA dynamic monitoring in the assessment of efficacy of neoadjuvant chemoradiotherapy (nCRT) in locally advanced colorectal cancer (LARC) and prognosis prediction. We have made two key designs to our study, the first being the use of full-exon large panels of 425 cancer-associated genes instead of small panels of few hot spot genes for ctDNA detection, which resulted in not only the correlation of single gene mutations with nrct efficacy, but also the correlation of gene variation for multiple signaling pathways with nrct efficacy. Secondly, a plurality of monitoring time points (4 time points) are set before the operation, 425 gene large panels are combined, and meanwhile, the elimination of the mutation in the ctDNA and the dynamic change of the acquired mutation are monitored. Demonstrates the value of ctDNA dynamic monitoring in nCRT efficacy prediction and presents new insights into patient selection for W & W strategies. The present invention also demonstrates the potential role of ctDNA detection in early prediction of the prognosis of LARC patients.

The 425 gene panel used for detection in the present invention is composed of the genes shown below:

ABCB1、ABCB4、ABCC2、ADH1A、ADH1B、ADH1C、AIP、AKT1、AKT2、AKT3、ALDH2、 ALK、AMER1、APC、AR、ARAF、ARID1A、ARID1B、ARID2、ARID5B、ASCL4、ASXL1、 ATF1、ATIC、ATM、ATR、ATRX、AURKA、AURKB、AXIN2、AXL、B2M、BAD、BAI3、 BAK1、BAP1、BARD1、BAX、BCL2、BCL2L11、BCR、BIRC3、BLM、BMPR1A、BRAF、 BRCA1、BRCA2、BRD4、BRIP1、BTG2、BTK、BUB1B、c11orf30、CASP8、CBL、CBLB、 CCND1、CCNE1、CD274、CD74、CDA、CDC73、CDH1、CDK10、CDK12、CDK4、CDK6、CDK8、CDKN1A、CDKN1B、CDKN1C、CDKN2A、CDKN2B、CDKN2C、CEBPA、CEP57、 CHD4、CHEK1、CHEK2、CREBBP、CRKL、CSF1R、CTCF、CTLA4、CTNNB1、CUL3、 CUX1、CXCR4、CYLD、CYP19A1、CYP2A13、CYP2A6、CYP2A7、CYP2B6*6、CYP2C19*2、 CYP2C9*3、CYP2D6、CYP3A4*4、CYP3A5、DAXX、DDR2、DENND1A、DHFR、DICER1、 DLL3、DNMT3A、DPYD、DUSP2、EGFR、EML4、EP300、EPAS1、EPCAM、EPHA2、 EPHA3、EPHA5、EPHB2、ERBB2、ERBB2IP、ERBB3、ERBB4、ERCC1、ERCC2、ERCC3、 ERCC4、ERCC5、ESR1、ETV1、ETV4、ETV6、EWSR1、EXT1、EXT2、EZH2、FANCA、 FANCC、FANCD2、FANCE、FANCF、FANCG、FANCI、FANCL、FANCM、FAT1、FBXW7、 FGF19、FGFR1、FGFR2、FGFR3、FGFR4、FH、FLCN、FLT1、FLT3、FLT4、FOXA1、FOXP1、 FRG1、GATA1、GATA2、GATA3、GATA4、GATA6、GNA11、GNAQ、GNAS、GRIN2A、 GRM3、GRM8、GSTM1、GSTM4、GSTM5、GSTP1、GSTT1、HDAC2、HDAC9、HGF、HLA-A、HNF1A、HNF1B、HRAS、HSD3B1、IDH1、IDH2、IFNG、IFNGR1、IGF1R、IGF2、 IKBKE、IKZF1、IL7R、INPP4B、IRF2、JAK1、JAK2、JAK3、JARID2、JUN、KDM5A、 KDM6A、KDR、KEAP1、KIF1B、KIF5B、KIT、KITLG、KLLN、KMT2A、KMT2B、KMT2C、 KMT2D、KRAS、LHCGR、LMO1、LRP1B、LYN、LZTR1、MAP2K1、MAP2K2、MAP2K4、 MAP3K1、MAP3K4、MAP4K3、MAX、MCL1、MDM2、MDM4、MECOM、MED12、MEF2B、 MEN1、MET、MGMT、MITF、MLH1、MLH3、MLLT1、MLLT3、MLLT4、MPL、MRE11A、 MSH2、MSH6、MTHFR、MTOR、MUTYH、MYC、MYCL、MYCN、MYD88、MYH9、NAT1、NBN、NCOR1、NF1、NF2、NFE2L2、NFKBIA、NKX2-1、NKX2-4、NOTCH1、 NOTCH2、NOTCH3、NPM1、NQO1、NRAS、NRG1、NSD1、NTRK1、NTRK2、NTRK3、 PAK3、PALB2、PALLD、PARK2、PARP1、PARP2、PAX5、PBRM1、PDCD1、PDCD1LG2、PDE11A、PDGFRA、PDGFRB、PDK1、PGR、PHOX2B、PIK3C3、PIK3CA、PIK3R1、PIK3R2、 PKHD1、PLAG1、PLK1、PMS1、PMS2、POLD1、POLD3、POLE、POLH、POT1、PPARD、 PPP2R1A、PRDM1、PRF1、PRKACA、PRKACG、PRKAR1A、PRKCI、PRKDC、PRSS1、 PRSS3、PTCH1、PTEN、PTK2、PTPN11、PTPN13、PTPRD、QKI、RAC1、RAC3、RAD50、 RAD51、RAD51B、RAD51C、RAD51D、RAD54L、RAF1、RARA、RARG、RASGEF1A、 RB1、RECQL4、RELN、RET、RHOA、RICTOR、RNF43、ROS1、RPTOR、RRM1、RUNX1、RUNX1T1、SBDS、SDC4、SDHA、SDHB、SDHC、SDHD、SEPT9、SETBP1、SETD2、SF3B1、 SGK1、SLC34A2、SLC3A2、SLC7A8、SMAD2、SMAD3、SMAD4、SMAD7、SMARCA4、 SMARCB1、SMO、SOS1、SOX1、SOX14、SOX2、SOX21、SPOP、SPRY4、SRC、SRY、 STAG2、STAT3、STK11、STMN1、STT3A、SUFU、TAP1、TAP2、TEK、TEKT4、TERC、 TERT、TET2、TGFBR2、THADA、TMEM127、TMPRSS2、TNFAIP3、TNFRSF11A、TNFRSF14、 TNFRSF19、TNFSF11、TOP1、TOP2A、TP53、TP63、TPMT、TSC1、TSC2、TSHR、TTF1、TUBB3、TUBB4A、TUBB4B、TUBB6、TYMS、U2AF1、UGT1A1、VAMP2、VEGFA、VHL、 WAS、WISP3、WRN、WT1、XPA、XPC、XRCC1、YAP1、ZNF2、ZNF217、ZNF703；

patient and sample collection

Based on a randomized controlled trial (Cinclar, NCT02605265), the department of radiation oncology, subsidiary Shanghai tumor center, university of Compound Dane, received 119 LARC patients (cT3-4/N0-2, M0) in total from 2016, month 2, 7, to 2017, month 10, 31. Patients received nCRT (50gy/25 fraction; capecitabine in combination with irinotecan regimen) and one cycle of interval chemotherapy (capecitabine in combination with irinotecan regimen) followed by total rectal resection (TME) and adjuvant chemotherapy (capecitabine in combination with oxaliplatin regimen). Plasma sample collection time was: pre-nrct treatment (time 1), 15 th (time 2) and 25 th (time 3) nrct radiation therapy, 0-1d pre-surgery (time 4) and 5-12d post-surgery (time 5). Post-operative rectal cancer tissue specimens were evaluated for tumor response according to the 2010 united states committee for cancer council (AJCC) TRG grading criteria, including pCR and Tumor Regression Grading (TRG). TRG0 was complete remission (pCR), with no residual tumor cells; TRG 1-3 belongs to non-pCR, TRG 1 is a good reaction, and only single or few tumor cells remain; TRG 2 is a tumor cell which has small tumor reaction and remains; TRG3 was poorly responsive to tumor, with little or no tumor cells killed.

High depth targeted sequencing of collected plasma samples was performed using 425 cancer-associated genes panel, with an average sequencing depth of about 4000X. Patients with mutations detected at baseline plasma and complete samples at 5 time points were analyzed for both mutation clearance and acquired mutations, and patients with no mutations detected at baseline were analyzed for only acquired mutations. Mutation clearance was analyzed simultaneously for the highest Variant Allele Frequency (VAF) mutation in baseline plasma and for changes in all detected mutations in baseline plasma at 2, 3, 4, 5 monitoring points. To reduce potential false positives, mutations that were not detected at baseline but were detected at 2, 3, 4, 5, at least 2 time points were defined as acquired mutations. The somatic mutation in the present invention includes two cases, point mutation and insertion deletion mutation; "Point mutation" refers to a mutation caused by a single base substitution, resulting in a change in the encoded amino acid; "indel mutations" means that one or more base insertions or deletions result in an increase or decrease in the encoded amino acid, and these types of mutations may be "in-frame" in the coding sequence of a protein, resulting in the addition or decrease of amino acids in the protein; or may result in a "frameshift", typically leading to premature truncation of the protein;

statistical analysis

The Fisher's exact test and the Cochran-Armitage test were used to compare the proportion of patients with certain clinical or genetic characteristics (e.g., genetic mutations in signaling pathways, clearance of genetic mutations, acquired mutation status, etc.) in different TRG groups.

And evaluating the risk factors influencing mutation zero clearing in the nCRT treatment process by adopting single-factor logistic regression analysis. Survival analyses included the Kaplan-Meier method and univariate or multivariate Cox proportional Risk model, using the "survivval" and "survivor" R packages for analysis. The pCR prediction model was constructed using the R "caret" software package. pCR predictors included gene features such as TP53, POLD1, FAT1 mutational status and mutational status of 4 KEGG signal pathways; status of acquired mutations and zero clearing of baseline mutations.

Patient characteristics

The median age of 119 patients was 57 years, of which 71% were male. The clinical stages of most patients are IIIB (66%) or IIIC (31%). 32% and 25% of patients are positive for exovenous invasion (EVMI) and positive for mesenteric fascia invasion (MRF). Postoperative pathology examination revealed 41 patients (34.5%) as pCR and 78 (65.5%) as non-pCR. Patients graded by TRG were 41 (34.5%), 12 (10.1%), 53 (44.5%) and 13 (10.9%) in grade 0 (pCR), grade 1, grade 2 and grade 3, respectively. The clinical characteristics were not statistically different when the pCR group was compared with non-pCR group.

Analysis of correlation between LARC patient baseline genomic features and TRG

Somatic mutations were detected in baseline plasma (plasma measured at time 1) in 100 of 119 patients (84%). Whether a mutation was detected in the baseline ctDNA was not associated with nrct treatment response or DFS (fig. 1). The most common mutant genes in LARC are TP53, APC and KRAS, other genes with higher mutation frequencies include KMT2B, NOTCH1 and POLD1 (region a of fig. 2). We found that the mutation frequencies of genes were different between the pCR group and non-pCR group, or different TRG groups. The frequencies of mutations of POLD1 and FAT1 genes in the pCR group were significantly higher than those in the non-pCR group (both p ═ 0.05, region B in fig. 2 and region a in fig. 3). The mutation frequencies of TP53 and APC genes increased from TRG0 to TRG3 (p 0.08 for TP53 and p 0.09 for APC, region B in fig. 2). In addition, the mutation frequency of TP53 and APC gene was higher in the non-pCR group than in the pCR group, but the difference was not significant (region A of FIG. 3). Next, we extend the analysis to the signal path level. We found changes in 4 KEGG signaling pathways, including Adhesion Junction (AJ), Histone Methyltransferase (HM), DNA Damage Repair (DR), and Osteoclast Differentiation (OD), that were associated with the patient's therapeutic response to nrct (mutant genes belonging to 4 signaling pathways are shown in table 1). The gene mutation frequency of all 4 signaling pathways decreased from TRG0 to TRG3, and particularly the DR pathway differed significantly (p ═ 0.005, region C in fig. 2). Also all 4 signal pathways mutated more frequently in the pCR group than in the non-pCR group (p <0.05, FIG. 3B region).

correlation of ctDNA dynamics with nCRT therapeutic response

It is hypothesized that clearance of baseline ctDNA mutations following nrct treatment may reflect patient response to treatment. The percentage of patients with ctDNA mutation clear ( time point 2 or 3 preoperative time points) was assessed separately using both the assessment method of the mutation with the highest Variant Allele Frequency (VAF) detected at baseline and all mutations detected at baseline, and both showed a downward trend from TRG0 to TRG3 (p <0.05, regions a and C of fig. 4). Among them, all mutations at 3 Time points were cleared (Time234_ all _ clear) with the most significant difference (p 0.001), and the clearing rate decreased from 87% in TRG0 to 33% in TRG3 group (region C in fig. 4). Previous ctDNA studies have focused mainly on the zeroing of baseline mutations, while little is known about acquired mutations during nrct treatment. We observed some acquired gene mutations such as TP53, PARP2, ESR1, CDK12, CHEK2, RECQL 4. Interestingly, the proportion of patients with acquired mutations increased gradually from TRG0 to TRG3 (p ═ 0.04, regions B and C of fig. 4). These results indicate that non-pCR patients not only have a low clearance rate of baseline mutations, but also acquire more mutations during nCRT treatment.

ctDNA monitoring as an early biomarker for predicting disease-free progression (DFS)

The invention further evaluates the potential of ctDNA monitoring for early prediction of DFS. TP53, APC and KRAS gene mutations were detected as risk factors for poor DFS in baseline plasma (TP53, APC and KRAS p ═ 0.00053, 0.03 and 0.02, respectively; log rank test, panel C of FIG. 5, FIG. 7). pCR patients had better DFS than non-pCR patients (p 0.0025, panel a of figure 5), with poorer DFS for higher TRG scores (p 0.018, panel F of figure 8). Some pathological features, such as lymph node metastasis (p 0.00049, area D of fig. 5), peri-neural infiltration (p 0.014, area a of fig. 8), tumor deposition (p 0.0016, area B of fig. 8) and vascular infiltration (p 0.003, area C of fig. 8) are associated with worsening DFS.

With respect to ctDNA dynamics, clearing mutations in baseline plasma predicts better DFS in time 2-5 compared to ctDNA mutation non-cleared patients (fig. 5B, fig. 6, and fig. 9). Of the 4 time points, the time2_ baseline highest VAF mutant clear was most significantly correlated with the better DFS (p 0.02, region B of fig. 5). Importantly, time2_ baseline maximum VAF mutation zero clearing is a favorable factor independent of pCR, TP53 mutation, and lymph node metastatic status (E-G region of fig. 6).

Effect of mutation nullification and acquired mutations in ctDNA on treatment sensitivity and drug resistance

We also analyzed the zero clearing rate of the mutations in the different signal paths in the baseline, with a total zero clearing rate of 85.6%. Mutations in some of these signaling pathways are more difficult to clear, such as TP53 and WNT in the cell cycle signaling pathway. In contrast, the non-zero rate of all 4 signal paths (DR, HM, AJ, OD) is lower than the total non-zero rate. In addition, the non-zero clearing rate of DNA repair (including DR genes) and SWI/SNF (chromatin remodeling) -associated mutations is low. The SWI/SNF signaling pathway interacts with Histone Methyltransferase (HM), playing a role in chromatin remodeling. This is consistent with our previous findings that changes in DR and HM signal pathways at baseline correlate with a higher proportion of pCR. These results strongly suggest that aberrant DNA repair and histone remodeling mechanisms in LARC may be sensitive factors for nrct.

The W & W strategy is a clinically feasible treatment strategy for cCR patients after the new adjuvant radiotherapy and chemotherapy of colorectal cancer. However, the lack of patient-selected criteria and the consistent definition of cCR led to inconsistent results from various studies involving the W & W strategy. Furthermore, the cCR-based W & W approach is considered to be inferior to the prognosis of patients undergoing surgery. Theoretically, if pCR can be accurately predicted and the choice of W & W patients is guided, the risk of relapse in W & W patients can be reduced and the prognosis improved. At present, the prediction of pCR is mainly dependent on clinical variables and imaging examination. Several individual molecular biomarkers, such as H2AX, were also used for pCR prediction. However, single markers have limited sensitivity, specificity and accuracy for predicting pCR, and the complexity of the assay also limits its clinical application.

In our study, we found several new genomic features that were significantly different between the pCR and non-pCR groups. Baseline mutations at baseline TP53 were found to be associated with adverse treatment responses and DFS as in this study. Previous studies have also shown that the mutation TP53 is an independent predictor of adverse effects of radiation or chemotherapy for rectal cancer. In contrast, FAT1 and POLD1 mutations occurred at a higher rate in the better-responding group. A recent study found that in patients with squamous cell carcinoma of the head and neck, the mutation in FAT1 was significantly enriched in cisplatin-responsive patients compared to non-responsive patients. POLD1 is a member of the DR pathway, and patients with DR pathway mutations are significantly enriched in the pCR group. A large number of studies have confirmed that DR deficiency can improve the response of radiotherapy and chemotherapy, and treatment strategies based on this have been clinically applied. Changes in 3 other KEGG signal pathways in addition to the DR pathway: HM, AJ and OD are also associated with better therapeutic response. In vitro and in vivo studies have shown that these signaling pathways affect the sensitivity of radiotherapy and chemotherapy. We also found that the 4 signal paths all have higher abrupt zero clearing rates than the average. All these results indicate that changes in the genes and signaling pathways described above at baseline inherently provide tumor cell sensitivity to radiation or chemotherapy. In addition to baseline genomic characteristics, the clearing of baseline mutations during nrct treatment also reflects sensitivity to treatment.

None of the baseline ctDNA mutations detected or not found in this study were correlated with treatment response or post-operative DFS, suggesting that residual tumors, but not primary tumors, were the source of recurrence. As expected, clearing the baseline detection mutation at all time points predicts better DFS, with mutation clearing at time2 being most significant, indicating that patient sensitivity to nrct is manifested at an early stage of treatment, consistent with the observations from other published studies. Non-clearing suggests that the patient has some malignant pathological features, such as perineural infiltration or tumor deposition that prevents ctDNA from clearing. Therefore, time2 ctDNA clearing can be an early detection indicator for high risk patients, and earlier more aggressive intervention can improve the prognosis for these patients. We observed that detection of the TP53 mutation at baseline and the acquired TP53 mutation at other time points predicts a high risk of progression. The relationship between TP53 gene mutation and colorectal cancer patient prognosis has been widely reported. Our analysis showed that several features were available preoperatively, including TP53 detected, KRAS and APC mutations in baseline ctDNA, baseline mutations that were not cleared, and males were risk factors for DFS. By combining the characteristics, high-risk people can be discovered as early as possible, so that the treatment scheme can be adjusted in time.

TABLE 1 basic characteristics of patients enrolled in this study

MRF:mesorectal fascia.EVMI:extramural vascular invasion.

TABLE 2 mutant genes in the Signal pathway

Claims (10)

1. Use of a mutation detection reagent for the POLD1 and/or FAT1 genes for the preparation of a reagent for predicting the therapeutic effect after treatment of a locally advanced rectal cancer patient.
2. 2. The use of claim 1, wherein in one embodiment, the treatment is nrct treatment;

   in one embodiment, the prediction of therapeutic effect is a distinction between complete pathological remission (pCR) or incomplete pathological remission (non-pCR);

   in one embodiment, the mutation detection reagent is used for detecting the frequency of gene mutation in plasma at time 1; the time 1 refers to before nCRT treatment;

   in one embodiment, the nrct treatment can be with capecitabine in combination with irinotecan.
3. 3. Use of a mutation detection reagent for Adhesion Junction (AJ), Histone Methyltransferase (HM), DNA Damage Repair (DR), and/or Osteoclast Differentiation (OD) signaling pathway-related genes for the preparation of a reagent for predicting the post-treatment therapeutic effect of a locally advanced rectal cancer patient;

   in one embodiment, the treatment is nrct treatment;

   in one embodiment, the prediction of therapeutic effect is a distinction between complete pathological remission (pCR) or incomplete pathological remission (non-pCR);

   in one embodiment, the mutation detection reagent is used for detecting the frequency of gene mutation in plasma at time 1; the time 1 refers to before nCRT treatment;

   in one embodiment, the nrct treatment can be with capecitabine in combination with irinotecan.
4. 4. Use of an agent for detecting mutations in ctDNA in plasma for the preparation of a Tumor Regression Grade (TRG) prognostic agent for differentiating locally advanced rectal cancer.
5. 5. The use of claim 4, wherein in one embodiment, the prediction of therapeutic effect is prediction of Tumor Regression Grade (TRG);

   in one embodiment, the treatment is nrct treatment;

   in one embodiment, the test agent is for the detection of the frequency of gene mutations or mutations at the highest variant allele frequency in plasma at time 1, time2, time 3, time 4; the time 1 refers to: before nCRT treatment; the time2 refers to: at time 15 of nrct radiotherapy; the time 3 refers to: nrct radiation therapy at time 25; the time 4 refers to: TME preoperative 0-1 d; the time 5 refers to: TME 5-12d after operation;

   in one embodiment, the application further comprises: the zero clearing rate of the ctDNA is judged by any one of the following two judgment methods:

   (1) determining whether the mutation with the highest Variant Allele Frequency (VAF) in the plasma ctDNA detected at time 1 is cleared at time2, 3 or 4;

   (2) it was determined whether all detected mutations in plasma ctDNA detected at time 1 were cleared at time2, 3, or 4.
6. 6. Use of a reagent for detecting mutations in TP53, APC or KRAS genes for the preparation of a prognostic reagent for disease-free survival (DFS) following treatment in patients with locally advanced rectal cancer.
7. 7. The use of claim 6, wherein the detection reagent is for detecting a gene mutation in plasma at time 1; the time 1 refers to before nCRT treatment;

   in one embodiment, the disease-free survival is disease-free survival following combination therapy of neoadjuvant chemoradiotherapy (nCRT) and total rectal resection (TME) in combination with adjuvant chemotherapy.
8. 8. Use of a reagent for detecting all detected mutations of plasma ctDNA for the preparation of a disease-free survival (DFS) prognostic reagent for locally advanced rectal cancer; in one embodiment, the reagents for detecting all detected mutations in plasma ctDNA are used for detection at time points 1, 2, 3, 4, and 5, respectively; the application also comprises: determining whether all detected mutations in plasma ctDNA at time 1 are cleared at times 2, 3, 4, or 5; the time 1 refers to: before nCRT treatment; the time2 refers to: at time 15 of nrct radiotherapy; the time 3 refers to: nrct radiation therapy at time 25; the time 4 refers to: TME preoperative 0-1 d; the time 5 refers to: TME 5-12d after operation.
9. 9. Use of a reagent for detecting the highest VAF mutation in plasma ctDNA detected at time 1 for the preparation of a disease-free survival (DFS) prognostic reagent for locally advanced rectal cancer; in one embodiment, the application comprises: obtaining plasma ctDNA samples at time 1, 2 and detecting mutations, determining whether the highest VAF mutation in baseline plasma ctDNA is cleared at time 2; the time 1 refers to: before nCRT treatment; the time2 refers to: at time 15 of nrct radiotherapy; the time 3 refers to: nrct radiation therapy at time 25; the time 4 refers to: TME preoperative 0-1 d; the time 5 refers to: TME 5-12d after operation.
10. 10. An apparatus for predicting the effectiveness of a treatment for a patient with locally advanced rectal cancer, comprising:

    the plasma ctDNA extraction module is used for extracting ctDNA of patient plasma;

    the sequencing module is used for sequencing the ctDNA obtained by the plasma ctDNA extraction module;

    and the gene mutation analysis module is used for counting the gene mutation in the sample obtained by the plasma ctDNA extraction module.

Images