CN112143810B

CN112143810B - Gene markers for predicting cancer immunotherapy effect and application thereof

Info

Publication number: CN112143810B
Application number: CN202011052653.6A
Authority: CN
Inventors: 唐海波; 李晓宁
Original assignee: TCM-INTEGRATED HOSPITAL OF SOUTHERN MEDICAL UNIVERSITY
Current assignee: TCM-INTEGRATED HOSPITAL OF SOUTHERN MEDICAL UNIVERSITY
Priority date: 2020-09-29
Filing date: 2020-09-29
Publication date: 2023-11-21
Anticipated expiration: 2040-09-29
Also published as: CN112143810A

Abstract

The invention discloses a group of gene markers for predicting the immune therapeutic effect of cancer and application thereof. Markers include BRD4, RPS6KA4, TET1, BCL10, FGFR2, RAD51C, TCF L2, TGFBR1, AR, BBC3, H3C11, TSC2, NSD1, PNRC1, IGF2, H3C3, SOCS1, TCF3, TSHR, PMS2, EGFR, CDKN2A. The model constructed by 22 genes in the invention can predict the immune therapeutic effect, better guide clinical medication, and has the predicted sensitivity and specificity superior to PD-L1 expression.

Description

Gene markers for predicting cancer immunotherapy effect and application thereof

Technical Field

The invention belongs to the technical field of biological medicines, and particularly relates to a group of gene markers for predicting the immune treatment effect of cancer and application thereof.

Background

Lung cancer is currently the most common malignancy, and common treatments include surgery, chemotherapy, targeting, and immunotherapy. Immunotherapy is a treatment method for comparing fire in recent years. But only 20-25% of patients benefit from immunotherapy, researchers have explored a variety of biological markers for screening the beneficiary population in order to more accurately predict and partition the best beneficiary population for immunotherapy.

The commonly used biological markers at present comprise PD-L1 expression, tumor mutation load (TMB), copy number variation, microsatellite instability, mismatch gene repair deletion and the like, wherein the PD-L1 is approved more, when the PD-L1 positive rate reaches more than 50%, the PD-1 immunotherapy effective rate is close to 50%, and the PD-L1 negative patient effective rate is only about 10%. Keynote-024 showed that PD-L1 positive patients (PD-L1. Gtoreq.50%) significantly improved the progression free survival and overall survival of patients over first-line standard chemotherapy, and based on these data NCCN guidelines (NCCN Guidelines Version 2.2020Non-Small Cell Lung Cancer) were clearly indicated from 2017 that detection of PD-L1 could be performed in patients with initial Non-small cell lung cancer (NSCLC) without significant driving gene mutations, and that if PD-L1 was > 50%, pamphlet Li Zhushan could be selected as first-line drug. This guideline has been the gold standard to date.

Due to the biological properties of PD-L1, PD-L1 also faces certain challenges as a biomarker. For example, tumors have heterogeneity, and PD-L1 expression is different between lesions at different locations on the same lesion. In addition, the expression of PD-L1 is inducible, dynamic, that is to say, different modes of treatment affect the expression of PD-L1 at different stages of treatment. The consistency of detecting the PD-L1 expression level among different platforms is poor, and the detection method is also under great dispute.

Following PD-L1, TMB gradually becomes a potential marker for predicting the efficacy of immune checkpoint therapy. As a broader spectrum of immunotherapy, clinical applications have been moving from clinical research hotspots and have been written in NCCN guidelines. However, TMB has a number of problems as a predictive marker, such as the cost of TMB detection is still higher than that of PD-L1 detection due to the reliance on gene sequencing. The detection standards of different platforms are not uniform, and the sample storage time has certain influence on the result judgment of TMB and the like.

Microsatellite highly unstable (MSI-H) and mismatch repair functional defects (dMMR) represent the results produced by two different detection methods, but they represent very similar clinical guidelines and MSI-H can be considered equivalent to dMMR. If a patient's tumor tissue lesion is detected as MSI-H, or dMMR, this means that the somatic cell is over mutated and the tumor expresses many new antigens, which make the tumor easily recognized by the immune system, thereby activating the opportunity to exert a killing function. Of colorectal cancer patients, approximately 10% are patients with MSH-H or dMMR. Nivolumab has been approved by the FDA for use in dMMR/MSI-H metastatic colorectal cancer (mCRC) patients following standard chemotherapy progression. MSI-H is not common in melanoma, lung cancer and squamous cell carcinoma.

The current various biomarkers have limitations and cannot completely and accurately predict the clinical efficacy and benefit of patients. Therefore, the exploration and development of biomarkers for more accurate and rapid prediction of cancer immunotherapy has important clinical significance.

Disclosure of Invention

A first object of the present invention is to provide, in view of the above drawbacks, a set of genetic markers for predicting the effect of cancer immunotherapy.

A second object of the present invention is to provide the use of a reagent for detecting mutation of the above gene marker for the preparation of a reagent for predicting the effect of immunotherapy of cancer.

A third object of the present invention is to provide a kit.

The technical scheme adopted by the invention is as follows:

in a first aspect of the invention, there is provided a set of gene markers for predicting the effect of cancer immunotherapy comprising the following two sets:

group A: BRD4, RPS6KA4, TET1, BCL10, FGFR2, RAD51C, TCF L2, TGFBR1, AR;

group B: BBC3, H3C11, TSC2, NSD1, PNRC1, IGF2, H3C3, SOCS1, TCF3, TSHR, PMS2, EGFR, CDKN2A.

The genetic marker according to the first aspect of the present invention, wherein the cancer is lung cancer.

Further, the genetic marker according to the first aspect of the present invention is characterized in that the cancer is lung adenocarcinoma.

According to the gene marker of the first aspect of the invention, the judgment standard for predicting the cancer immunotherapy effect is as follows:

a high support immunotherapeutic group if at least one gene mutation in group a or at least one gene mutation in group a is accompanied by at least one gene mutation in group B;

if the genes in the A group and the B group have no mutation, the genes are the middle support immunotherapy group;

if one or more of the genes in group B are mutated simultaneously, then the group is a low support immunotherapy group.

In a second aspect, the invention provides the use of an agent for detecting a mutation in a gene marker according to the first aspect of the invention in the preparation of an agent for predicting the effect of immunotherapy on cancer.

Preferably, according to the use of the second aspect of the invention, the reagent is a primer for detecting a mutation in the gene marker.

Preferably, according to the use of the second aspect of the invention, the reagents are primers and probes for detecting mutations in the gene markers.

Further, according to the use of the second aspect of the present invention, the reagent is a gene chip.

According to the application of the second aspect of the invention, the predicted cancer immunotherapy effect is judged by detecting the mutation condition of the gene marker according to the first aspect of the invention, and the judgment standard is as follows:

if the genes of the A group and the B group have no mutation, the genes are the middle support immunotherapy group;

if one or more of the genes of group B are mutated simultaneously, then the group is a low support immunotherapeutic.

According to a second aspect of the invention, the detection agent detects at the nucleic acid level.

According to a second aspect of the invention, the method for performing the detection agent comprises: polymerase chain reaction, denaturing gradient gel electrophoresis, whole exon sequencing, second generation sequencing, in situ hybridization, high performance liquid chromatography, biological mass spectrometry, gene chip detection, pyrosequencing, or single strand conformational isomerism polymorphism analysis techniques.

According to a second aspect of the invention, the reagent further comprises a sample processing reagent.

According to a second aspect of the invention, the sample is selected from fresh tissue, paraffin-embedded tissue, blood, serum, plasma, saliva, urine, cerebrospinal fluid, semen, sputum or a puncture fluid of a cancer patient.

In a third aspect of the invention there is provided a kit comprising reagents for detecting a gene marker according to the first aspect of the invention,

wherein, the mutation condition of the gene marker in the first aspect of the invention is detected to predict the cancer immunotherapy effect, and the judgment standard is as follows:

The beneficial effects of the invention are as follows:

the gene markers screened by the invention can predict the cancer immunotherapy effect, can efficiently screen out target crowd of immunotherapy, and has the advantages of rapid detection, accurate prediction result, low cost and the like. Compared with the two traditional methods of PD-L1 and TMB, the detection of the genetic variation state is more reliable, whether patients benefit and distinguish more detailed, the prediction efficiency is higher, the medicine taking can be better guided, and the application prospect is wide.

Drawings

FIG. 1 is a progression-free survival (PFS) comparison between the high, medium and low support immunotherapeutic groups.

FIG. 2 is the area under the ROC curve for the prediction of cancer immunotherapy effect using the construction model and PD-L1 expression as predictors, respectively, in example 1.

Detailed Description

The present invention will be described in further detail with reference to the following examples, which are only for the purpose of illustrating the invention and are not to be construed as limiting the scope of the invention.

PD-1: programmed death receptor-1 (programmed death 1).

PD-L1: programmed death ligand-1 (programmed death ligand 1).

TMB: tumor mutation load (Tumour mutation burden), which is the total number of somatic gene coding errors, base substitutions, gene insertions or deletions detected per million bases, is a quantitative biomarker reflecting the total number of mutations carried by tumor cells.

PFS: progression-free survival (progress-free survival), which refers to the time from the start of a randomized clinical trial to the progression of tumorigenesis (in any respect) or death due to any cause, is the time required for tumor regrowth under drug control.

OS: total Survival (Overall survivinal) refers to the time from the start of randomization to death due to any cause.

CR: complete remission (complete response), meaning that all target lesions disappeared, no new lesions appeared, and tumor markers were normal, for at least 4 weeks.

PR: partial remission (partial response), meaning that the sum of the maximum diameters of target lesions is reduced by 30% or more, for at least 4 weeks.

SD: disease stability (stable disease), which refers to the sum of the maximum diameters of target lesions, is reduced by less than PR, or increased by less than PD.

PD: disease progression (progressive disease) refers to at least an increase of 20% or more in the sum of the maximum diameters of the target lesions, or the appearance of new lesions.

ORR: objective remission rate (objective response rate), which refers to the proportion of patients with tumor shrinkage up to a certain amount and for a certain period of time, contains cases of CR and PR.

DCR: disease control rate (disease control rate), which refers to the proportion of patients whose tumors shrink or stabilize and remain for a certain period of time, includes cases of CR, PR and SD.

DCB: sustained clinical benefit (Durable Clinical Benefit), refers to complete remission CR/partial remission PR or stable disease SD lasting >6 months.

NDB: no sustained clinical benefit (No Durable Benefit), meaning sustained disease progression PD or SD less than or equal to6 months

AUC: the Area is the Area enclosed by the coordinate axis Under the ROC Curve. The ROC curve, which is a curve plotted on the ordinate with true positive rate (sensitivity) and false positive rate (1-specificity) on the abscissa, is fully referred to as the subject's working characteristic curve (receiver operating characteristic curve) according to a series of different classifications (demarcation values or decision thresholds). The closer the AUC is to 1.0, the higher the detection method authenticity; when the value is equal to 0.5, the authenticity is the lowest, and the application value is not provided.

Sensitivity: in the population where clinical treatment benefits (positive) by imaging judgment, the probability of benefit is predicted by predictors.

Specificity: in the population where clinical treatment does not benefit (negative) by imaging, the chance of not benefiting is predicted by a predictor.

Positive predictive value: the term "accuracy of a predicted positive result" refers to the proportion of the number of patients actually benefiting from the predicted factor test result, i.e. the percentage of the positive result actually benefiting from the predicted factor test result.

Negative predictive value: the term "accuracy of a predicted negative result" refers to the proportion of the number of cases that the predictor predicts that the actual immune therapy cannot benefit from, among the total number of cases that the predictor predicts that the predictor does not benefit from.

Example 1 target screening to predict the Effect of cancer immunotherapy

Samples were derived from 29 cases of advanced lung adenocarcinoma patients who received immunotherapy at the souvenir Stoneketeline cancer center in 2012-2013. The treatment scheme of the patient is as follows: the pembrolizumab is injected once every 2-3 weeks with the dosage of 10mg/kg. Tumor specimens were taken from primary foci, metastases and lymph nodes, including fresh tissue, paraffin-embedded tissue, and control groups were taken from the patient's blood. Tumor tissue was confirmed by HE staining prior to sequencing of tumor specimens.

DNA was extracted using the DNeasy kit (Qiagen) and then the exon sequences were enriched for capture using SureSelect Human All Exon v 2.0.2.0 (Agilent, 44 Mb) according to the instructions of the product. The enriched exome library was then bi-directionally sequenced using the Illumina HiSeq2000 sequencing system to an average sequencing depth of 150X. 376 target genes were then re-sequenced using an A mpliseq (Life Technologies) sequencing platform to a depth of 500X. Finally, the original sequencing data was aligned with the hg37 genomic sequence using a Burrows-Wheeler Aligner (BWA) V0.7.10. Raw data were quality controlled using Genome Analysis Toolkit (GATK) 3.2.2.

The evaluation of a gene mutation, including one of a point mutation and a fusion mutation, is defined as the gene mutation as long as it is Amplified in copy number (mainly comprising AMP: amplified, HOMDEL: homozygou sly Deleted). Insertion/deletion (insertion/deletion. Indel) identification was performed with VarScan v2.2.3 (52), strelka V1.0.13 (53) by comparing sequencing reads of tumors with matched normal genomes, genomic copy number variation (copy number variation, CNV) and Fusion gene identification was performed using GATK. Mutant annotation was then performed using snpefect v3.5d.

Out of 29 patients, 12 cases of patients can benefit from immunotherapy, 15 cases of patients cannot benefit from immunotherapy, and 22 genes most relevant to the effect of predicting cancer immunotherapy are screened by comparing the mutation status of genes in two groups of patients (clinical benefit group and clinical non-benefit group) and combining related documents, and detailed information of the genes is shown in table 1.

TABLE 1 22 Gene information summary tables

The model is constructed by combining the model quantification standard, and the genetic composition and algorithm of the model are as follows:

group a genes 9: BRD4, RPS6KA4, TET1, BCL10, FGFR2, RAD51C, TCF L2, TGFBR1, AR

Group B genes 13: BBC3, H3C11, TSC2, NSD1, PNRC1, IGF2, H3C3, SOCS1, TCF3, TSHR, PMS2, EGFR, CDKN2A.

Table 2 algorithm for modeling

Example 2 further validation of the build model in example 1

A total of 188 patients with advanced lung adenocarcinoma and receiving anti-PD- (L) 1 treatment participated in this study. The patient characteristics are shown in Table 3.

Table 3: clinical characterization of 188 patients

	N＝188(％)
		Middle position age (scope)	64.5(32-92)
Sex (sex)
		Female	103(54.79)
Man's body	85(45.21)
		Smoking status
At present or before	132(70.21)
		Previously or never	56(29.79)
Treatment timing
		1 line	34(18.09)
2-wire	91(48.40)
		3 or more lines	63(33.51)
Therapeutic response
		Partial mitigation	37
Disease stabilization	61
		Disease progression	90
Disease control rate	52.13％
		Complete remission, partial remission, disease stabilization	98
Disease progression	90
		Objective remission rate	19.68％
Complete remission + partial remission	37
		Disease progression + disease stabilization	151
Sustained clinical benefit rate	29.44％
		Sustained clinical benefit	53
Non-persistent clinical benefit	127

Patients were divided into 3 groups according to the algorithm of the model, and clinical indexes of the patients were counted respectively, and specific results are shown in table 4.

Table 4: clinical benefit cases between 3 groups

It can be seen that the population of the high support immunotherapy group (High Support Immunotherapy, HSI) is about 14.36% of the total population, the population of the low support immunotherapy group (Low Support Immunotherapy, LSI) is about 31.91%, and the population of the low support immunotherapy group (Low Support Immunotherapy, LSI) is about 53.72%.

Three groups of patients had significant differences in disease control rate (disease control rate, DCR) (P < 0.001), with the highest efficacy in the HSI group, 88.89% (95% ci confidence interval, 71.12to 96.97), followed by 55.45% in the MSI group (95% ci,45.73to 64.76), the worst in the LSI group, and 30.00% (95%CI,19.84to 42.57). Also, there was a significant difference in objective remission rates (objective response rate, ORR) (p=0.001), with the highest efficacy of the HSI group being 33.33% (95%CI,18.53to 52.29), followed by the worst of the MSI group 24.75% (95%CI,17.32to 34.04), LSI group, only 5.00% (95%CI,1.17to 14.25). The clinical benefit rate (the rate of DCB) was significantly higher in the HSI group than in the MSI group, LSI group (85.19%, 95%CI,66.90to 94.70;27.96%, 95%CI,19.81to 37.85;6.67%, 95%CI,2.16to 16.39;respectively,P < 0.001).

The PFS comparisons between the three groups are shown in fig. 1. As can be seen from a combination of fig. 1 and table 3, there was a significant difference in progression free survival time (P < 0.001) after three groups of patients received anti-PD-L1 treatment, with the best efficacy in the HSI group, with a 1-year PFS rate of 36% (95%CI,16.4to 55.6), a median progression free survival time of 10.400 months (95%CI,9.563to 11.237), followed by a 3.170 month PFS in the MSI group (95%CI,1.851to 4.489), a 18% 1-year PFS rate (95%CI,10.16to 25.84), the worst efficacy in the LSI group, a 5% 1-year PFS rate (95% ci,0to 10.88), and a median PFS of only 2.100 months (95%CI,1.730to 2.470).

Example 3 comparison of the model constructed in example 1 with the predicted efficacy of PD-L1

Based on the gold standard that the prior literature has PD-L1 expression of more than or equal to 50% as the predicted immune therapy benefit, in order to further verify the effect of the model prediction immune therapy constructed in the embodiment 1, two markers, namely PD-L1 expression and the model constructed in the embodiment 1, are respectively used for predicting the benefit, and the specific conditions are shown in Table 5.

TABLE 5 construction of models and prediction of PD-L1 expression benefits from example 1

It can be seen that when PD-L1 expression predicts the benefit of immunotherapy, the benefit is 16, wherein the clinical true benefit is 11, and the clinical non-benefit is 5; the number of persons who did not benefit is 62, the number of persons who did not benefit clinically truly is 14, and the number of persons who did not benefit clinically is 48.

When the model constructed in example 1 predicts the benefits of immunotherapy, the number of benefits is 27, the number of clinically true benefits is 23, and the number of clinically non-benefits is 4; the number of persons who do not benefit is 60, the number of persons who do not benefit clinically truly is 4, and the number of persons who do not benefit clinically is 56.

The predictive efficiency is then assessed by sensitivity, specificity, positive predictive value, and negative predictive value. The sensitivity is the probability of predicting the benefit through the predictor in the population benefiting from the clinical real benefit situation; the specificity is that the probability of not benefiting is predicted by a predictor in the population not benefiting clinically; the positive predictive value is the proportion of the number of the patients truly clinically benefiting from the number of the benefited cases judged by the predictive factor test result; the negative predictive value is the proportion of the number of patients truly clinically not benefiting from the total number of not benefiting cases predicted by the predictive factor.

It can be derived that the sensitivity was 44% (95% CI,25.02to 64.73), the specificity was 90.56% (95%CI,78.58to 96.47), the positive predictive value was 68.75% (95% CI,41.48to 87.87) and the negative predictive value was 77.42% (95%CI,64.72to 86.68) when the PD-L1 expression was used to predict the benefit of immunotherapy.

The sensitivity was 85.18% (95% CI, 65.3995.14), the specificity was 93.33% (95%CI,82.99to 97.84), the positive predictive value was 85.18% (95% CI, 65.3995.14), the negative predictive value was 93.33% (95%CI,82.99to 97.84), the sensitivity and negative predictive value were both significantly higher than PD-L1 by 50%, and the specificity and positive predictive value were no different from PD-L1 by 50% when the model constructed in example 1 was used to predict the benefits of immunotherapy.

The sensitivity and specificity of the index were further evaluated by using the area AUC enclosed by ROC curve and coordinate axis, as shown in fig. 2, and the specific statistical results are shown in table 6.

TABLE 6 predictive efficacy of the model constructed in example 1 and PD-L1 as a predictor

It can be seen that in this example, acu=0.673 (95%CI,0.535to 0.811) for PD-L1 expression, whereas auc=0.893 (95%CI,0.806to 0.979), p=0.008, for the model constructed in example 1, had significant statistical differences.

These data demonstrate that the predictive effect of the model constructed in example 1 is superior to that of PD-L1, with patients who benefit more than PD-L1, and with greater efficacy in predicting whether patients would benefit from immunotherapy.

The above examples merely represent a few embodiments of the present invention and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention.

Claims

1. One group of gene markers for predicting the anti-PD-L1 immunotherapy effect of lung adenocarcinoma is the following two groups:

group A: BRD4, RPS6KA4, TET1, BCL10, FGFR2, RAD51C, TCF L2, TGFBR1, AR;

2. The genetic marker of claim 1, wherein the criterion for predicting the effect of anti-PD-L1 immunotherapy of lung adenocarcinoma is:

3. Use of a reagent for detecting mutation of a gene marker in the preparation of a reagent for predicting the anti-PD-L1 immunotherapeutic effect of lung adenocarcinoma, wherein the gene marker is as defined in claim 1.

4. The use according to claim 3, wherein the reagent is a primer for detecting mutation of the gene marker.

5. The use according to claim 3, wherein the reagents are primers and probes for detecting mutations in the gene markers.

6. The use according to claim 3, wherein the reagent is a gene chip.

7. The use according to claim 3, wherein the criterion for predicting the effect of an anti-PD-L1 immunotherapy of lung adenocarcinoma is:

8. The application according to claim 3, wherein the method for performing detection comprises: polymerase chain reaction, denaturing gradient gel electrophoresis, whole exon sequencing, second generation sequencing, in situ hybridization, high performance liquid chromatography, biological mass spectrometry, gene chip detection, pyrosequencing, or single strand conformational isomerism polymorphism analysis techniques.

9. A kit comprising a reagent for detecting a mutation of the gene marker according to claim 1,

wherein, the lung adenocarcinoma anti-PD-L1 immunotherapy effect is predicted by detecting the mutation condition of the gene marker of claim 1, and the judgment standard is as follows: