The biomarkers related to immune infiltration to predict distant metastasis in patients with breast cancer

doi:10.21203/rs.3.rs-2239727/v1

Download PDF

Research Article

The biomarkers related to immune infiltration to predict distant metastasis in patients with breast cancer

https://doi.org/10.21203/rs.3.rs-2239727/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background:The development of distant metastasis (DM) results in poor prognosis of breast cancer (BC) patients, however, it is difficult to predict the risk of distant metastasis.

Methods: differentially expressed gene (DEGs) were screened out using GSE184717 and GSE183947.GSE20685 were randomly assigned to the training and the internal validation cohort. A signature was developed according to the results of univariate and multivariate Cox regression analysis,which was validated by using internal and external(GSE6532) validation cohort. Gene set enrichment analysis (GSEA) was used for functional analysis.Finally, a nomogram was constructed and calibration curves and concordance index (C-index) were compiled to determine predictive and discriminatory capacity.The clinical benefit of this nomogram was revealed by decision curve analysis (DCA). Finally, we explored the relationships between candidate genes and immune cell infiltration, and the possible mechanism.

Results: A signature containing CD74 and TSPAN7 was developed according to the results of univariate and multivariate Cox regression analysis,which was validated by using internal and external(GSE6532) validation cohort. Mechanistically,the signature reflect the overall level of immune infiltration in tissues, especially myeloid immune cells. The expression of CD74 and TSPAN7 is heterogeneous, low expression levels of TSPAN7 is caused by methylation modification in breast cancer cells, which is negatively correlated with CD74 expression level.CD74 is mainly derived from myeloid immune cells and do not affect the proportion of CD8+T cells. This signature could act as an independent predictive factor in patients with BC (P = 0.01, HR = 0.63), and it has been validated in internal (P = 0.023, HR = 0.58) and external (P = 0.0065, HR = 0.67) cohort. Finally, we constructed a individualized prediction nomogram based on our signature. The model showed good discrimination in training, internal and external cohort, with a C-index of 0.742, 0.801, 0.695 respectively, and good calibration. DCA demonstrated that the prediction nomogram was clinically useful.

Conclusion: A new immune infiltration related signature developed for predicting metastatic risk will improve the treatment and management of BC patients

distant metastasis

gene signature

nomogram

immune infiltration

myeloid cells

Currently, breast cancer (BC) has been the highest cause of cancer death for women ¹. Approximately 90% of patient death come from metastatic complications although only 20–30% of patients suffer from metastatic recurrence, which makes the treatment schedules of metastatic patients different from those of non-metastatic patients ^2,3. Therefore, accurate identification of distant metastasis (DM) in each individual patient with BC is crucial for determining individualized follow-up strategy and the optimal treatment regimen. The further exploration of new biomarkers is of great significance.

BC is a heterogeneous disease at molecular levels, which is associated with distinct patterns of metastatic spread ⁴. As a result of different molecular subtypes, accurate prediction of DM is still a great challenge. Notably, gene expression-based molecular subtyping appears to have clinical implications for the treatment of patients with BC ^5,6. Moreover, previous studies have shown that molecular subtypes of breast cancer could be considered as a risk factor for distant recurrence ^7,8 ,This suggests that gene expression profiling has a great value in predicting the probability of DM ^9,10 .However, what pattern of gene expression causes this difference still remains unclear. In this study, we aim to find gene biomarker associated with metastasis and construct a gene signature that could accurately predict distant metastasis–free survival (DMFS).

Nomogram, a simple devices for predicting the likelihood of disease, is widely used in the field of oncology ¹¹ .However, there is no literature that has applied gene signature to a nomogram for predicting DMFS in BC, although multiple predictive nomograms have been constructed for patients with BC ^12,13,14. Therefore, the aim of this study was to provides a reliable nomogram to predict the risk of metastasis development based on gene signature in patients with BC.

Data Collection and Pre-processing

Data of the cancer and adjacent normal tissues of samples with metastasis were downloaded from the Gene Expression Omnibus (GEO) database:GSE183947 and GSE184717. Meanwhile,GSE20685,GSE6532 were singled out to construct gene signature and nomogram. Samples without complete clinical information or based on different platform were regarded as substandard samples in the present study. Subsequently, batch effects were removed using Combat from R package SVA ¹⁵ .

Acquisition of DEGs

An R package limma ¹⁶ was applied to identify DEGs between cancer and adjacent normal tissues. The threshold was set to |logFC| >2 and the adjusted p < 0.05. Then, we use Venn diagrams to find the intersection of DEGs that simultaneously upregulated or downregulated in both metastatic tumor and primary tumor.

Screening of optimal predictive biomarkers and development of signature

GSE20685 was randomed 1: 1 and splitted into a training cohort and an internal validation cohort. For reproducibility, the random seed was used and set to 3. To find optimal predictive biomarkers, univariate and multivariate Cox regression analysis was performed in the training cohort with a p-value cutoff of 0.05.Then,the regression coefficient was defined according to the multivariate Cox regression model and the formulas was described as follows:

Finally, results were visualized using the “forestplot” package in R.

Construction and assessment of the Nomogram

In order to make better use of the signature, a nomogram was constructed using the “RMS” package. The Harrell’s Concordance index (C-index) values range from 0–1, which is positively correlated to the predictive performance of the nomogram ¹⁷ The nomogram was subjected to bootstrapping validation (1,000 bootstrap resamples) to calculate a relatively corrected C-index. To assess the consistency of DMFS at 3-,5-,10-years between the nomogram predicted probabilities and observed rates, calibration curves were plotted.

Evaluation of predictive value

Gene Expression Profiling Interactive Analysis (GEPIA) (http://gepia.cancer-pku.cn/) is an online tool, which contains massive RNA sequencing data from TCGA and Genotype-Tissue Expression (GETx) ¹⁸. We analyze the differential gene expression and correlation between breast cancer tissues(n = 1085) and normal tissues(n = 291) using GEPIA.

In terms of the signature, the optimal cutoff value was calculated according to the median value of the signature in training cohort. Then we use it to divide patients into two groups and predict the DMFS in the training, internal validation, and external validation cohort by means of plotting survival curves based on the Kaplan-Meier method.To date fifty-five dataset have been included in Breast Cancer Gene-Expression Miner v4.7(bc-GenExMiner)(http://bcgenex.centregauducheau.fr/),which can be used to improve gene prognostic analysis performance ¹⁹ .candidate biomarkers were validated again using bc-GenExMiner.

Compared with ROC curves, decision curve analysis (DCA) can integrate patient and doctor preference into analysis, which is increasingly being utilized in clinical practice ²⁰. To evaluate the clinical utility of models, DCA curves was developed using the “stdca.R” package in R.

Gene Set Enrichment Analysis

The degree of differential gene expression was reordered by high- and low-score groups instead of definite differential gene thresholds, which was used to screen significantly enriched KEGG pathways. This method helps minimize losses of original gene expression data ²¹. We performed Gene set enrichment analysis (GSEA) to analyze the deference in anti-metastatic potential between two groups.

Immune Infiltration Analysis

The relationship between optimal predictive biomarkers and Immune infiltrating levels was analyzed using Timer(https://cistrome.shinyapps.io/timer/) ²² .An algorithm named quanTIseq was used to estimate the fraction of immune cell subsets infiltrating the tissue from GSE20685 and GSE6532 ²³. Student’s t-test was used to test for significance between high-score group and low-score group.

Immunohistochemistry

28 tissue samples of breast cancer were collected in ShunDe Hospital of Guangzhou University of Chinese Medicine to analyze the correlation between CD74 and myeloid cells. To reduce error, Immunohistochemistry(IHC) for CD74(1:200 dilution; EPR4064, Abcam)and CD33(1:200 dilution; EPR23051-101, Abcam) were performed by an autostrainer system(Lumatas Titan, LumatasBiosystem Inc.) on 3-µm-thick,formalin-fixed and parrffin-embedded (FFPE) Sections. Counterstaining was done with hematoxylin. The average optical density (AOD) was calculated with Image Pro Plus6.0 to determine the protein expression level.

Cell cuture

Three cell lines(MCF-7,ZR-75-1, MDA-MB-231)were maintained in DMEM medium with 10% fetal bovine serum (FBS). Cells were seeded in 6-well plates and,12 hours after plating, demethylation was induced with 10µM 5-aza for 24 hours. Stable expression of TSPAN7 was confirmed by RT-qPCR.Briefly, Total RNA was extracted using TRIzol (Invitrogen) and a two-step reverse transcription-quantitative PCR (RT-qPCR) protocol was performed using PrimeScript RT Master Mix(Takara) and TB Green Premix Ex Taq(Takara). The experiment was repeated three times biologically for statistical analysis.

Screening of DEGs

A flow chart of the study design is shown in Figure 1. The R package “limma” was used to screen DEGs between tumor and normal tissues in GSE184717 and GSE183947, where a total of 1967 (1494 upregulated and 473 downregulated) and 351 (180 upregulated and 171downregulated) DEGs were obtained, respectively. The distribution of each gene were visualized by volcano plots (Figure 2A,B). The resulting list of DEGs was the intersection between the above datasets and a total of 62 overlapping DEGs were obtained(Figure 2C).

TABLE 1: The relationship between the siganture and patient characteristics.

Two genes Were Screened Out as Potential Predictive Biomarkers

The clinicopathologic characteristics are summarized in Table 1. To identify DEGs that associated with metastasis, univariate Cox regression analysis was performed.A total of nine candidate genes(CD74,TSPAN7,COL11A1,FLG,MMP11,CHRDL1,FNDC1,MELK,PITX1)with an adjusted P value < 0.05 were screened out(Figure 3A). Detailed information for each gene was listed in Table 2.

TABLE 2.Details of the candidate genes

Gene Symbol	Gene ID	Full Name	Cytogenetic locations	Protein Function
CD74	972	CD74 molecule	5q33.1	CD markers; Disease related genes; Cancer-related genes:Candidate cancer biomarkers
TSPAN7	7102	tetraspanin 7	Xp11.4	Transporters:Accessory Factors Involved in Transport; CD markers; Disease related genes; Potential drug targets
COL11A1	1301	collagen type XI alpha 1 chain	1p21.1	Predicted intracellular proteins; Disease related genes; Predicted secreted proteins; Cancer-related genes:Candidate cancer biomarkers
FLG	2312	filaggrin	1q21.3	Cancer-related genes:Mutated cancer genes; Predicted intracellular proteins; Disease related genes
MMP11	4320	matrix metallopeptidase 11	22q11.23	Peptidases:Metallopeptidases; Cancer-related genes:Candidate cancer biomarkers; FDA approved drug targets:Small molecule drugs; Predicted secreted proteins; Enzymes
CHRDL1	91851	chordin like 1	Xq23	Disease related genes; Predicted secreted proteins
FNDC1	84624	fibronectin type III domain containing 1	6q25.3	Predicted intracellular proteins; Predicted secreted proteins
MELK	9833	maternal embryonic leucine zipper kinase	9p13.2	Disease related genes; ENZYME proteins:Transferases; Predicted intracellular proteins; Potential drug targets; Kinases:CAMK Ser/Thr protein kinases; Enzymes
PITX1	5307	paired like homeodomain 1	5q31.1	Predicted intracellular proteins; Disease related genes; Transcription factors:Helix-turn-helix domains

we further analyze the differential gene expression between breast cancer tissues(n = 1085) and normal tissues(n = 291) using GEPIA. As shown in Figure 4,Eight genes were found to be differentially expressed in breast cancer tissues, including six upregulated genes(CD74,MMP11,MELK,COL11A1,PITX1,FNDC1) and two downregulated genes(TSPAN7,CHRDL1).

Interestingly, breast tissues with low CD74 expression were related to poorer DMFS although CD74 commonly upregulated in breast cancer patients.
Additionally, bc-GenExMiner was utilized to validate the relationship between predictive biomarkers and DMFS. The expression levels of seven genes,including TSPAN7(HR:0.75; 95%CI:0.69-0.83;P < 0.0001),CD74(HR:0.88;95%CI:0.81-0.97;P = 0.0094),MMP11(HR:1.33; 95%CI:1.21-1.46;P < 0.0001),MELK(HR:1.87;95%CI:1.70-2.06; P < 0.0001),COL11A1(HR:1.13; 95%CI:1.03-1.24; P = 0.0108),CHRDL1(HR:0.81; 95%CI:0.73-0.89; P < 0.0001),PITX1(HR:1.17; 95%CI:1.07-1.29; P = 0.0010), were found able to estimate DMFS(Figure 5).

Finally, CD74 （HR:0.61;95%CI:0.4-0.92;P = 0.019）and TSPAN7（HR:0.51;95%CI:0.29-0.9;P = 0.021）were screened out to develop a signature according to the results of multivariate Cox analysis (Figure 3B).

Development and Validation of the Signature

We calculated the signature based on the expression of CD74 and TSPAN7 as follows:
Signature = （0.7697433 X expression of CD74）+（0.6633031 X expression of TSPAN7）
Of note, the expression of CD74 and TSPAN7 was negatively correlated with DMFS, suggesting that patients with low scores have a higher probability of distant metastasis.
To be better applied in clinical diagnosis, a constant cutoff value was determined by the median of the training cohort (15.488). Survival analysis, using the Kaplan-Meier method, indicated that the low score group portends a worse DMFS(HR:0.63;95%CI:0.49-0.82; P = 0.01; Figure 6A). Subsequently, survival analysis was performed twice with the same results in the internal validation cohort (HR:0.58;95%CI:0.36-0.96; P = 0.023; Figure 6B), and the external validation cohort (HR:0.67;95%CI:0.47-0.95; P = 0.0065; Figure 6C), respectively. To individually show the differences between low score and high score groups, we visualized the scores, distant metastasis, and gene expression profiles in the three cohorts（Figure 6D,E,F）.The above results indicated that the signature could predict the risk of metastasis development as an independent risk feature.

Gene Set Enrichment Analysis

GSEA was performed to analyze the causes of breast cancer metastasis risk difference between high and low score groups. Figure 7A showed upregulated genes in the high score group were significantly enriched in immune related signal pathways, such as T cell receptor signaling and chemokine signaling pathway, suggesting that our signature may be an immune infection related signature(IIRS).

Immune Infiltration Analysis

we further utilized TIMER to analyze possible correlation between CD74, TSPAN7 expression and levels of immune infiltration in breast cancer (Figure 7B). Both CD74 and TSPAN7 are associated with multiple immune cell infiltration, especially T cells. To identify the relationship between our cohort and immune infiltration, we calculate the fraction of immune cell subsets using quanTIseq.As shown in Figure 7C, The high score group had a higher fraction of B cell(P < 0.001),M2Macrophage(P < 0.001),Neutrophil(P = 0.009),CD8+T cell(P < 0.001),Tregs(P < 0.001). On the contrary, the fraction of NK cell(P = 0.012) and uncharacterized cell was found to increase significantly in the low score group. It is worth noting that the analysis showed that there was the most significant difference in CD8+T cells between groups.

In view of myeloid cells play an important role in the recruitment and concentration of CD8+ T cells ²⁴.we attempted to evaluate the correlation between myeloid cell levels and the signature in breast cancer patients. CD33 is widely used as a marker for tumor infiltrating myeloid cells^25,26,so we collected FFPEs from 28 consecutive pairs of breast cancer patients and performed IHC for CD74 and CD33, respectively.Interestingly, CD74 protein mainly existed in some CD33+ cells（Figure 8A）, and were positively correlated with CD33 levels in three cohorts(Figure 8B), GEPIA(Figure 8C) and IHC(Figure 8D), respectively. In addition, there is a significant positive correlation between CD74 and CD8 in breast cancer(Figure 8E), which means that CD74+ myeloid cells will not affect the recruitment of CD8+T cells.
Since TSPAN7 is significantly downregulated in breast cancer tissues, we tried to identify the cause of low expression of TSPAN7 by detecting the level of methylation. we hypothesized that TSPAN7 methylation occurs in breast cancer cells and treated three TSPAN7-silenced cells (MCF-7, ZR-75-1, MDA-MB-231) with demethylation agent 5-Aza, respectively.TSPAN7 expression was subsequently restored in all treated cells(MDA-MB-231,ZR-75-1,MCF-7)( Figure 8F). Importantly, we found that the level of TSPAN7 was significantly correlated with the level of CD74 (p=8.3e-11), which has been proved to exist mainly in immune cells(Figure 8G). Therefore, breast cancer cells undergo TSPAN7 methylation, which is related to the infiltration of CD74 positive immune cells.

Construction and Assessment of Predictive Nomogram

In order to increase clinical utility,a predictive nomogram was constructed(FIGURE 9A) based on the risk feature verified by univariate and multivariate Cox analysis, including the IIRS, age, T stage, and N stage (Figure 3C, D). Interestingly, we find that younger patients are at higher risk of metastasis, which are consistent with previous findings²⁷. Subsequently, calibrate curves were plotted to identify the consistency between ideal outcome and actual observation in prediction of 3-,5- 10-year distant metastasis-free survival times.The calibration curves showed good performance in training cohort (Figure 9B), internal validation cohort (Figure 9C), and external validation cohort (Figure 9D), especially for long term survival rate（10-year DMFS）.

To assess the predictive performance of this model, the nomogram was internally validated by 1,000 bootstrap resamples in three cohorts. Pleasantly, and surprisingly, the nomogram yielded a C-index of 0.742 (95% CI, 0.715 to 0.769) for training cohort ,0.801 (95% CI, 0.754 to 0. 0.848) for internal validation cohort and 0.695(95% CI, 0.643 to 0. 0.747) for external validation cohort. Therefore, we demonstrated that the nomogram could predict the DMFS of BC patients effectively.

We subsequently performed a decision curve analysis to evaluate the clinical net benefit in predicting the probability of 10- year DMFS.As shown in Figure 9E, if the threshold probability of a patient or doctor is < 48% or >63%, using the nomogram with IIRS to predict DMFS adds more benefit than the other without IIRS.

It is well known that almost all metastatic breast cancer has poor overall survival and incurable nature ²⁸. In recent years, it is gratifying to note that the median survival time after diagnosis of metastatic breast cancer has been increasing due to improved treatment ²⁹. Therefore, early assessment and diagnosis of distant metastasis are still meaningful strategies for improving the prognosis of breast cancer patients.

Nowadays multiple biomarkers, such as circulating miRNA ³⁰, circulating tumor DNA ³¹ and circulating tumor cell ³², can be used to detect metastasis of breast cancer. What’s more, several serum biomarkers related to metastasis,including soluble POSTN ³³, PTHrP ³⁴, S100P ³⁵ ,have been found. Nonetheless, when these biomarkers increase, it is very likely that breast cancer has undergone distant metastasis ³⁶. Therefore, improved methods based on gene expression to predict the risk of breast cancer metastasis in advance are needed.

Unfortunately, there is no effective way to achieve this goal. Although PAM50 signature has proven to be able to provide prognostic information from the lymph node metastasis of breast cancer patients, only a few subtypes are associated with metastasis ^8,12, suggesting that individualized accurate prediction according to PAM50 subtypes is still difficult.

In this study, we sought to identify metastasis-related genes to predict the probability of distant metastasis in breast cancer patients. A total of 62 genes were screened out in GSE184717 and GSE183947.7 genes(CD74,TSPAN7,COL11A1, MMP11, CHRDL1, MELK,PITX1)proved to be associated with DMFS in breast cancer patients, and seven of them (TSPAN7,CD74,MMP11,MELK,COL11A1,CHRDL1,PITX1) were verified by bc-GenExMiner. Two potential biomarkers (CD74, TSPAN7) of metastasis development were used to construct a gene signature by multivariate analysis. The signature proved to be associated with distant metastasis–free survival in three cohort.

The CD74 gene, responsible for producing a protein associate with class II major histocompatibility complex (MHC) is implicated in an effective intratumor immune response ³⁷. It also serves as cell surface receptor for the cytokine macrophage migration inhibitory factor (MIF), which may play a pro-oncogenic role in promoting breast cancer cell-stroma interactions ³⁸. Of note,CD74 was observed to be related to triple-negative breast cancer,which is the most aggressive subtype of breast cancer ³⁹. The TSPAN7 is a cell-surface protein coding gene, and the coding protein of which plays a role in the regulation of cell development, activation, growth and motility ⁴⁰. For this reason, TSPAN7 is also known as CD231. Our study is the first to link CD74 and TSPAN7 expression with distant metastasis–free survival in breast cancer, highlighting gene signature based on CD74,TSPAN7 as predictor of metastasis development in BC, with a strong effect on patients’ distant metastasis–free survival.

In past studies, multiple prognostic models have been constructed because of the differential expression of CD74 ⁴¹ or TSPAN7 ⁴² between tumor and normal tissues. However, the reason for this difference is still unclear. As a result, much effort has been expended in trying to explore differential expression of CD74 or TSPAN7 in cancer cell strains, ignoring the influence of immune infiltration ^{43,44, 45,46}. Notably, heterogeneity of CD74 expression has been confirmed in tumor tissues ⁴⁷, indicating that characterization of immune landscape cannot be discounted. In this study, GSEA revealed a statistically enrichment of immune-related signaling pathways in the high score group. Thus we speculated that the signature we constructed may reveal the levels of immune infiltration. The results of TIMER database confirm this conjecture. CD74 and TSPAN7 expression was negatively correlated with tumor purity while positively correlated with the level of multiple immune cells. Furthermore, the fraction of immune cells in high score group was significantly higher than that in low score group, especially CT8 + T cell. According to the results presented above, we identified a novel cause of differential expression of CD74 and TSPAN7.The levels of CD74 and TSPAN7 reflect the ability of immune cells to infiltrate tumors. To the best of our knowledge, this is the first study to show that high CD74 and TSPAN7 expression is associated with tumor-infiltrating immune cells. Further studies on Intra-tumoral heterogeneity are warranted, in order to analyze the relationship between survival time and the level of immune cell infiltration.

Our study has some potential limitations. First, pathologic features were unavailable, so the correlation between our signature and pathological features could not be evaluated. Second, quanTIseq is a prediction tool of immune cell fraction based on deconvolution algorithm, as a result, it will predict a minimal amount of immune cells even though they are absent ⁴⁸ .

In conclusion, we identified seven genes related to distant metastasis–free survival, namely TSPAN7,CD74,MMP11,MELK,COL11A1,CHRDL1,PITX1,and a signature based on CD74 and TSPAN7 expression that may have potential of predicting DMFS. We found that methylation of TSPAN7 in breast cancer cells inhibits the recruitment of CD74 positive immune cells, which may be associated with a lower risk of metastasis. The present study was the first to propose that high CD74 expression may be derived from tumor-infiltrating immune cells. Given the favorable discrimination of the signature, we developed a nomogram for clinical applications. This is the first nomogram based on gene signature that can be used to facilitate the individualized prediction of DMFS in breast cancer patients.

Acknowledgements

Special thanks to the researchers that collected, curated, and maintain the publicly database GEPIA and bc-GenExMiner.

funding

This work was supported by the basic and applied basic research program of Guangdong province (2019A1515011960);. This research was also supported by Guangzhou Science and Technology Plan Project(202102010178).

ethics statement

This study was conducted in compliance with the declaration of Helsinki.
Informed consent was obtained from all subjects. The ethics approval statements for human subjects were provided by the Ethnic Committee of Shunde Hospital of Guangzhou University of Chinese Medicine (reference number KY-2022085).

author contributions

All authors listed have made direct intellectual contributions to the work, and approved it for its publication. Ren Chengsi made a substantial contribution when drafting the manuscript.

Availability of data and materials

The datasets we analyzed during the current study are available in the Gene Expression Omnibus (GEO) repository with the accession numbers GSE184717,GSE183947, GSE20685 and GSE6532. These data analysed during the current study are available in GEO repository https://www.ncbi.nlm.nih.gov/geo/, GEPIA http://gepia.cancer-pku.cn/ and bc-GenExMiner http://bcgenex.centregauducheau.fr/.

Competing interests

All authors listed declare that they have no competing interest.

Consent for publication

Not applicable.

Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I,et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021 May;71(3):209-249.
Medeiros B, Allan AL. Molecular Mechanisms of Breast Cancer Metastasis to the Lung: Clinical and Experimental Perspectives. Int J Mol Sci. 2019 May 8;20(9):2272.
Chen W, Hoffmann AD, Liu H, Liu X. Organotropism: new insights into molecular mechanisms of breast cancer metastasis. NPJ Precis Oncol. 2018 Feb 16;2(1):4.
Liang Y, Zhang H, Song X, Yang Q. Metastatic heterogeneity of breast cancer: Molecular mechanism and potential therapeutic targets. Semin Cancer Biol. 2020 Feb;60:14-27.
Prat A, Pineda E, Adamo B, Galván P, Fernández A,et al. Clinical implications of the intrinsic molecular subtypes of breast cancer. Breast. 2015 Nov;24 Suppl 2:S26-35.
Masuda H, Baggerly KA, Wang Y, Zhang Y, Gonzalez-Angulo AM,et al. Differential response to neoadjuvant chemotherapy among 7 triple-negative breast cancer molecular subtypes. Clin Cancer Res. 2013 Oct 1;19(19):5533-40.
Gnant M, Filipits M, Greil R, Stoeger H, Rudas M,et al. Predicting distant recurrence in receptor-positive breast cancer patients with limited clinicopathological risk: using the PAM50 Risk of Recurrence score in 1478 postmenopausal patients of the ABCSG-8 trial treated with adjuvant endocrine therapy alone. Ann Oncol. 2014 Feb;25(2):339-45.
Tobin NP, Lundberg A, Lindström LS, Harrell JC, Foukakis T,et al. PAM50 Provides Prognostic Information When Applied to the Lymph Node Metastases of Advanced Breast Cancer Patients. Clin Cancer Res. 2017 Dec 1;23(23):7225-7231.
Ellis MJ, Suman VJ, Hoog J, Lin L, Snider J,et al. Randomized phase II neoadjuvant comparison between letrozole, anastrozole, and exemestane for postmenopausal women with estrogen receptor-rich stage 2 to 3 breast cancer: clinical and biomarker outcomes and predictive value of the baseline PAM50-based intrinsic subtype--ACOSOG Z1031. J Clin Oncol. 2011 Jun 10;29(17):2342-9.
Filipits M, Nielsen TO, Rudas M, Greil R, Stöger H,et al. The PAM50 risk-of-recurrence score predicts risk for late distant recurrence after endocrine therapy in postmenopausal women with endocrine-responsive early breast cancer. Clin Cancer Res. 2014 Mar 1;20(5):1298-305.
Balachandran VP, Gonen M, Smith JJ, DeMatteo RP. Nomograms in oncology: more than meets the eye. Lancet Oncol. 2015 Apr;16(4):e173-80.
Wang Y, Yang Y, Chen Z, Zhu T, Wu J,et al.Development and validation of a novel nomogram for predicting distant metastasis-free survival among breast cancer patients. Ann Transl Med. 2019 Oct;7(20):537.
Huang ZZ, Hua X, Song CG, Xia W, Bi XW,et al. The Prognostic Prediction Value of Systemic Inflammation Score and the Development of a Nomogram for Patients With Surgically Treated Breast Cancer. Front Oncol. 2020 Oct 20;10:563731.
Rouzier R, Pusztai L, Delaloge S, Gonzalez-Angulo AM, Andre F,et al. Nomograms to predict pathologic complete response and metastasis-free survival after preoperative chemotherapy for breast cancer. J Clin Oncol. 2005 Nov 20;23(33):8331-9.
ohnson WE, Li C, Rabinovic A. Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics. 2007 Jan;8(1):118-27.
Ritchie ME, Phipson B, Wu D, Hu Y, Law CW,et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015 Apr 20;43(7):e47.
Harrell FE Jr, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med. 1996 Feb 28;15(4):361-87.
Tang Z, Li C, Kang B, Gao G, Li C, Zhang Z. GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res. 2017 Jul 3;45(W1):W98-W102.
Jézéquel P, Campone M, Gouraud W, Guérin-Charbonnel C, Leux C,et al.bc-GenExMiner: an easy-to-use online platform for gene prognostic analyses in breast cancer. Breast Cancer Res Treat. 2012 Feb;131(3):765-75.
Huang YQ, Liang CH, He L, Tian J, Liang CS,et al. Development and Validation of a Radiomics Nomogram for Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer. J Clin Oncol. 2016 Jun 20;34(18):2157-64.
Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL,et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005 Oct 25;102(43):15545-50.
Li T, Fan J, Wang B, Traugh N, Chen Q,et al. TIMER: A Web Server for Comprehensive Analysis of Tumor-Infiltrating Immune Cells. Cancer Res. 2017 Nov 1;77(21):e108-e110.
Finotello F, Mayer C, Plattner C, Laschober G, Rieder D,et al. Molecular and pharmacological modulators of the tumor immune contexture revealed by deconvolution of RNA-seq data. Genome Med. 2019 May 24;11(1):34.
Jiang X, Wang J, Zheng X, Liu Z, Zhang X,et al. Intratumoral administration of STING-activating nanovaccine enhances T cell immunotherapy. J Immunother Cancer. 2022 May;10(5):e003960.
Toor SM, Taha RZ, Sasidharan Nair V, Saleh R, Murshed K,et al. Differential gene expression of tumor-infiltrating CD33+ myeloid cells in advanced- versus early-stage colorectal cancer. Cancer Immunol Immunother. 2021 Mar;70(3):803-815.
Choi JW, Kim YJ, Yun KA, Won CH, Lee MW,et al. The prognostic significance of VISTA and CD33-positive myeloid cells in cutaneous melanoma and their relationship with PD-1 expression. Sci Rep. 2020 Sep 1;10(1):14372.
Purushotham A, Shamil E, Cariati M, Agbaje O, Muhidin A,et al. Age at diagnosis and distant metastasis in breast cancer--a surprising inverse relationship. Eur J Cancer. 2014 Jul;50(10):1697-1705.
Gobbini E, Ezzalfani M, Dieras V, Bachelot T, Brain E,et al.Time trends of overall survival among metastatic breast cancer patients in the real-life ESME cohort. Eur J Cancer. 2018 Jun;96:17-24.
Mariotto AB, Etzioni R, Hurlbert M, Penberthy L, Mayer M. Estimation of the Number of Women Living with Metastatic Breast Cancer in the United States. Cancer Epidemiol Biomarkers Prev. 2017 Jun;26(6):809-815.
Baldasici O, Pileczki V, Cruceriu D, Gavrilas LI, Tudoran O,et al. Breast Cancer-Delivered Exosomal miRNA as Liquid Biopsy Biomarkers for Metastasis Prediction: A Focus on Translational Research with Clinical Applicability. Int J Mol Sci. 2022 Aug 19;23(16):9371.
Page K, Martinson LJ, Fernandez-Garcia D, Hills A, Gleason KLT,et al. Circulating Tumor DNA Profiling From Breast Cancer Screening Through to Metastatic Disease. JCO Precis Oncol. 2021 Nov 24;5:PO.20.00522.
Ring A, Campo D, Porras TB, Kaur P, Forte VA,et al. Circulating Tumor Cell Transcriptomics as Biopsy Surrogates in Metastatic Breast Cancer. Ann Surg Oncol. 2022 May;29(5):2882-2894.
Jia L, Li G, Ma N, Zhang A, Zhou Y,et al. Soluble POSTN is a novel biomarker complementing CA153 and CEA for breast cancer diagnosis and metastasis prediction. BMC Cancer. 2022 Jul 12;22(1):760.
Washam CL, Byrum SD, Leitzel K, Ali SM, Tackett AJ,et al. Identification of PTHrP(12-48) as a plasma biomarker associated with breast cancer bone metastasis. Cancer Epidemiol Biomarkers Prev. 2013 May;22(5):972-83.
Peng C, Chen H, Wallwiener M,et al. Plasma S100P level as a novel prognostic marker of metastatic breast cancer. Breast Cancer Res Treat. 2016 Jun;157(2):329-338.
Al-Mahmood S, Sapiezynski J, Garbuzenko OB, Minko T. Metastatic and triple-negative breast cancer: challenges and treatment options. Drug Deliv Transl Res. 2018 Oct;8(5):1483-1507.
Wang ZQ, Milne K, Webb JR, Watson PH. CD74 and intratumoral immune response in breast cancer. Oncotarget. 2017 Feb 21;8(8):12664-12674.
Verjans E, Noetzel E, Bektas N, Schütz AK, Lue H,et al. Dual role of macrophage migration inhibitory factor (MIF) in human breast cancer. BMC Cancer. 2009 Jul 14;9:230.
Tian B, Zhang Y, Li N, Liu X, Dong J. CD74: a potential novel target for triple-negative breast cancer. Tumour Biol. 2012 Dec;33(6):2273-7.
Perot BP, Ménager MM. Tetraspanin 7 and its closest paralog tetraspanin 6: membrane organizers with key functions in brain development, viral infection, innate immunity, diabetes and cancer. Med Microbiol Immunol. 2020 Aug;209(4):427-436.
Wang J, Yang Z, Zhang C, Ouyang J, Zhang G, Wu C. A four-gene signature in the tumor microenvironment that significantly associates with the prognosis of patients with breast cancer. Gene. 2020 Nov 30;761:145049.
Wu M, Li X, Liu R, Yuan H, Liu W, Liu Z. Development and validation of a metastasis-related Gene Signature for predicting the Overall Survival in patients with Pancreatic Ductal Adenocarcinoma. J Cancer. 2020 Aug 29;11(21):6299-6318.
Greenwood C, Metodieva G, Al-Janabi K, Lausen B, Alldridge L,et al. Stat1 and CD74 overexpression is co-dependent and linked to increased invasion and lymph node metastasis in triple-negative breast cancer. J Proteomics. 2012 Jun 6;75(10):3031-40.
Wang X, Lin M, Zhao J, Zhu S, Xu M, Zhou X. TSPAN7 promotes the migration and proliferation of lung cancer cells via epithelial-to-mesenchymal transition. Onco Targets Ther. 2018 Dec 13;11:8815-8822.
Yu X, Li S, Pang M, Du Y, Xu T,et al. TSPAN7 Exerts Anti-Tumor Effects in Bladder Cancer Through the PTEN/PI3K/AKT Pathway. Front Oncol. 2021 Jan 8;10:613869.
Qi Y, Li H, Lv J, Qi W, Shen L,et al. Expression and function of transmembrane 4 superfamily proteins in digestive system cancers. Cancer Cell Int. 2020 Jul 16;20:314.
Richard V, Kindt N, Decaestecker C, Gabius HJ, Laurent G,et al. Involvement of macrophage migration inhibitory factor and its receptor (CD74) in human breast cancer. Oncol Rep. 2014 Aug;32(2):523-9.
Sturm G, Finotello F, Petitprez F, Zhang JD, Baumbach J,et al. Comprehensive evaluation of transcriptome-based cell-type quantification methods for immuno-oncology. Bioinformatics. 2019 Jul 15;35(14):i436-i445.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

The biomarkers related to immune infiltration to predict distant metastasis in patients with breast cancer

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Data Collection and Pre-processing

Acquisition of DEGs

Screening of optimal predictive biomarkers and development of signature

Construction and assessment of the Nomogram

Evaluation of predictive value

Gene Set Enrichment Analysis

Immune Infiltration Analysis

Immunohistochemistry

Cell cuture

Results

Discussion

Declarations

References

Additional Declarations

Status:

Version 1