Prognostic model and immunomicroenvironment analysis of combined thyroid and breast cancer in females based on cuproptosis related miRNA

doi:10.21203/rs.3.rs-2619126/v1

Download PDF

Research Article

Prognostic model and immunomicroenvironment analysis of combined thyroid and breast cancer in females based on cuproptosis related miRNA

https://doi.org/10.21203/rs.3.rs-2619126/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Previous studies have shown that breast cancer and thyroid cancer are more common in females . Among them, the incidence of breast cancer in the female cancer research ranks first. Thyroid cancer is the most common endocrine malignancy in women. cuproptosis is a new type of programmed cell death discovered recently. The objective of this study was to evaluate the prognostic significance of cuproptosis related miRNA (CRM) in female breast and thyroid cancers and to explore potential associations between the two cancers.

Methods：

We extracted transcriptomic data and clinicopathological features of women with THCA and BRCA from the Cancer Genome Atlas (TCGA) database. Univariate Cox regression and LASSO analysis were used to establish the prognostic characteristics of CRM. In order to verify the accuracy of the model, Kaplan-Meyer (K-M) and transient receiver operating characteristics (ROC) analysis were used. We drew a column graph that included clinical features and miRNA features to refine the prediction of the patient prognosis model. Finally, we performed immunoinfiltration correlation analysis.

Results：

In this study, we constructed a prognostic profile of CRM containing 15 miRNAs. This CRM feature was an independent predictor of overall survival. In addition, risk score was a better independent prognostic factor than traditional clinicopathological features. The correlation and differentiation analysis of immune invasion found a strong positive correlation among immune cells such as aDCs and DCs, while Macrophages showed significant differences among the risk group. The study revealed that there was strong positive correlation between immune functions such as APC co stimulation and Check-point. Furthermore, indicators of APC co inhibition, APC co stimulation, Check-point, and Inflammation-promoting showed significant differences between risk groups. Based on risk score and immune score, we finally screened out 6 differential expression genes (DEGs) : such as PCOLCE, SV2C. These DEGs were significantly correlated with one or more immune cells and their functions during immune invasion.

Conclusion：

CRM features can be used as novel biomarkers to predict the prognosis of patients with breast cancer and thyroid cancer, and to predict the clinical outcome and treatment response of patients, thus providing basic insights for further research.

Cuproptosis

Immunotherapy

Tumor immune microenvironment

miRNA，Thyroid cancer

Breast cancer

Breast cancer (BC) is the most diagnosed cancer among females worldwide and the leading cause of cancer-related death among women ^[1]. Improvements in breast cancer screening, early detection and treatment have resulted in improved long-term survival^[2]. Previous studies have shown that when most women are diagnosed with early BC, the 5-year survival rate is > 95%^[3]. Through follow-up, it was found that a significant proportion of long-term breast cancer survivors may have an increased risk of developing secondary primary malignancies, including thyroid cancer^[4–5]. Thyroid cancer (TC) is considered to be the most common endocrine malignancy in women ^[6]. The incidence of TC has been increasing at the fastest rate, mainly for differentiated thyroid cancer^[7]. Among them, about 75% of differentiated thyroid cancers occur in females^[8]. Breast cancer and thyroid cancer are primary tumors that can occur simultaneously or not simultaneously in patients, and their incidences are higher than expected in studies^{[2, 9–10]}. The probability of being diagnosed with invasive follicular thyroid cancer might be higher in breast cancer survivors than in thyroid cancer patients^[11]. In addition, non-malignant thyroid nodules are more common in patients with breast cancer than in women without breast tumors^[12]. Previous studies have shown that risk factors for BC and TC were first described in 1987^[13].Howerver, the possible link between them may remain unclear^{[9, 14]}. Etiology and risk factors have also been investigated, but no firm conclusions have yet been reached.

Copper ions are commonly found in various organisms, which have functions such as hormone maturation and coagulation promotion^[15]. A recent study published in the journal Science found that copper can directly bind to lipoacylated mitochondrial enzymes,triggering protein toxic stress and instability of iron-sulfur (Fe-S) cluster proteins, leading to cell death^[16]. Copper poisoning can induce a distinct type of cell death that might be distinct from regulatory cell death (e.g.,ferroptosis^[17], autophagy, apoptosis, necrosis apoptosis^[18]). Tsvetkov et al. referred to this previously unidentified cell death as "cuproptosis"^[19] and identified several genes involved in copper-induced cell killing. For example, FDX1 was a key regulator of cuproptosis and an upstream regulator of protein lipolysis. As a mitochondrial reductase, FDX1 was a direct target of Ilomo^[19]. High intracellular copper ion concentration can affect multiple types of cancer, such as thyroid^[20], breast ^[21], lung cancer^[22] and prostate cancer^[23]. The copper chelating agents and ion carriers have been recommended as anti-tumor drugs^[24].Cui et al. found through triple-negative breast cancer mouse model that copper-depleting nanoparticle administration could inhibit tumor growth and significantly improve survival rate^[25]. Copper can become a participant in the regulation of physiological and pathological processes of cancer^[26]. Therefore, inhibitors and modulators of copper-related variables as potential cancer treatments have attracted increasing attention in the scientific community.

MicroRNAs (miRNAs) can be regarded as non-coding Rnas, about 21–24 nucleotides long, that can play a crucial role in cell differentiation, development, apoptosis, and cell cycle regulation. They can act as tumor suppressor genes or oncogenes, and can also be regulated by tumor suppressor genes and oncogenes^[27]. Therefore, miRNAs might be involved in multiple pathological processes and might be crucial in cancer progression. They could be key biomarkers for various carcinomas^[28].

In this study, based on the expression statistics and clinical data of female patients with THCA and BRCA in the TCGA database, we characterized the miRNA (CRM) associated with prognostic cuproptosis. We applied multiple Cox regression to determine the independent prognostic effect of this model, and developed a histogram of cuproptosis miRNA characteristics and clinical variables. Finally, we conducted a tumor immune microenvironment analysis to explore its potential mechanism of action. In addition, we aimed to explore the association between breast cancer and thyroid cancer and find key signaling pathways for the treatment of breast cancer and thyroid cancer.

2.1. Data downloading and organizing

Based on TCGA database (https://portal.gdc.cancer.gov/), we download the TCGA THCA and TCGA - BRCA gene expression about women 1609 set of microRNAs - Seq and 1627 set of RNA - Seq data (including tumor group and normal group). Meanwhile, clinicopathological information of 1456 female patients was integrated, including age, survival time, survival state, T, M, N and TMNstage.

2.2. Identification of miRNA associated with cuproptosis

We identified 19 cuproptosis related genes (CRG) from recently published candidate gene literature, Including FDX1, LIPT1, LIAS, DLD, DBT, GCSH, DLST, DLAT, PDHA1, PDHB, SLC31A1, ATP7A, ATP7B, NFE2L2, NLRP3, LIPT2, MTF1, GLS, CDKN2A^[16]. Furthermore, 25703 miRNAs were extracted from the 19 genes, and differentially expressed miRNAs (DEmiRNA) were obtained by differential analysis with the "edgeR" package with normal tissues, THCA and BRCA tissues. The standard was | logFC | 1.0 or higher and the false discovery rate (FDR) < 0.05. Subsequently, DEmiRNA was used to identify potential prognostic miRNAs.

2.3 Construction of cuproptosis related miRNA prognostic model

Univariate Cox regression analysis of differentially expressed miRNAs was performed using survival package in R language to screen out copper-dying miRNAs associated with prognosis. Subsequently, in order to prevent overfitting of cuproptosis-related prognostic miRNAs, minimum absolute contraction and selection operator (LASSO) Cox regression was used to reduce the number of miRNAs based on the "glmnet" package^[29]. Multivariate Cox regression analysis was then performed, and a prognostic model was constructed based on Akaike information criteria (AIC) values. Finally, an optimal prognosis model was constructed based on 15 miRNAs associated with cuproptosis. We named this model "CRM." The following formula was used to evaluate the risk score: Risk score = coef (miRNA1) *expr (miRNA1) + coef (miRNA2) *expr (miRNA2) +...... +coef (miRNAn) * expr (miRNAn), where coef was the coefficient, coef (miRNAn) was the miRNA coefficient related to survival, expr (miRNAn) was the expression of miRNA.

2.4 Validation of the prognostic risk model

We divided the samples in the risk score model into high - and low-risk groups based on median values to assess the prognosis of the characteristics. Based on the constructed risk model, principal component analysis (PCA) was performed to observe the expression patterns of the two groups. In addition, we performed a Kaplan-Meier survival analysis and a "survival" package using R language to calculate the clinical best cutoff and area under the curve (AUC) values for the 3-year receiver operating characteristic (ROC) curve (1 -, 3 -, 5-).

2.5. Prognostic independence analysis of clinicopathological parameters and risk scores

We integrated and classified relevant clinicopathological information, including age ( < = 60, > 60), T (T1-2, T3-4), M (M0, M1), N (N0-1, N2-3) and Stage (I-II, III-IV). Simultaneously, we applied the "limma" package in R language to analyze the differences in the expression of risk scores in different clinical groups.

Based on the "survival" package, we used univariate and multivariate Cox regression analyses to explore whether the predictive power of risk scores could be independent of clinicopathological parameters.

2.6. Building and evaluating the column graph model

Based on risk scores and various clinicopathological parameters, we used the R language packages "rms", "replot" and "survival" to draw column plots and calibrate curves. In this study, the column graph was applied to predict the three-year probability of prognosis (1, 3, and 5 years), and the calibration curve was used to assess whether the survival predicted by the column graph was consistent with the observed probability.

2.7. Extraction and quantification of immunoinfiltrating cells and their functions

We used the R language "limma" and "edgeR" to integrate the risk score of the prognostic model with the mRNA-Seq data to obtain the standardized gene expression matrix and differential gene set. In order to extract and quantify the immune-infiltrating cells and functions of the transformed gene expression matrix, we entered the keyword "IMMUNITY" in the MsigDB database (https://www.gsea-msigdb.org/gsea/msigdb/) and searched the corresponding gmt files. Then, the transformed expression matrix was scored by single sample gene set enrichment analysis (ssGSEA) using the "GSVA" package in R. ssGSEA is an analytical tool that includes a total of 29 immune scores to estimate the degree of immune infiltration in each sample and to create a heatmap using the software package pheatmap.

2.8. Analysis of correlation and differences between immune cells and immune functions

In order to analyze the correlation between immune cells and functions, based on the results of ssGSEA expression matrix, the corrplot software package in R was used for correlation analysis and heat map was drawn. In addition,for the sake of comparing the differences of immune cells and function in the two groups, the grouping of samples and the results of ssGSEA expression matrix were integrated, and the rank sum test was conducted by R "ggpubr" package, and the box diagram was drawn.

2.9. Identification of DEG based on risk score and immunity score

We applied the "edgeR" package to identify risk-related DEG between high-risk and low-risk groups of THCA and BRCA patients. We downloaded Thyroid papillary carcinoma and Breast Na-seq-v2 data (immunorelated score) of carcinoma from MD Anderson database (https://bioinformatics.mdanderson.org/estimate/disease.html) to represent the invasion level of immune cells in tumor tissue. Meanwhile, according to the immune score, THCA and BRCA patients were divided into two groups, high immunity group and low immunity group, and immune-related DEG was obtained. The R "VennDiagram" package was used to compare the immune score DEG with the risk score DEG, obtain the intersection gene set and draw the Venn diagram.

2.10. Correlation analysis between DEG and immune infiltration

We applied R "survival" package for univariate Cox regression analysis of the intersection genes to screen out the final DEG. In order to analyze the correlation of DEG on immune cells and function, the two result files of ssGSEA and standardized gene expression matrix were analyzed using the "psych" R language software package, and the correlation heat map was constructed.

2.11. Statistical analysis

In the study,statistical analysis was performed using R software (version 4.2.0). Differentially expressed the FRG cut-off value is set to | logFC | 1.0 or higher and the false discovery rate (FDR) < 0.05. t test and Chi-square test were used to calculate whether the difference was significant. P < 0.05 was considered the critical standard for statistical significance.

3.1. Identification of cuproptosis miRNA

Clinicopathological information of 1456 female THCA and BRCA patients was downloaded from the TCGA database. MiRNAs targeting cuproptosis associated genes (CRG) were implemented from TargetScanHuman database. The results showed that 75 cuproptosis related miRNAs were differentially expressed between tumor and non-tumor tissues. Among them, 52 mRNAs in tumor tissues were significantly up-regulated, while 23 were down-regulated (Fig. 1A and B).

3.2. Constructing a prognosis model of miRNA associated with cuproptosis

In this study, we performed univariate Cox regression analysis to obtain 35 DE-CrMs (Fig. 1C). To minimize potential overfitting, 35 DE-CrMs were submitted to LASSO Cox regression analysis to establish prognostic features(Fig. 1D and E). Finally, 15 miRNAs associated with cuproptosis stood out in establishing prognostic models. Multivariate Cox regression analysis was used (Fig. 1F), according to the following formula: risk score = e{(-1.813)*expression level of miR-30e+(-0.002)*expression level of miR-1266+(-0.004)*expression level of miR-651+(-0.009)*expression level of miR-2110+(-0.021)*expression level of miR-627+(-0.026)*expression level of miR-1275+(-0.073)*expression level of miR-346+(-0.092)*expression level of miR-548aw+(-0.126)*expression level of miR-3188+(-0.192)*expression level of miR-4453+(0.022)* expression level of miR-4501+(0.028)*expression level of miR-4788+ (-0.329)* expression level of miR-3907+(0.180)*expression level of miR-466+(-0.037)* expression level of miR-765}.

In this study, patients were divided into a high risk group (n = 717) and a low risk group (n = 718) according to the median cut-off value. The distribution of risk scores and survival status supported the miRNA risk model to divide THCA and BRCA patients into two groups (Fig. 2A and B). Principal component analysis (PCA) showed that the expression patterns of the two groups could be significantly separated (Fig. 2C).Kaplan-Meier survival curve showed that risk stratification could represent different survival states of THCA and BRCA patients. The high-risk subgroup exhibited significantly poorer OS compared to the low-risk subgroup (Fig. 2D,P < 0.0001). Time-dependent ROC and AUC analysis showed that the prognostic ability of the miRNA model was 0.727 (1 year), 0.738 (3 years) and 0.686 (5 years), respectively, indicating that this risk profile has accurate predictive ability for the outcome of THCA and BRCA patients (Fig. 2E).

3.3. Validation of the prognostic risk model by clinicopathological features

To determine the accuracy of the CRM prognostic model, we correlated clinicopathological parameters with risk scores (Fig. 3A ~ E). According to the statistics, age, T stage and N stage were significantly correlated with risk score. To further evaluate the independent risk factors for THCA and BRCA, univariate and multivariate Cox regression analyses were performed combining clinicopathological parameters and risk scores (Fig. 3F and G). Univariate and multivariate analysis of risk score, age and N stage showed that they were independent prognostic factors and could effectively predict Overall Survival of patients (P < 0.05). No significant correlation was found between other clinical parameters and OS of THCA and BRCA patients.

3.4. Establishment and evaluation of line graph model

The establishment of a column graph can be used as a reliable and effective tool to predict OS in THCA and BRCA. We combined miRNA risk characteristics with crucial clinical variables including age, total, T, N, and M stages (Fig. 4A). The survival rate of THCA and BRCA patients can be accurately predicted by the graph, as shown in the calibration curve (Fig. 4B). These results indicated that the histogram can be used as a quantitative method for the prognosis of THCA and BRCA patients, which was of great significance in clinical practice.

3.5. Degree of immune cell and functional infiltration

In the study, we applied ssGSEA to analyze the degree of immunoinfiltration of 29 immune cells and functions in the THCA and BRCA data sets (Fig. 5A). The summarized gene set included aDCs, B cells, CD8 + T cells, DCs, iDCs, Macrophages, Mast cells, Neutrophils, NK cells, pDCs, T helper cells, Tfh, Th1 cells, Th2 cells, TIL, Treg; Immune function included APC co inhibition, APC co stimulation, CCR, Check-point, Cytolytic activity, HLA, Inflammation-promoting, MHC class I, Parainflammation, T cell co-inhibition, T cell co-stimulation, Type I IFN Reponse, Type II IFN Reponse.

3.6. Correlation and difference analysis of immune infiltration

Immune cell and functions correlation heat map (Fig. 5B and C) showed a strong positive correlation between aDCs and T fh (r = 0.83), DCs and T fh pointed a strong positive correlation between aDCs and DCs (r = 0.70), pDCs and Th2 cells showed a strong positive correlation (r = 0.73). A strong positive correlation was demonstrated for APC co stimulation and CCR (r = 0.82), and for CCR and Check-point (r = 0.86). Check-point and T cell co-inhibition revealed strong positive correlation (r = 0.91), and Check-point and T cell co-stimulation showed strong positive correlation (r = 0.82). "HLA and MHC class I may have a very strong positive correlation (r = 0.83) and HLA and Parainflammation may have a very strong positive correlation (r = 0.83).

We analyzed the difference of immunoinfiltrating cells and function in peripheral blood of high-risk group and low-risk group through box diagram, as shown in the Fig. 5D and E(* : P < 0.05, ** : P < 0.01, *** : P < 0.001). The results showed that aDCs, B cells, DCs, Macrophages, Mast cells, Neutrophils, Tfh, Th1 cells, Th2 cells and Treg were significantly different in the immune infiltrated cells. APC co inhibition, APC co stimulation, Check-point, Inflammation-promoting, Parainflammation, T cell co-inhibition, APC co stimulation, check-point, inflammation -promoting, parainflammation, T cell co-inhibition, T cell co-stimulation, Type I IFN Reponse, Type II IFN Reponse showed significant difference between the two groups.

3.7. Identification of DEG based on risk score and immunity score

In view of the above indispensable relation between immune rating and risk score, according to the THCA, BRCA risk score of patients, we identified 6810 DEG, including 1706 raised genetic genes and 5104 down-regulated (| log2FC | > 1 and FDR < 0.05). Simultaneously, 929 DEGs were obtained from the immune subgroup. 406 intersections of DEG were compared between the risk and immune groups. Finally, 118 DEGs were obtained by univariate Cox regression analysis of the intersection genes.

3.8. Correlation analysis between DEG and immune infiltration

Based on two result files of ssGSEA and gene expression matrix, our study obtained the correlation score map of DEG with immune cells and function (Fig. 5F). Six differential expression genes, including PCOLCE, SV2C, FBXO45, TSPY4, LIPH and CBLN3, were found to be significantly correlated with one or more immune cells and their functions in the immune infiltrate (P < 0.001).

Breast cancer and thyroid tumore can be regarded as one of the most common cancers in femlaes, and their incidence might be still on the rise, seriously affecting female's healthy life ^[30–31]. Previous studies have confirmed that thyroid cancer, as the second primary malignant tumor of breast cancer patients (and vice versa), has a high incidence ^{[4, 32]}. Multiple studies have shown an association between breast cancer and thyroid cancer. Whereas, the mechanisms involved remained elusive, which can lead to the search for more effective diagnosis and treatment options. At present, pathological assessment and AJCC TNM staging might be still the main diagnostic and prognostic methods for patients with thyroid and breast cancer, but they may not be accurate and sensitive enough for individualized diagnosis and treatment. The patients with the same carcinoma pathological stage may often have different therapeutic effects and clinical outcomes. Furthermore, accurately predicting the prognosis of cancer patients can accelerate the clinician's ability to determine individual treatment strategies by identifying patients' risk status. Therefore, finding new molecular biomarkers to improve patients' prognosis and quality of life can be an urgent task.

Recently, Tsvetkov et al. announced a new form of copper-induced cell death that has become the focus of the scientific community, called cuproposis^[16]. It was characterized by the accumulation of intracellular free copper and protein lipidation, triggering cytotoxic stress, which induced cell death. As a program of tumor immunotherapy, cell death attracted increasing attention, including ferroptosis ^[17], autophagy, apoptosis, necrosis apoptosis ^[18], as well as cuproptosis induced by ionic carriers acting on copper supplementation. Besides, copper depletion can reduce the activity of angiogenic factors to inhibit vascular development, and copper chelation can be recommended for cancer therapy due to its role as an anti-angiogenic factor ^[33]. Studies have also shown that copper metabolism is a key factor promoting breast cancer ^[34–35]. So a treatment aimed at Cuproposis could be a potential strategy for treating cancer. Nevertheless, the value of cuproptosis in cancer prognosis, especially in breast cancer and thyroid cancer, remained unclear. MicroRNA (miRNA) was a class of short non-coding Rnas with about 19 to 25 nucleotides. By binding with complementary targeted mRNAs, the mRNA can be inhibited or degraded, thus affecting cell cycle, metastasis, metabolism and cell death, which were essential regulatory factors in tumor development. Dysregulation of miRNA was often found in the development of malignant tumors, and it can be applied as a potential biomarker for a variety of cancers ^[36], especially in the exploration of TC and BC therapy, to construct the prognosis characteristics of miRNA targeting CRG.

In this work, we systematically studied the expression of 19 CRGs targeting miRNAs in THCA and BRCA tumor tissues, and the results showed that 35 miRNAs had prognostic value in female patients with THCA and BRCA. We then constructed prognostic features of 15 CrG-targeting miRNAs and demonstrated that these features were independent predictors by multivariate Cox regression analysis. Based on this characteristic, a risk score for each patient was determined. The sample was divided into high-risk and low-risk groups based on the median risk score.The ROC curves and PCA were used to assess the predictive power of constructing this feature. In addition, risk scores and other clinicopathological features were analyzed by univariate and multivariate analyses. The results indicated that risk score can be used as an independent prognostic indicator for THCA and BRCA females. By combining the risk score with these clinicopathological features (age, stage, T stage, N stage, M stage), a novel histogram was constructed to provide personalized predictions.Happily, the charts accurately predicted one -, three - and five-year survival rates for females with thyroid and breast carcinoma. In addition, ssGSEA scores were performed based on THCA and BRCA data sets, so as to estimate the degree of immune infiltration of each sample, and to analyze the correlation and difference of immune infiltration.previous studies have revealed that the immune score is a biomarker to estimate the overall survival rate of TC and BC ^[37–38]. The patients with a high immune score performed better in chemotherapy and immunotherapy. Based on the significant correlation between the immune score and the risk score, 118 DEGs were obtained by univariate Cox regression analysis compared the immune group and the risk group. Finally, correlation analysis was conducted by combining DEGs and immune infiltration, and six differential expression genes, PCOLCE, SV2C, FBXO45, TSPY4, LIPH and CBLN3, were screened out. The results showed that they were closely related to one or more immune cells and their functions during immune infiltration, which may play a key role in the treatment progress of TC and BC.

Among the prognostic features constructed in this study, most miRNAs have appeared in cancer-related studies. For instance, miR-30e acts as an inhibitor of bone metastasis in breast tumor. In addition, overexpression of miR-30 family restores bone homeostasis by reversing the effects of tumor cell conditioning medium on osteoclast and osteoblast generation ^[39].miR-346 knockdown inhibited cell growth in colorectal cancer cells by upregulating LHX6 ^[40]. miR-466 declined the growth and metastasis of prostate cancer by directly targeting the bone-associated transcription factor RUNX2 ^[41]. miR-765 regulated the growth and metastasis of breast cancer by regulating the Mir-765-ING4 negative feedback loop ^[42]. These literature studies suggested that miRNA markers associated with cuproptosis may be associated with the progression of TC and BC, and reveal expected targets for the treatment of TC and BC.

Currently, our study may have some limitations, including insufficient sample size and lack of experimental validation. On the one hand, our study relied on publicly available TCGA information, and this miRNA prognostic model associated with cuproptosis needed to be further validated with multi-center, prospective, real-world data. On the other hand, although our study preliminarily identified an association between cuproptosis related miRNAs and anti-tumor immunity, the underlying mechanisms must be further explored through other experiments.

In summary, the study identified reliable prognostic characteristics based on cuproptosis-related miRNAs. It was used as a reliable prognostic model and a highly available column graph was constructed, including independent prognostic factors such as risk scores. The results of our study may be beneficial for personalized prediction and treatment strategies for cancer patients.Nevertheless, inccreasing large-sample studies might be needed to verify the feasibility of risk characteristics. The potential mechanism between miRNA labeling and TC and BC development may warrant further investigation.

CRM : Cuproptosis related miRNA

TCGA : The Cancer Genome Atlas

K-M : Kaplan-Meyer

ROC : Receiver operating characteristics

DEGs : Differential expression genes

BC : Breast cancer

TC : Thyroid cancer

AIC : Akaike information criteria

PCA : Principal component analysis

AUC : Area under the curve

FDR : False discovery rate

CRG : Cuproptosis associated genes

Acknowledgments

We are grateful to the TCGA database for providing the platform and to the contributors for uploading their meaningful datasets.

Funding

Not applicable

Data Availability Statement

The public datasets for this study can be found in the TCGA database (https://portal.gdc.cancer.gov/).

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Consent for publication

All authors consent to the publication of this study.

.Author contributions

XS designed the overall study and revised the paper. JZ and YT drafted the manuscript and performed the data analysis. YZ and YL participated in the data collection. All authors contributed to the article and approved the submitted version.

Author details

¹The Second Affiliated Hospital of Nanjing University of Chinese Medicine, Jiangsu Provincial Second Hospital of Chinese Medicine. Nanjing, China.

Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, Parkin DM, Forman D, Bray F. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015 Mar 1;136(5):E359-86. doi: 10.1002/ijc.29210. Epub 2014 Oct 9. PMID: 25220842.
Morrow M, Burstein HJ, Harris JR. Malignant tumors of the breast. 2015.
Howlader, N., Noone, A.M., Krapcho, M., et al. (2019) SEER Explorer. Breast Cancer-Stage Distribution of SEER Incidence Cases, 2007-2016 by Sex. National Cancer Institute, Bethesda.
Nielsen SM, White MG, Hong S, Aschebrook-Kilfoy B, Kaplan EL, Angelos P, Kulkarni SA, Olopade OI, Grogan RH. The Breast-Thyroid Cancer Link: A Systematic Review and Meta-analysis. Cancer Epidemiol Biomarkers Prev. 2016 Feb;25(2):231-8. doi: 10.1158/1055-9965.EPI-15-0833. PMID: 26908594; PMCID: PMC4770576.
Mellemkjær L, Christensen J, Frederiksen K, Pukkala E, Weiderpass E, Bray F, Friis S, Andersson M, Olsen JH. Risk of primary non-breast cancer after female breast cancer by age at diagnosis. Cancer Epidemiol Biomarkers Prev. 2011 Aug;20(8):1784-92. doi:10.1158/1055-9965.EPI-11-0009. Epub 2011 Jun 8. PMID: 21653644.
Chereau N, Trésallet C, Noullet S, Godiris-Petit G, Tissier F, Leenhardt L, Menegaux F. Prognosis of papillary thyroid carcinoma in elderly patients after thyroid resection: A retrospective cohort analysis. Medicine (Baltimore). 2016 Nov;95(47):e5450. doi: 10.1097/MD.0000000000005450. PMID: 27893690; PMCID: PMC5134883.
Wiltshire JJ, Drake TM, Uttley L, Balasubramanian SP. Systematic Review of Trends in the Incidence Rates of Thyroid Cancer. Thyroid. 2016 Nov;26(11):1541-1552.doi:10.1089/thy.2016.0100. PMID: 27571228.
United States Cancer Statistics: Data Visualizations—Demographics. [(accessed on 16 December 2021)]
Bolf EL, Sprague BL, Carr FE. A Linkage Between Thyroid and Breast Cancer: A Common Etiology? Cancer Epidemiol Biomarkers Prev. 2019 Apr;28(4):643-649. doi: 10.1158/1055-9965.EPI-18-0877. Epub 2018 Dec 12. PMID: 30541751.
Huang NS, Chen XX, Wei WJ, Mo M, Chen JY, Ma B, Yang SW, Xu WB, Wu J, Ji QH, Guo XM, Liu GY, Shao ZM, Wang Y. Association between breast cancer and thyroid cancer: A study based on 13 978 patients with breast cancer. Cancer Med. 2018 Dec;7(12):6393-6400. doi: 10.1002/cam4.1856. Epub 2018 Nov 27. PMID: 30480382; PMCID: PMC6308067.
Kuo JH, Chabot JA, Lee JA. Breast cancer in thyroid cancer survivors: An analysis of the Surveillance, Epidemiology, and End Results-9 database. Surgery. 2016 Jan;159(1):23-9. doi: 10.1016/j.surg.2015.10.009. Epub 2015 Oct 29. PMID: 26522696.
Shi Y, Li X, Ran L, Arshad B, Li H, Xu Z, Zhao C, Wu Y, Wu H, Chen H, Li HY, Wu KN, Kong LQ. Study on the status of thyroid function and thyroid nodules in chinese breast cancer patients. Oncotarget. 2017 Aug 24;8(46):80820-80825. doi: 10.18632/oncotarget.20542. PMID: 29113346; PMCID: PMC5655241.
McTiernan A, Weiss NS, Daling JR. Incidence of thyroid cancer in women in relation to known or suspected risk factors for breast cancer. Cancer Res. 1987 Jan 1;47(1):292-5. PMID: 3791213.
Dong L, Lu J, Zhao B, Wang W, Zhao Y. Review of the possible association between thyroid and breast carcinoma. World J Surg Oncol. 2018 Jul 5;16(1):130. doi: 10.1186/s12957-018-1436-0. PMID: 29976206; PMCID: PMC6034293.
Bisaglia M, Bubacco L. Copper Ions and Parkinson's Disease: Why Is Homeostasis So Relevant? Biomolecules. 2020 Jan 29;10(2):195. doi: 10.3390/biom10020195. PMID: 32013126; PMCID: PMC7072482.
Tsvetkov P, Coy S, Petrova B, Dreishpoon M, Verma A, Abdusamad M, Rossen J, Joesch-Cohen L, Humeidi R, Spangler RD, Eaton JK, Frenkel E, Kocak M, Corsello SM, Lutsenko S, Kanarek N, Santagata S, Golub TR. Copper induces cell death by targeting lipoylated TCA cycle proteins. Science. 2022 Mar 18;375(6586):1254-1261. doi: 10.1126/science.abf0529. Epub 2022 Mar 17. Erratum in: Science. 2022 Apr 22;376(6591):eabq4855. PMID: 35298263; PMCID: PMC9273333.
Jiang X, Stockwell BR, Conrad M. Ferroptosis: mechanisms, biology and role in disease. Nat Rev Mol Cell Biol. 2021 Apr;22(4):266-282. doi: 10.1038/s41580-020-00324-8. Epub 2021 Jan 25. PMID: 33495651; PMCID: PMC8142022.
Bedoui S, Herold MJ, Strasser A. Emerging connectivity of programmed cell death pathways and its physiological implications. Nat Rev Mol Cell Biol. 2020 Nov;21(11):678-695. doi: 10.1038/s41580-020-0270-8. Epub 2020 Sep 1. PMID: 32873928.
Tsvetkov P, Detappe A, Cai K, Keys HR, Brune Z, Ying W, Thiru P, Reidy M, Kugener G, Rossen J, Kocak M, Kory N, Tsherniak A, Santagata S, Whitesell L, Ghobrial IM, Markley JL, Lindquist S, Golub TR. Mitochondrial metabolism promotes adaptation to proteotoxic stress. Nat Chem Biol. 2019 Jul;15(7):681-689. doi: 10.1038/s41589-019-0291-9. Epub 2019 May 27. Erratum in: Nat Chem Biol. 2019 Jun 4;: PMID: 31133756; PMCID: PMC8183600.
Baltaci AK, Dundar TK, Aksoy F, Mogulkoc R. Changes in the Serum Levels of Trace Elements Before and After the Operation in Thyroid Cancer Patients. Biol Trace Elem Res. 2017 Jan;175(1):57-64. doi: 10.1007/s12011-016-0768-2. Epub 2016 Jun 6. PMID: 27263537.
Jouybari L, Kiani F, Islami F, Sanagoo A, Sayehmiri F, Hosnedlova B, Doşa MD, Kizek R, Chirumbolo S, Bjørklund G. Copper Concentrations in Breast Cancer: A Systematic Review and Meta-Analysis. Curr Med Chem. 2020;27(37):6373-6383. doi:10.2174/0929867326666190918120209. PMID: 31533596.
Dìez M, Cerdàn FJ, Arroyo M, Balibrea JL. Use of the copper/zinc ratio in the diagnosis of lung cancer. Cancer. 1989 Feb 15;63(4):726-30. doi: 10.1002/1097-0142(19890215)63:4<726::aid-cncr2820630421>3.0.co;2-p. PMID: 2914279.
Saleh SAK, Adly HM, Abdelkhaliq AA, Nassir AM. Serum Levels of Selenium, Zinc, Copper, Manganese, and Iron in Prostate Cancer Patients. Curr Urol. 2020 Mar;14(1):44-49. doi: 10.1159/000499261. Epub 2020 Mar 20. PMID: 32398996; PMCID: PMC7206590.
Ge EJ, Bush AI, Casini A, Cobine PA, Cross JR, DeNicola GM, Dou QP, Franz KJ, Gohil VM, Gupta S, Kaler SG, Lutsenko S, Mittal V, Petris MJ, Polishchuk R, Ralle M, Schilsky ML, Tonks NK, Vahdat LT, Van Aelst L, Xi D, Yuan P, Brady DC, Chang CJ. Connecting copper and cancer: from transition metal signalling to metalloplasia. Nat Rev Cancer. 2022 Feb;22(2):102-113. doi: 10.1038/s41568-021-00417-2. Epub 2021 Nov 11. PMID: 34764459; PMCID: PMC8810673.
Cui L, Gouw AM, LaGory EL, Guo S, Attarwala N, Tang Y, Qi J, Chen YS, Gao Z, Casey KM, Bazhin AA, Chen M, Hu L, Xie J, Fang M, Zhang C, Zhu Q, Wang Z, Giaccia AJ, Gambhir SS, Zhu W, Felsher DW, Pegram MD, Goun EA, Le A, Rao J. Mitochondrial copper depletion suppresses triple-negative breast cancer in mice. Nat Biotechnol. 2021 Mar;39(3):357-367. doi: 10.1038/s41587-020-0707-9. Epub 2020 Oct 19. PMID: 33077961; PMCID: PMC7956242.
Li H, Chan L, Bartuzi P, Melton SD, Weber A, Ben-Shlomo S, Varol C, Raetz M, Mao X, Starokadomskyy P, van Sommeren S, Mokadem M, Schneider H, Weisberg R, Westra HJ, Esko T, Metspalu A, Kumar V, Faubion WA, Yarovinsky F, Hofker M, Wijmenga C, Kracht M, Franke L, Aguirre V, Weersma RK, Gluck N, van de Sluis B, Burstein E. Copper metabolism domain-containing 1 represses genes that promote inflammation and protects mice from colitis and colitis-associated cancer. Gastroenterology. 2014 Jul;147(1):184-195.e3. doi: 10.1053/j.gastro.2014.04.007. Epub 2014 Apr 13. PMID: 24727021; PMCID: PMC4086320.
St Laurent G 3rd, Seilheimer B, Tackett M, Zhou J, Shtokalo D, Vyatkin Y, Ri M, Toma I, Jones D, McCaffrey TA. Deep Sequencing Transcriptome Analysis of Murine Wound Healing: Effects of a Multicomponent, Multitarget Natural Product Therapy-Tr14. Front Mol Biosci. 2017 Aug 17;4:57. doi: 10.3389/fmolb.2017.00057. PMID: 28879183; PMCID: PMC5572416.
Wang H. Predicting cancer-related MiRNAs using expression profiles in tumor tissue. Curr Pharm Biotechnol. 2014;15(5):438-44. doi: 10.2174/1389201015666140519121255. PMID: 24846064.
Tibshirani R. The lasso method for variable selection in the Cox model. Stat Med. 1997 Feb 28;16(4):385-95. doi: 10.1002/(sici)1097-0258(19970228)16:4<385::aid-sim380>3.0.co;2-3. PMID: 9044528.
Ahmad A. Breast Cancer Statistics: Recent Trends. Adv Exp Med Biol. 2019;1152:1-7. doi: 10.1007/978-3-030-20301-6_1. PMID: 31456176.
Davies L, Welch HG. Current Thyroid Cancer Trends in the United States. JAMA Otolaryngol Head Neck Surg. 2014;140(4):317–322. doi:10.1001/jamaoto.2014.1
Joseph KR, Edirimanne S, Eslick GD. The association between breast cancer and thyroid cancer: a meta-analysis. Breast Cancer Res Treat. 2015 Jul;152(1):173-181. doi: 10.1007/s10549-015-3456-6. Epub 2015 Jun 10. PMID: 26058757.
Chan N, Willis A, Kornhauser N, Ward MM, Lee SB, Nackos E, Seo BR, Chuang E, Cigler T, Moore A, Donovan D, Cobham MV, Fitzpatrick V, Schneider S, Wiener A, Guillaume-Abraham J, Aljom E, Zelkowitz R, Warren JD, Lane ME, Fischbach C, Mittal V, Vahdat L. Correction: Influencing the Tumor Microenvironment: A Phase II Study of Copper Depletion Using Tetrathiomolybdate in Patients with Breast Cancer at High Risk for Recurrence and in Preclinical Models of Lung Metastases. Clin Cancer Res. 2020 Sep 15;26(18):5051. doi: 10.1158/1078-0432.CCR-20-3177. Erratum for: Clin Cancer Res. 2017 Feb 1;23(3):666-676. PMID: 32934032.
Ramchandani D, Berisa M, Tavarez DA, Li Z, Miele M, Bai Y, Lee SB, Ban Y, Dephoure N, Hendrickson RC, Cloonan SM, Gao D, Cross JR, Vahdat LT, Mittal V. Copper depletion modulates mitochondrial oxidative phosphorylation to impair triple negative breast cancer metastasis. Nat Commun. 2021 Dec 15;12(1):7311. doi: 10.1038/s41467-021-27559-z. PMID: 34911956; PMCID: PMC8674260.
Laws K, Bineva-Todd G, Eskandari A, Lu C, O'Reilly N, Suntharalingam K. A Copper(II) Phenanthroline Metallopeptide That Targets and Disrupts Mitochondrial Function in Breast Cancer Stem Cells. Angew Chem Int Ed Engl. 2018 Jan 2;57(1):287-291. doi: 10.1002/anie.201710910. Epub 2017 Dec 5. PMID: 29144008.
Rupaimoole R, Slack FJ. MicroRNA therapeutics: towards a new era for the management of cancer and other diseases. Nat Rev Drug Discov. 2017 Mar;16(3):203-222. doi: 10.1038/nrd.2016.246. Epub 2017 Feb 17. PMID: 28209991.
Wu P, Shi J, Wang Z, Sun W, Zhang H. Evaluate the immune-related eRNA models and signature score to predict the response to immunotherapy in thyroid carcinoma. Cancer Cell Int. 2022 Oct 10;22(1):307. doi: 10.1186/s12935-022-02722-8. PMID: 36217201; PMCID: PMC9549686.
Wang S, Zhang Q, Yu C, Cao Y, Zuo Y, Yang L. Immune cell infiltration-based signature for prognosis and immunogenomic analysis in breast cancer. Brief Bioinform. 2021 Mar 22;22(2):2020-2031. doi: 10.1093/bib/bbaa026. PMID: 32141494.
Croset M, Pantano F, Kan CWS, Bonnelye E, Descotes F, Alix-Panabières C, Lecellier CH, Bachelier R, Allioli N, Hong SS, Bartkowiak K, Pantel K, Clézardin P. miRNA-30 Family Members Inhibit Breast Cancer Invasion, Osteomimicry, and Bone Destruction by Directly Targeting Multiple Bone Metastasis-Associated Genes. Cancer Res. 2018 Sep 15;78(18):5259-5273. doi: 10.1158/0008-5472.CAN-17-3058. Epub 2018 Jul 24. PMID: 30042152.
Li X, Zhou Y, Wen P, Yuan Y, Xiao Z, Shi H, Zhou H. Tumor suppressor LHX6 upregulation contributes to the inhibitory effect of miR-346 knockdown on colorectal cancer cell growth. Environ Toxicol. 2022 Mar;37(3):435-445. doi: 10.1002/tox.23410. Epub 2021 Nov 13. PMID: 34773443.
Colden M, Dar AA, Saini S, Dahiya PV, Shahryari V, Yamamura S, Tanaka Y, Stein G, Dahiya R, Majid S. MicroRNA-466 inhibits tumor growth and bone metastasis in prostate cancer by direct regulation of osteogenic transcription factor RUNX2. Cell Death Dis. 2017 Jan 26;8(1):e2572. doi: 10.1038/cddis.2017.15. PMID: 28125091; PMCID: PMC5386393.
Jiao Y, Yuan C, Wu H, Li X, Yu J. Oncogenic microRNA-765 promotes the growth and metastasis of breast carcinoma by directly targeting ING4. J Cell Biochem. 2019 Nov 13. doi: 10.1002/jcb.29545. Epub ahead of print. PMID: 31724215.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Prognostic model and immunomicroenvironment analysis of combined thyroid and breast cancer in females based on cuproptosis related miRNA

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials And Methods

2.1. Data downloading and organizing

2.2. Identification of miRNA associated with cuproptosis

2.3 Construction of cuproptosis related miRNA prognostic model

2.4 Validation of the prognostic risk model

2.5. Prognostic independence analysis of clinicopathological parameters and risk scores

2.6. Building and evaluating the column graph model

2.7. Extraction and quantification of immunoinfiltrating cells and their functions

2.8. Analysis of correlation and differences between immune cells and immune functions

2.9. Identification of DEG based on risk score and immunity score

2.10. Correlation analysis between DEG and immune infiltration

2.11. Statistical analysis

3. Results

3.1. Identification of cuproptosis miRNA

3.2. Constructing a prognosis model of miRNA associated with cuproptosis

3.3. Validation of the prognostic risk model by clinicopathological features

3.4. Establishment and evaluation of line graph model

3.5. Degree of immune cell and functional infiltration

3.6. Correlation and difference analysis of immune infiltration

3.7. Identification of DEG based on risk score and immunity score

3.8. Correlation analysis between DEG and immune infiltration

4. Discussion

5. Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1