Journal Pre-proof

The androgen receptor expression and association with patient’s survival in different cancers

GENOMICS

Chao Hu, Dan Fang, Haojun Xu, Qianghu Wang, Hongping Xia

PII:	S0888-7543(19)30392-1
DOI:	https://doi.org/10.1016/j.ygeno.2019.11.005
Reference:	YGENO 9398
To appear in:	Genomics
Received date:	28 June 2019
Revised date:	4 November 2019
Accepted date:	11 November 2019

Please cite this article as: C. Hu, D. Fang, H. Xu, et al., The androgen receptor expression and association with patient’s survival in different cancers, Genomics (2019), https://doi.org/10.1016/j.ygeno.2019.11.005

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

C 2019 Published by Elsevier.

The androgen receptor expression and association with patient’s survival in different cancers

Chao Hu1,2,3, Dan Fang1,2,3, Haojun Xu 1,2,3, Qianghu Wang4, Hongping Xia1,2,3,* xiahongping@njmu.edu.cn

1State Key Laboratory of Reproductive Medicine, Nanjing Medical University, Nanjing 211166, China

2Key Laboratory of Antibody Technique of National Health Commission, Nanjing Medical University Nanjing 211166, China

3Department of Pathology, School of Basic Medical Scie ices & The Affiliated Sir Run Run Hospital, Nanjing Medical University, Nanjing 21116, China

4Department of Bioinformatics, School of Bior co ical Engineering and Informatics, Nanjing Medical University, 211116, Nanjing, Chi -.

Jelen well Predicar lure Of

Corresponding author at: State Kov Laboratory of Reproductive Medicine & Department of Pathology, School of Basic Medical Sciences & The Affiliated Sir Run Run Hospital, Nanjing Medical University Nanjing 21116, China.

Abstract

To understand the androgen receptor (AR) in different human malignancies, we conducted a pan-cancer analysis of AR in different tumor tissues and association with patient survival and obtained AR expression data from The Cancer Genome Atlas. Pan-Cancer Analysis of AR indicated that 12 tumor types had decreased AR expression in the tumor, while glioblastoma multiforme has overexpressed AR. The survival analysis showed that high AR mRNA is

associated with poor survival of stomach adenocarcinoma and low-grade glioma, but better survival of adrenocortical carcinoma, kidney renal clear cell carcinoma, acute myeloid leukemia, liver hepatocellular carcinoma, ovarian serous cystadenocarcinoma, and skin cutaneous melanoma based on AR mRNA, protein or AR-score. AR was associated with different clinical characteristics and AR correlated genes enriched in cancer-related pathways. These data indicate that AR signaling may be strongly associated with some cancer development and patients’ survival, which is promising for potential treatment using antiandrogen therapies.

Keywords: Pan-cancer analysis; androgen receptor; patient survival; mRNA; protein

1. Introduction

The androgen receptor (AR), also known as nuclear receptor subfamily 3 group C member 4(NR3C4), belongs to the puchar receptor superfamily and is encoded by the AR gene located at Xq11-12 of the X chromosome. AR was first discovered and characterized in the late 1960s by several independent groups[1]. There are two prominent isoforms of the androgen receptors AR-A (87 kDa) and AR-B (110 kDa), with three major functional domains the N-terminal domain, DNA-binding domain, and androgen-binding domain [2]. AR is a type of steroid and nuclear receptor that is activated by the androgenic hormones like testosterone and dihydrotestosterone. As a zinc finger DNA-binding transcription factor, AR is activated by phosphorylation and dimerization and translocated from the cytoplasm to the nucleus. Testosterone and dihydrotestosterone are responsible for the male sexual characteristics by activating AR. Activation of AR regulates androgen related genes expression and is important for the development and maintenance of the male sexual phenotype [3]. AR is expressed in a diverse range of tissues and mediated androgen actions to

play important biological actions in bone, muscle, prostate, adipose tissue, cardiovascular, immune, neural, hemopoietic systems, and both male and female reproductive development and function[4]. AR deficiencies will cause androgen insensitivity syndrome[5], Kennedy’s disease[6], and other additional disorders[7].

AR has been investigated extensively in hormone-dependent cancers such as prostate and breast cancer. Cancer cells that are AR positive may need androgens to grow and progression. Emerging evidence also suggested the potential ir aportance of the AR signaling in other hormone-independent human malignancies. For example, sex differences are associated with various cancers according to the cancer ej idemiology data. The incidence of liver and stomach cancer has been shown higher in nale. than in females [8]. Comprehensive characterization of molecular differences between male and female tumor tissues hasshown the extensive sex-biased gene expression sion. tures in some cancer types[9]. Besides lifestyle and genetics, sex hormones may also play an important role in contributing to the cancer incidence differences between men and women in some cancers. Increasing evidence support La the role of AR signaling in ot iei hormone-independent tumor types, including bladder, kidney, pancreatic, liver, enon trial, mantle cell lymphoma, and salivary gland cancers, etc.[10]. However, the critical role of AR signaling in other hormone-independent human malignancies is poorly u derstood. Here, we conducted a pan-cancer analysis of AR mRNA and protein expression in different tumor tissues and the association with patient survival. This study explores the potential importance of AR signaling in different malignancies and the potential treatment implications of using antiandrogen therapies in these malignancies.

2. Materials and Methods

2.1. Data Sources

The data we used for all analysis were obtained from TCGA (The Cancer Genome Atlas), RNA-Seq expression data (level 3) were downloaded from Broad GDAC Firehose (http://gdac.broadinstitute.org/), include genes, isoform, and Z-score data. RPPA protein expression data (level 4) were downloaded from TCPA (https://tcpaportal.org/). Clinical data we used was TCGA-Clinical Data Resource (CDR) Outcome, a curated resource of the clinical annotations for TCGA data and provides recommendations for the use of clinical endpoints [11], downloaded from TCGA Pan-Cancer Atlas (https://gdc.cancer.gov/about- data/publications/pancanatlas).

2.2. AR-score analysis

For each sample in each cancer type, the signal ng or activation of AR-pathway is indicated by AR-score. The AR-score is derived from the mRNA expression of genes that are experimentally validated AR transcrip’10. al targets[12]. AR-Score was calculated by composite expression of a list of20 genes. The AR output score for each sample is computed as the average of the Z-scores of the A signaling gene signature. Z-score was computed for the expression ofeach gene in each sample by subtracting the mean of the mRNA expression values and dividing by the standard deviation.

2.3. AR mRNA/Prote .. . /Iscforms expression and AR activity in different cancer types

Normalized RNA-Seq data (mRNA and Isoforms data) downloaded from TCGA were transformed by log2(x+1) before subsequent analysis. For different data types, AR expression between tumor and normal samples was compared for each cancer type by Wilcoxon rank-sum test, and the expression difference between tumor and matched normal from the same patient were compared by paired Wilcoxon rank-sum test. AR expression differences between males and females was compared by the Wilcoxon rank-sum test for cancers that include both genders. Pearson correlation coefficients among mRNA, protein,

and AR-score were calculated for all cancers. AR expression across different clinical characteristics was compared, p-value was calculated by the Wilcoxon rank-sum test for every two categories of every characteristic.

2.4. Survival Analysis

For each cancer type, patients were separated into two groups based on median values of AR mRNA expression, protein expression, AR-score, or isoforms expression. Differences in overall survival and progression-free interval between high- pression and low-expression groups were compared using Kaplan-Meier curves, with p-va. “es calculated via log-rank test, using the Survival package in R. Univariate Cox regression analysis and Multivariate Cox analysis was used to compare the influence of cor tinuous AR expression on survival along with other clinical characteristics, such as stage, grade, age, histology, gender, race, tumor status, and new tumor events.

2.5. Pathway enrichment analysis

For pathway enrichment a na’Sis, we firstly generated a list of AR corrected genes for each TCGA tumors. Pearson con lation coefficients between AR expression and that of every gene and adjusted p-value (q val., adjusted by the Bonferroni method) were calculated. The genes with an absolute value of con lation coefficient over 0.5 and q value less than 0.05 were applied to subsequent pathway enrichment analysis. Pathway enrichment analysis was performed by cluster profile R package.

3. Results

We obtained AR gene expression data by RNA-seq and protein expression determined by Reverse Phase Protein Array (RPPA) that have corresponding clinical

information over 10000 patients across 33 cancer types published by The Cancer Genome Atlas (TCGA). The basic characteristics of each TCGA cancer were reported in Table 1. Clinical variables, including age, gender, race, stage, grade, new tumor events, tumor status, histology, were considered in our analysis. And outcome endpoints of overall survival (OS), defined as the period from date ofdiagnosis until death from any cause, and progression-free interval (PFI), defined as the period from date of diagnosis until the occurrence ofan event in which the patient with or without the tumor does not get worse, were considered as the endpoints event for each cancer type (excluded PFI for LAML)

3.1. A pan-cancer analysis of the AR mRNA expression in . ifferent cancer types

To understand the critical role of AR signaling in human malignancies, we firstly conducted a pan-cancer analysis of AR mRNA expression in different tumor tissues. Among the 33 tumor types profiled by TCGA, the .k. TA expression difference of AR in 23 tumor types with both tumor and normal tissues was shown in Figure 1A. 13 of 23 tumor types showed significantly different expression (with Wilcoxon rank-sum test p-value < 0.05 and absolute log2 fold change > 1) of AR mRNA between tumor and normal tissues. Among them, 12 tumor types showed significantly low expression of AR mRNA in tumor than normal tissues, while on. v or e tumor type (GBM) showed higher expression of AR mRNA in tumor than normal tissue (Fig.1). We showed the expression difference between tumor and matched normal from the same patient for 22 tumor types in Figure 1B. Moreover, we also compared the mRNA expression difference of AR in each cancer type between females and males with p-value calculated by the Wilcoxon rank-sum test. There are three tumor types (DLBC, SARC, and THYM) showed statistically significant differences between male and female tumor tissues (Fig.1C).

3.2. A pan-cancer analysis of the AR-score in different cancer types

To further understand the critical role of AR signaling and activation of AR-pathway in human malignancies, we computed AR-score for each sample in TCGA used a 20-gene signature. And the difference of AR-score between tumor and normal was compared (Fig.2A). Fifteen cancer types showed significantly low AR-score in tumor tissues than normal only PRAD showed higher AR-score. And the similar result was observed in the AR-score difference between paired tumor and normal, except for UCEC (Fig.2B). We also compared AR-score differences between male and female, and BLCA and THYM showed a statistically significant difference (Fig.2C).

3.3. A pan-cancer analysis of the AR protein expression in . lifferent cancer types

We also conducted a pan-cancer analysis of 2 R r rotein expression in different tumor tissues. Protein expression of AR in differer cancer types was showed in Figure 3A. Meanwhile, we also compared the proteinexpression difference of AR in each cancer type between females and males in Figure 3B. The result showed that BLCA, BRCA, KIRC, KIRP, LGG, LIHC, LUAD, MESO, HPG, READ, SKCM, and THCA have statistically significant difference (P-value 0.05). AR protein expression was higher in male tumor tissues compared to female tun - tissues of these 12 tumor types.

3.4. AR expression le, aland association with clinical characteristics

For each cancer type, the AR expression data across different patient clinical characteristics were analyzed. As shown in Figure 4, the AR mRNA expression was associated with one or more clinical characteristics in 15 tumors. The AR expression was different between age categories in BRCA, LIHC, and SARC. Tumor free patients showed high AR expression levels in ACC, KIRC, KIRP, UCEC, while LGG and STAD showed low AR expression. Decreased expression of AR correlated significantly with the tumor stage in BRCA, KIRC, THCA, as opposed to STAD. High AR expression was associated with high

tumor grade in ESCA, HNSC, STAD, on the contrary of KIRC, LIHC, UCEC. Besides, AR expression was significantly associated with race in BRCA, LIHC, LGG, STAD, significantly associated with histological type in COAD, LGG, THCA, UCEC, UVM, and significantly associated with new tumor event in CHOL, HNSC, STAD. We also investigated AR protein expression and AR-score across clinical characteristics, and in the BRCA, COAD, HNSC, KIRC, LGG, LIHC, SARC, STAD, THCA, UCEC, UVM patients, we observed similar AR protein or AR-score expression patterns to mRNA in age, race, stage, grade, tumor status and histological type (Fig.5). We investigated the correlation between AR mRNA expression, protein expression, and AR-score. Ou’ result showed that the AR mRNA was a significant positive correlation with protein expression in most tumor types (except DLBC, PAAD, PRAD, THYM, UVM), mRNA was generally correlation with AR-score in a low correlation coefficient, while protein ard AP .- score have variable correlation among different tumors (Fig.S1-S3).

3.5. The association of AR mRNA Love with the survival of patients in different cancer types

We next investigated th association of AR mRNA expression level with the survival of patients in different cancer types. Cases were assigned into two groups (High group and Low group) using medina expression value as the cutoff for each tumor type. The difference in OS and PFI between High group and Low group were compared using Kaplan-Meier survival curves, and statistical significance was calculated by log-rank test for each cancer type (Fig.6). The result showed that high AR mRNA expression was associated with good OS and PFS in KIRC, while high AR mRNA expression was associated with poor OS and PFS in STAD (Fig.6 C and G). High AR mRNA expression was also associated with favorable OS of ACC, LIHC, LAML, OV (Fig.6 A, B, D, F). Although there are no significant differences between High and Low groups in LGG, we observed that numeric AR expression was

significantly associated with patient survival that higher AR expression patients had a worse prognosis (Fig.6E). The rest of the cancer types did not show statistically significantly differences in high versus low AR mRNA expression.

3.6. The association of AR-score with the survival of patients in different cancer types

We applied the same survival analysis for AR-score as the mRNA used in different cancer types. The result showed that high AR protein expression was associated with good OS and PFS in KIRC and STAD, while high AR protein expres ion was associated with poor OS and PFS in LGG (Fig.7 A-C). The rest of the cancer De, did not show statistically significantly differences in high versus low AR-score.

3.7. The association of AR protein level with the su. Ival of patients in different cancer types

We further investigated the association of AR protein levels with the survival of patients in different cancer types (Fjø 8, Kaplan-Meier survival curves and log-rank test p- values for both OS and PFS of difierein cancer types showed that high AR protein expression was statistically significantly associated with good OS and PFS in KIRC and SKCM, while high AR protein expression was significantly associated with poor OS and PFS in LGG (Fig.8 A, C, E). High A.` protein expression was also significantly associated with good OS of OV and LIHC (Fig.8 B, D). The rest of the cancer types did not show statistically significant differences in high versus low AR protein expression.

3.8. AR isoforms analysis in different cancer types

The firehose database provided splicing variants expression data for all TCGA cancers, and we extracted the five different AR isoform (uc004dwu.1, uc004dwv.1, uc011mpd.1, uc011mpe.1, uc011mpf.1) expression data (Table S2). The expression level of

the five iso forms was compared between the tumor and normal (Fig.9A, Fig.S4). uc004dwu.1 and uc004dwv.1 were the predominant splicing variant of AR in TCGA cancers and highly positive correlation with total AR mRNA expression. And uc011 mpf.1 (known as AR-V7) have relatively high expression in BRCA and PRAD compared to other tumors. Besides, survival analysis showed that high uc011mpd.One expression was associated with worse survival in GBM and high uc011mpf.One expression was associated with worse survival in PRAD (Fig.9B and C).

3.9. The Cox regression analysis of AR expression level in din. rent cancer types

The cox analysis of AR expression in different «ance. types was further investigated. Continuous AR expression was used in Cox more sion analysis, and other clinical characteristics (age, gender, stage, grade, rice, new tumor event, tumor status, and histological types) were included in nu tiveriate cox regression. The univariate and multivariate cox regression analysis showed that AR mRNA level has independent prognostic variable for patients overall survival (n +SC, KIRC, LAML and LGG (table 2), AR protein level has independent prognostic variable in KIRC for overall survival and progression-free interval, AR-score has indepe dent prognostic variable in LGG for overall survival and progression-free interval (tab e S1).

3.10. Pathway enrichment analysis of AR expression correlated genes

For each TCGA cancers, the Pearson correlation coefficients between AR mRNA expression and other genes was calculated, significantly AR correlated genes were selected with conditions that absolute value of correlation coefficient over 0.5 and q value less than 0.05. The number of genes significantly correlated with AR varies greatly in different TCGA tumors (from 0 to 3718) (Fig.10A). KEGG pathway enrichment analysis was performed on selected significantly correlated genes for each tumor, respectively. The genes were

significantly enriched in 35 cancer-associated pathways in 14 TCGA cancers, and the significantly enriched pathways could be divided into four groups: Cellular Processes, Cell Signaling, Cancer-related, and Drug Resistance, according to the pathway function (Fig.10 B).

4. Discussion

The main aim of the current investigation is based on the hypothesis that androgen receptor plays a critical role in the development of different tumo. types. The AR differential expression analysis across Pan-Cancer data sets indicates + AR signaling may be strongly associated with some cancer development and patients survival. Previous studies reported AR to play an import role in prostate cancer develop nent through regulation of not only transcription networks but also genomic stability and DNA repair[13]. AR promoted invasion and angiogenesis in bladder c nce : through regulating CD24 and TSP1[14]. Pan- Cancer Analysis of AR differential expression across TCGA data sets indicated that 13 tumor types (BLCA, CESC, CHOL, COAL DESCA, HNSC, KIRP, LIHC, LUSC, READ, STAD, THCA, UCEC) have decrease. A. mRNA in tumor compared to normal or paired normal tissues, while 1 tumor (GB).^) have overexpressed AR mRNA in tumor compared to normal tissues.

The previous study found that estrogenic, not testosterone, immunoactivity CHOL lesions[15], and our analysis showed that AR expression lower in CHOL and patients with new tumor events, suggesting potential protective roles of AR in CHOL progression. The previous study showed that the expression and activation of ARs in colon tumors results in induction of anti-tumor responses and extensive reduction of tumor incidence [16], increasing number of AR CAG repeats was directly associated with colon cancer among men, and AR gene may modulate tumorigenesis of vitamin D, vitamin D receptor[17, 18]. Our analysis

showed that AR mRNA level was decreased in COAD and READ, and AR expression correlated genes were enriched in multiple tumor-related pathways, including Focal adhesion, Wnt signaling pathway, PI3K-Akt signaling pathway, and MAPK signaling pathway et. The studies have shown high AR expression based on immunohistochemistry in salivary duct carcinoma[19] and laryngeal carcinoma[20]. AR-positive salivary duct carcinoma may be promising for androgen deprivation therapy[21]. In kidney cancer, KIRP showed AR mRNA overexpression in paired tumor and normal. However, high AR mRNA and protein level are associated with better survival of KIRC, and AR expression was negatively correlated with tumor stage, grade and tumor status. The similar conclu,10. s using different cohorts were reported by other studies[22, 23]. The AR correlated gern enrichment analysis indicated that AR not only related to multiple cell signaling and can er-related pathways, also related to anti-tumor drug resistance pathways, EGFR t ro irs kinase inhibitor resistance pathway and Endocrine resistance, while an early study reported increased expression of AR in renal cell carcinoma resulted in acquired res. tance to the receptor tyrosine kinase inhibitor sunitinib[24]. Other studies reported that AR might play positive roles in promoting RCC initiation, progression, and i. vasion via modulation of HIF2a-VEGF signals and AR degradation enhancer ASC-J, suppression of RCC progression[25]. In addition, AR increases hematogenous metastas.’s Dut decreases lymphatic metastasis of kidney cancer through the regulation of miR-185 and VEGF isoforms[26]. Previous studies reported confusing consequence of AR function in HCC. aforetime research found that hepatic AR could up- regulate hepatitis B virus RNA and promotes HBV-induced hepatocellular carcinoma [27], and overexpression may increase oxidative stress and DNA damage lead to hepatocarcino genesis[28]. The recent study showed that AR is overexpressed in the nucleus of HCC tumors and associated with poor survival[29]. However, studies also report that AR suppressed tumor cell migration and increase cell adhesion by activating AR-ß1-integrin-

AKT signaling[30] and AR could suppress hepatocellular carcinoma cell migration and increases anoikis[31]. our analysis based on the TCGA dataset showed that AR mRNA was decreased in HCC, high AR mRNA level associated with better survival of LIHC and AR expression was reverse correlated with tumor grade. The previous and our own study both suggest that AR in HCC play dual yet opposite roles, and the further investigation may need to clarify the prognosis and therapeutic role of AR in HCC. LUSC showed decreased expression of AR mRNA and AR correlated genes enriched in Focal adhesion, ECM-receptor cGMP-PKG signaling and PI3K-Akt signaling pathways, while one study showed that around 11% NSCLC have positive AR by immunohistochemical amning. AR signaling may be different based on the KRAS state of NSCLCs[32], and another study report AR and EGFR cross-talk could regulate p38MAPK-dependent activaun of the mTOR/CD1 pathway[33]. The previous study showed that the positive rate of AR in gastric cancer tissues was around 42.4% (59/139) and AR was assoc.te, with poor progress free survival[34], and downregulation of AR suppressed the migration and invasion of gastric cancer cell lines and inhibited the epithelial-mesenchyma. +ansition pathways[35]. While our research showed low AR expression in STAD umor, but higher AR expression was associated with poor prognosis and correlated with tumor stage and grade, most AR correlated genes were positively correlated a1.1 enriched in the cell cycle, focal adhesions, and some important tumor signaling pathways. One study showed that around 20.5% women and 23.1% men thyroid cancer patients expressed AR and AR(+) tumors showed more frequent capsular invasion than AR(-) tumors[36], while TCGA dataset showed AR decreased in THCA, and correlated genes were enriched in Wnt signaling pathways. AR has been shown overexpressed by 54% (27/50) of endometrial carcinoma [37], and AR-positive was associated with good prognosis and favorable clinicopathological features[38]. A similar conclusion was observed in tumor grade and tumor status in UCEC. We observed significant

overexpression of AR in GBM. Although we didn’t observe survival differences in GBM, we found high AR expression was associated with poor prognosis in LGG and bad treatment outcomes. While other studies reported, AR to play a promoting role in gliomas[39-41].

AR played an important role in the development of prostate cancer and was considered as the driver of castration-resistant prostate cancer. Antiandrogens inhibit the androgen receptor signaling and have an important role in the treatment of prostate cancer. AR has an impact on prostate cancer development throug’s the regulation of not only transcription networks but also genomic stability and DNA repair[42]. AR splice variants have been implicated in the development and progression of metastatic prostate cancer [43]. Recent studies reported CHD1 loss drives tumorigensis by altering androgen receptor binding at lineage-specific enhancers [44], AR direct transcriptional control of the translation inhibitor 4EBP1 to negatively regulate prote n synthesis[45]. We observed high AR-score in PRAD tumor, although the mRNA without a fference, and high expression AR-V7 was associated with poor prognosis. Ablation of the AR in I. man breast cancer cell line suppressed cell proliferation, and transfection of AR led to increa. t.cogen-activated protein kinase activation suggest a positive role of AR in breast cancer[46]. However, a study found that low AR expression was correlated with high clinical stage and low nuc’aa . grade, AR expression correlated with good prognosis[47], while the other found AR expression was not associated with prognosis [48]. These may suggest a dual function of AR in tumor development as reported in prostate and liver cancer[31, 49], and further investigation may need to clarify it.

Although there are some tumors we couldn’t know the different expression of AR in tumor without normal expression data, the survival analysis based on AR expression showed that high AR mRNA expression is associated with poor survival of LGG, while high AR mRNA expression is associated with better survival of ACC, KIRC, LAML, OV. Furthermore, the survival analysis-based AR protein expression showed that high AR protein

level is associated with poor survival of LGG, while high AR protein level is associated with better survival of KIRC, SKCM, OV. In some cancers as ACC, SKCM, and STAD, the survival result of AR high group and the low group were inconsistent, such as STAD has a significant survival difference in AR mRNA data, but no difference was observed in protein expression data. And correlation analysis revealed variable mRNA and protein correlation coefficients across all cancers, while the poor correlation between AR-score and mRNA or protein was observed. Some studies also reported poor correlations between mRNA and protein expression of some genes[50-52]. Except for some of these genes are related to other omics data such as copy numbers, the post-transcriptional egulatory mechanisms such as protein translation and degradation are important reasons for this result. Besides, the AR signals were regulated by a complex biological process, including the regulation of miRNA or IncRNA[53-55]. AR-score is inferred from 20 4’x target genes reported in other literature, experimentally validated from several carcer cell lines, may not completely reflect AR activity in all cancers, and more effective methods to evaluate AR activity is necessary.

The pathway enrichmen a. alysis result showed that AR correlates genes were enriched in pathways related cellular processes, cell signaling, cancer-related, drug resistance. Many path way are common to multiple tumors, and we found that gastrointestinal tumors, n cluding COAD, ESCA, PAAD, READ, STAD, have similar enrich pathways suggest AR may play a similar role in the five tumors. Therefore, AR signaling is important in different malignancies and the potential treatment implications of using antiandrogen therapies in these malignancies in LGG.

Acknowledgments

This work was supported by grants from the National Young 1000 Talents Program of China, Jiangsu Province Education Department Grant, Jiangsu Province “Innovative and

Entrepreneurial Team” and “Innovative and Entrepreneurial Talent” Grant and Southeast University-Nanjing Medical University Cooperative Research Project.

Conflicts of interest

There are no conflicts of interest.

References:

[1] E.D. Crawford, P.F. Schellhammer, D.G. McLeod, J.W. Moul, C.S. Higano, N. Shore, L. Denis, P. Iversen, M.A. Eisenberger, F. Labrie, Androgen Receptor-Targ ter Treatments for Prostate Cancer: 35 Years’ Progress with Antiandrogens, The Journal of urolog) , (2018).

[2] C.M. Wilson, M.J. McPhaul, A and B forms of the amarogen receptor are present in human genital skin fibroblasts, Proceedings of the National Arac my of Sciences, 91 (1994) 1234-1238.

[3] R.A. Davey, M. Grossmann, Androgen re. otor structure, function and biology: from bench to bedside, The Clinical Biochemist Review _, 3/ (2016) 3.

[4] C. Chang, S. Yeh, S.O. Lee, T .-. n. Chang, Androgen receptor (AR) pathophysiological roles in androgen-related diseases in s!in, kone/musde, metabolic syndrome and neuron/immune systems: lessons learned from mice ’ acting AR in specific cells, Nuclear receptor signaling, 11 (2013).

[5] T.R. BROWN, Human an, drogen insensitivity syndrome, Journal of andrology, 16 (1995) 299-303.

[6] Z. Yu, N. Dadgar, M. Albertelli, K. Gruis, C. Jordan, D.M. Robins, A.P. Lieberman, Androgen- dependent pathology demonstrates myopathic contribution to the Kennedy disease phenotype in a mouse knock-in model, The Journal of clinical investigation, 116 (2006) 2663-2672.

[7] R. Singh, L. Singh, K. Thangaraj, Phenotypic heterogeneity of mutations in androgen receptor gene, Asian journal of andrology, 9 (2007) 147-179.

[8] R. Siegel, K. Miller, A. Jemal, Cancer statistics, 2018 CA: a cancer. J Clin 68: 7-30, in, 2017.

[9] Y. Yuan, L. Liu, H. Chen, Y. Wang, Y. Xu, H. Mao, J. Li, G.B. Mills, Y. Shu, L. Li, Comprehensive characterization of molecular differences in cancer between male and female patients, Cancer cell, 29 (2016) 711-722.

[10] M.T. Schweizer, E.Y. Yu, AR-signaling in human malignancies: prostate cancer and beyond, Cancers, 9 (2017) 7.

[11] J. Liu, T. Lichtenberg, K.A. Hoadley, L.M. Poisson, A.J. Lazar, A.D. Chemiack, A.J. Kovatich, C.C. Benz, D.A. Levine, A.V. Lee, L. Omberg, D.M. Wolf, C.D. Shriver, V. Thorsson, N. Cancer Genome Atlas Research, H. Hu, An Integrated TCGA Pan-Cancer Clinical Data Recource to Drive High-Quality Survival Outcome Analytics, Cell, 173 (2018) 400-416 e411.

[12] H. Hieronymus, J. Lamb, K.N. Ross, X.P. Peng, C. C: ment, A. Rodina, M. Nieto, J. Du, K. Stegmaier, S.M. Raj, K.N. Maloney, J. Clardy, W.C. Hahn., G. Chiosis, T.R. Golub, Gene expression signature-based chemical genomic prediction ident fir.s a novel dass of HSP90 pathway modulators, Cancer cell, 10 (2006) 321-330.

[13] I.G. Mills, Maintaining and reprogramming genomic androgen receptor activity in prostate cancer, Nature Reviews Cancer, 14 (20 4) 127.

[14] P. Li, J. Chen, H. Miyamoto, And rogen receptor signaling in bladder cancer, Cancers, 9 (2017) 20.

[15] A.K. Liossi, K.G. Aroni, K.A. ,wrkou, C. Kittas, S.P. Markaki, Immunohistochemical study of sex steroid hormones in prima ry liv er cancer, Cancer detection and prevention, 13 (1988) 195-201.

[16] S. Gu, N. Papadopoulou, E .- M. Gehring, O. Nasir, K. Dimas, S.K. Bhavsar, M. Föller, K. Alevizopoulos, F. Lang, C. Stournaras, Functional membrane androgen receptors in colon tumors trigger pro-apoptotic responses in vitro and reduce drastically tumor incidence in vivo, Molecular cancer, 8 (2009) 114.

[17] M.L. Slattery, C. Sweeney, M. Murtaugh, K.N. Ma, R.K. Wolff, J.D. Potter, B.J. Caan, W. Samowitz, Associations between ERalpha, ERbeta, and AR genotypes and colon and rectal cancer, Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology, 14 (2005) 2936-2942.

[18] M.L. Slattery, C. Sweeney, M. Murtaugh, K.N. Ma, B.J. Caan, J.D. Potter, R. Wolff, Associations between vitamin D, vitamin D receptor gene and the androgen receptor gene with colon and rectal cancer, International journal of cancer, 118 (2006) 3140-3146.

[19] M.G. Dalin, P.A. Watson, A.L. Ho, L.G. Morris, Androgen receptor signaling in salivary gland cancer, Cancers, 9 (2017) 17.

[20] A.K. Goulioumis, J. Varakis, P. Goumas, H. Papadaki, Androgen receptor in laryngeal carcinoma: could there be an androgen-refractory tumor?, ISRN oncology, 2011 (2011).

[21] H.C. Jaspers, B.M. Verbist, R. Schoffelen, V. Mattijssen, P.J. Slouw.veg, W.T. van der Graaf, C.M. van Herpen, Androgen receptor-positive salivary duct carcinoma. ~ disease entity with promising new treatment options, Journal of Clinical Oncology, 29 (201: ) e4/3-e476.

[22] C. Langner, M. Ratschek, P. Rehak, L. Schips, R. Zigeuner, Steroid hormone receptor expression in renal cell carcinoma: an immunohistochemical ar alysis of 182 tumors, The Journal of urology, 171 (2004) 611-614.

[23] G. Zhu, L. Liang, L. Li, Q. Dang, W. Song, S. . eh, D. He, C. Chang, The expression and evaluation of androgen receptor in human renal cell carcinoma, Urology, 83 (2014) 510 e519-524.

[24] R. Adelaiye-Ogala, N.P. Damayanti, A.R. Orillion, S. Arisa, S. Chintala, M.A. Titus, C. Kao, R. Pili, Therapeutic Targeting of Sunitin b-induced AR Phosphorylation in Renal Cell Carcinoma, Cancer research, 78 (2018) 2886-2 396.

[25] D. He, L. Li, G. Zhu, L D’ang, Z. Guan, L. Chang, Y. Chen, S. Yeh, C. Chang, ASC-J9 suppresses renal cell carcinoma progression by targeting an androgen receptor-dependent HIF2alpha/VEGF signaling pathway, Cancer research, 74 (2014) 4420-4430.

[26] Q. Huang, Y. Sun, X. Ma, Y. Gao, X. Li, Y. Niu, X. Zhang, C. Chang, Androgen receptor increases hematogenous metastasis yet decreases lymphatic metastasis of renal cell carcinoma, Nature communications, 8 (2017) 918.

[27] M.H. Wu, W.L. Ma, C.L. Hsu, Y.L. Chen, J.H. Ou, C.K. Ryan, Y.C. Hung, S. Yeh, C. Chang, Androgen receptor promotes hepatitis B virus-induced hepatocarcinogenesis through modulation of hepatitis B virus RNA transcription, Science translational medicine, 2 (2010) 32ra35.

[28] C.L. Ma, C.L. Hsu, M.H. Wu, C.T. Wu, C.C. Wu, J.J. Lai, Y.S. Jou, C.W. Chen, S.Y. Yeh, C.S. Chang, Androgen receptor is a new potential therapeutic target for the treatment of hepatocellular carcinoma, Gastroenterology, 135 (2008) 947-955.

[29] H. Zhang, X.X. Li, Y. Yang, Y. Zhang, H.Y. Wang, X.S. Zheng, Significance and mechanism of androgen receptor overexpression and androgen receptor/mechanist. - target of rapamycin cross - talk in hepatocellular carcinoma, Hepatology, 67 (2018) 2271-2206

decreases cell migration via modulating beta 1-integrin-, . signaling in hepatocellular carcinoma cells, Cancer Lett, 351 (2014) 64-71.

[31] W.L. Ma, C.L. Hsu, C.C. Yeh, M.H. Wu C.k Huang, L.B. Jeng, Y.C. Hung, T.Y. Lin, S.Y. Yeh, C.S. Chang, Hepatic androgen receptor oppresses hepatocellular carcinoma metastasis through modulation of cell migration and anoil is H :patology, 56 (2012) 176-185.

[32] A.R. Wang, H. Beyer, S. Brerna. S. Stiles, D. Wiese, D. Buehler, A. Saeed, A.M. Baschnagel, G. Iyer, Androgen receptor drives a. “ferential gene expression in KRAS-mediated non-small cell lung cancer, in, AACR, 2018

[33] A.G. Recchia, A.M. M. ti, M. Lanzino, M.L. Panno, E. Turano, R. Zumpano, A. Belfiore, S. Ando, M. Maggiolini, A cross-talk between the androgen receptor and the epidermal growth factor receptor leads to p38MAPK-dependent activation of mTOR and cyclinD1 expression in prostate and lung cancer cells, The international journal of biochemistry & cell biology, 41 (2009) 603-614.

[34] W. Tang, R. Liu, Y. Yan, X. Pan, M. Wang, X. Han, H. Ren, Z. Zhang, Expression of estrogen receptors and androgen receptor and their clinical significance in gastric cancer, Oncotarget, 8 (2017) 40765.

[35] W. Tang, R. Liu, Y. Yan, X. Pan, M. Wang, X. Han, H. Ren, Z. Zhang, Expression of estrogen receptors and androgen receptor and their clinical significance in gastric cancer, Oncotarget, 8 (2017) 40765-40777.

[36] F. Magri, V. Capelli, M. Rotondi, P. Leporati, L. La Manna, R. Ruggiero, A. Malovini, R. Bellazzi, L. Villani, L. Chiovato, Expression of estrogen and androgen receptors in differentiated thyroid cancer: an additional criterion to assess the patient’s risk, Endocrine-related cancer, 19 (2012) 463-471.

[37] S.L. Zadeh, L.R. Duska, A.M. Mills, Androgen receptor expression in endometrial carcinoma, International Journal of Gynecological Pathology, 37 (2018) 167-173.

[38] A.M. Kamal, J.N. Bulmer, S.B. DeCruze, H.F. Stringfellow D. Martin-Hirsch, D.K. Hapangama,

subsequent loss in endometrial cancer is associated with po’ survival, Brit J Cancer, 114 (2016) 688- 696.

[39] N. Zalcman, T. Canello, H. Ovadia, H. Charb t, B. Zelikovitch, A. Mordechai, Y. Fellig, S. Rabani, T. Shahar, A. Lossos, I. Lavon, Androgen recepto .: a potential therapeutic target for glioblastoma, Oncotarget, 9 (2018) 19980-19993.

[40] C. Liu, Y. Zhang, K. Zhang, C. RI. n, Y. Zhao, J. Zhang, Expression of estrogen receptors, androgen receptor and steroid receptor activator-3 is negatively correlated to the differentiation of astrocytictumors, Canrer pide miol, 38 (2014) 291-297.

[41] X. Yu, Y. Jiang, W. Wei, P. Cong, Y. Ding, L. Xiang, K. Wu, Androgen receptor signaling regulates growth of glioblastoma multiforme in men, Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine, 36 (2015) 967-972.

[42] I.G. Mills, Maintaining and reprogramming genomic androgen receptor activity in prostate cancer, Nature reviews. Cancer, 14 (2014) 187-198.

[43] A. Paschalis, A. Sharp, J.C. Welti, A. Neeb, G.V. Raj, J. Luo, S.R. Plymate, J.S. de Bono, Alternative splicing in prostate cancer, Nature reviews. Clinical oncology, 15 (2018) 663-675.

[44] Altered Androgen Receptor Activity Drives CHD1-Null Prostate Cancer, Cancer discovery, 9 (2019) OF7.

[45] Y. Liu, J.L Horn, K. Banda, A.Z. Goodman, Y. Lim, S. Jana, S. Arora, A.A. Germanos, L. Wen, W.R. Hardin, Y.C. Yang, I.M. Coleman, R.G. Tharakan, E.Y. Cai, T. Uo, S.P.S. Pillai, E. Corey, C. Morrissey, Y. Chen, B.S. Carver, S.R. Plymate, S. Beronja, P.S. Nelson, A.C. Hsieh, The androgen receptor regulates a druggable translational regulon in advanced prostate cancer, Science translational medicine, 11 (2019).

[46] S. Yeh, Y.C. Hu, P.H. Wang, C. Xie, Q. Xu, M.Y. Tsai, Z. Dong, A.S. Wang, T.H. Lee, C. Chang, Abnormal mammary gland development and growth retardation , emale mice and MCF7 breast cancer cells lacking androgen receptor, The Journal of experimental medicine, 198 (2003) 1899-1908. [47] T. Maeda, Y. Nakanishi, Y. Hirotani, F. Fuchinoue, K. L.com oto, K. Sakurai, S. Amano, N. Nemoto, Immunohistochemical co-expression status of cy o’.e atin 5/6, androgen receptor, and p53 as prognostic factors of adjuvant chemotherapy or ti ple negative breast cancer, Medical molecular morphology, 49 (2016) 11-21.

[48] K.H. Kensler, M.M. Regan, Y.J. Her g, 6 M/. Baker, M.E. Pyle, S.J. Schnitt, A. Hazra, R. Kammler, B. Thurlimann, M. Colleoni, G. Viale, M. Brown, R.M. Tamimi, Prognostic and predictive value of androgen receptor expression in postmenopausal women with estrogen receptor-positive breast cancer: results from the Breast International Group Trial 1-98, Breast cancer research : BCR, 21 (2019) 30.

[49] Y. Niu, S. Altuwaijri, K.P. Lai, C.T. Wu, W.A. Ricke, E.M. Messing, J. Yao, S. Yeh, C. Chang, Androgen receptor is a tumor suppressor and proliferator in prostate cancer, Proc Natl Acad Sci U S A, 105 (2008) 12182-12187.

[50] R. Akbani, P.K. Ng, H.M. Werner, M. Shahmoradgoli, F. Zhang, Z. Ju, W. Liu, J.Y. Yang, K. Yoshihara, J. Li, S. Ling, E.G. Seviour, P.T. Ram, J.D. Minna, L. Diao, P. Tong, J.V. Heymach, S.M. Hill, F. Dondelinger, N. Stadler, L.A. Byers, F. Meric-Bernstam, J.N. Weinstein, B.M. Broom, R.G. Verhaak, H.

Liang, S. Mukherjee, Y. Lu, G.B. Mills, A pan-cancer proteomic perspective on The Cancer Genome Atlas, Nature communications, 5 (2014) 3887.

[51] S. Koplev, K. Lin, A.B. Dohlman, A. Ma’ayan, Integration of pan-cancer transcriptomics with RPPA proteomics reveals mechanisms of epithelial-mesenchymal transition, PLoS Comput Biol, 14 (2018) e 1005911.

[52] S. Myhre, O.C. Lingjaerde, B.T. Hennessy, M.R. Aure, M.S. Carey, J. Alsner, T. Tramm, J. Overgaard, G.B. Mills, A.L. Borresen-Dale, T. Sorlie, Influence of DNA copy number and mRNA levels on the expression of breast cancer related proteins, Molecular oncology. 7 (2013) 704-718.

[53] L. Yang, C. Lin, C. Jin, J.C. Yang, B. Tanasa, W. Li, D. Merkuriev. ” “. Ohgi, D. Meng, J. Zhang, C.P. Evans, M.G. Rosenfeld, IncRNA-dependent mechanisms of androgen-receptor-regulated gene activation programs, Nature, 500 (2013) 598-602.

[54] A. Parolia, E. Venalainen, H. Xue, R. Mather, D. jr., [. Wu, P. Pucci, J. Rogalski, J.R. Evans, F. Feng, C.C. Collins, Y. Wang, F. Crea, The long ion oding RNA HORAS5 mediates castration-resistant prostate cancer survival by activating the and. ogen receptor transcriptional program, Molecular oncology, 13 (2019) 1121-1136.

[55] C. Coarfa, W. Fiskus, V.K. Eec.‘nun, K. Rajapakshe, C. Foley, S.A. Chew, S.S. Shah, C. Geng, J. Shou, J.S. Mohamed, B.W. O’Mal. v, N. Mitsiades, Comprehensive proteomic profiling identifies the androgen receptor ayis Und ther signaling pathways as targets of microRNAs suppressed in metastatic prostate cancer Oncogene, 35 (2016) 2345-2356.

Figure 1 Pan-cancer analysis of the AR mRNA expression in different cancer types. (A) mRNA expression difference of AR in each cancer type between Tumor and Normal. (B) mRNA expression difference of AR between tumor and matched normal from the same patients in different cancer types. (C) mRNA expression difference of AR in each cancer type between females and males. *: p-value <0.05; **: p-value < 0.01; ***: p-value < 0.001.

Figure 2 Pan-cancer analysis of the AR-score in different cancer types. (A) The AR-score difference in each cancer type between Tumor and Normal. (B) The AR-score difference across all TCGA tumors and normal sthe a sThe ss. *: p-value < 0.05; **: p-value < 0.01; ***: p-value < 0.001.

Figure 3 Pan-cancer analysis of the AR protein in diff rent cancer types. (A) AR protein expression in each cancer type Tumor. (B) AF. protein expression in each cancer type between female and male. *: p value <0.>>p value <0.01; ***: p value < 0.001. Figure 4 The significant association of AR mRNA expression with clinical characteristics, including age, race, stage, grade. 116 w tumor events, histological type and first treatment outcome in ACC(A), BRCA(8), CHOL(C), COAD(D), ESCA(E), KIRC(F), HNSC(G), LIHC(H), LGG(I), KIRP(J), STAD(K), THCA(L), UCEC(M), SARC(N), UVM(O). Figure 5 The significant association of AR protein or AR-score with clinical characteristics, including age, race, stage, grade, new tumor events, histological type and first treatment outcome in BRCA(A), HNSC(B), KIRC(C), COAD(D), LGG(E), LIHC(F), SARC(G), STAD(H), UCEC(I), THCA(J), UVM(K).

Figure 6 The significant association of AR mRNA expression with OS and PFI of patients in different cancer types. (A) ACC, (B) LAML, (C) KIRC, (D) LIHC, (E) LGG, (F) OV and (G) STAD; HR: Hazard Ratio.

Figure 7 The significant association of AR-score with OS and PFI of patients in different cancer types. (A) LGG, (B) SKCM, (C) STAD; HR: Hazard Ratio.

Figure 8 The significant association of AR protein expression with OS and PFI of patients in different cancer types. (A) KIRC, (B) LIHC, (C) LGG, (D) OV, (E) SKCM; HR: Hazard Ratio.

Figure 9 AR isoforms analysis for TCGA cancers. (A) Different AR isoforms expression of each cancer in tumor and normal, (B) uc011 mpd.1 expression w is associated with survival in GBM, (C) uc011mpf.1 expression was associated with surviva’ in PRAD; HR: Hazard Ratio. Figure 10 AR expression correlated genes and pathway enrichment analysis. (A) The number of AR Significant correlated genes in each tumor, (B) Significantly enriched cancer-related KEGG pathways in different TCGA tumors.

Journal Přes, og elin peterson

Expression of mRNA log2 (RSEM+1)

ACC

Expression of mRNA log2 (RSEM+1)

BLCA

Tumor

ACC

BRCA

Expression of mRNA log2 (RSEM+1)

BLCA

CESC

Tumor

ACC

Tumo

BRCA

CHOL

Tumor

BLCA

Norma

Tumor

FEMAL

CESC

MALE

Tumor

N=36

BRCA

…

COAD

CHOL

FEMAL

Mionnay

MALE

Tumor

CESC

DLBC

Norma?

FEMALE

fumar

Nag

COAD

hora

MALE

CHOL

…

ESCA

Normal

DLBC

FEMAL

Normal

COAD

GBM

MALE

Tumor

FEMALE

No23

ESCA

HNSC

MALE

Tumor

DLBC

Normal

hlormaj

Nos

FEMALE

GBM

ESCA

KICH

Tumor

HNSC

MALE

FEMALE

KIRC

MALE

Furor

Tumor

FEMALE

Nu3T

GBM

Normal

;

KICH

MALE

Fumar

”

HNSC

KIRP

FEMALE

KIRC

LAML

MALE

102

Tumor

KICH

FEMALE

KIRP

Figure 1

LGG

MALE

KIRC

Norma

Tumor

FEMALE

LAMI

Journal

MALE

LIHC

KIRP

Normal

LGG

FEMALE

4.208

LUAD

MALE

Tumor

LAML

Norma

FEMALE

LIHC

hora

LUSC

MALE

Tumor

Nº515

LGG

Norma

FEMALE

LUAD

MESO

LIHC

Norma

FEMALE

USC

MALET

Tumor

LUAD

Norma

Nase

MESO

PAAD

MALE

Tumor

USC

FEMALE

Normal

PCPG

MAL

Tumor

MESO

Normal

PAAD

…

FEMALE

PRAD

MALE

Tumor

N=179

Normal

Tumor

FEMALE

Nº16

PCPG

Alomma?

PAAD

READ

MALE

Nommal

Tumor

FEMALE

PRAD

Norma

SARC

MALE

Tum

PCPG

Tumor

No82

FEMALE

READ

Normal

PRAD

SKCM

MALE

₹

. …

Tumor

¥4102

FEMAL

SARC

Normal

STAD

MALE

Tumor

READ

Noma

Tumor

FEMALE

SKCM

Alarma?

TGCT

MALE

Tumor

Nagys

SARC

Normal

FEMALE

STAD

MALE

Tumor

SKCM

THCA

Normal

Tumor

FEMALE

** 142

TGCT

THYM

MALE

Tumor

STAD

Norma

Tumor

…

FEMAL

** 180

THCA

Alomma

UCEC

MALE

Tumor

TGCT

Tumor

FEMALE

¥9159

THYM

Nugo

…

THCA

UCS

MALE

Nomma

Tumor

FEMALE

UCEC

UVM

MALE

Tumor

Nº57

THYM

FEMALE

UCS

Tumor

N=23

MALE

152

JCEC

Norma

Tumor

FEMALE

UVM

Normal

MALE

Tumor

UCS

Normal

FEMALE

MALE

Tumor

UVM

Normal

FEMALE

MALE

FEMALE

MALE

Score

ACC

BLCA

Tumor

No79

… .

BRCA

Score

Normal

ACC

Nº408

.. .

Tumor

BLCA

CESC

Norma

Tumor

… .

CHOL

Tumor

Noma

Normal

Tumor

N=304

COAD

Tumor

Norma

Tumor

CHOL

DLBC

Tumor

Nomar

Weg

Nº457

Tumor

ESCA

Tumor

Norma

Ne48

Tumor

DLBC

GBM

Tumor

Normal

…

Norma

N=184

Tumor

HNSC

Tumor

Normal

Norma

N=157

Tumor

GBN

КІСН

Tumor

Journal Pre-proof

Norma

Nº520

Tumor

HNIS

KIRC

Tumor

Norma

Tumor

KIRP

Tumor

Normal

No25

…

Noma

Ne72

KIR

LAMIL

Tumor

Normal

Norma

N×290

LGG

Figure 2

Tumor

Alomma

N=173

Tumor

AML

Normal

LIHC

Norma

N=516

Tumor

LUAD

Tumor

Normal

…

Norma

Nº371

Tumor

LIHC

LUSC

Normal

Norma

Tumor

Journal Pre-p

LUAD

MESO

Tumor

Nomar

Norma

14=501

Tumor

LUSC

Noma

Norma

Tumor

No87

PAAD

Tumor

MESO

Norma

Tumor

NY-303

PCPG

Tumor

Normal

Norma

N=178

Tumor

PAAD

PRAD

Tumor

Nommal

Normal

N=179

Tumor

CPG

READ

Tumor

Noma

Tumor

PRAD

SARC

Tum

Noma

N=166

Tumor

READ

SKCM

Norma

Nº259

Tumor

SARC

STAD

Tumor

Normal

Tumor

TGCT

Tumor

Nomar

Norma

Tumor

STAD

Normal

…

THCA

Norma

N=150

Tumor

THYM

Tumor

Norma

Nº501

Tumor

THICA

UCEC

Tumor

Nomma

Norma

N=120

₹

Tumor

THYM

UCS

Tumor

Nomar

Norma

Tumor

… .

Tumor

CEC

Norma

UVM

Normal

Tumor

UCS

Norma

Normal

Tumor

UVM

Normal

Noma

Tumor

Normal

Tumor

Normal

Expression of RPPA

ACC

BLCA

Expression of RPPA

Neds

BRCA

Tumor

ACC

BLCA

CESC

Tumor

Nº874

BRCA

CHOL

Tumor

FEMAL

MALE

. …

Tumor

Netz, 18171

N-85

FEMAL

CESC

COAD

MALE

Nº30

CHOL

DLBC

Tumor

FEMAL

MALE

N=384

FEMALE

N=171

COAD

ESCA

Tumor

MALE

DLBC

Tumor

GBM

FEMAL

MALÉ

N=189

¥$125

HNSC

PAAD PCPG PRAD READ SARC SKCM STAD TGCT E THCA A 1 1 Journal Pre-proof -

FEMAL

ESCA

MALE

Tumor

GBM

KICH

FEMAL

Nim15

MALE

Tumor

Nº348

HNSC

KIRC

FEMALE

MALE

N-82

FEMAL

Tumor

KICH

KIRP

MALE

Tumor

LGG

Figure 3

FEMALE

KIRC

MALE

LIHC

FEMALE

N=25

Tumor

KIRP

MALE

FEMAL

Tumor

/0427

LGG

LUAD

MALE

No56

Tumor

V#184

LIHC

LUSC

FEMALE

MALE

V=362

MESO

FEMAL

LUAD

MALE

Tumor

FEMALE

Journal Pre-pr

LUSC

MALE

FEMAL

N=195

MESO

Tumor

MALE

FEMALE

Tumor

MALE

Nº105

FEMALE

N=12

PAAD

Tumor

MALE

N=79

FEMALE

1-410

PCPG

Tumor

MALE

Tumor

351

FEMALE

PRAD

FEMALI

READ

Tumor

~#130

MAI

=221

ALA

Tumor

SARC

MALE

351

₦4352

FEMAL

SKCM

Tumor

MALE

-357

FEMAL

Tumor

MALE

** 105

STAD

Natte N=118

TGCT

THYM

Nu144

FEMAL

Tumor

MALÉ

Nº372

FEMAL

W-121

THCA

UCEC

Tumor

MALE

235

FEMALE

Tumor

Nº90

UCS

Na10g

THYM

MALÉ

FEMAL

N=264

UCEC

Tumor

UVM

MALE

Tumor

FEMAL

UCS

MALE

Tumor

No12

FEMAL

UVM

MALE

FEMALE

Nº48

MALE

FEMALE

Nº13

MALE

Journal Pre-proof

Figure 4

ACC

BRCA

CHOL

6.049

1ª

0.00053

0.04

0.022

0.012

P<2224-18

0.0027

0.0039

7.5

0 002

0009

0.0075

AR_ARNA

5.0

AR_MRNA

AR_MIRNA

AR_MRNA

AM_MIRNDA

2.6

Micomon, p = 0.049

Kruskti Walis, p < 2.20-16-

Kruskal-Wallis, p = 8-0019

B-

Kruskal-Wallis, p — 6019

Wicoxon, p = 0.012

TUMOR FREE

WOW_TUMOR

-04-60

SAGKYN

Tumar_Status

Race

Age_Type

Stage

New_Tumor_Event

COAD

ESCA

KIRC

0.0002

0.029

0.0017

8 24-06

1.8e-09

3.10-05

0.0048

0.017

1.90-07

AR_MRNA

AR_ERNA

AR_MRNA

2.4

Willcoxon, p = 0.0082

Cobo Muctroca Alorscarcinoma

Kruskal-Wallis, p = 0.039 .

Kruskal-Wallis, p = 3:54-06

Kruska

P= 0.00022

Wilcoxon, p = 1.80-09

Histological_Type

soge_N

FONICA FREE

WITH TUMOR

Grade

Stage

Tumar_Status

HINSC

HNSC

LIHC

HIC

LIHC

0 0033

0.0022

10097

0.0014

0 0079

0 0024

0.048

0.011

0 00015

0.017

”

0.021

8.56-05

0.002

AR_MRNA

AR_ITRNA

AR_MIRNA,

. .

Kruskal-Wallis, p =00017

Widexon, p = 0.0022

…

Kruskal-Wallis, p = 0 018

Kruskal-Wallis, p = 3-59-05

Kruskal-Wallis, p = 0.00017

Grade

New_Tumor_Event

Grade

Age_T

Race

LGG

19-09

0.0076

0.024

0 041

KIRP

0015

p < 2.229-16

0 003

0.0018

7.5

AR_MRNA

AR_MRINA

AR_ARNA

AR_MRNA

2.6

Kruskal-Wallis, p < 2.2-16

Kruskal-Wallis, p = 0.0063

uskas-Walla. p = 0.054

Wildaxon, p = 0.041

Wilcoxan. p = 0.015

TUMOR FREE

WITH_ TUMOR

YOUR FREE

WWWW TUMOR

Histological_Type

Race

ent_outcome_first_course

Tumor_Status

STAD

0.00008

0.00020

le-06

0 002

0044

0.00028

12.5

0.0023

0.0015

AR_MRNA

AR_ARNA

AR MRNA

AR_RNA

AR_ITRINA

2.6

26-

Kruskal-Wallis, p = 0.0013

WWWcion. p = 0.000.

Kruskal-Wallis, p = 0.00082

Wildemon, p = 0.002

Kruskal Walls, p = 0.0029

gunge !

gange V

TUMOR JOUR

WITH TUMOR

Tummer_

Grade

Ni0

Stage

Now_Tumor_Event

Race

THCA

UCEC

2.10-05

0.0000

0 00014

0 0353

0.008

0.014

0.00015

0 00051

0.011

AR_TRNA

AR_MIRNA

ARR_mRUNGA,

AR_mRINA

AR_ARNA

2.5

Kruskal-Wallis, p = 3 50-05

Kruskal-Wallis, p = 0.832

Kruskal-Wallis. p = 1.30-06

Kruskal-Walla, p = 0.00066

Wilcomon, p = 0.0053

14 Cl

TUMOR_TRES

WITH_TUMOR

Histological_Type

Stage

Grade

Histological_Type

Tumar_Status

SARC

UVM

0 011

0 014

0.0033

9.023

0.048

10-

AR_MRNA

Kruskal-Wallis, p = 0#043

Kruskal-Wallis, p =

Elhiloid_Cel

Spindin_Cel

Age_Type

Histological_Type

Journal Pre-proof

Figure 5

BRCA

HNSC

0.0078

0.019

0.04

0.011

0.00013

1.66-05

0.047

0.031

0.012

0.0026

0.0002

0 0016

0.013

0.001

0.00048

AR_RPPA

AR_Score

. .

AR_Score

·-

Kruskal-Wallis, p = 7.46-09

Kruskal-Wallis, p = D.011

Kruskal-Wallis, p = 0.0032

Kruskal-Wallis, p = 0.01

Kruskal-Wallis, p = 8.00036

guga Jv

Black

Race

Age_Type

Stage

Race

Grade

KIRC

COAD

0 01

1.6

0.0071

1.5

0.00027

5 09:05

00003

1.70-07

te-06

1.0

0.5

AR_RPPA

AR RPPA

AR_RPPA

AR_Score

0.0

wir

20, p = 5:90-05

Wilcoxon, p = 30-04

Kruskal-Wallis, p = 8.7e-07

-. 0

Kruskal-Wallis, p = 0.027

Willcoxon, p = 0.00027

Com_Adellouue

Colon_Adandcasinoma

TUMOR PREE

WITH TUMOR

Grade

Stage

Tumor_Status

al_Type

Hatological_Type

LGG

4.10.05

0.045

0.0005

9.001

100012

0 0054

8.1e-09

47e-13

0.0074

0.00042

0.5

AR_RPPA

AR_Score

AR_Scom

AR_Score

4.5

Kruskal-Wallis, p = 4:26-13

Kruskal-Wallis, p = 0.19

Wilcomon, p = de-04

al Wallis, p = 10-00

Kruskal-Walis, p = 8.50-08

Hatslogical_Type

treatment_outcome_fint_course

TUMOR FREE

WITH TUMOR,

Tumor_Status

Histological_Type

treatment_outcome_Srt_counta

LIHC

SARC

0012

0.023

0.0017

0.0013

0.047

0 028

0 045

0.021

0.017

0.018

AR_RPPA

AR_Score

AR_RPPA

’

Kruskal-Wallis. p = 0. 829

Kruskal-Wallis, p = 0.036

Kuska

p= 0.0019

Kruskal-Wallis, p = 0.0048

Kruskal-Wals, p = 0.062

Konto

Age_Type

Grade

Race

Bạch

Race

Age_Type

STAD

UCEC

10029

0100224

0 042

0.0008

0.03

9.011

0.0013

AR_RPPA

AR_Score

AR_Scom

AR_RPPA

AR RPPA

”

Kruskal-Wallis, p = 0:047

. P= 0.0024

-4

UMOR

Kruskal-Wallis, p = 0.002 G

-4

Kruskal-Wallis, p = 0.00017

Wilodkon, p = 0.03

gange.36

Suge.M

NOR FREE

G Grade

GÅ

G2 Grade

TUMOR FREE

WITH_TUMOR

Stage

mor_Star

Tumor_Status

THCA

0.00021

30-

10064

4.40-06

0.00038

0.0079

AR_RPPA

AR_Score

Kruskal-Wallis, p = 1.60-05

Kruskal-Wallis, p = 0.0017

Epitelu_Coffinde_Col

TV COM

Spode_Cel

Histological_Type

Figure 6

ACC mRNA OS

ACC mRNA PFI

LAML mRNA OS

Group:

Low

High

Group: + Low + High

Group:

Low + High

1.00

0,75

0.75

Survival probability

0.50

9.50

N(High) = 39

N(High) = 83

0.25

N(Low) = 40

0.25

N[Low) = 40

0.25

N(Low) = 78

HR = 0.42(0.185~0.955)

HR = 0.568(0.303-1.066)

HR = 0.621(0.420-0.918)

Log-Rank = 0.0328

Log-Rank = 0.0744

Log-Rank = 0.0158

HR(numeric) = 0.775(0.643-0.934)

HR(numeric) = 0.871(0.762-0.995)

HR(numeric) = 0.925(0.870-0.984)

0.00

HR(numeric) p-value = 0.00737

0.00

HR(numeric) p-value = 0.0426

0.00

HR(numeric) p-value = 0.0136

100

150

100

150

100

Survival Months

KIRC mRNA OS

KIRC mRNA PFI

LIHC mRNA OS

Group:

Low

High

Group: + Low + High

Group:

Low + High

1.00

1.0

0,75

0.75

Survival probability

0,50

0.50

N(High) = 266

N(High) = 185

0.25

N(Low) = 267

0.25

N(Low) = 265

0.25

N(Low) = 185

HR = 0.415(0.302-0.570)

HR = 0.477(0.345-0.661)

HR = 0.681(0.481-0.964)

Log-Rank = 2.19e-08

Log-Rank = 5.2e-06

Log-Rank = 0.029

HR(numeric) = 0.803(0.750~0.858)

HR(numeric) = 0.827(0.769-0.889)

HR(numeric) = 0.933(0.877-0.993)

0.00

HR(numeric) p-value = 1.426-10

0.00

HR(numeric) p-value = 2.83e-07

0.00

HR(numeric) p-value = 0.0289

100

150

100

150

120

Survival Months

“vival I

th5

Survival Months

LGG mRNA OS

GG mRNA PFI

OV mRNA OS

Group:

Low

High

Group:

Low

High

Group:

Low

High

1.00

0.75

Survival probability

Phility

val p. OF

Survival probability

0.50

9.50

N(High) = 257

N(High) = 151

0.25

N(Low) = 257

0.25

N(Low) = 151

HR = 1.35(0.94-1.93)

HR = 1.27(0.954-1.692)

HR = 0.732(0.547-0.980)

Log-Rank = 0.104

Log-Rank = 0.101

Log-Rank = 0.0354

HR(numeric) = 1.18(1.06~1.31)

HR(numeric) = 1.11(1.03~1.20)

HR(numeric) = 0.948(0.878-1.024)

0.00

HR(numeric) p-value = 0.0029

0.00

HR(numeric) p-value = 0.00957

0.00

HR(numeric) p-value = 0.172

100

150

200

120

180

100

150

200

Survival Months

STAD mRNA OS

STAD mRNA PFI

Group:

Low

High

Group:

Low

High

1.00

0.75

Survival probability

0.50

.50

N(High) = 205

N(High) =205

0.25

N(Low) = 204

0.25

N(Low) = 206

HR = 1.69(1.22-2.32)

HR = 1.95(1.38~2.77)

Log-Rank = 0.00123

Log-Rank = 0.000134

HR(numeric) = 1.12(1.04-1.21)

HR(numeric) = 1.14(1.05-1.23)

0.00

HR(numeric) p-value = 0.0045

0.00

HR(numeric) p-value = 0.00271

120

Survival Months

Figure 7

LGG AR Score OS

LGG AR Score PFI

Group: + Low + High

1.00

0.75

Survival probability

0.50

N(High) = 256

0.25

N(Low) =258

0.25

N(Low) = 258

HR = 2.53(1.74~3.68)

HR = 2.12(1.58~2.84)

Log-Rank = 4.71e-07

Log-Rank = 2.71e

HR(numeric) = 1.08(1.05~1.11)

HR(numeric) = 1. 7(1.05~

09)

0.00

HR(numeric) p-value = 5.27e-10

0.00

HR(numeric) ~ value

de-10

100

150

200

120

160

Survival Months

SKCM AR Score OS

SKCM AR Score PFI

Group: + Low + High

Group: + Low ++ High

1.00

0.75

3.75

Survival probability

0.50

N(High) = 226

0.25

N(Low) = 224

0.25

N(Low) = 225

HR = 0.706(0.539~0.926)

HR = 0.739(0,589~0.928)

Log-Rank = 0.0114

Log-Rank = 0.00901

HR(numeric) = 0.973(0.950~0.996)

HR(numeric) = 0.981(0.963~0.999)

0.00

HR(numeric) p-value = 0.023

0.00

HR(numeric) p-value = 0.0371

100

300

400

100

200

300

400

Survival .

Survival Months

“TAL . .. Score OS

STAD AR Score PFI

Group:

Low

High

Group: + Low + High

1.00

0.75

Survival probability

0.50

N(High) = 202

N(High) = 203

0.25

N(Low) =207

0.25

N(Low) = 208

HR = 1.76(1.28~2.41)

HR = 1.45(1.03~2.03)

Log-Rank = 0.000445

Log-Rank = 0.0317

HR(numeric) = 1.04(1.01~1.08)

HR(numeric) = 1.05(1.01~1.08)

HR(numeric) p-value = 0.0196

0.00

HR(numeric) p-value = 0.0128

120

Survival Months

Journal Pre-proof

Figure 8

KIRC RPPA OS

KIRC RPPA PFI

LIHC RPPA OS

Group:

LOW

High

Group: + Low + High

Group:

Low

High

1.00

0.75

Survival probability

0.50

N(High) =222

N(High) = 221

N(High) =92

0.25

N(Low) = 223

0.25

N[Low) = 223

0.25

N(Low) = 92

HR = 0.457(0.328-0.637)

HR = 0.534(0.382-0.747)

HR = 0.49(0.323~0.744)

Log-Rank = 2.02e-06

Log-Rank = 2e-04

Log-Rank = 0.000634

HR(numeric) = 0.29(0.196~0.428)

HR(numeric) = 0.375(0.247-0.568)

HR(numeric) = 0.633(0.354~1.132)

0.00

HR(numeric) p-value = 5.19e-10

0.00

HR(numeric) p-value = 3.87e-06

0.00

HR(numeric) p-value = 0.123

100

150

100

150

120

Survival Months

LGG RPPA OS

LGG RPPA PFI

OV RPPA OS

Group:

+ Low + High

Group: + Low + High

Group:

Low ++ High

1.00

1.0.

0.75

Survival probability

.50

0.50

N(High) = 213

N[High) = 213

N(High) = 201

0.25

N(Low) =212

0.25

N[Low) = 212

0.25

N(Low) = 205

HR = 1.55(1.03~2.33)

HR = 1.4(1.01-1.94)

HR = 0.727(0.565~0.936)

Log-Rank = 0.0345

Log-Rank = 0.0418

Log-Rank = 0.0131

HR(numeric) = 1.89(1.01~3.52)

HR(numeric) = 1.4(0.848~2.316)

HR(numeric) = 0.76(0.622~0.930)

0.00

HR(numeric) p-value = 0.0452

0.00

HR(numeric) p-value = 0.188

0.00

HR(numeric) p-value = 0.00756

100

150

200

180

100

150

200

Survival Months

Survv

Months

Survival Months

SKCM RPPA OS

SKI

RPPA PFI

Group:

+ Low + High

Group:

+ Low + High

1.00

0,75

0.75

Survival probability

Survi al probability

3,50

N(High) = 171

Ng/figh) = 172

0.25

N(Low) = 170

25-

N(Low) = 170-

HR = 0.626(0.452-0.865)

HR = 0.738(0.563 — 0.968)

Log-Rank = 0.00427

Log-Rank = 0.0278

HR(numeric) = 0.597(0.315-1.132)

HR(numeric) = 0.737(0.433-1-255)

0.00

HR(numeric) p-value = 0.114

0.00

HR(numeric) p-value = 0.261

100

200

30/

400

100

200

300

400

Survival Months

Figure 9

Mean expression level (Tumor)

ACC

BLCA

BRCA

CESC.

CHOL

-COAD -

- DLBC

ESCA

GBM

-HNSC-

KICH

KIRC

KIRP

LAML

LGG

LIHC

- LUAD

- LUSC.

-MESO-

- PAAD

-PCPG-

PRAD

-READ-

SARC

SKCM-

- STAD

TGCT

THCA.

THYM-

- UCEC

UCS

UVM

Mean expression level (Normal)

uc004dwu.1

uc004dwv.1

uc011m - 1.1

uc011mpe.1

uc011mpf.1

GBM uc011mpd.1 OS

GBM uc011mpd.1 PFI

Group:

Low + High

Group: : ++ Low + High

1.00

Survival probability

0.75

Survival probability

0.75

0.50

N(High) = 10

0.25

N(Low) = 146

HR = 2.21(1.12-4,38)

0.25

N(Low) # 146

Log-Rank + 0.0203-1h_

HR = 2.34(1.18~4.66)

Log-Rank = 0.0121

HR(numeric) = 2.08(1.45~2 98)

HR(numeric) = 1.74(1.23~2.44)

0.00

HR(numeric) p-value = f 1/6 5

0.00

HR(numeric) p-value = 0.00154

Su vival Months

Survival Months

Group:

Low

High

PRAD uc011mpf.1 (AR-V7) OS

1.00

Survival probability

0.75

0.50

N(High) = 248

0.25

N(Low) = 249

HR = 5.07( 1.02~25.21)

Log-Rank = 0.0303

HR(numeric) = 1.9(1.21~2.97)

0.00

HR(numeric) p-value = 0.00495

120

160

Survival Months

PRAD uc011mpf.1(AR-V7) PFI

Group: + Low + High

1.00

Survival probability

0.75

0.50

N(High) = 248

0.25

N(Low) = 249

HR = 1.99(1.30~3.05)

Log-Rank = 0.0012

HR(numeric) = 1.32(1.13~1.54)

0.00

HR(numeric) p-value = 0.000446

120

160

Survival Months

Figure 10

2000

TYPE

Negative

Positive

2036

1853

1706

1682

AR significantly correlated gene number

1500

1399

1000

1021

960

776

819

639

660

603

543

550

544

500

487

431

447

454

384

318

294

244

179

193

186

223

147

137

104

131

ACC BLCA

BRCA

CESC

CHOL

COAD

DLBC

ESCA

GBM

HNSC

KICH

KIRC

KIRP

LAML

LGG

LIHC

LUJU

MESO

PAAD

PCPG

PRAD

READ

SARC

SKCM

STAD

TGCT

THCA

THYM

UCEC

UCS

UVM

Drug Resistance

Endocrin resis ance

EGFR tyrosine kinase inhibi. - resistance

Cancer Related

Renal cell . cinoma

Prostate cancer

Non-smallIl lung cancer

Melanoma

astric cancer

Er ion etrial cancer

Corectal cancer

Chronic myeloid leukemia

Breast cancer

KEGG Pathways

Prowdoglycans in cancer

qvalue

PD-L1 expression and PD -? chec. Point pathway in cancer

Cell Signaling

Wnt signaling pathway

0.04

TOF-beta signaling pathway

- 0.03

Ras signaling pathway

- 0.02

Rap1 signaling pathway-

0.01

PI3K-Akt signaling pathway

Phospholipase D signaling pathway

mTOR signaling pathway

MAPK signaling pathway

Count

Hippo signaling pathway - multiple species

HIF-1 signaling pathway

Hedgehog signaling pathway

FoxO signaling pathway

ErbB signaling pathway

cGMP-PKG signaling pathway

cAMP signaling pathway

Calcium signaling pathway

ECM-receptor interaction

Cell adhesion molecules (CAMs)

Cellular Processes

Gap junction

Focal adhesion

Adherens junction

Cell cycle

CESC

COAD

ESCA

HNSC

KIRC

KIRP

LGG

LUSC

PAAD

PRAD

READ

SARC

STAD

THCA

Table 1 Patient characteristics of TCGA cohort.

									Tum
								Ne
									or
								w
									Stat
								Tu
									us
				Gend				mo
			Age		Race		Grade		(Tu
	Can	Nu		er		Stage		r
			(Me		(White\|Bl		(G1\|G2\|		mor
Disease Name	cer	mb		(Mal				Ev
Disease Name			an+S		ack\|Othe		G3\|G4\|		free \|
	Type	er		e \|Fe		IV\|NAs)		ent
			D)		r\|NAs)		NAs)		Wit
				male)				(Y	h
								es\|
									tum
								No	or\|N
					are-proof			)
									As)


			46.7
Adrenocortical									39 \|
Adrenocortical			±		66 \|1\|1\|	0 \| 9 \| 37 \|	0\| 0\| 0	37 \|
	ACC	79							38 \|
carcinoma			15 7	48	11	16 \| 15 \| 2	\| 0\| 79	42
									2
			7
			68.1					14
						0 \| 2 \| 131	21 \| 0 \|		222 \|
Bladder urothelial	BIC	41	±	304	327 \| 23 \|			2 \|
Bladder urothelial						\|141\|	388 \| 0\|		165
carcinoma	A	2	10.5	108	44 \| 18			27
						136 \| 2	3		25
			8					0
			58.4					10
						0 \| 183 \|	0\|0\|0		935
Breast invasive	BRC	10	5±	12 \|	757 \| 183			2 \|
						621 \| 249	\| 0\|		126 \|
carcinoma	A	97	13.2	1085	62 \| 95			99
						\| 20 \| 24	1097		36
			1					5
Cervical and			48.2			0 \| 163	18 \| 136	47	199
Cervical and	CES	30		0 \|	210 \| 30 \|
			4±			70 \| 45 \|	\| 119 \| 1	25	80

	Journal Pre-proof
endocervical cancers	C	6	13.8	306	30 \| 36	21 \| 7	32	9	27
			63.0
							1 \| 15 \|		15
	CHO		3 ±	16 \|	31 \|2\|3\|	0\| 19 \| 9\|		19
Cholangiocarcinoma		36					18 \| 2 \|		19
	L		12.8	20	0	1\| 7\| 0		17
							0		2
			5
			66.9
						0 \| 76 \|		15	283
Colon	COA	45	5±	241	213 \| 59 \|		0\|0\| 0
						177 \| 128		44	145
adenocarcinoma	D	7	13.0	216	12 \| 173		\| 0\| 457
			6			\| 6: \| 11		2	29

Lymphoid			56.2		Preroga
Neoplasm Diffuse	DLB		7 ±	22 \|	29'1\|18	0 \| 8 \| 17 \|	0\| 0\| 0	0 \|	37 \|
		48
Large B-cell	C		13.9	26	\|0	5 \| 12 \| 6	\| 0\| 48	48	9\| 2
Lymphoma			5
			62.4
						0 \| 18 \| 79	19 \| 77 \|	71 \|	100
Esophageal	ESC	18	5	158	114 \| 5 \|
						\| 56\| 9\|	49 \| 0 \|	11	78
carcinoma	A	4	:1.8	27	46 \| 20
		Journa				23	40	4	7
			9
			59.4					16
									18 \|
Glioblastoma	GB	31	2±	198	260 \| 33	0\|0\|0\|0	0\| 0\| 0	7 \|
									266 \|
multiforme	M	7	13.9	119	7 \| 17	\| 0\| 317	\| 0\| 317	15
									33
			5					0
			60.8
Head and Neck						0 \| 27 \| 72	62 \| 305	69	330
	HNS	52	8 ±	385	445 \| 48 \|
squamous cell						\| 81 \| 266	125 \| 7	45	162
	C	1	11.8	136	13 \| 15
carcinoma						75	\| 22	2	29
			6

Kidney			51.5						53
Kidney	KIC			39 \|	58 \|4\|2\|	0\|21 \|25	0\|0\|0	0\|
		66	2±						12
Chromophobe	H			27	2	\|14\|6\|0	\|0\|66	66
			14.3						1
			60.5
						0 \| 269	14 \| 230	0 \|	353
Kidney renal clear	KIR	53	9±	345	465 \| 56 \|
						57 \| 124 \|	\| 206 \|	53	159
cell carcinoma	C	6	12.1	191	8 \| 7
						83 \| 3	78 \| 8	6	24
			6

			61.4
Kidney renal						0 \| 173 \|		0 \|	229
	KIR	29	9±	214 \|	207 \| 61 \|		0\| 0\| 0
papillary cell						21 \| 52 \|		29	47
	P	1	12.0	77	8 \| 1.		\| 0\| 291
carcinoma						15 \| 30		1	15
			7
			55.2
								0 \|
Acute Myeloid	LA	17	8 ±	9	156 \| 13 \|	0\|0\|0\|0	0\|0\|0		0\| 0
								17
Leukemia	ML	3	16.1	80	2 \| 2	\|0\| 173	\| 0\| 173		\| 173
								3
			4
			12.9
							0 \| 249 \|	0 \|	212
Brain Lower Grade		51	4±	285	475 \| 21	0\|0\|0\|0
	LCJ						265 \| 0 \|	51	269
Glioma		5	13.3	230	9 \| 10	\| 0\| 515
							1	5	34
			6
			59.5					17
						0 \| 173 \|	55 \| 180		201
Liver hepatocellular	LIH	37	5 ±	252	187 \| 17 \|			5 \|
						86 \| 86 \| 5	\| 122 \|		152 \|
carcinoma	C	4	13.5	122	160 \| 10			19
						\| 24	12 \| 5		21
			2					9
Lung	LUA	52	65.3	241	392 \| 53	0 \| 279 \|	0\|0\|0	13	292 \|
adenocarcinoma	D	0	2±	279	9 \| 66	123 \| 84 \|	\| 0\| 520	1 \|	171

			Journal Pre-proof

			10.0			26 \| 8		38	57
			1					9
			67.2			0 \| 245		81	319
Lung squamous cell	LUS	50		373	351 \| 31 \|		0\|0\|0
			8 ±			163 \| 85		42	114
carcinoma	C	4		131	9 \| 113		\| 0\| 504
			8.61			7\| 4		3	71
			62.9			0 \| 10 \| 16			16
	MES			71 \|	85 \|1\|1\|		0\|0\| 0	44 \|
Mesothelioma		87	9±			\| 45\| 16\|			50 \|
	O			16	0		\|0\| 87	43
			9.76			L'			21
								27
			59.3			0 \| 15 \| 28	5 \| 58 \|		99
Ovarian serous		49		0 \|	408 \| 32)			1 \|
	OV		8 ±			\| 377 \| 65\|	415 \| 1		327
cystadenocarcinoma		0		490	23\|27			21
			11.6			5	11		64
								9
			64.8
						0 \| 21 \|	32 \| 97 \|	84	58 \|
Pancreatic	PAA	18	6±	102	162 \| 7 \|
						152 \| 4 \| 5	51 \| 2 \|	10	105
adenocarcinoma	D	5	11.0	83	11 \| 5
						\|3	3	1	22
			5
			47.3
								18	166
Pheochromocytoma	I	17	3±	78 \|	148 \| 20 \|	0\|0\|0\|0	0\|0\|0
								16	13 \|
and Paraganglioma	G	9	15.1	101	7 \| 4	\| 0\| 179	\| 0\| 179
								1	0
			2
			61.0					91 \|	345
Prostate	PRA	49		498	147 \|7\|2	0\|0\|0\|0	0\|0\|0
			1±					40	89 \|
adenocarcinoma	D	8		0	\| 342	\| 0\| 498	\| 0\| 498
			6.82					7	64
Rectum	REA	16	64.3	91	82 \|6\| 1\|	0 \| 31 \| 51	0\|0\|0	8 \|	101
adenocarcinoma	D	8	3±	77	79	\| 52 \| 25\|	\| 0\| 168	16	55

			Journal Pre-proof
			11.9			9		0	12
			60.8					12
									129 \|
	SAR	26	7 ±	119 \|	228 \| 18 \|	0\|0\|0\|0	0\| 0\| 0	5 \|
Sarcoma									123 \|
	C	1	14.6	142	6\| 9	\|0\|261	\|0\|261	13
									9
			5					6
			58.2					24
						7 \| 77 \|			214
Skin Cutaneous	SKC	47	2±	290	447 \| 1 \|		0\|0\|0	3 \|
						11) \| 171			247 \|
Melanoma	M	0	15.7	180	12 \| 10		\| 0\| 470	22
						\| 2 \| 52			9
			3					7
			65.7					10
						0 \| 59 \|	12 \| 156		249 \|
Stomach	STA	43	±	281	27°\|2\|			1 \|
						129 \| 180	\| 259 \| 0		139 \|
adenocarcinoma	D	6	10.7	155	$9\| 62			33
						\| 42 \| 26	\|9		48
			1					5
			31.9			0 \| 101		33
Testicular Germ Cell	TGC	13		134	119 \| 6 \| 4		0\| 0\| 0		126 \|
			9:			12 \| 14 \| 0		10
Tumors	T	4		0	\| 5		\| 0\| 134		7\| 1
			2.31			\|7		1
			47.2
						0 \| 284 \|		43 \|	449 \|
	INC	50	7 ±	135	331 \| 27 \|		0\| 0\| 0
Thyroid carcinoma						51 \| 111 \|		46	42 \|
	A	3	15.7	368	53 \| 92		\| 0\| 503
						55 \| 2		0	12
			9
			58.1
	THY	12	1 ±	64	102 \| 6 \|	0 \| 38 \| 61	0\| 0\| 0	20 \|	113 \|
Thymoma	M	3	13.0	59	13 \| 2	\| 15 \| 7\| 2	\|0\| 123	10	9\| 1
								3
			5
Uterine Corpus	UCE	54	63.9	0 \|	373 \| 109	0 \| 341	99 \| 121	85	426

Table 2 The univariate and multivariate analysis of AR mRNA in different cancer types
Endometrial	C	7	4±	547	33 \|32	52 \|124 \|	\| 327 \| 0	46	101 \|
Carcinoma			11.1			30 \| 0	\|0	2	20
			5
			69.7						17
Uterine					44 \|9\|3\|	0\|22\|5\|	0\|0\|0	31\|
Uterine	UCS	57	4±	0 \| 57					37
Carcinosarcoma					1	20\|10\|0	\|0\|57	26
			9.3						3
			61.6
									55 \|
	UV		5±	45	55 \|0\|0\|	010,39\|	0\|0\|0	22\|
Uveal Melanoma		80							25 \|
Uveal Melanoma	M		13.9	35	25	74\|0	\|0\|80	58
									0
			5		soundmal Pre-pro

Cancer	mRNA Of Univariate	mRNA OS Multivariate Cox	mRNA PFI Univariate Cox	mRNA PFI Multivariate Cox
ACC	0.775(0.643-0.934)	0.802(0.643-0.999) 0.049	0.871(0.762-0.995) 0.0426	0.913(0.765-1.089) 0.3119

	0.0074
BLCA	0.982(0.924-1.044)	0.959(0.894-1.028)	1.02(0.96-1.084)	0.952(0.889-1.02)
BLCA	0.5601	0.237	0.5149	0.1643
BRCA	1.022(0.958-1.089)	0.947(0.872-1.028)	0.975(0.918-1.036)	0.912(0.84-0.991)
BRCA	0.5131	0.1962	0.4103	0.0296
CESC	1.043(0.932-1.167)	1.023(0.881-1.188)	1.053(0.943-1.175)	0.914(0.776-1.076)
CESC	0.465	0.7659	0.3579	0.2802

Journal Pre-proof
CHOL	0.978(0.801-1.195)	1.674(1.04-2.694)	0.867(0.708-1.062)	1.086(0.667-1.769)
CHOL	0.8285	0.0338	0.1692	0.7405
COAD	1.079(0.949-1.227)	1.055(0.842-1.322)	1.106(0.986-1.24)	1.119(0.916-1.368)
COAD	0.2474	0.6441	0.0846	0.2709
DLBC	0.852(0.429-1.691)	0.64(0.152-2.696)	0.999(0.583-1.711)	NA
DLBC	0.6467	0.5431	0.9959
ESCA	0.913(0.787-1.058)	0.97(0.759-1.239)	1.002(0.876-1.146)	0.647(0.493-0.85)
ESCA	0.2245	0.8045	0.9736	0.0018
GBM	0.876(0.777-0.988)	0.887(0.774-1.017)	0.918(0.81 -1.034)	0.873(0.763-0.998)
GBM	0.0304	0.0852	C 1595	0.0472
HNSC	0.952(0.886-1.024)	0.95(0.864-1.045)	0 9- 7(0.8 77-1.023)	1.01(0.915-1.115)
HNSC	0.1861	0.2911	0.1698	0.8482
KICH	1.177(0.888-1.561)	1.094(0.875-1.367)	061(0.832-1.353)	1.004(0.808-1.249)
KICH	0.2557	0.4308	0.6333	0.9695
KIRC	0.803(0.75-0.858)	0.822(0.735-).89.)	0.827(0.769-0.889)	0.906(0.822-0.999)
KIRC	0	0	0	0.0481
KIRP	0.933(0.812-1.071)	1.020.863-1.189)	0.905(0.802-1.021)	1.028(0.911-1.16)
KIRP	0.3254	J.8749	0.1047	0.656
LAML	0.925(0.87-0.984)	0.>09(0.853-0.968)	NA	NA
LAML	0.0136	0.003		NA
LGG	1.175(1.057-1.307)	1.135(1.001-1.286)	1.112(1.026-1.205)	1.061(0.967-1.164)
LGG	0.0029	0.0477	0.0096	0.2095
LIHC	0.933(0.877-0.993)	0.959(0.892-1.032)	0.951(0.9-1.005)	0.958(0.895-1.025)
LIHC	0.0289	0.2631	0.0749	0.2138
LUAD	0.969(0.895-1.05) 0.4463	0.935(0.848-1.03) 0.1722	1.008(0.933-1.088)	1.009(0.923-1.103) 0.8443
LUAD	0.969(0.895-1.05) 0.4463	0.935(0.848-1.03) 0.1722	0.8483	1.009(0.923-1.103) 0.8443
LUSC	1.036(0.964-1.114)	1.044(0.951-1.145)	1.038(0.951-1.132)	1.156(1.041-1.284)
LUSC	0.332	0.364	0.4042	0.0068
MESO	0.932(0.808-1.074)	0.842(0.682-1.039) 0.1091	0.944(0.808-1.103) 0.468	0.932(0.734-1.183) 0.5606
MESO	0.3302	0.842(0.682-1.039) 0.1091	0.944(0.808-1.103) 0.468	0.932(0.734-1.183) 0.5606

Journal Pre-proof
OV	0.948(0.878-1.024) 0.1715	0.984(0.9-1.077)	0.956(0.888-1.03)	0.938(0.867-1.014)
OV	0.948(0.878-1.024) 0.1715	0.7295	0.2376	0.1068
PAAD	1.076(0.951-1.217)	0.955(0.823-1.107)	1.093(0.967-1.236)	0.906(0.787-1.044) 0.1724
PAAD	0.2469	0.5392	0.1555	0.906(0.787-1.044) 0.1724
PCPG	0.772(0.42-1.421)	1.041(0.589-1.839)	0.929(0.694-1.244)	0.942(0.729-1.216)
PCPG	0.4065	0.8909	0.6199	0.6456
PRAD	1.208(0.784-1.862)	1.182(0.697-2.004)	1.008(0.88-1.156)	1.013(0.855-1.2)
PRAD	0.3922	0.5359	0.9063	0.8817
READ	0.777(0.555-1.088)	0.767(0.371-1.585)	0.907(0.7 173)	1.232(0.708-2.147)
READ	0.1419	0.4738	C 4563	0.4603
SARC	0.962(0.895-1.034)	0.929(0.835-1.033)	0 9. 4(0.9 26-1.045)	0.861(0.782-0.949)
SARC	0.2896	0.1741	0.5932	0.0025
SKCM	0.985(0.907-1.07)	0.9(0.819-0.988)	957(0.891-1.028)	0.899(0.833-0.971)
SKCM	0.7175	0.0263	0.2299	0.0069
STAD	1.12(1.036-1.212)	1.123(0.0 ,6- . 26,7	1.136(1.045-1.235)	0.974(0.88-1.079)
STAD	0.0045	0.058€	0.0027	0.62
TGCT	0.813(0.42-1.573) 0.5383	NA	0.912(0.736-1.13)	0.861(0.685-1.081)
TGCT	0.813(0.42-1.573) 0.5383	NA	0.4007	0.1977
THCA	1.206(0.905-1.607)	3.>14(1.271-8.639)	1(0.85-1.177)	0.815(0.668-0.995)
THCA	0.202	0.0143	0.9987	0.044
THYM	1.092(0.723-1.65)	0.821(0.518-1.301)	0.947(0.701-1.28)	0.759(0.477-1.206)
THYM	0.6766	0.4018	0.7239	0.2432
UCEC	0.969(0.881-1.066)	1.019(0.914-1.136)	0.952(0.879-1.032)	0.945(0.861-1.038)
UCEC	0.5163	0.7364	0.2334	0.2374
UCS	0.952(0.803-1.129)	0.915(0.759-1.103)	0.924(0.785-1.087)	0.848(0.687-1.045)
UCS	0.5721	0.3499	0.3408	0.1224
UVM	1.157(0.782-1.713)	0.709(0.366-1.375)	1.345(0.914-1.98)	1.024(0.57-1.84)
UVM	0.465	0.3091	0.1323	0.9375

The upper part represents the hazard ratio (95% confidence interval) and under part represents p-value, p-value less than 0.05 were bold.

Journal Pre-proof

Highlights:

· The first comprehensive pan-cancer analysis of AR in different tumor types.

. Most of tumor types have decreased AR but only GBM has overexpressed AR mRNA.

. AR signaling is associated with some cancer development and patients’ survival.

· AR is promising for antiandrogen therapies in AR related malignancies.

Journal Pre-proof

Expression of mRNA log2 (RSEM+1)

ACC

Expression of mRNA log2 (RSEM+1)

BLCA

.- 8

Tumor

N=79

BRCA

Normal

.-…

ACC

W=408

Tumor

BLCA

. … … …

CESC

Normal

N=19

W=1093

… ..

Expression of mRNA log2 (RSEM+1)

Tumor

BRCA

Normal

N=112

CHOL

Tumor

Normal

W=304

ACC

Tumor

··

CESC

Normal

COAD

N=19

N=3

Tumor

N=19

N=36

BLCA

Normal

Tumor

CHOL

DLBC

W=112

Normal

N=9

Tumor

FEMALE

N=48

N=112

BRCA

Normal

N=457

Tumor

COAD

ESCA

N=31

N=3

N=41

MALE

Tumor

Normal

FEMALE

N=116

CESC

Normal

N=3

N=48

Tumor

DLBC

GBM

MALE

N=311

N=9

Normal

… … ..

Tumor

N=9

W=184

FEMALE

N=1192

CHOL

Normal

Tumor

HNSC

MALE

N=13

N=11

Tumor

N=41

COAD

ESCA

Normal

N=307

Normal

N=41

W=157

FEMALE

KICH

Tumor

MALÉ

Tumor

GBM

Normal

N=5

FEMALE

N=23

… …

DLBC

Normal

W=520

N=22

Tumor

KIRC

N=44

MALE

N=11

Tumor

Normal

W=236

Normal

N=11

FEMALE

ESCA

HNSC

N=66

Tumor

KICH

KIRP

N=260

N=25

MALE

Tumor

Normal

N=26

N=533

FEMALE

N=22

-8-8-89

GBM

Normal

Tumor

N=43

N=72

LAML

MALÉ

Tumor

Normal

N=43

FEMALE

N=31

HNSC

KIRC

Normal

W=290

N=164

Tumor

KIRP

Normal

N=32

LGG

MALE

N=25

N=54

Normal

N=25

W=173

FEMALE MALÉ

KICH

Tumor

LIHC

N=102

8-0-0-0

Tumor

N=72

LAML

Normal

N=150

Normal

N=72

N=516

FEMALE

Tumor

. … …

KIRC

LUAD

MALÉ

V=414

…-…

Tumor

N=32

Normal

. … …

Normal

N=32

LGG

N=39

W=371

FEMALE

KIRP

Tumor

LIHC

LUSC

MALE

N=52

Tumor

Normal

N=50

W=515

FEMALE

W=208

V=397

LAML

Normal

Tumor

LUAD

MESO

Normal

N=59

MALE

Tumor

N=501

FEMALE

N=86

N=236

LGG

Normal

Tumor

N=51

LUSC

MALE

N=50

Tumor

Normal

N=80

N=50

N=87

FEMALE

N=93

LIHC

Normal

Tumor

PAAD

MALE

N=58

Tumor

Normal

N=58

W=303

FEMALE

…

V=285

LUAD

MESO

N=230

Normal

Tumor

PCPG

MALE

. …

Tumor

N=51

Normal

N=51

…

N=143

W=178

FEMALE

Figure 1

V=278

LUSC

Normal

Tumor

N=4

PAAD

PRAD

MALE

Tumor

Normal

FEMALE

N=311

MESO

Normal

W=179

Tumor

READ

V=263

…

MALE

Tumor

PCPG

Normal

N=3

—

FEMALE

W=144

Normal

W=497

Tumor

SARC

N=408

N=52

MALE

N=4

Normal

—.

Tumor

PRAD

FEMALE

N=16

N=4

W=166

N=71

PAAD

Normal

Tumor

SKCM

MALE

N=3

PCPG

READ

Normal

N=10

Tumor

N=303

Normal

N=3

W=259

FEMALE

Tumor

SARC

STAD

MALÉ

N=52

Tumor

Normal

N=2

W=469

FEMALE

N=82

N=52

N=100

. …

PRAD

Normal

Tumor

MALE

Tumor

N=9

SKCM

TGCT

Normal

FEMALE

N=102

N=9

…

READ

W=415

… .

Normal

Tumor

STAD

THCA

MALÉ

N=80

Tumor

N=2

Normal

N=35

N=2

… .

W=150

FEMALE

SARC

Normal

Tumor

THYM

MALE

V=549

… .

Tumor

TGCT

Normal

.-…

…

FEMALE

N=82

SKCM

Normal

W=501

N=93

Tumor

N=59

MALE

N=32

Tumor

FEMALE MALE

STAD

THCA

Normal

UCEC

W=142

N=32

=120

… ..

Normal

Tumor

N=2

THYM

UCS

N=119

Tumor

Normal

FEMALE

W=180

N=290

TGCT

Normal

W=545

Tumor

N=59

N=35

MALE

Tumor

THCA

UCEC

UVM

Normal

N=159

Normal

N=59

FEMALE

N=291

… .

Tumor

N=57

MALE

Tumor

N=2

Normal

N=2

Tumor

N=80

MALE

V=134

THYM

UCS

FEMALE

Tumor

N=23

UCEC

UVM

Normal

FEMALE

N=408

Normal

MALE

V=152

Tumor

FEMALE

N=59

N=63

. … …

UCS

Normal

MALE

Tumor

FEMALE MALÉ

N=568

UVM

Normal

FEMALE MALE

N=57

₿

FEMALE

N=35

N=45

MALÉ

AR Score

ACC

BLCA

N=79

BRCA

Tumor

… .

AR Score

ACC

Normal

…

N=408

CESC

Tumor

N=19

BLCA

Normal

… .

W=1093

CHOL

Tumor

N=112

BRCA

Normal

Tumor

N=304

Normal

Tumor

N=3

CESC

COAD

Normal

Tumor

N=36

DLBC

Normal

Tumor

CHOL

Normal

N=9

Tumor

N=457

COAD

ESCA

Normal

Tumor

N=41

Normal

Tumor

N=48

Normal

Tumor

GBM

DLBC

Normal

Tumor

N=184

HNSC

Normal

Tumor

N=11

ESCA

Normal

N=157

Tumor

KICH

Normal

Tumor

N=5

GBM

Normal

Tumor

N=520

Tumor

KIRC

Normal

…

N=44

HNSC

Normal

N=66

Tumor

… …

Tumor

KIRP

Normal

N=25

KICH

Normal

Tumor

N=533

Tumor

LAML

Normal

N=72

KIRC

Normal

Tumor

N=290

Tumor

LGG

Normal

N=32

KIRP

Normal

Tumor

W=173

Tumor

LIHC

Normal

LAML

Normal

Tumor

N=516

Tumor

… .

Normal

LUAD

LGG

Normal

…

Tumor

N=371

LUSC

Tumor

Normal

N=50

LIHC

Normal

…

N=515

”

Tumor

MESO

Tumor

Normal

N=59

LUAD

Normal

Tumor

N=501

Tumor

Normal

N=51

LUSC

Normal

Tumor

N=87

PAAD

Tumor

Figure 2

Normal

MESO

Normal

Tumor

.. .

N=303

PCPG

.. .

Tumor

Normal

. …

Tumor

N=178

Tumor

. .

Normal

N=4

PRAD

PAAD

Normal

Tumor

N=179

Normal

N=3

PCPG

. …

READ

Tumor

… .

Normal

Tumor

N=497

SARC

Tumor

Normal

N=52

PRAD

Normal

…

Tumor

N=166

… ..

SKCM

Tumor

Normal

READ

Normal

N=10

Tumor

N=259

STAD

Tumor

Normal

SARC

Normal

N=2

Tumor

N=465

TGCT

Tumor

Normal

SKCM

Normal

Tumor

N=415

Tumor

THCA

Normal

N=35

…

STAD

Normal

Tumor

W=150

THYM

Tumor

Normal

—

TGCT

Normal

…

Tumor

N=501

UCEC

Tumor

N=59

Normal

THCA

Normal

Tumor

N=120

…

Tumor

UCS

Normal

N=2

THYM

Normal

Tumor

N=545

Tumor

UVM

… .

Normal

N=35

…

Normal

UCEC

Tumor

N=57

Tumor

Normal

UCS

Normal

Tumor

”

N=80

Tumor

Normal

—

UVM

Normal

Tumor

Normal

Tumor

Normal

Tumor Normal

Expression of RPPA

ACC

BLCA

Expression of RPPA

N=46

BRCA

Tumor

ACC

N=344

CESC

Tumor

BLCA

N=874

CHOL

N=28

BRCA

Tumor

FEMALE

MALÉ

N=18

. …

N=171

COAD

FEMALE MALÉ

N=85

CESC

Tumor

V=259

N=30

DLBC

FEMALÉ

N=865

CHOL

Tumor

MALÉ

N=8

N=354

ESCA

N=171

COAD

Tumor

NUL

FEMALÊ

MALÉ

N=33

FEMALE

N=17

Tumor

DLBC

… .

GBM

MALÉ

N=13

N=126

HNSC

Tumor

FEMALE

N=169

ESCA

MALÉ

V=183

N=205

KICH

N=15

…

Tumor

GBM

… . …

FEMALE

MALÉ

N=18

N=346

KIRC

N=18

HNSC

Tumor

FEMALE MALÉ

…

N=108

N=63

KIRP

N=82

Tumor

FEMALE

W=122

… …

KICH

MALÉ

N=445

LGG

Tumor

FEMALE

N=98

V=248

MALE

KIRC

N=207

LIHC

N=25

Tumor

FEMALE MALÉ

N=38

KIRP

N=427

LUAD

W=148

Tumor

FEMALE

W=297

LGG

MALE

N=184

LUSC

N=56

Tumor

FEMALÉ

N=151

LIHC

MALÉ

N=362

LUAD

MESO

Tumor

FEMALE

W=189

N=237

MALÉ

N=325

N=70

Tumor

FEMALÊ MALE

LUSC

N=114

…

w … .

N=61

Figure 3

MESO

PAAD

N=195

Tumor

FEMALE

V=167

MALE

N=411

PCPG

N=79

Tumor

… .

FEMALE

N=246

MALE

N=105

PRAD

N=12

PAAD

Tumor

FEMALE

… .

MALÉ

N=49

N=79

. … …

PCPG

READ

N=410

Tumor

FEMALE

MALE

N=351

SARC

N=52

Tumor

FEMALE

PRAD

MALE

N=53

… .

N=130

SKCM

N=40

Tumor

FEMALE

…-.

READ

N=39

. …

MALÉ

N=221

STAD

Tumor

FEMALE

SARC

T … .

MALÉ

V=351

N=352

N=61

…

SKCM

TGCT

Tumor

FEMALE

MALE

N=68

N=357

W=116

Tumor

FEMALE

STAD

THCA

MALÉ

N=105

… .

THYM

…

… .

N=118

… …

N=144

Tumor

FEMALE

TGCT

N=208

MALÉ

N=372

THCA

UCEC

N=121

Tumor

FEMALE

N=236

MALÉ

…

N=90

UCS

Tumor

FEMALE

MALÉ

N=104

… . ..

… …

THYM

N=404

N=264

UCEC

UVM

Tumor

… .

FEMALE

MALE

V=108

N=48

Tumor

FEMALE

N=43

MALÉ

N=47

UCS

N=12

Tumor

FEMALE

N=404

UVM

MALÉ

…

FEMALE

N=48

MALÉ

.-.

FEMALE

MALE

N=12

ACC

0.049

0.00053

0.04

0.022

0.012

B-

0.033

0.0075

AR_mRNA

AR_MRNA

Wilcoxon, p = 0.012

Yes

New_Tumor_Event

COAD

0.0082

Colon Mucinous_Adenocarcinoma

Histological_Type

HNSC

0.0033

0.0024

0.017

10-

0.021

AR_mRNA

Kruskal-Wallis, p = 0.0017

Grade

LGG

0.0076

0.024

0.0018

0.029

7.5

AR_mRNA

Wilcoxon, p = 0.015

TUMOR_FREE

WITH_TUMOR

Tumor_Status

STAD

0.00068

0.00028

0.0015

7,5

7.5

AR_mRNA

2.5

Kruskal-Wallis, p = 0.9013

grage _!

Stage_M

Stage_0%

Stage_N

Stage

Tumor_Status

Grade

UCEC

0.0002

0.00015

AR_MRNA

Kruskal-Wallis, p = 1.3e-05

Grade

UCEC

0.00014

10.0-

7.5

5.0

”

2.5

0.0

Kruskal-Wallis, p = 0.00066

Endometrioid

Mixed

Serous

Histological_Type

UCEC

0.0053

10.0

7.5-

AR_mRNA

5.0

2.5

0.0

Wilcoxon, p = 0.0053

TUMOR_FREE

WITH_TUMOR

Tumor_Status

SARC

0.011

0.0033

0.048

10-

AR_mRNA

Kruskal-Wallis, p = 0.0043

Age<340

40<Age ← 60

60-Age <380

Ag680

Age_Type

STAD

0.00039

Outve

Complete_Response

Partial_Response

Progressive_Disease

Stable_Disease

TUMOR_FREE

WITH_TUMOR

Tumor_Status

STAD

0.044

10.0

0.0023

AR_MRNA

1.0

5.0-

6.0

AR_mRNA

5.0

2.5

0,0

VAicoxon, p = 0.00039

0.0

Kruskal-Wallis, p = 0.00032

0.0

Wilcoxon, p = 0.002

0.0

Kruskal-Wallis, p = 0.0029

Write

Black

Other

Race

L THCA

12-

2.10-05

0.008

0.00051

AR_mRNA

Kruskal-Wallis, p = 3.50-05

Classicalhasural

Follicular

Cities

Tall_Cell

Histological_Type

THCA

0.026

0.044

0.011

Kruskal-Wallis, p = 0.032

Stage_

Stage_W

Stage_1

Stage_N

Stage

STAD

80-05

New_Tumor_Event

KIRP

0.015

AR_MRNA

AR_mRNA

5.0

2.5

Kruskal-Wallis, p = 0.014

0,0

Wilcoxon, p = 0.041

Kruskal-Wallis, p = 0.0063

Astrocytoma

Oligoastrocytoma

Oligodendroglioma

Histological_Type

BRCA

CHOL

p < 2.22e-16

7.5

5.0

AR_mRNA

25-

0.0

Wilcoxon, p = 0.049

TUMOR_FREE

WITH_TUMOR

Tumor_Status

Race

Age_Type

KIRC

0.0017

12-

3.1e-05

0.0048

1.9e-07

7.5

B-

AR_mRNA

AR_MRNA

5.0

2.5

0.0

Wilcoxon, p = 1.8-09

Kruskal-Wallis, p = 0.00022

Kruskal-Wallis, p = 3.50-06

Stage-

Stage_0

glage_00

Stage_IV

TUMOR_FREE

WITH_TUMOR

Tumor_Status

LIHC

0.0037

0.048

AR_mRNA

Age ⇐ 40

4DeAgoe 60

80€Aggres80

A00-80

Age_Type

KIRC

10.0

0.017

7.5

AR_MRNA

AR_mRNA

10-

”

25-

0.0

Kruskal-Wallis, p = 0.039

Grade

ESCA

0.029

Black

COM

Age-40

40<Age+160

60<Agecx80

Age80

Stage_)

Stage_D

stage_V

Stage_IV

Stage

Kruskal-Wallis, p = 0:0019

Kruskal-Wallis, p

0:019

Kruskal-Wallis, p < 2.2e-16-

White

Kruskal-Wallis, p < 2.2e-16

2.98-09

p < 2.22e-16

0.003

www

Black

Race

treatment_outcome_first_course

LGG

10.0

0.041

LIHC

0.0014

0.011

0.046

6.50-05

0.002

Kruskal-Wallis, p = 2:50-05

Grade

LIHC

0.0079

0.00015

Kruskal-Wallis, p = 0.00017

white

Brack

Oliver

Race

Kruskal-Wallis, p = 0.018

0.0039

0.0027

0.022

Grade

Stage

Figure 4

HNSC

10.0

0.0022

7.5

5.0

2.5

0.0

Wilcoxon, p = 0.0022

Yes

New_Tumor_Event

AR_mRNA

AR_MRNA

Wilcoxon, p = 0.0082

Colon_Adenocarcinoma

UVM

0.014

0.023

AR_mRNA

.”

Kruskal-Wallis, p = 0.934

Epitelioid_Cel

Epithelold_Cel(Spindle_Cell

Spindle_Cel

Spindle_Cel/Epitvetold_Cell

Histological_Type

AR_mRNA

TUMOR_FREE

WITH_TUMOR

LGG

AR_mRNA

0.002

Yes

6.2e-05

1.8e-09

BRCA

0.04

0.031

0.012

0.0015

AR_RPPA

AR_Score

-1

Kruskal-Wallis, p = 0.0032

Kruskal-Wallis, p = 0.01

white

Black

Ofbet

Race

COAD

5.9e-05

0,5

0.0-

-0.5

AR_RPPA

-1.0-

-1.5-

Wilcoxon, p = 5.9e-05

-2.0

Colon_Adenocarcinoma

Colon_Mucinous_Adenocarcinoma

Histological_Type

LGG

0.00012

8.1e-09

0.0074

0.00042

AR_Score

Kruskal-Wallis, p = 8.5e-08

Partial_Remission/Response

Progressive_Disease

Stable_Disease

treatment_outcome_first_course

SARC

0.047

0.017

AR_RPPA

-2

Kruskal-Wallis, p = 0.052

Age ⇐ 40

40-Aged-60

60-Age — 80

Age>80

Age_Type

STAD

0.029

0.5-

0.011

0.0

AR_RPPA

0.5

-1.0

Kruskal-Wallis, p = 0.047

Stage

stage_0

Stage_Y

Stage_IV

Stage

STAD

AR_Score

-10

Wilcoxon, p = 0.0024

TUMOR_FREE

Tumor_Status

UVM

0.0054

0.00038

0.0079

AR_Score

-10

Kruskal-Wallis, p = 0.0017

Epithelicid_Ceil

EpithelHold_CellSpindle_Cell.

Spindle_Cell

Spindie_Cel/Epanelloid_Cel)

Histological_Type

UCEC

0.0008

0.0013

-1

Kruskal-Wallis, p = 0.00017

Grade

UCEC

AR_RPPA

-1

KIRC

0.01

1.7e-07

1e-06

0.5-

0.5

AR_RPPA

0.0

-0.5

-1.0

Wilcoxon, p = 0.00027

-1

Kruskal-Wallis, p = 8.7e-07

Grade

Stage

LGG

0.046

0.0006

0.5

0.0

AR_RPPA

0.5

-10-

Wilcoxon, p = 6e-04

TUMOR_FREE

WITH_TUMOR

Tumor_Status

Histological_Type

HNSC

0.00013

0.0026

0.013

0.00048

AR_Score

Kruskal-Wallis, p = 0.00036

Grade

COAD

0.0003

10-

AR_Score

-10

Wilcoxon, p = 3e-04

Colon_Adenocarcinoma

Colon_Mucinous_Adenocarcinoma Histological_Type

LGG

1.5

4.1e-06

0.0054

1.0

4.7e-13

0.5

0,0

-0.5

-1.0

Kruskal-Wallis, p = 4.2e-13

Astrocytorna

Cligoastrocytoma

Oligodendrogioma

Histological_Type

treatment_outcome_first_course

KIRC

1.5

0.00027

TUMOR_FREE

Tumor_Status

BRCA

0.011

0.047

0.0092

0.021

AR_RPPA

Kruskal-Wallis, p = 0.011

Age ⇐ 40

40<Age ⇐ 60

60-Aged-80

Age>80

Stage_

Stage_0

Stage_01

Stage_I

Race

Age_Type

KIRC

1.5

0.0078

stage_

Stage_N

Stage_M

Stage_I

WITH_TUMOR

LGG

0.001

3.5e-07

Kruskal-Wallis, p = 1e-08

Astrocytoma

Ougeastrocytoma

Oligodendroglioma

LIHC

0.012

0.028

AR_RPPA

Kruskal-Wallis, p = 0.029

Age — 40

40-Age-80

60<Age ⇐ 80

Age>80

Age_Type

LIHC

0.023

0.045

Kruskal-Wallis, p = 0.036

Grade

AR_RPPA

AR_Score

LIHC

0.0013

-10

Kruskal-Wallis, p = 0.0048

while

Black

other

Race

LIHC

0.0017

0.021

0.018

Kruskal-Wallis, p = 0.0019

www

Black Race

citver

AR_RPPA

-1.0

Kruskal-Wallis, p = 0.19

Complete_Response

Partial_Response

Progressive_Disease

Stable_Disea50

1.0-

AR_Score

AR_RPPA

0.0

-0.5

-1.0

Kruskal-Wallis, p = 0.027

0.019

0.0078

1.6e-05

AR_RPPA

-1.

Kruskal-Wallis, p = 7.4e-05

while

Black

Ofniet

THCA

0.00021

4.40-06

AR_RPPA

Kruskal-Wallis, p = 1.6e-05

Classical/usual

Folicular

CiÉVET

Tab_Col

Histological_Type

0.0024

WITH_TUMOR

STAD

0.042

AR_Score

Kruskal-Wallis, p = 0.062

Grade

AR_RPPA

Figure 5

0.03

Wilcoxon, p = 0.03

TUMOR_FREE

WITH_TUMOR

Tumor_Status

Stage

Remission/Response

ACC mRNA OS

ACC mRNA PFI

LAML mRNA OS

Group: + Low + High

1.00

0.75

Survival probability

0,50

N(High) = 39

0.25

N(Low) =40

HR = 0.42(0.185~0.955)

Log-Rank = 0.0328

HR(numeric) = 0.775(0.643~0.934)

0.00

HR(numeric) p-value = 0.00737

100

150

Survival Months

1.00

0.75

Survival probability

0.50

N(High) = 39

0.25

N(Low) =40

HR = 0.568(0.303~1.066)

Log-Rank = 0.0744

HR(numeric) = 0.871(0.762~0.995)

HR(numeric) p-value = 0.0426

0.00

100

150

Survival Months

1.00

0.75

Survival probability

0.50

N(High) = 83

0.25

N(Low) =78

HR = 0.621(0.420~0.918)

Log-Rank = 0.0158

HR(numeric) = 0.925(0.870~0.984)

HR(numeric) p-value = 0.0136

0.00

100

Survival Months

KIRC mRNA OS

KIRC mRNA PFI

Group: + Low + High

Survival probability

1.00

0.75

0.50

N(High) = 266

0.25

N(Low) = 265

HR = 0.477(0.345~0.661)

Log-Rank = 5.2e-06

HR(numeric) = 0.827(0.769~0.889)

0.00

HR(numeric) p-value = 2.83e-07

100

150

Survival Months

LGG mRNA PFI

Group: + Low + High

1.00

0.75

0.50

N(High) = 257

0.25

N(Low) = 257

HR = 1.27(0.954~1.692)

Log-Rank = 0.101

HR(numeric) = 1.11(1.03~1.20)

0.00

HR(numeric) p-value = 0.00957

120

160

Survival Months

OV mRNA OS

Group: + Low + High

1.00

0,75

Survival probability

0.50

N(High) = 151

0.25

N(Low) = 151

HR = 0.732(0.547~0.980)

Log-Rank = 0.0354

HR(numeric) = 0.948(0.878~1.024)

0.00

HR(numeric) p-value = 0.172

100

150

200

Survival Months

STAD mRNA OS

Group:

+ Low

High

1.00

0.75

Survival probability

0.50

N(High) = 205

0.25

N(Low) = 204

HR = 1.69(1.22~2.32)

Log-Rank = 0.00123

HR(numeric) = 1.12(1.04~1.21)

HR(numeric) p-value = 0.0045

0.00

120

Survival Months

STAD mRNA PFI

Group: + Low

High

1.00

0.75

Survival probability

0.50

N(High) = 205

0.25

N(Low) = 206

HR = 1.95(1.38~2.77)

Log-Rank = 0.000134

HR(numeric) = 1.14(1.05~1.23)

0.00

HR(numeric) p-value = 0.00271

120

Survival Months

LIHC mRNA OS

Group:

Low

+ High

1.00

0.75

Survival probability

0.50

N(High) = 185

0.25

N(Low) = 185

HR = 0.681(0.481~0.964)

Log-Rank = 0.029

HR(numeric) = 0.933(0.877~0.993)

0.00

HR(numeric) p-value = 0.0289

120

Survival Months

oup: + Low + High

1.00

0.75

Survival probability

0.50

N(High) = 266

0.25

N(Low) =267

HR = 0.415(0.302~0.570)

Log-Rank = 2.19e-08

HR(numeric) = 0.803(0.750~0.858)

0.00

HR(numeric) p-value = 1.42e-10

100

150

Survival Months

LGG mRNA OS

Group: + Low + High

Figure 6

1.00

0.75

Survival probability

0.50

N(High) = 257

0.25

N(Low) =257

HR = 1.35(0.94~1.93)

Log-Rank = 0.104

HR(numeric) = 1.18(1.06~1.31)

0.00

HR(numeric) p-value = 0.0029

100

150

200

Survival Months

Survival probability

Group: + Low + High

LGG AR Score PFI

Group: + Low + High

1.00

0.75

Survival probability

0.50

~. 0.50

N(High) = 256

0.25

N(Low) =258

0.25

N(Low) =258

HR = 2.53(1.74~3.68)

HR = 2.12(1.58~2.84)

Log-Rank = 4.71e-07

Log-Rank = 2.71e-07

HR(numeric) = 1.08(1.05~1.11)

HR(numeric) = 1.07(1.05~1.09)

HR(numeric) p-value = 5.27e-10

0.00

HR(numeric) p-value = 1.68e-10

100

150

200

120

160

Survival Months

SKCM AR Score OS

SKCM AR Score PFI

Group: + Low + High

Group:

Low

High

1.00

0.75

Survival probability

0.50

N(High) = 226

0.25

N(Low) =224

HR = 0.706(0.539~0.926)

Log-Rank = 0.0114

HR(numeric) = 0.973(0.950~0.996)

HR(numeric) p-value = 0.023

0.00

100

200

300

400

Survival Months

STAD AR Score OS

Group: + Low + High

1.00

0.75

Survival probability

0.50

N(High) = 226

0.25

N(Low) =225

HR = 0.739(0.589~0.928)

Log-Rank = 0.00901

HR(numeric) = 0.981(0.963~0.999)

0.00

HR(numeric) p-value = 0.0371

100

200

300

400

Survival Months

STAD AR Score PFI

Group: + Low + High

1.00

0.75

Survival probability

0.50

N(High) = 202

0.25

N(Low) =207

HR = 1.76(1.28~2.41)

Log-Rank = 0.000445

HR(numeric) = 1.04(1.01~1.08)

HR(numeric) p-value = 0.0196

0.00

120

Survival Months

Figure 7

1.00

0.75

Survival probability

0.50

N(High) = 203

0.25

N(Low) =208

HR = 1.45(1.03~2.03)

Log-Rank = 0.0317

HR(numeric) = 1.05(1.01~1.08)

0.00

HR(numeric) p-value = 0.0128

120

Survival Months

KIRC RPPA PFI

Group: + Low + High

Group:

+ Low

+ High

Group: +

Low

+ High

1.00

0.75

Survival probability

0.50

N(High) = 222

N(High) = 221

0.25

N(Low) =223

0.25

N(Low) = 223

HR = 0.457(0.328~0.637)

HR = 0.534(0.382~0.747)

Log-Rank = 2.02e-06

Log-Rank = 2e-04

HR(numeric) = 0.29(0.196~0.428)

HR(numeric) = 0.375(0.247~0.568)

HR(numeric) p-value = 5.19e-10

0.00

HR(numeric) p-value = 3.87e-06

100

150

100

150

Survival Months

LGG RPPA OS

LGG RPPA PFI

Group:

Low

High

Group:

+ Low +

High

1.00

0.75

Survival probability

0.50

N(High) = 213

0.25

N(Low) =212

HR = 1.55(1.03~2.33)

Log-Rank = 0.0345

HR(numeric) = 1.89(1.01~3.52)

0.00

HR(numeric) p-value = 0.0452

100

150

200

Survival Months

SKCM RPPA OS

Group:

Low

High

1.00

0.75

Survival probability

.0.50

N(High) = 171

0.25

N(Low) = 170

HR = 0.626(0.452~0.865)

Log-Rank = 0.00427

HR(numeric) = 0.597(0.315~1.132)

HR(numeric) p-value = 0.114

0.00

100

200

300

400

Survival Months

SKCM RPPA PFI

Group:

Low

High

Figure 8

1.00

0.75

Survival probability

0.50

N(High) = 172

0.25

N(Low) = 170

HR = 0.738(0.563-0.968)

Log-Rank = 0.0278

HR(numeric) = 0.737(0.433~1.255)

0.00

HR(numeric) p-value = 0.261

100

200

300

400

Survival Months

1.00

0.75

Survival probability

0.50

N(High) = 92

0.25

N(Low) = 92

HR = 0.49(0.323~0.744)

Log-Rank = 0.000634

HR(numeric) = 0.633(0.354~1.132)

HR(numeric) p-value = 0.123

0.00

120

Survival Months

OV RPPA OS

Group: + Low + High

1.00

0.75

Survival probability

0.50

N(High) = 201

0.25

N(Low) = 205

HR = 0.727(0.565~0.936)

Log-Rank = 0.0131

HR(numeric) = 0.76(0.622~0.930)

HR(numeric) p-value = 0.00756

0.00

100

150

200

Survival Months

1.00

0.75

Survival probability

0.50

N(High) = 213

0.25

N(Low) = 212

HR = 1.4(1.01~1.94)

Log-Rank = 0.0418

HR(numeric) = 1.4(0.848~2.316)

0.00

HR(numeric) p-value = 0.188

120

160

Survival Months

LIHC RPPA OS

Mean expression level (Tumor)

ACC

BLCA

BRCA

CESC

- CHOL

COAD

DLBC

ESCA

GBM

- HNSC

KICH

KIRC

KIRP

LAML

LGG

LIHC

- LUAD

LUSC

MESO .

- PAAD

- PCPG

- PRAD

READ

SARC

SKCM-

STAD

TGCT

THCA

THYM

- UCEC

UCS

UVM

Mean expression level (Normal)

uc004dwu.1

uc004dwv.1

uc011mpd.1

uc011mpe.1

uc011mpf.1

GBM uc011mpd.1 OS

GBM uc011mpd.1 PFI

Group: + Low + High

1.00

Survival probability

0.75

Survival probability

0.75

0.50

N(High) = 10

0.25

N(Low) = 146

HR = 2.21(1.12~4.38)

0.25

N(Low) # 146

Log-Rank # 0.0203

HR = 2.34(1.18~4.66)

Log-Rank = 0.0121

HR(numeric) = 2.08(1.45~2.98)

0.00

HR(numeric) p-value = 6.17e-05

HR(numeric) = 1.74(1.23~2.44)

0.00

HR(numeric) p-value = 0.00154

Survival Months

PRAD uc011mpf.1 (AR-V7) OS

Group: + Low + High

PRAD uc011mpf.1(AR-V7) PFI

Group: + Low + High

Figure 9

1.00

Survival probability

0.75

Survival probability

0.75

0.50

N(High) = 248 N(Low) = 249

N(High) = 248

0.25

HR = 5.07( 1.02~25.21)

0.25

N(Low) = 249

Log-Rank = 0.0303

HR = 1.99(1.30~3.05)

Log-Rank = 0.0012

HR(numeric) = 1.9(1.21~2.97)

0.00

HR(numeric) p-value = 0.00495

HR(numeric) = 1.32(1.13~1.54)

0.00

HR(numeric) p-value = 0.000446

120

160

120

160

Survival Months

TYPE

☐ Negative

☐ Positive

2036

2000

1853

1706

1682

AR significantly correlated gene number

1500

1399

1000

1021

960

819

776

639

660

603

543

550

544

500

487

431

447

454

384

318

294

244

223

179

193

186

137

104

131

147

ACC

BLCA

BRCA

CESC

CHOL

COAD

DLBC

ESCA

GBM

HNSC

KICH

KIRC

KIRP

LAML

LGG

LIHC

LUAD

LUSC

MESO

PAAD

PCPG

PRAD

READ

SARC

SKCM

STAD

TGCT

THCA

THYM

UCEC

UCS

UVM

KEGG Pathways

Figure 10

Drug Resistance

Endocrine resistance

EGFR tyrosine kinase inhibitor resistance

Cancer Related

Renal cell carcinoma

Prostate cancer

Non-small cell lung cancer

Melanoma

Gastric cancer

Endometrial cancer

Colorectal cancer

Chronic myeloid leukemia

Breast cancer

Proteoglycans in cancer

qvalue

PD-L1 expression and PD-1 checkpoint pathway in cancer

Cell Signaling

Wnt signaling pathway

0.04

TGF-beta signaling pathway

0.03

Ras signaling pathway

0.02

Rap1 signaling pathway

0.01

PI3K-Akt signaling pathway

Phospholipase D signaling pathway

mTOR signaling pathway

MAPK signaling pathway

Count

Hippo signaling pathway - multiple species

HIF-1 signaling pathway

Hedgehog signaling pathway

FoxO signaling pathway

ErbB signaling pathway

cGMP-PKG signaling pathway

cAMP signaling pathway

Calcium signaling pathway

ECM-receptor interaction

Cell adhesion molecules (CAMs)

Cellular Processes

Gap junction

Focal adhesion

Adherens junction

Cell cycle

CESC

COAD

ESCA

HNSC

KIRC

KIRP

LGG

LUSC

PAAD

PRAD

READ

SARC

STAD

THCA

ACC

number = 46 ; Cor = 0.673 ; p-value = 2.93e-07

1.0-

0.5

AR_RPPA

0.0-

-0.5

-1.0-

AR_mRNA

COAD

number = 352 ; Cor = 0.376 ; p-value = 2.89e-13

0.0-

AR_RPPA

0.5

1.0

-1.5

-2.0

AR_mRNA

KICH

number = 63 ; Cor = 0.589 ; p-value = 3.84e-07

0.4

AR_RPPA

0.0-

-0.4-

0.0

2.5

5.0

7.5

AR_mRNA

10.0

LIHC

number = 181 ; Cor = 0.682 ; p-value = 3.88e-26

1.0

AR_RPPA

0.5

0.0

-0.5

AR_mRNA

PAAD

number = 98 ; Cor = 0.187 ; p-value = 0.0656

0.5

AR_RPPA

0.0

-0.5-

-1.0

AR_mRNA

SKCM number = 347 ; Cor = 0.305 ; p-value = 6.83e-09

STAD

number = 336 ; Cor = 0.245 ; p-value = 5.34e-0€

0.4

0.0-

AR_RPPA

-0.4

-0.8

-1.2

0.0

2.5

5.0

7.5

AR_mRNA

UCS

number = 48 ; Cor = 0.4 ; p-value = 0.00489

1.0

0.5

AR_RPPA

0,0

-0,5

-1.0-

0.0

2.5

5.0

7.5

AR_mRNA

BRCA

number = 869 ; Cor = 0.718 ; p-value = 1.09e-138

AR_RPPA

-1

AR_mRNA

ESCA

number = 125 ; Cor = 0.181 ; p-value = 0.0439

0.0

-0.5-

AR_RPPA

-1.0

-1.5

-2.0

AR_mRNA

KIRP

number = 206 ; Cor = 0.39 ; p-value = 7.02e-09

AR_RPPA

AR_mRNA

LUSC

number = 322 ; Cor = 0.386 ; p-value = 7.12e-13

0.5

0.0 :

AR_RPPA

0.5

-1.0-

AR_mRNA

PRAD

number = 350 ; Cor = 0.067 ; p-value = 0.21

AR_RPPA

AR_mRNA

TGCT

number = 118 ; Cor = 0.407 ; p-value = 4.72e-06

AR_RPPA

-2

AR_mRNA

UVM

number = 12 ; Cor =- 0.367 ; p-value = 0.24

0.00

-0.25

AR_RPPA

-0.50-

-0.75

AR_mRNA

CESC

number = 169 ; Cor = 0.436 ; p-value = 3.21e-09

AR_RPPA

-1

-2

0.0

2.5

5.0

7.5

AR_mRNA

GBM

number = 67 ; Cor = 0.624 ; p-value = 1.73e-08

1.0-

0.5-

AR_RPPA

0.0

-0.5-

-1.0-

2.5

5.0

7.5

AR_mRNA

AR_RPPA

1.0

0.5-

0.0

-0.5

-1.0

0.0

2.5

5.0

7.5

AR_mRNA

LGG

number = 427 ; Cor = 0.574 ; p-value = 9.94e-39

0.5

AR_RPPA

0.0

0.5

-1.0

0.0

2.5

5.0

7.5

AR_mRNA

number = 227 ; Cor = 0.833 ; p-value = 9.95e-60

AR_RPPA

1 -

-1

AR_mRNA

SARC

number = 219 ; Cor = 0.698 ; p-value = 2.5e-33

AR_RPPA

-2

AR_mRNA

THYM number = 87 ; Cor=0.209 ; p-value = 0.0516

0.5

0.0-

AR_RPPA

0.5

-1.0

-1.5

AR_mRNA

UCEC number = 403 ; Cor = 0.661 ; p-value = 5.89e-52

AR_RPPA

-1

0,0

2.5

5.0

7.5

10.0

AR_mRNA

BLCA

number = 340 ; Cor = 0.631 ; p-value = 3.11e-39

AR_RPPA

-1

0.0

2.5

5.0

7.5

AR_mRNA

DLBC number = 33 ; Cor = 0.225 ; p-value = 0.209

0.5

0.0

AR_RPPA

-0.5

-1.0

-1.5

AR_mRNA

KIRC number = 442 ; Cor = 0.558 ; p-value = 1.69e-37

1.0-

AR_RPPA

0.5

0.0

-0.5

-1.0

0.0

2.5

5.0

7.5

10.0

AR_mRNA

LUAD

number = 357 ; Cor = 0.463 ; p-value = 2.18e-20

AR_RPPA

-1

0.0

2.5

5.0

7.5

AR_mRNA

PCPG

MESO

number = 80 ; Cor = 0.223 ; p-value = 0.0467

1.0-

0.5

AR_RPPA

0.0

-0.5

-1.0

0.0

2.5

5.0

7.5

AR_mRNA

number = 61 ; Cor = 0.501 ; p-value = 3.99e-05

1.5-

1.0-

AR_RPPA

0.5

0.0

-0.5-

-1.0

0.0

2.5

5.0

7.5

AR_mRNA

READ

number = 127 ; Cor = 0.367 ; p-value = 2.15e-0

-0.25-

0.50

AR_RPPA

-0.75

€

-1.00

-1.25

AR_mRNA

THCA

number = 366 ; Cor = 0.467 ; p-value = 3.49e-21

AR_RPPA

0.0

2.5

5.0

7.5

AR_mRNA

AR_RPPA

-2

Figure 11

-1-

AR_mRNA

CHOL

number = 30 ; Cor = 0.427 ; p-value = 0.0185

AR_RPPA

0,5

0.0-

-0.5

-1.0

-1.5

0.0

2.5

5.0

7.5

10.0

AR_mRNA

HNSC

number = 339 ; Cor = 0.335 ; p-value = 2.61e-10

ACC

number = 79 ; Cor = 0.288 ; p-value = 0.00999

AR_Score

-10-

AR_mRNA

COAD number = 500 ; Cor = 0.355 ; p-value = 2.82e-16

AR_Score

-10

AR_mRNA

KICH

number = 91 ; Cor=0.107 ; p-value = 0.311

AR_Score

-5

-10-

0.0

2.5

5.0

7.5

10.0

AR_mRNA

LIHC

number = 423 ; Cor = 0.242 ; p-value = 4.83e-07

AR_Score

-10

AR_mRNA

PAAD

number = 183 ; Cor = 0.281 ; p-value = 0.000114

AR_Score

-10

AR_mRNA

SKCM

number = 473 ; Cor = 0.177 ; p-value = 0.000109

AR_Score

AR_mRNA

UCEC number = 581 ; Cor = 0.071 ; p-value = 0.0873

BLCA

number = 427 ; Cor = 0.136 ; p-value = 0.00477

AR_Score

Ã.

0.0

2.5

5.0

7.5

AR_mRNA

BRCA

number = 1212 ; Cor = 0.147 ; p-value = 2.9e-07

50-

AR_Score

AR_mRNA

ESCA

number = 196 ; Cor = 0.363 ; p-value = 1.76e-07

AR_Score

-10

AR_mRNA

KIRP number = 323 ; Cor = 0.09 ; p-value = 0.107

AR_Score

-10-

AR_mRNA

LUSC

number = 552 ; Cor = 0.276 ; p-value = 4.43e-11

AR_Score

-10

AR_mRNA

PRAD

number = 550 ; Cor = 0.27 ; p-value = 1.22e-10

AR_Score

-10

-20

AR_mRNA

TGCT

number = 156 ; Cor = 0.301 ; p-value = 0.000138

AR_Score

-10

AR_mRNA

UVM

number = 80 ; Cor = 0.268 ; p-value = 0.0163

AR_Score

-10

AR_mRNA

CHOL

number = 309 ; Cor = 0.155 ; p-value = 0.00646

-10

0.0

2.5

5.0

7.5

AR_mRNA

AR_Score

number = 45 ; Cor = 0.547 ; p-value = 0.000102

-10

0.0

2.5

5.0

7.5

10.0

AR_mRNA

HNSC

number = 566 ; Cor = 0.339 ; p-value = 1.05e-16

-10

0.0

2.5

5.0

7.5

AR_mRNA

LGG

number = 530 ; Cor = 0.38 ; p-value = 1.29e-19

AR_Score

-10

0.0

2.5

5.0

7.5

AR_mRNA

number = 307 ; Cor = - 0.196 ; p-value = 0.00054

-10

AR_mRNA

SARC

number = 265 ; Cor = - 0.017 ; p-value = 0.783

-10

AR_mRNA

AR_Score

THYM

number = 122 ; Cor = 0.388 ; p-value = 9.82e-06

AR_Score

-10

AR_mRNA

AR_Score

-10

0.0

2.5

5.0

7.5

10.0

AR_mRNA

DLBC

number = 48 ; Cor = 0.378 ; p-value = 0.00815

AR_Score

-5

AR_mRNA

KIRC

number = 606 ; Cor = 0.028 ; p-value = 0.492

AR_Score

-10

0.0

2.5

5.0

7.5

10.0

AR_mRNA

LUAD

number = 576 ; Cor = 0.157 ; p-value = 0.00015

AR_Score

-10

0.0

2.5

5.0

7,5

AR_mRNA

PCPG

number = 187 ; Cor = 0.323 ; p-value = 6.75e-06

AR_Score

-10

0.0

2.5

5.0

7.5

AR_mRNA

STAD

number = 450 ; Cor = 0.18 ; p-value = 0.000126

AR_Score

-10

0.0

2.5

5.0

7.5

AR_mRNA

UCS

number = 57 ; Cor = 0.485 ; p-value = 0.00013

Figure 12

10-

AR_Score

-5

-10

0.0

2.5

5.0

7.5

AR_mRNA

AR_Score

-10-

2.5

5.0

7.5

AR_mRNA

LAML

number = 173 ; Cor = 0.099 ; p-value = 0.193

AR_Score

-10

0.0

2.5

5.0

7.5

AR_mRNA

MESO

number = 87 ; Cor = - 0.08 ; p-value = 0.464

AR_Score

-10

0.0

2.5

5,0

7.5

AR_mRNA

READ

number = 177 ; Cor = 0.511 ; p-value = 3.77e-13

AR_Score

AR_mRNA

THCA

number = 568 ; Cor = 0.355 ; p-value = 2.54e-18

40-

AR_Score

20-

0.0

2.5

5.0

7.5

AR_mRNA

AR_Score

CESC

GBM

number = 171 ; Cor = - 0.046 ; p-value = 0.549

AR_Score

Figure 13

ACC number = 46 ; Cor = 0.174 ; p-value = 0.249

BLCA

BRCA

CESC

CHOL

number = 340 ; Cor = 0.054 ; p-value = 0.323

number = 869 ; Cor = 0.058 ; p-value = 0.0889

number = 169 ; Cor = 0.056 ; p-value = 0.471

number = 30 ; Cor = - 0.183 ; p-value = 0.332

0.5

1.0-

0.0-

0.5-

AR_RPPA

0.5

0.0-

-0.5

-1

-1.0-

-1

-1.0

-1

-2

-1.5

-10

AR_Score

COAD number = 352 ; Cor = 0.07 ; p-value = 0.189

DLBC

ESCA

GBM

HNSC

number = 33 ; Cor =- 0.21 ; p-value = 0.242

number = 125 ; Cor = 0.013 ; p-value = 0.889

number = 67 ; Cor = 0.065 ; p-value = 0.6

number = 339 ; Cor = 0.108 ; p-value = 0.0463

0.5

0.0

1.0

0.0

-0.5

0.5

AR_RPPA

-0.5

AR_RPPA

0.5

1.0

0.0

1.0

-0.5

-1.0-

-1.5

-0.5-

-1.0

-2.0

-1.5

-2.0

-1.0

-10

-5

-10

AR_Score

KICH

KIRC

KIRP

LGG

number =63 ; Cor =- 0.145 ; p-value = 0.256

number = 442 ; Cor =- 0.148 ; p-value = 0.00185

number = 206 ; Cor = - 0.343 ; p-value = 4.66e-07

number = 427 ; Cor = 0.307 ; p-value = 8.87e-11

0.4

1.0-

0.5

AR_RPPA

0.0

AR_RPPA

0.5

AR_RPPA

0.0

-0.5

-0.4-

-0.5-

-1.0-

-1.0

-10

-5

-10

AR_Score

LIHC

LUAD

LUSC

MESO

number = 181 ; Cor = 0.001 ; p-value = 0.994

number = 357 ; Cor = 0.112 ; p-value = 0.0346

number = 322 ; Cor = - 0.044 ; p-value = 0.432

number = 61 ; Cor = - 0.198 ; p-value = 0.127

number = 227 ; Cor =- 0.235 ; p-value = 0.000363

1.5

0.5

1.0-

AR_RPPA

0.0

0.5

AR_RPPA

0.5

AR_RPPA

0.0

0.0-

0.5

-0.5

-1.0-

-1

-10

AR_Score

PAAD

PCPG

PRAD

READ

SARC

number = 98 ; Cor = - 0.004 ; p-value = 0.967

number = 80 ; Cor = - 0.188 ; p-value = 0.0958

number = 350 ; Cor = 0.206 ; p-value = 0.000102

number = 127 : Cor = 0.298 ; p-value = 0.00067

number = 219 ; Cor = - 0.152 ; p-value = 0.0249

0.5

1.0

-0.25-

0.5

-0.50

AR_RPPA

0.0

AR_RPPA

0.0

-0.75

-0.5

-0.5-

-1.00

-1.0

-1.25

-2

-1.0

-10

-20

-10

AR_Score

SKCM

STAD

TGCT

THCA

THYM

number = 347 ; Cor = - 0.059 ; p-value = 0.274

number = 336 ; Cor = - 0.065 ; p-value = 0.234

number = 118 ; Cor = 0.066 ; p-value = 0.476

number = 366 ; Cor = - 0.02 ; p-value = 0.706

number = 87 ; Cor = - 0.091 ; p-value = 0.4

0.4

0.5

0.0

0.0-

AR_RPPA

0.5

-0.4

-1

-1.0-

-0.8

-1.5

-2

-1.2

-2

-10

AR_Score

UCEC

UCS

UVM

number = 403 ; Cor =- 0.142 ; p-value = 0.00423

number = 48 ; Cor =- 0.154 ; p-value = 0.295

number = 12 ; Cor = 0.065 ; p-value = 0.842

1.0

0.5

-0.25

AR_RPPA

0.0-

0.50

-0.5

-0.75-

-1.0-

-1

-10

-5

-10

AR_Score

ACC

BLCA

BRCA

CESC

COAD

DLBC

ESCA

KICH

KIRC

KIRP

LAML

LGG

LIHC

LUAD

LUSC

SKCM

STAD

TGCT

THCA

THYM

UCEC

UCS

UVM

Type

Tumor

Normal

uc004dwu.

uc004dwv.1

uc011mpd.1

uc011mpe.

uc011mpf.1

uc004dwu.1

uc004dwv.1

uc011mpd.1

uc011mpe.

uc011mpf.1

uc004dwu.1

uc004dwv.1

uc011mpd.1

uc011mpe.1

uc011mpf.

uc004dwu.1

uc004dwv.1

uc011mpd.1

uc011mpe.1

uc011mpf.1

uc004dwu.

uc004dwv.1

uc011mpd.1

uc011mpe.

uc011mpf.1

Figure 14

READ

SARC

PAAD

PCPG

PRAD

MESO

Expresssion Isoform log2(RSEM + 1)

CHOL
* *	* *

GBM

HNSC
* *	***

Quartz 4

Explorer

31759122

Journal Pre-proof

The androgen receptor expression and association with patient’s survival in different cancers

Abstract

1. Introduction

2. Materials and Methods

2.1. Data Sources

2.2. AR-score analysis

2.3. AR mRNA/Prote .. . /Iscforms expression and AR activity in different cancer types

2.4. Survival Analysis

2.5. Pathway enrichment analysis

3. Results

3.1. A pan-cancer analysis of the AR mRNA expression in . ifferent cancer types

3.2. A pan-cancer analysis of the AR-score in different cancer types

3.3. A pan-cancer analysis of the AR protein expression in . lifferent cancer types

3.4. AR expression le, aland association with clinical characteristics

3.5. The association of AR mRNA Love with the survival of patients in different cancer types

3.6. The association of AR-score with the survival of patients in different cancer types

3.7. The association of AR protein level with the su. Ival of patients in different cancer types

3.8. AR isoforms analysis in different cancer types

3.9. The Cox regression analysis of AR expression level in din. rent cancer types

3.10. Pathway enrichment analysis of AR expression correlated genes

4. Discussion

Acknowledgments

Conflicts of interest

References:

Journal Pre-proof

Highlights:

Graph View

Table of Contents

Backlinks