LAPTM5 Plays a Key Role in the Diagnosis and Prognosis of Testicular Germ Cell Tumors

Hindawi International Journal of Genomics Volume 2021, Article ID 8816456, 18 pages https://doi.org/10.1155/2021/8816456

Research Article LAPTM5 Plays a Key Role in the Diagnosis and Prognosis of Testicular Germ Cell Tumors

Xiunan Li,1 Yu Su,2 Jiayao Zhang,3 Ye Zhu,4 Yingkun Xu ,5 and Guangzhen Wu 1

1Department of Urology, The First Aﬃliated Hospital of Dalian Medical University, Dalian, Liaoning 116011, China 2Department of Endocrinology, Shandong Provincial Hospital, Cheeloo College of Medicine, Shandong University, Jinan, Shandong 250021, China 3Department of Hepatobiliary Surgery and Center of Organ Transplantation, Shandong Provincial Hospital, Cheeloo College of Medicine, Shandong University, Jinan, Shandong 250021, China 4Department of Orthopedic Surgery, The First Aﬃliated Hospital of Dalian Medical University, Dalian, Liaoning 116011, China 5Department of Urology, Shandong Provincial Hospital, Cheeloo College of Medicine, Shandong University, Jinan, Shandong 250021, China

Correspondence should be addressed to Yingkun Xu; [email protected] and Guangzhen Wu; [email protected]

Received 22 September 2020; Revised 11 November 2020; Accepted 22 December 2020; Published 13 January 2021

Academic Editor: Xingshun Xu

Copyright © 2021 Xiunan Li et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Objective. Testicular germ cell tumors (TGCT) are a serious malignant tumor with low early diagnosis rates and high mortality. Methods. To investigate novel biomarkers to predict the diagnosis and prognosis of this cancer, bioinformatics analysis was used as an accurate, efficient, and economical method. Results. Our study detected 39 upregulated and 589 downregulated differentially expressed genes (DEGs) using the GEO and TCGA databases. To identify the function of DEGs, GO functional analysis, three pathway analysis (KEGG, REACTOME, and PANTHER), and protein-protein interaction network were performed using the KOBAS website, as well as the String database. After a series of analyses in GEPIA and TIMER, including differential expression, we found one candidate gene related to the prognosis and diagnosis of TGCT. LAPTM5 was also associated with CD8+ T cell and PDCD1 expression, which suggests that it may affect immune infiltration. Conclusions. LAPTM5 was identified as a hub gene, which could be used as a potential biomarker for TGCT diagnosis and prognosis.

1. Introduction detected in most patients with TGCT [6]. The upregulation of cellular tumor antigen (FGFR3) [7] and proliferation Testicular germ cell tumors (TGCT) are clinically rare malig- marker protein (BCL10), [8, 9] are thought to be involved nant endocrine tumors [1], with an incidence of 0.002‰– in the development of multiple tumors, including TGCT. 0.005% in the worldwide. This cancer has various clinical Additionally, Wilkie et al. also verified that the expression manifestations and is prone to invasion and metastasis [2]. of FGFR3 and HRAS was increased in TGCT compared to Due to the low rate of early diagnosis and high mortality, normal tissues [10]. However, it is still necessary to discover the survival period is generally no more than three years more representative candidate biomarkers for high-risk [3]. Complete surgical resection is regarded as the only TGCT assessment and inadequate prognosis predictions. potentially curative treatment for TGCT [4, 5]. Therefore, Currently, the discovery of crucial tumor biomarkers finding novel biomarkers for effective screening in the earlier using bioinformatics analysis has become a reliable and prof- stages of TGCT can be a powerful tool to improve the long- itable method [11–13], providing a variety of basic informa- term survival. tion about gene functions, regulatory pathways, cellular Many studies have shown that prognosis is related to the network, and prognostic results of diverse diseases, which levels of multiple molecules. Overexpression of KIT can be play an important role in practical research. In our study, 2 International Journal of Genomics

Volcano GSE3218 GSE3218

0 logFC

–5

0204060 80 Adj.P.val

(a) (b)

TCGA TCGA e diﬀerentially expressed genes on chromosomes 123456789101112 13 14 15 16 17 18 19 20 21 22 X Y GSE3218 Up-regulated DEGs 10 39 795

e gene positions are based on GRCh38.p2(NCBI).6127 genes.

Over-expressed genes Under-expressed genes (c) (d)

TCGA

GSE3218

497 589 4691 Down-regulated DEGs

(e)

Figure 1: Identification of differentially expressed genes in TGCT. (a) Heatmap of GSE3218. (b) Volcano plot of GSE3218. (c) The location of differentially expressed genes of TCGA on chromosomes. Red represents overexpressed genes, and blue represents underexpressed genes. (d, e) Venn diagram of DEGs among GSE3218 and TCGA data. we first downloaded raw data from GEO and TCGA data- Tumor Immune Estimation Resource (TIMER) was used bases to obtain differentially expressed genes (DEGs) in to determine the distribution in pan cancers, pathway TGCT. We performed gene ontology, pathway enrichment enrichment, features in pathological parameters, and their analysis, and a protein-protein interaction (PPI) network. relationship with other genes. We attempted to identify GEPIA was adopted to observe differential expression precise hub genes that may serve as novel biomarkers for and prognostic characteristics of these genes. In addition, TGCT. International Journal of Genomics 3

Pathway enrichment –log10 (Pvalue) Spermatogenesis Single organism reproductive process 120 Single-organism process Single-organism developmental process Single-organism cellular process Single-multicellular organism process Sexual reproduction Reproductive process Reproduction Regulation of cellular process Regulation of biological process Protein binding Primary metabolic process 100 Organic substance metabolic process Organelle part Organelle Nucleus Non-membrane-bounded organelle Multicellular organismal reproductive process Multicellular organismal process Multicellular organism reproduction Multicellular organism development Multi-organism reproductive process Multi-organism process 80 Metabolic process Membrane-bounded organelle Male gamete generation Intracellular part Pathway name Intracellular organelle part Intracellular organelle Intracellular non-membrane-bounded organelle Intracellular membrane-bounded organelle Intracellular Gamete generation Developmental process 60 Cytoplasmic part Cytoplasm Cilium Cellular process Cellular metabolic process Cellular developmental process Cellular component organization or biogenesis Cellular component organization Cell projection Cell part 40 Cell diﬀerentiation Cell Biological regulation Binding Anatomical structure development

–2.30500e-34 1.03725e-33 2.30500e-33 3.57275e-33 4.84050e-33

Pvalue

Count 100 400 200 500 300 (a)

Figure 2: Continued. 4 International Journal of Genomics

REACTOME Pathway enrichment –log10 (Pvalue) Synthesis of PE Signaling by hedgehog Signal transduction Reproduction PIWI-interacting RNA (piRNA) biogenesis Peptide hormone metabolism 6 Organelle biogenesis and maintenance MHC class II antigen presentation Metabolism of lipids and lipoproteins Metabolism Meiotic synapsis 4 Meiosis Pathway name Pathway Intraﬂagellar transport Glutathione conjugation Fertilization Biological oxidations

Assembly of the primary cilium 2 Anchoring of the basal body to the plasma membrane @@Pyruvate metabolism @@Disease

0.000 0.025 0.050 0.075 0.100

Pvalue Count 20 40 60 (b)

Figure 2: Continued. International Journal of Genomics 5

PANTHER Pathway enrichment –log10 (Pvalue)

2.2

p53 pathway feedback loops 2

2.0

Huntington disease 1.8 Pathway name Pathway

1.6

CCKR signaling map

1.4

0.01 0.02 0.03 0.04

Pvalue Count 4 8 5 9 6 7 10 (c) KEGG Pathway enrichment –log10 (Pvalue) Synthesis and degradation of ketone bodies 4.0 Salivary secretion Pyruvate metabolism Protein processing in endoplasmic reticulum Ovarian steroidogenesis 3.5 Metabolic pathways Longevity regulating pathway - multiple species Longevity regulating pathway Leukocyte transendothelial migration 3.0 Inﬂuenza A Huntington's disease Glycolysis / Gluconeogenesis 2.5 Pathway name Pathway Glutathione metabolism Glucagon signaling pathway Fc gamma R-mediated phagocytosis Epstein-Barr virus infection 2.0 Endocytosis Carbon metabolism Bacterial invasion of epithelial cells Adherens junction 1.5

0.00 0.01 0.02 0.03

Pvalue Count 10 30 20 40 (d)

Figure 2: Function analysis and pathway analysis of DEGs. (a–d) Bubble plots of GO, REACTOME, PANTHER, and KEGG analysis. 6 International Journal of Genomics

FDNC3 DEGs ⁎ OS 8

491 9 619 6 (TPM+1)

2 4 Expression−log 2

TGCT (num(T) = 137; num(N) = 165) (a) (b) SYNGR4 FSCN3 ⁎ ⁎ 8

6 (TPM+1) (TPM+1)

2 2 4 4 Expression−log Expression−log 2 2

0 0

TGCT TGCT (num(T) = 137; num(N) = 165) (num(T) = 137; num(N) = 165) (c) (d)

Figure 3: Continued. nentoa ora fGenomics of Journal International

Expression−log (TPM+1) 2 Expression−log2 (TPM+1) 3 4 5 6 7 2 3 4 5 6 7 (num(T) =137;num(N)165) (num(T) =137;num(N)165) (g) (e) VAMP2 GLB1L TGCT TGCT ⁎ ⁎ Figure :Continued. 3:

Expression−log2 (TPM+1) Expression−log2 (TPM+1)

10 0 2 4 6 8 2 4 6 8 (num(T) =137;num(N)165) (num(T) =137;num(N)165) (h) (f) LAPTM5 GSTA1 TGCT TGCT ⁎ ⁎ 7 8 International Journal of Genomics

MAMLD1 PRM1 ⁎ ⁎ 15 6

10 (TPM+1)

4 (TPM+1) 2

5 Expression−log Expression−log 2

0 0 TGCT TGCT (num(T) = 137; num(N) = 165) (num(T) = 137; num(N) = 165) (i) (j)

Figure 3: Cancer-related risk genes in TGCT. (A) Venn diagram of DEGs and survival-related genes in TGCT. (b–j) Gene expression in cancerous and normal samples. Red represents TGCT samples, and blue represents normal samples. ∗P <0:05.

2. Materials and Methods 2.4. PPI Network and Identification of Hub Genes. The PPI network between target genes was built using the String data- 2.1. Data Selection. The GEO database (http://www.ncbi.nlm base (http://stringdb.org/) [22]. String is an online database for .nih.gov/geo/) was used for identifying the data sets of gene searching known proteins and predicting protein interactions. expression between TGCT tissues and normal testis tissues This interaction includes not only the direct interaction [14]. Complete information regarding the relevant data sets between proteins but also the correlation of the indirect func- ff was then evaluated. Finally, in line with the A ymetrix tions of proteins. First, we entered the target genes into the Human Genome (GPL570) platform, one data set, GSE3218, database and set the confidence score as ≥0.7. Then, the was chosen for subsequent analysis, which contained 101 unlinked proteins were removed, and the remaining protein TGCT samples and 6 normal samples. Moreover, TGCT data- interaction data and panoramic images were obtained. sets in TCGA were also used in the analysis, which contained 137 TGCT samples and 165 normal samples. 2.5. Gene Expression Analysis and Survival Analysis. GEPIA (http://gepia.cancerpku.cn/detail.php) is a newly developed 2.2. Differential Expression Analysis. The R language was interactive web server for analyzing the RNA sequencing used to analyze GEO data and draw volcano plots and heat- expression data of 9736 tumors and 8587 normal samples > P < maps [15]. |Log2FC| 1, adjusted value 0.05 were consid- in TCGA and GTEx projects [23]. Based on the analysis in ered the cutoff criterion to identify the differentially GEPIA database, all genes associated with survival of TGCT expressed genes (DEGs). To ensure more accurate results, were found. Furthermore, we identified genes related to we also analyzed DEGs in GSE3218 to verify TCGA data DEGs and survival-related genes as TGCT risked genes. [16]. An online tool, Bioinformatics & Evolutionary Geno- mics Bioinformatics & Evolutionary Genomics (http:// 2.6. Mutation Analysis in Risked Genes. The TGCT (TCGA, bioinformatics.psb.ugent.be/webtools/Venn/) (21079755), Provisional) dataset was selected, comprising data from 154 was used to draw the Venn diagram for upregulated and pathology reports. These genes were further analyzed using downregulated DEGs. cbioportal (http://www.cbioportal.org/index.do) [24]. Geno- mic analysis was covered with mutations and coexpression 2.3. Gene Ontology and Pathway Enrichment Analysis. The analysis. Coexpression and networking were calculated based upregulated and downregulated DEGs were integrated into on the online instructions provided for the cbioportal. The the KOBAS website [17]; we performed Gene Ontology adjusted P value < 0.05 was considered statistically significant. (GO) functional annotation enrichment analysis [18] and KEGG [19], REACTOME [20], and PANTHER [21] path- 2.7. Immune Infiltration Analysis. TIMER (https://cistrome way enrichment analysis. The adjusted P value of <0.05 was .shinyapps.io/timer/) was used to analyze immune infiltra- considered statistically significant. tion among pan-cancer [25]. In this study, we used this International Journal of Genomics 9

Overall survival Overall survival 1.0 1.0

Logrank P = 0.041 Logrank P = 0.04 0.8 HR(high) = 1.6e−09 0.8 HR(high) = 1.6e−09 P(HR) = 1 P(HR) = 1 n(high) = 66 n(high) = 68 0.6 n(low) = 63 0.6 n(low) = 62

0.4 0.4 Percent survival Percent survival 0.2 0.2 FNDC8 FSCN3 0.0 0.0 0 50 100 150 200 250 0 50 100 150 200 250 Months Months

Low FNDC8 group Low FSCN3 group High FNDC8 group High FSCN3 group (a) (b) Overall survival Overall survival 1.0 1.0

Logrank P = 0.043 Logrank P = 0.019 0.8 HR(high) = 1.6e−09 0.8 HR(high) = 1.1e−09 P(HR) = 1 P(HR) = 1 n(high) = 68 n(high) = 68 0.6 n(low) = 68 0.6 n(low) = 68

0.4 0.4 Percent survival Percent survival 0.2 0.2 GLB1L GSTA1 0.0 0.0 0 50 100 150 200 250 0 50 100 150 200 250 Months Months

Low GLB1L group Low GSTA1 group High GLB1L group High GSTA1 group (c) (d) Overall survival Overall survival 1.0 1.0

Logrank P = 0.022 Logrank P = 0.043 0.8 HR(high) = 9.1e+08 0.8 HR(high) = 6.1e+08 P(HR) = 1 P(HR) = 1 n(high) = 68 n(high) = 68 0.6 n(low) = 67 0.6 n(low) = 68

0.4 0.4 Percent survival Percent survival 0.2 0.2 LAPTM5 MAMLD1 0.0 0.0 0 50 100 150 200 250 0 50 100 150 200 250 Months Months

Low LAPTM5 group Low MAMLD1 group High LAPTM5 group High MAMLD1 group (e) (f)

Figure 4: Continued. 10 International Journal of Genomics

Overall survival Overall survival 1.0 1.0

Logrank P = 0.034 Logrank P = 0.04 0.8 HR(high) = 1.5e−09 0.8 HR(high) = 1.6e−09 P(HR) = 1 P(HR) = 1 n(high) = 66 n(high) = 67 0.6 n(low) = 66 0.6 n(low) = 67

0.4 0.4 Percent survival Percent survival 0.2 0.2 PRM1 SYNGR4 0.0 0.0 0 50 100 150 200 250 0 50 100 150 200 250 Months Months

Low PRM1 group Low SYNGR4 group High PRM1 group High SYNGR4 group (g) (h) Overall survival 1.0

Logrank P = 0.042 0.8 HR(high) = 6.1e+08 P(HR) = 1 n(high) = 68 0.6 n(low) = 68

0.4 Percent survival 0.2 VAMP2 0.0 0 50 100 150 200 250 Months

Low VAMP2 group High VAMP2 group (i)

Figure 4: Survival plots for the nine risk-related genes. (a–i) FNDC8, FSCN3, GLB1L, GSTA1, LAPTM5, MAMLD1, PRM1, SYNGR4, and VAMP2, respectively. Red represents the high gene expression group, and blue represents the low gene expression group. P values are shown. online website to analyze the association between hub genes 3.2. Functional Enrichment of DEGs. GO functional enrich- and immune infiltration. ment analysis was performed on the identified DEGs, dem- onstrating that most genes are related to the regulation of cellular processes, metabolic processes, and biological regula- 3. Results tion (Figure 2(a)). The detailed statistical information of GO analysis is shown in Supplementary materials Table S3. In 3.1. DEGs in TGCT. Using the limma package, DEGs in addition, using three pathway databases (KEGG, GSE3218 were identified (Figure 1(a)). As shown in REACTOME, and PANTHER) revealed that TGCT-related Figure 1(b), 49 overexpressed genes and 1086 low- DEGs were mainly concentrated in metabolic pathways, expressed genes were considered as DEGs. Moreover, DEGs salivary secretion, and the TP53 pathway (Figures 2(b)– in TCGA datasets were also identified. The distribution of 2(d)). The detailed statistical information of the above three these DEGs, identified using TCGA, on different chromo- pathway analysis is shown in Supplementary materials somes is shown in Figure 1(c). Finally, DEGs in TGCT were Table S4-S6. identified by taking the intersection between the GEO datasets and TCGA data. A total of 628 DEGs consisting of 39 3.3. Identification of TGCT-Associated Risk Genes. Based on upregulated genes and 589 downregulated genes were the analysis of TCGA data, we identified genes related to obtained in our study (Figures 1(d) and 1(e)). Among them, prognosis. To better determine the TGCT-associated risk the detailed information of these genes involved in the above genes, we crossed the DEGs with those related to prognosis. two Venn diagrams are organized in Supplementary mate- As a result, nine genes were identified (Figure 3(a)), which rials Table S1-S2. are SYNGR4, FNDC8, FSCN3, GLB1L, GSTA1, LAPTM5, International Journal of Genomics 11

Ampliﬁcation Homozygous deletion Copy number Gain Hemizygous deletion ick border: seed gene in border: linker gene mRNA Expression Up-regulated Mutation Down-regulated Mutated

Alteration frequency (%) 0 100

Not FDA approved FDA approved drug targets gene GENE drug targets gene

Figure 5: Protein-Protein Interaction (PPI) network. (a) PPI network diagram between nine risk-related genes and their related genes.

VAMP2, MAMLD1, and PRM1. The detailed information KIRC. In contrast, LAPTM5 has a lower expression in about this Venn diagram can be obtained through Supple- other cancer types, such as LUAD, HNSC, and LUSC. This mentary materials Table S7. Till now, we have not suggests that the LAPTM5 gene plays different biological distinguished whether these DEGs are highly or roles in different cancer types (Figure 6(c)). Then, using underexpressed. Therefore, we subsequently used the LAPTM5 and its most related genes in the string dataset GEPIA website based on TCGA database to explore their (Figure 6(d)), pathway analysis was used to identify the expression levels (Figures 3(b)–3(j)) and their correlation LAPTM5 pathway function (Figure 6(e)). The results with overall survival (Figures 4(a)–4(i)). Above all, we indicate that there may be a correlation between these found genes associated with high expression (poor genes and immunoregulatory interactions between a prognosis), such as LAPTM5. Similarly, we also found Lymphoid and a non-Lymphoid cell. The relationship genes with low expression (good prognosis), such as between the coexpression of LAPTM5 and its related genes MAMLD1 and VAMP2. Moreover, coexpression analysis is shown in Figure 6(f). was used for the TGCT-associated risk genes. As shown in Figure 5(a), most of the genes were correlated with VAMP2. 3.5. Analysis of the Correlation between LAPTM5 and Immunity. In order to explore the correlation between 3.4. Hub Gene Identification. To further identify the associ- LAPTM5 and immunity, we performed GSEA of the ated hub gene, we searched for the top 40 hub genes using LAPTM5 gene associated with TGCT. The analysis results the cytohubba plug-in in the Cytoscape software show that LAPTM5 can activate antigen-activated B cell (Figure 6(a)) and then overlapped the overall survival- receptor (BCR), leading to the generation of second messen- related genes intersections (Figure 6(b)) to find LAPTM5 as gers (Figure 7(b)), complement cascade (Figure 7(c)), costi- the TGCT-related hub gene. The detailed information related mulation by the CD28 family (Figure 7(d)), creation of C4 to this Venn diagram is shown in Supplementary materials and C2 activators (Figure 7(e)), immunoregulatory interac- Table S8. Combined with the previous findings that tions between a Lymphoid and a non-Lymphoid cell LAPTM5 is abnormally highly expressed in TGCT, and the (Figure 7(g)), initial complement triggering (Figure 7(h)), prognosis of the LAPTM5 high expression group is and signaling using the BCR biological pathway significantly worse than that of the low expression group. (Figure 7(i)). Secondly, LAPTM5 inhibited adherens junc- Therefore, we determined that LAPTM5 is a hub gene in tion interactions (Figure 7(a)), FGFR1 ligand binding, and TGCT. In addition, we found that LAPTM5 is highly activation of biological pathways (Figure 7(f)). Furthermore, expressed in some cancer types, such as ESCA, KIRP, and the correlation between LAPTM5 and immune infiltration 12 International Journal of Genomics

Top 40 hub genes by degree

LDHC

SYCP1 SPA17 PRM1 DKKL1 CREM SPO11

CAMK4 TEX12

TNP2 STAG3

ACR SYCP2

SPACA1 FKBP6 OS

ODF1 PIWIL1

CRISP 2 DDX25 PRM2 PAPOLB GAPDHS AKAP4 Top 40 hub genes PGK2

CCL2 MMP9 499 1 39 CD74

CXCR4 SPP1

CCL19 MMP12

C1QB IL13

CDH1 HLA-DRB1 LAPTM5 HLA-DRA KRT19 TYROBP C1QALAPTM5

(a) (b)

17.5 ⁎⁎⁎ ⁎⁎⁎ ⁎⁎ ⁎⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎ ⁎⁎⁎ ⁎⁎ ⁎⁎⁎

15.0

12.5

10.0

7.5

5.0 LAPTM5 expression level (log2 RSEM) OV.Tumor UCS.Tumor LGG.Tumor ACC.Tumor GBM.Tumor KIRP.Tumor KIRC.Tumor LIHC.Tumor UVM.Tumor CESC.Tumor ESCA.Tumor LUSC.Tumor KICH.Tumor SARC.Tumor BLCA.Tumor KIRP.Normal PCPG.Tumor STAD.Tumor DLBC.Tumor BRCA.Tumor KIRC.Normal LIHC.Normal PRAD.Tumor TGCT.Tumor READ.Tumor UCEC.Tumor HNSC.Tumor LAML.Tumor MESO.Tumor PAAD.Tumor ESCA.Normal LUAD.Tumor LUSC.Normal SKCM.Tumor CHOL.Tumor KICH.Normal THCA.Tumor BLCA.Normal STAD.Normal COAD.Tumor BRCA.Normal PRAD.Normal READ.Normal THYM.Tumor UCEC.Normal HNSC.Normal LUAD.Normal CHOL.Normal THCA.Normal COAD.Normal SKCM.Metastasis BRCA−Her2.Tumor BRCA−Basal.Tumor BRCA−Luminal.Tumor HNSC−HPVpos.Tumor HNSC−HPVneg.Tumor (c)

ICAM1 CORO1A

LAPTM4A

ITGB2 ITGAL

CD53 LAPTM4B LAPTM5

HCLS1 ITGAM ITGAX

CTSS SYK CLEC5A SPI1

PLEK

TREM2 TYROBP

CD300LB

CD300E TREM1

(d)

Figure 6: Continued. International Journal of Genomics 13

ITGAM

Staphylococcus aureus infection

CLEC5A Viral myocarditis Immunoregulatory interactions between a Integrin cell surface Lymphoid and a TREM1 interactions non-Lymphoid cell TYROBP DAP12 interactions CD300E

ITGAX TREM2

ITGAL CD300LB ITGB2 ICAM1 SYK

DAP12 signaling Malaria (e)

LAPTM5 vs. CD300E LAPTM5 vs. CD300LB LAPTM5 vs. ICAM1 LAPTM5 vs. ITGAL 9 13 14 4 8 Spearman: 0.17 Spearman: 0.29 Spearman: 0.85 13 Spearman: 0.87 (P = 0.0391) 7 (P = 3.725e-4) 12 (P = 2.46e-42) (P = 4.92e-46) 2 y = 0.23x + -2.41 Pearson: 0.09 Pearson: 0.30 y = 0.76x + 0.97 Pearson: 0.86 12 y = 1.06x + -2.65 Pearson: 0.87 (P = 0.249) 6 (P = 2.051e-4) 11 (P = 4.10e-44) (P = 1.52e-46) y = 0.34x + 0.03 11 0 5 10 10 4 –2 9 9 3 –4 8 2 8 7 –6 1 7 0 6 –8 6 5 mRNA expression (RNA Seq V2 RSEM): ICAM1 (log2) mRNA expression (RNA Seq V2 RSEM): CD300E (log2) mRNA expression (RNA Seq V2 RSEM): ITGAL (log2) 8 9 10 11 12 13 14 15 16 mRNA expression (RNA Seq V2 RSEM): CD300LB (log2) 8 9 10 11 12 13 14 15 16 8 9 10 11 12 13 14 15 16 8 9 10 11 12 13 14 15 16 mRNA expression (RNA Seq V2 RSEM): LAPTM5 (log2) mRNA expression (RNA Seq V2 RSEM): LAPTM5 (log2) mRNA expression (RNA Seq V2 RSEM): LAPTM5 (log2) mRNA expression (RNA Seq V2 RSEM): LAPTM5 (log2)

LAPTM5 vs. ITGB2 LAPTM5 vs. TREM1 LAPTM5 vs. TREM2 LAPTM5 vs. TYROBP

15 16 12 10 Spearman: 0.89 Spearman: 0.19 Spearman: 0.60 14 Spearman: 0.94 15 (P = 1.18e-52) (P = 0.0234) (P = 3.46e-16) (P = 3.31e-68) 8 10 14 Pearson: 0.90 Pearson: 0.24 Pearson: 0.62 13 Pearson: 0.95 y = 1.08x + -1.17 6 y = 0.8x + -2.43 (P = 4.22e-56) y = 0.53x + -2.75 (P = 3.598e-3) (P = 3.31e-17) y = 0.99x + -1.43 (P = 4.50e-74) 8 12 13 4 12 11 2 6 11 0 10 4 10 –2 9

9 –4 2 8 8 –6 7 0 7 –8 6 mRNA expression (RNA Seq V2 RSEM): ITGB2 (log2) mRNA expression (RNA Seq V2 RSEM): TREM1 (log2) mRNA expression (RNA Seq V2 RSEM): TREM2 (log2) 8 9 10 11 12 13 14 15 16 8 9 10 11 12 13 14 15 16 8 9 10 11 12 13 14 15 16 mRNA expression (RNA Seq V2 RSEM): TYROBP (log2) 8 9 10 11 12 13 14 15 16 mRNA expression (RNA Seq V2 RSEM): LAPTM5 (log2) mRNA expression (RNA Seq V2 RSEM): LAPTM5 (log2) mRNA expression (RNA Seq V2 RSEM): LAPTM5 (log2) mRNA expression (RNA Seq V2 RSEM): LAPTM5 (log2)

ITGB2 mutated Neither mutated (f)

Figure 6: Schematics of the hub genes. (a) Top 40 hub genes by degree. (b) Venn diagram for the hub gene intersection. (c) The pan-cancer expression for LAPTM5. ∗P <0:05, ∗∗P <0:01, ∗∗∗P <0:001. (d) Protein-protein interaction network for LAPTM5 and its related gene. (e) Pathway analysis for LAPTM5 and its related gene. (f) The relationship between the co-expression of LAPTM5 and its related genes. P values are shown. was analyzed using the TIMER website. As shown in treatments, and improving the ineffective prognosis of Figures 8(a) and 8(b), LAPTM5 was associated with purity TGCT. and CD8+ T cells. Moreover, the immune checkpoint genes In our research, GSE3218 (6 normal testis tissues and 101 were related to LAPTM5 expression. As shown in TGCT tissues) was selected from the GEO database. After the Figures 8(c)–8(f), PDCD1 showed coexpression with results obtained from using the R language were correlated LAPTM5. with data from TCGA, 39 upregulated genes and 589 downregulated genes were identified in our study. These genes may 4. Discussion play a critical role in the development of TGCT. We also carried out GO functional analysis and pathway analysis In the past 20 years, molecular biology studies on TGCT have (KEGG, REACTOME, and PANTHER) using the KOBAS made great progress [26–29], but the main pathogenesis of website to learn the gene function and regulatory processes this cancer is still unclear. As a highly malignant tumor, there of these candidates. In the functional analysis, the TP53 path- is an urgent need to find effective diagnostic and prognostic way was associated with DEGs, which is consistent with the markers for identifying early-stage patients, developing valid results of previous research [30]. The TP53 pathway plays 14 International Journal of Genomics

Enrichment plot: REACTOME_ADHERENS_JUNCTIONS_INTERACTIONS 0.0 –0.1 –0.2 0.8 –0.3 0.6 –0.4 0.4 –0.5 –0.6 0.2 Enrichment score (ES) score Enrichment

Enrichment score (ES) score Enrichment 0.0

1.0 ‘h’ (positively correlated) 1.0 ‘h’ (positively correlated) 0.5 0.5 Zero cross at 23059 Zero cross at 23059 0.0 0.0 –0.5 ‘T’ (negatively correlated) –0.5 ‘T’ (negatively correlated) 0 10,000 20,000 30,000 40,000 50,000 0 10,000 20,000 30,000 40,000 50,000 Ranked list metricRanked (signal2noise) Rank in ordered dataset

Ranked list metricRanked (signal2noise) Rank in ordered dataset Enrichment profile Enrichment profile Hits Hits Ranking metric scores Ranking metric scores (a) (b)

Enrichment plot: REACTOME__COMPLEMENT_CASCADE 0.8 0.7 0.8 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2 0.1 0.1 Enrichment score (ES) score Enrichment 0.0 (ES) score Enrichment 0.0

‘h’ (positively correlated) 1.0 ‘h’ (positively correlated) 1.0 0.5 0.5 Zero cross at 23059 Zero cross at 23059 0.0 0.0 –0.5 ‘T’ (negatively correlated) –0.5 ‘T’ (negatively correlated) 0 10,000 20,000 30,000 40,000 50,000 0 10,000 20,000 30,000 40,000 50,000

Ranked list metricRanked (signal2noise) Rank in ordered dataset

Ranked list metricRanked (signal2noise) Rank in ordered dataset Enrichment profile Enrichment profile Hits Hits Ranking metric scores Ranking metric scores (c) (d) Enrichment plot: REACTOME_FGFR1_LIGAND_BINDING_AND 0.8 _ACTIVATION 0.7 0.0 0.6 –0.1 0.5 0.4 –0.2 0.3 –0.3 0.2 –0.4 0.1 –0.5

Enrichment score (ES) score Enrichment –0.6

0.0 (ES) score Enrichment –0.7

1.0 ‘h’ (positively correlated) 1.0 ‘h’ (positively correlated) 0.5 0.5 Zero cross at 23059 Zero cross at 23059 0.0 0.0 –0.5 ‘T’ (negatively correlated) –0.5 ‘T’ (negatively correlated) 0 10,000 20,000 30,000 40,000 50,000 0 10,000 20,000 30,000 40,000 50,000 Rank in ordered dataset Ranked list metricRanked (signal2noise)

Ranked list metricRanked (signal2noise) Rank in ordered dataset Enrichment profile Enrichment profile Hits Hits Ranking metric scores Ranking metric scores (e) (f)

Figure 7: Continued. International Journal of Genomics 15

0.8 0.7 0.8 0.6 0.6 0.5 0.4 0.4 0.3 0.2 0.2 0.1

Enrichment score (ES) score Enrichment 0.0 0.0 Enrichment score (ES) score Enrichment

1.0 ‘h’ (positively correlated) 1.0 ‘h’ (positively correlated) 0.5 0.5 Zero cross at 23059 Zero cross at 23059 0.0 0.0

–0.5 ‘T’ (negatively correlated) –0.5 ‘T’ (negatively correlated) 0 10,000 20,000 30,000 40,000 50,000 0 10,000 20,000 30,000 40,000 50,000

Ranked list metricRanked (signal2noise) Rank in ordered dataset Rank in ordered dataset Ranked list metricRanked (signal2noise)

Enrichment profile Enrichment profile Hits Hits Ranking metric scores Ranking metric scores (g) (h)

0.7 0.6 0.5 0.4 0.3 0.2

Enrichment score (ES) score Enrichment 0.1 0.0

1.0 ‘h’ (positively correlated) 0.5 Zero cross at 23059 0.0

–0.5 ‘T’ (negatively correlated) 0 10,000 20,000 30,000 40,000 50,000 Rank in ordered dataset Ranked list metricRanked (signal2noise)

Enrichment profile Hits Ranking metric scores (i)

Figure 7: GSEA. (a) Adherens junctions interactions. (b) Antigen activates B cell receptor BCR leading to generation of second messengers. (c) Complement cascade. (d) Costimulation by the CD28 family. (e) Creation of C4 and C2 activators. (f) FGFR1 ligand binding and activation. (g) Immunoregulatory interactions between a Lymphoid and a non-Lymphoid cell. (h) Initial triggering of complement. (i) Signaling by the B cell receptor BCR. an important role in TGCT. Previous studies have shown tive splicing results in multiple transcript variants. Many that TP53 pathway is abnormally expressed in TGCT [31]. studies have shown the relationship between GSTA1 and In addition, microRNA-214 can regulate the progress of cancer, including breast cancer, hepatocellular carcinoma TGCT by regulating the activity of the TP53 pathway [32]. [33], and colorectal cancer [34]. Mastermind-like domain Moreover, we identified the DEGs and survival- containing 1 (MAMLD1) encodes a mastermind-like associated genes. Furthermore, nine genes (FNDC8, domain-containing protein. This protein may function as a SYNGR4, FSCN3, GLB1L, GSTA1, LAPTM5, MAMLD1, transcriptional coactivator. Mutations in this gene cause X- VAMP2, and PRM1) were identified as both DEGs and linked hypospadias type 2 and alternate splicing results in survival-related genes. Glutathione S-transferase alpha 1 multiple transcript variants. Using mass spectrometry, (GSTA1) encodes a member of the enzyme family whose Mohammed et al. found that some phosphorylated proteins, function is to add glutathione to therapeutic drug carcino- including MAMLD1, affect tumorigenesis [35]. Protamine 1 gens and oxidative stress products. This action is an essential (PRM1) was found to affect colon cancer proliferation, inva- step in the detoxification of these compounds. This family of sion, migration, diagnosis, and prognosis [36]. Other genes, enzymes plays a special role in protecting cells from reactive similar to galactosidase beta 1 like (GLB1L), fibronectin type oxygen species and peroxidation products. The polymor- iii domain containing 8 (FNDC8), synaptogyrin 4 phism of this gene affects the individual’s ability to metabo- (SYNGR4), MAMLD1, and fascin actin-bundling protein 3 lize different drugs. This gene is located in a cluster of (FSCN3), which is a novel paralog of action binding protein similar genes and pseudogenes on chromosome 6. Alterna- fascin, especially expressed in slender sperm heads, have 16 International Journal of Genomics

Purity B Cell CD8 + T Cell CD4 + T Cell Macrophage Neutrophil Dendritic Cell 16 cor = −0.71 Partial.cor = 0.329 Partial.cor = 0.018 Partial.cor = 0.537 Partial.cor = 0.321 Partial.cor = 0.639 Partial.cor = 0.44 P = 5.33e−24 P = 4.77e−05 P = 8.32e−01 P = 2.73e−12 P = 7.32e−05 P = 3.10e−18 P = 2.75e−08 14

12 TGCT 10

8 0.25 0.50 0.75 1.00 0.0 0.1 0.2 0.3 0.4 0.1 0.2 0.3 0.4 0.1 0.2 0.3 0.4 0.5 0.00 0.05 0.10 0.15 0.20 0.10 0.15 0.20 0.25 0.25 0.50 0.75 1.00 LAPTM5 expression level (log2 RSEM) Inﬁltration Level (a) TGCT LAPTM5 ⁎ ⁎⁎⁎ ⁎ ⁎ ⁎ 10.0 cor = 0.816 ⁎ P = 0e + 00

7.5 1.0

5.0 PDCD1

0.5 2.5 Expression level (log2 RSEM) Inﬁltration level 8 101214 Expression level (log2 RSEM) 0.0 B Cell CD8+T cell CD4+T cell Macrophage Neutrophil Dendritic cell Copy number Arm−level deletion Diploid/Normal Arm−level gain (b) (c)

LAPTM5 LAPTM5 LAPTM5 cor = 0.888 cor = 0.667 10.0 cor = 0.823 P = 0e + 00 P = 0e + 00 P = 0e + 00 9 7.5 7.5

5.0 6 5.0 CD274 CTLA4

PDCD1LG2 2.5 3 2.5 Expression level (log2 RSEM) Expression level (log2 RSEM) 0.0 Expression level (log2 RSEM)

8 101214 8 101214 8101214 Expression level (log2 RSEM) Expression level (log2 RSEM) Expression level (log2 RSEM) (d) (e) (f)

Figure 8: Relationship between LAPTM5 and immune responses. (a, b) Relationship between LAPTM5 and immune infiltration. ∗P <0:05, ∗∗P <0:01, ∗∗∗P <0:001.(c–f) Relationship between LAPTM5 and immune checkpoints. P values are shown. not been reported to be associated with tumors. These genes methylation and genes [39]. In TGCT, we found abnormal should be investigated in-depth in TGCT in the future. high expression of LAPTM5 in cancer samples. Moreover, We further identified hub genes affecting TGCT through LAPTM5 can also affect tumor prognosis. However, at pres- core protein network and prognosis analysis. As a result of ent, laboratory and clinical studies of LAPTM5 in TGCT this analysis, we found that LAPTM5 encodes a transmem- have not been carried out, and its potential mechanism needs brane receptor that is associated with lysosomes. The to be confirmed by further studies. encoded protein, also known as E3 protein, may play a role in hematopoiesis. Previous studies have shown that LAPTM5 is a protein preferentially expressed in immune cells and that 5. Conclusions it interacts with the Nedd4 family of ubiquitin ligases. LAPTM5 is an active regulator of the proinflammatory sig- In summary, based on the results of a series of bioinformatic naling pathways in macrophages [37]. In addition, we per- analyses, our study concluded that LAPTM5 is particularly formed GSEA in TGCT for LAPTM5, the results of which important for the high-risk assessment and inadequate prog- showed that there is a correlation between LAPTM5 and nosis of TGCT, suggesting that LAPTM5 could be a very use- multiple important immune pathways. In bladder cancer ful biomarker for determining the accurate diagnosis and cells, Chen et al. reported that low expression of LAPTM5 prognosis of this disease. However, more efforts should be suppresses cell cycle changes [38]. LAPTM5 can also be a invested in clinical experiments to learn the biological func- biomarker in lung cancer, which shows a correlation between tions and pathological importance of these genes in TGCT. International Journal of Genomics 17

In addition, we believe that our research can provide valuable Supplementary Materials data for future scientific research. Table S1: the detailed information on Venn diagrams related to upregulated DEGs. Table S2: the detailed information on Abbreviations Venn diagrams related to downregulated DEGs. Table S3: the detailed statistical information related to GO analysis. LAPTM5: Lysosomal protein transmembrane 5 Table S4: the detailed statistical information related to GEO: Gene expression omnibus REACTOME analysis. Table S5: the detailed statistical infor- TCGA: The Cancer Genome Atlas mation related to PANTHER analysis. Table S6: the detailed GO: Gene Ontology statistical information related to KEGG analysis. Table S7: KEGG: Kyoto Encyclopedia of Genes and Genomes the detailed information about the Venn diagram between TGCT: Testicular germ cell tumors OS-related genes and DEGs. Table S8: the detailed informa- PDCD1: Programmed cell death 1 tion about the Venn diagram between OS-related genes and BCL10: BCL10 immune signaling adaptor Top 40 hub genes. (Supplementary materials) FGFR3: Fibroblast growth factor receptor 3 PPI: Protein-protein interaction References DEGs: Differentially expressed genes FNDC8: Fibronectin type III domain containing 8 [1] K.-P. Dieckmann and U. Pichlmeier, “Clinical epidemiology of SYNGR4: Synaptogyrin 4 testicular germ cell tumors,” World Journal of Urology, vol. 22, FSCN3: Fascin actin-bundling protein 3 no. 1, pp. 2–14, 2004. GLB1L: Galactosidase beta 1 like [2] L. Tremeau and N. Mottet, “Management of residual masses of GSTA1: Glutathione S-transferase alpha 1 testis germ cell tumors,” Bulletin du Cancer, vol. 107, no. 2, VAMP2: Vesicle associated membrane protein 2 pp. 215–223, 2020. MAMLD1: Mastermind like domain containing 1 [3] K. M. Saad, R. S. Abdelrahman, and E. Said, “Mechanistic per- PRM1: Protamine 1 spective of protective effects of nilotinib against cisplatin- ESCA: Esophageal carcinoma induced testicular injury in rats: Role of JNK/caspase-3 signal- KIRP: Kidney renal papillary cell carcinoma ing inhibition,” Environmental Toxicology and Pharmacology, KIRC: Kidney renal clear cell carcinoma vol. 76, p. 103334, 2020. LUAD: Lung adenocarcinoma [4] M. Cullen, R. Huddart, J. Joffe et al., “The 111 Study: A Single- HNSC: Head and neck squamous cell carcinoma arm, Phase 3 Trial Evaluating One Cycle of Bleomycin, Etopo- LUSC: Lung squamous cell carcinoma. side, and Cisplatin as Adjuvant Chemotherapy in High-risk, Stage 1 Nonseminomatous or Combined Germ Cell Tumours of the Testis![](https://ars.els- cdn.com/content/image/1-s2.0- Data Availability S0302283819308954-eulogo1.jpg),” European Urology, vol. 77, no. 3, pp. 344–351, 2020. The data used to support the findings of this study are avail- [5] H. Tsunezuka, T. Nakamura, K. Fujikawa et al., “Prediction able from the corresponding author upon request. models for the viability of pulmonary metastatic lesions after chemotherapy in nonseminomatous germ cell tumors,” Inter- national Journal of Urology, vol. 27, no. 3, pp. 206–212, 2020. Conflicts of Interest [6] Q. Tian, H. F. Frierson, G. W. Krystal, and C. A. Moskaluk, “ fl Activating _c-kit_ Gene Mutations in Human Germ Cell The authors declare that they have no con icts of interest. Tumors,” The American Journal of Pathology, vol. 154, no. 6, pp. 1643–1647, 1999. Authors’ Contributions [7] P. L. Tavormina, R. Shiang, L. M. Thompson et al., “Thanato- phoric dysplasia (types I and II) caused by distinct mutations Guangzhen Wu, Yingkun Xu, and Xiunan Li designed the in fibroblast growth factor receptor 3,” Nature Genetics, research methods and analyzed the data. Yingkun Xu, Yu vol. 9, no. 3, pp. 321–328, 1995. Su, and Jiayao Zhang participated in data collection. Xiunan [8] T. G. Willis, D. M. Jadayel, M.-Q. Du et al., “_Bcl10_ Is Li and Ye Zhu drafted and revised the manuscript. All Involved in t(1;14)(p22;q32) of MALT B Cell Lymphoma ” authors approved the version to be released and agreed to and Mutated in Multiple Tumor Types, Cell, vol. 96, no. 1, – be responsible for all aspects of the work. pp. 35 45, 1999. [9] S. H. Lee, M. S. Shin, H. S. Kim et al., “Point mutations and deletions of the Bcl10 gene in solid tumors and malignant lym- Acknowledgments phomas,” Cancer Research, vol. 59, no. 22, pp. 5674–5677, 1999. [10] A. Goriely, R. M. S. Hansen, I. B. Taylor et al., “Activating We thank the cancer genome atlas (TCGA) for providing mutations in _FGFR3_ and _HRAS_ reveal a shared genetic publicly available data. This project is supported by the Liao- origin for congenital disorders and testicular tumors,” Nature ning Provincial Department of Education’s 2020 scientific Genetics, vol. 41, no. 11, pp. 1247–1252, 2009. research project (LZ2020071). In addition, Yingkun Xu [11] J. Rung and A. Brazma, “Reuse of public genome-wide gene would also like to thank Jiayi Li for the useful discussions sur- expression data,” Nature Reviews. Genetics, vol. 14, no. 2, rounding this research. pp. 89–99, 2013. 18 International Journal of Genomics

[12] Y. Xu, G. Wu, J. Li et al., “Screening and Identification of Key ticular germ cell tumors,” Transl Androl Urol, vol. 9, Suppl 1, Biomarkers for Bladder Cancer: A Study Based on TCGA and pp. S31–S35, 2020. ” GEO Data, BioMed Research International, vol. 2020, Article [29] M. Azizi, A. M. Aydin, S. K. Cheriyan et al., “Therapeutic strat- ID 8283401, 20 pages, 2020. egies for uncommon testis cancer histologies: teratoma with [13] Y. Xu, X. Li, Y. Han et al., “A new prognostic risk model based malignant transformation and malignant testicular sex cord on PPAR pathway-related genes in kidney renal clear cell car- stromal tumors,” Transl Androl Urol, vol. 9, Suppl 1, cinoma,” PPAR Research, vol. 2020, Article ID 6937475, 13 pp. S91–s103, 2020. pages, 2020. [30] N. Singla, J. T. Lafin, R. A. Ghandour, S. Kaffenberger, J. F. [14] T. Barrett, D. B. Troup, S. E. Wilhite et al., “NCBI GEO: min- Amatruda, and A. Bagrodia, “Genetics of testicular germ cell ing tens of millions of expression profiles–database and tools tumors,” Current Opinion in Urology, vol. 29, no. 4, pp. 344– update,” Nucleic Acids Research, vol. 35, no. Database, 349, 2019. – pp. D760 D765, 2007. [31] L. Boublikova, V. Bakardjieva-Mihaylova, K. Skvarova Kra- [15] R. C. Team, R: a language and environment for statistical com- marzova et al., “Wilms tumor gene 1 (WT1), TP53, RAS/- puting, Vienna, Austria, 2013. BRAF and KIT aberrations in testicular germ cell tumors,” – [16] K. Tomczak, P. Czerwinska, and M. Wiznerowicz, “The Can- Cancer Letters, vol. 376, no. 2, pp. 367 376, 2016. cer Genome Atlas (TCGA): an immeasurable source of knowl- [32] B.-F. Chen, Y.-K. Suen, S. Gu, L. Li, and W.-Y. Chan, “A miR- edge,” Contemp Oncol (Pozn), vol. 19, no. 1A, pp. A68–A77, 199a/miR-214 Self-Regulatory Network via PSMD10, TP53 2015. and DNMT1 in Testicular Germ Cell Tumor,” Scientific [17] H. Ge, Y. Yan, D. Wu, Y. Huang, and F. Tian, “Potential role of Reports, vol. 4, 2015. LINC00996 in colorectal cancer: a study based on data mining [33] X. Liu, X. Sui, C. Zhang et al., “Glutathione S-transferase A1 and bioinformatics,” Oncotargets and Therapy, vol. Volume suppresses tumor progression and indicates better prognosis 11, pp. 4845–4855, 2018. of human primary hepatocellular carcinoma,” Journal of Can- – [18] J. Wang, D. Duncan, Z. Shi, and B. Zhang, “WEB-based GEne cer, vol. 11, no. 1, pp. 83 91, 2020. SeT AnaLysis Toolkit (WebGestalt): update 2013,” Nucleic [34] J. Beyerle, A. N. Holowatyj, M. Haffa et al., “Expression Pat- Acids Research, vol. 41, no. W1, pp. W77–W83, 2013. terns of Xenobiotic-Metabolizing Enzymes in Tumor and [19] M. Kanehisa, M. Furumichi, M. Tanabe, Y. Sato, and Adjacent Normal Mucosa Tissues among Patients with Colo- ” K. Morishima, “KEGG: new perspectives on genomes, path- rectal Cancer: The ColoCare Study, Cancer Epidemiology Bio- – ways, diseases and drugs,” Nucleic Acids Research, vol. 45, markers & Prevention, vol. 29, no. 2, pp. 460 469, 2020. no. D1, pp. D353–D361, 2017. [35] H. Zhou, S. Di Palma, C. Preisinger et al., “Toward a Compre- [20] B. Jassal, L. Matthews, G. Viteri et al., “The reactome pathway hensive Characterization of a Human Cancer Cell Phospho- ” knowledgebase,” Nucleic Acids Research, vol. 48, no. D1, proteome, Journal of Proteome Research, vol. 12, no. 1, – pp. D498–D503, 2020. pp. 260 271, 2012. “ [21] H. Mi, B. Lazareva-Ulitsky, R. Loo et al., “The PANTHER [36] Z. Chen, C. Shi, S. Gao, D. Song, and Y. Feng, Impact of prot- database of protein families, subfamilies, functions and path- amine I on colon cancer proliferation, invasion, migration, ” ways,” Nucleic Acids Research, vol. 33, no. Database issue, diagnosis and prognosis, Biological Chemistry, vol. 399, – pp. D284–D288, 2005. no. 3, pp. 265 275, 2018. [22] D. Szklarczyk, J. H. Morris, H. Cook et al., “The STRING data- [37] W. K. Glowacka, P. Alberts, R. Ouchida, J. Y. Wang, and “ base in 2017: quality-controlled protein-protein association D. Rotin, LAPTM5 protein is a positive regulator of proin- fl ” networks, made broadly accessible,” Nucleic Acids Research, ammatory signaling pathways in macrophages, The Journal – vol. 45, no. D1, pp. D362–D368, 2017. of Biological Chemistry, vol. 287, no. 33, pp. 27691 27702, 2012. [23] Z. Tang, C. Li, B. Kang, G. Gao, C. Li, and Z. Zhang, “GEPIA: a “ web server for cancer and normal gene expression profiling [38] L. Chen, G. Wang, Y. Luo et al., Downregulation of LAPTM5 and interactive analyses,” Nucleic Acids Research, vol. 45, suppresses cell proliferation and viability inducing cell cycle ” no. W1, pp. W98–W102, 2017. arrest at G0/G1 phase of bladder cancer cells, International Journal of Oncology, vol. 50, no. 1, pp. 263–271, 2017. [24] J. Gao, B. A. Aksoy, U. Dogrusoz et al., “Integrative Analysis of “ Complex Cancer Genomics and Clinical Profiles Using the [39] R. Cortese, O. Hartmann, K. Berlin, and F. Eckhardt, Correl- fi cBioPortal,” Science Signaling, vol. 6, no. 269, p. pl1, 2013. ative gene expression and DNA methylation pro ling in lung development nominate new biomarkers in lung cancer,” The [25] T. Li, J. Fan, B. Wang et al., “TIMER: a web server for compre- International Journal of Biochemistry & Cell Biology, vol. 40, hensive analysis of tumor-infiltrating immune cells,” Cancer no. 8, pp. 1494–1508, 2008. Research, vol. 77, no. 21, pp. e108–e110, 2017. [26] B. A. Siddiqui, M. Zhang, L. L. Pisters, and S. M. Tu, “Systemic therapy for primary and extragonadal germ cell tumors: prognosis and nuances of treatment,” Transl Androl Urol, vol. 9, Suppl 1, pp. S56–S65, 2020. [27] S. K. Cheriyan, M. Nicholson, A. M. Aydin et al., “Current management and management controversies in early- and intermediate-stage of nonseminoma germ cell tumors,” Transl Androl Urol, vol. 9, Suppl 1, pp. S45–S55, 2020. [28] R. P. Werntz and S. E. Eggener, “Defining risk of micrometa- static disease and tumor recurrence in patients with stage I tes-