ORIGINAL ARTICLE
Ahead of print publication  

Characterization of progression-related alternative splicing events in testicular germ cell tumors


1 Department of Urinary Surgery, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai 200025, China
2 Department of Medical Oncology, Gaozhou People's Hospital, Gaozhou 525200, China
3 First Clinical Medical College of Nanjing Medical University, Nanjing 210000, China

Date of Submission15-Aug-2019
Date of Acceptance26-Apr-2020
Date of Web Publication02-Oct-2020

Correspondence Address:
Dan-Feng Xu,
Department of Urinary Surgery, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai 200025, China

Lu Chen,
Department of Urinary Surgery, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai 200025, China

Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/aja.aja_30_20

PMID: 33037172

  Abstract 


Accumulating evidence supports the significance of aberrant alternative splicing (AS) events in cancer; however, genome-wide profiling of progression-free survival (PFS)-related AS events in testicular germ cell tumors (TGCT) has not been reported. Here, we analyzed high-throughput RNA-sequencing data and percent-spliced-in values for 150 patients with TGCT. Using univariate and multivariate Cox regression analysis and a least absolute shrinkage and selection operator method, we identified the top 15 AS events most closely associated with disease progression. A risk-associated AS score (ASS) for the 15 AS events was calculated for each patient. ASS, pathological stage, and T stage were significantly associated with disease progression by univariate analysis, but only ASS and pathological stage remained significant by multivariate analysis. The ability of these variables to predict 5-year progression was assessed using receiver operating characteristic curve analysis. ASS had stronger predictive value than a combination of age, pathological stage, and T stage (area under the curve = 0.899 and 0.715, respectively). Furthermore, Kaplan–Meier analysis of patients with low and high ASS demonstrated that high ASS was associated with significantly worse PFS than low ASS (P = 1.46 × 10−7). We also analyzed the biological functions of the PFS-related AS-related genes and found enrichment in pathways associated with DNA repair and modification. Finally, we identified a regulatory network of splicing factors with expression levels that correlated significantly with AS events in TGCT. Collectively, this study identifies a novel method for risk stratification of patients and provides insight into the molecular events underlying TGCT.

Keywords: alternative splicing events; network; progression-free survival-related model; testicular germ cell tumor


Article in PDF

How to cite this URL:
Zhang CJ, Li ZT, Shen KJ, Chen L, Xu DF, Gao Y. Characterization of progression-related alternative splicing events in testicular germ cell tumors. Asian J Androl [Epub ahead of print] [cited 2020 Oct 29]. Available from: https://www.ajandrology.com/preprintarticle.asp?id=297175

Chuan-Jie Zhang, Zong-Tai Li
These authors contributed equally to this work.



  Introduction Top


Testicular cancer (TC), in which malignant cells form in the tissues of one or both testicles, has an annual incidence of approximately 1% among all newly diagnosed cancers in males.[1] The most common form of TC is testicular germ cell tumors (TGCT), accounting for >95% of cases, which consist of seminoma and nonseminoma subtypes.[2] The overall mortality rate of TGCT remains high due to its propensity to recur and form metastases. Over the past few decades, the main treatment strategy for decreasing the risk of relapse of TGCT has been retroperitoneal lymph node dissection;[3] however, in recent years an advanced multidisciplinary approach combining surgical intervention with adjuvant chemotherapy or radiotherapy has improved the prognosis of patients with TC significantly, resulting in a 5-year survival rate of >95%.[4],[5] Nevertheless, the overall incidence of TC is still increasing worldwide.

Patients with a history of TC have a 2% increased risk of advanced tumor in the contralateral testis within 15 years of diagnosis.[6] Other studies have identified a number of risk factors, including environmental, hormonal, and genetic factors, that contribute alone or in combination to the development or recurrence of TC.[7],[8] However, there is a pressing need to understand the underlying molecular basis of TGCT, not only to elucidate the potential carcinogenic mechanisms but also to identify effective biomarkers for monitoring the risk for TGCT or its progression and prognosis. In recent years, increasing attention has been paid to the influence of aberrant epigenetic regulation, which integrates both environmental and genetic factors, on the risk of cancer development and progression.

More than 95% of human genes undergo alternative splicing (AS) as a normal physiological process to generate protein diversity.[9] As a vital post-transcriptional regulatory process, AS has the potential to generate mRNA isoforms that could play a potential pathogenic role in many diseases, including cancer.[10] Indeed, emerging evidence suggests a close relationship between dysregulation of AS and tumor progression and recurrence, treatment resistance, and other oncogenic mechanisms.[11] Recently, Xing et al.[12] identified an oncogenic role for the exonuclease DIS3L2, which contributed to the progression of liver cancer via regulation of heterogeneous nuclear ribonucleoprotein U (hnRNPU)-mediated AS. Moreover, the development of advanced high-throughput sequencing technology has allowed AS events to be profiled and successfully identified as PFS-related markers in many malignancies, including lung, ovarian, hepatocellular, and kidney cancer.[13],[14],[15],[16] Nevertheless, no genome-wide screening of progression-related AS events have been performed in TGCT, and little is known about the activity of potential TGCT-related splice variants.

In this study, we reported the first genome-wide profiling of AS events in TGCT. Using a 150-patient dataset from The Cancer Genome Atlas (TCGA) and the TCGA SpliceSeq dataset, we systematically analyzed the association between TGCT-specific AS events and disease progression and survival outcomes. We identified a robust AS score (ASS) based on the top 15 significant progression-free survival (PFS)-related AS events, and demonstrated its ability to predict the risk of 5-year progression. We also performed gene enrichment analysis and established a TGCT-specific regulatory network of AS events and the associated splicing factors (SF). Thus, this study sheds light on the molecular events underlying TGCT and identifies several PFS-related AS events that play potentially important roles in the progression of TGCT.


  Materials and Methods Top


Data acquisition and preprocessing

We obtained a dataset for 150 TGCT patients with corresponding clinical data and transcriptome profiles from TCGA (https://portal.gdc.cancer.gov/). Since the whole sequencing data of patients was obtained from the TCGA dataset, and it was unnecessary to provide the relevant ethics profiles. The expression data were quantified and normalized using the edgeR package. The TCGA SpliceSeq tool, run on a Java platform, was used to provide a comprehensive view of AS profiles in the TGCT patient cohort.[17] The percent-spliced-in (PSI) value were derived to quantify AS events in each patient sample and was calculated as: PSI = splice_in/(splice_in + splice_out), with a value range of 0 to 1. We defined the filter cutoff that the percentage of samples with PSI should be more than 75% and correctly calculated the seven types of AS events, including alternate acceptor site (AA), alternate donor site (AD), alternate promoter (AP), alternate terminator (AT), exon skip (ES), mutually exclusive exons (ME), and retained intron (RI). Each AS event was annotated using the gene symbol, the AS_id number in the SpliceSeq database, and the splicing category. The clinical data for 134 patients were extracted using Perl scripts and consisted of age, gender, pathological stage, and American Joint Committee on Cancer Tumor-Node-Metastasis stages.

Identification of PFS-related AS events and functional gene enrichment analysis

The PSI data from TGCT cohort were transformed into a single matrix and merged with the survival data. To comprehensively illustrate interactions between the seven AS types, we generated UpSet plots with UpSetR package (https://github.com/hms-dbmi/UpSetR) to display five or more interactive sets.[18] The impute R package was then used to interpolate the missing values using the K-nearest neighbor (KNN) algorithm.[19] We then designed a “for cycle” script based on the R survival package to conduct univariate Cox analysis for each AS event, with statistical significance defined as P < 0.01. Total PFS-related AS events were displayed in UpSet plots and Volcano plots created using the ggplot2 package (https://ggplot2.tidyverse.org/). A pie chart of AS event frequency for each AS category was constructed. The top 20 individual AS events in each category were presented as dot plots. The biological functions of the parent genes with progression-associated AS events were investigated using gene ontology (GO) analysis with the terms biological process, cellular component, and molecular function. Genes with a false discovery rate of <0.05 were considered to be significantly enriched. The network of enriched terms was constructed using the Metascape tool (https://metascape.org).[20]

Construction of progression-related AS signature for TGCT patients

The least absolute shrinkage and selection operator (LASSO) method was used to further screen the significant progression-related AS events using the glmnet and survival packages (https://www.rdocumentation.org/packages/glmnet/versions/3.0-2). The overall ASS was calculated using a multivariate Cox regression method: ASS = ∑(SPIi ×βi), where βi represents the coefficient of each of the 15 AS events. Receiver operating characteristic (ROC) curves were generated and the area under the curve (AUC) was calculated to assess the predictive value of the ASS for 5-year tumor progression. The ASS was calculated for each patient, and the median score was used to dichotomize the 150-patient cohort into high and low ASS groups. The PFS survival of the two groups was compared using Kaplan–Meier analysis with a log-rank test. The predictive significance of ASS was also compared with that of other clinical variables; namely, age, pathological stage, and T stage, using univariate and multivariate Cox regression analysis, and the predictive utility of a combination of age, pathological stage, and T stage was evaluated by ROC curve analysis. Because 75 and 15 patients were missing information on N and M stages, respectively, these variables were excluded from the analyses. Hazard ratios (HRs) with 95% confidence intervals were calculated.

Generation of a potential splicing factors-alternative splicing (SF-AS) regulatory network

We downloaded the list of 404 human SF genes from the SpliceAid2 database (http://www.introni.it/splicing.html)[21] and extracted the expression data for the SF genes from the 150-patient TGCT cohort dataset. We then evaluated Spearman's correlation coefficient for the expression of SF genes and the 15 PFS-related AS events. A significant correlation was defined as: Correlation >0.3 with adjusted P < 0.05, where Correlation >0.3 was considered to be positive regulation. The potential SF-AS regulatory network was constructed to illustrate significantly correlated SF-AS pairs using Cytoscape version 3.71 software (https://cytoscape.org/).

Statistical analyses

The Wilcoxon rank-sum test was used to compare ranked data with two categories. LASSO and Cox regression modeling were performed using glmnet and survival packages. Spearman's correlation analysis was conducted to estimate the correlation between AS events SF gene expression. All statistical analyses were conducted in RStudio version 3.6.1 (https://rstudio.com/), and P < 0.05 was considered statistically significant.


  Results Top


Genome-wide profiling of AS events in TGCT patients

The seven classes of AS are shown in [Figure 1]a. We obtained a total of 422 415 splicing events in 10 332 genes from 149 TGCT patients. The most frequent AS event was ES (15 879 events in 6443 genes), followed by AT (8721 events in 3813 genes), AP (8431 events in 3389 genes), AA (3441 events in 2441 genes), AD (2992 events in 2098 genes), RI (2790 events in 1858 genes), and ME (179 events in 176 genes), as shown in [Figure 1]b. All seven AS types occurred in some genes, highlighting the contribution of AS to transcriptome diversity.
Figure 1: Illustration of AS types and identification of all AS events in TGCT. (a) The seven types of AS events are ES, AP, AT, AD, AA, ME, and RI. (b) UpSet graph showing gene intersections for the seven types of AS events (n = 422 415). Red lines indicate multiple AS events occurring in a single gene. AS: alternative splicing; TGCT: testicular germ cell tumors; ES: exon skip; AP: alternate promoter; AT: alternate terminator; AD: alternate donor site; AA: alternate acceptor site; ME: mutually exclusive exon; RI: retained intron.

Click here to view


Of the 150-patient cohort, complete clinical information was available for 134 patients. As shown in [Supplementary Table 1 [Additional file 1]], the majority of patients (81.3%) were in the 20- to 40-year age group, and the proportion of patients with pathological stages in situ, I, II, and III was 34.3%, 41.0%, 9.0%, and 10.5%, respectively.

Profiling of progression-related AS events and functional analysis of parent genes in TGCT patients

To determine the relationship between AS events and patient survival, the AS matrix and survival data were merged together. A total of 300 PFS-related AS events with P < 0.01 were screened by univariate Cox regression analysis. Among them, 229 AS events were adverse PFS-related events (HR <1, P < 0.01) and 72 were considered risk factors (HR <1, P < 0.01). Several genes, including serine/threonine-protein phosphatase 2A regulatory subunit 4 (PPP2R4), katanin catalytic subunit A1 like 2(KATNAL2), F-box protein 7 (FBXO7), and chromodomain helicase DNA-binding protein 6 (CHD6), were found to be processed by multiple progression-related AS events. Accordingly, we generated UpSet plots to demonstrate the subset of interacting AS events in the TGCT cohort. Not surprisingly, ES related events were the most frequent [Figure 2]a. The distribution of the significance and the number of progression-related AS events are shown in the Volcano plot and pie charts, respectively, in [Figure 2]b and [Figure 2]c. The top 20 most significant PFS-related events for each AS category are shown in [Figure 3]; notably, none of the top 300 PFS-related events was ME type.
Figure 2: Profiling of progression-related AS events in TGCT. (a) UpSet plot showing gene intersections for the seven types of progression-related AS events in TGCT. (b) Volcano plot showing the distribution of significant and nonsignificant progression-related AS events (P < 0.001). (c) Pie chart showing the number of progression-related AS events in each category. ES: exon skip; AP: alternate promoter; AT: alternate terminator; AD: alternate donor site; AA: alternate acceptor site; ME: mutually exclusive exon; RI: retained intron; AS: alternative splicing; PFS: progression-free survival; TGCT: testicular germ cell tumors.

Click here to view
Figure 3: Subgroup analysis of progression-associated AS events in TGCT. (a) The top 20 progression-related AS events for AA in TGCT. (b) The top 20 progression-related AS events for AD in TGCT. (c) The top 20 progression-related AS events for AP in TGCT. (d) The top 20 progression-related AS events for AT in TGCT. (e) The top 20 progression-related AS events for ES in TGCT. (f) The top 20 progression-related AS events for RI in TGCT. There were no significant PFS-related ME events among the top 300 events. The color of each circle indicates that the P value and the Z-score value are strongly correlated in PFS.. AS: alternative splicing; PFS: progression-free survival; TGCT: testicular germ cell tumors; ME: mutually exclusive exon; ES: exon skip; AP: alternate promoter; AT: alternate terminator; AD: alternate donor site; AA: alternate acceptor site; RI: retained intron.

Click here to view


To understand the potential biological functions of the PFS-related AS events, we performed gene enrichment and GO analysis for the 268 parent genes associated with PFS-related AS events [Supplementary Table 2 [Additional file 2]]. A total of 20 cellular component or molecular function GO terms were significantly enriched among the genes, including DNA repair, microtubule-based process, regulation of GTPase activity, and DNA modification [Figure 4]a. Based on what we found, a network diagram was constructed, showing the crosstalk between the significantly enriched GO terms [Figure 4]b.
Figure 4: Functional analysis and network construction for parent genes associated with PFS-related AS events. (a) Enrichment analysis for the genes processed by the top significant progression-related AS events. (b) Functional nodes in the corresponding gene network of the most significant progression-related AS events. AS: alternative splicing; PFS: progression-free survival.

Click here to view


Derivation and assessment of a predictive ASS for TGCT patients

Having identified 300 PFS-related AS events, we used the LASSO regression method to further select a 15-event signature for disease progression in the TGCT patient cohort [Supplementary Table 3 [Additional file 3]] and [Supplementary Table 4 [Additional file 4]]. We calculated the ASS for each patient from the results of the multivariate Cox regression analysis and used the median ASS as a cutoff to dichotomize the patients into high and low ASS groups. ROC curve analysis of the predictive value of ASS for 5-year progression gave an AUC of 0.899, indicating high predictive accuracy [Figure 5]c. Indeed, patients with high ASS values had a significantly higher risk than the low ASS group of poor PFS according to Kaplan–Meier analysis (P = 1.462 × 10−7, log-rank test; [Figure 5]d and [Supplementary Figure 1 [Additional file 5]]. Univariate Cox regression analysis revealed that ASS, pathological stage, and T stage, but not age, were significant predictors of poor PFS, but only ASS and pathological stage remained significant in multivariate analysis [Supplementary Figure 2 [Additional file 6]]a. Finally, we compared the predictive value of ASS with that of the other significant clinicopathological variables (age, pathological stage, and T stage) by ROC analysis. The combination of the three variables gave an AUC of 0.715, which indicates that these traditional clinical variables have poorer predictive power than ASS, which had an AUC of 0.899 [Supplementary Figure 2]b and [Figure 5]c, respectively).
Figure 5: Identification of the 15-AS-event score for predicting TGCT progression. (a) The LASSO regression model was conducted to screen the pivotal hazard AS events and we illustrated the convergence curve, in which Log(Lambda) represented the horizontal axis, and coefficients represented the vertical axis. (b) Accordingly, the LASSO method selected 15 events from 300 potentially PFS-related AS events. (c) Receiver operating characteristic curve of the ability of the ASS to predict 5-year progression. (d) Kaplan–Meier analysis of PFS-related survival of TGCT patients according to the ASS. High and low ASS scores represent more than or less than the median ASS, respectively. P = 1.462 × 10−7 by log-rank test. AS: alternative splicing; ASS: AS score; PFS: progression-free survival; TGCT: testicular germ cell tumors; ROC: receiver operating characteristic; AUC: area under the curve; LASSO: least absolute shrinkage and selection operator.

Click here to view


Construction of a regulatory network of SFs and progression-related AS events

To determine whether expression of SFs correlated with specific PFS-related AS events in TGCT, we examined the transcriptome profiles of 404 splicing factors in the TGCT cohort dataset. Using a cutoff for significance with Spearman's correlation analysis of | Correlation | >0.3 and adjusted P < 0.05, we identified a total of 149 SFs that were significantly associated with PFS-related AS events in TGCT. The network of 431 potential SF-AF regulatory pairs generated from these correlations is shown in [Supplementary Figure 2]c. Of the 431 pairs, 77 and 354 represented positive and negative regulatory processes, respectively. The selected AS events were also annotated with two colors, in which the pink in ellipses represented the risk AS events with HR >1 while the blue color represented the adverse correlation with PFS endpoints.


  Discussion Top


Although TGCT can be cured with combination therapy consisting of surgery, chemotherapy/radiotherapy, tumor progression and recurrence remain a concern.[22] Currently, risk stratification of TGCT patients is mainly based on tumor size, pathological subtype, and serum biomarkers such as α-fetoprotein and lactate dehydrogenase.[23],[24],[25] However, these factors are of limited use for predicting progression of TGCT, highlighting the need for more accurate PFS-related biomarkers. Previous studies have indicated that AS plays a crucial role in the biology of tumors, which prompted us to focus on the potential PFS-related value of aberrant AS events in TGCT.[26]

Several studies have identified AS-related factors or other gene signatures, consisting of SF1, the histone variant macroH2A.1 histone (MacroH2A1), and RNA-binding protein of fox homologs(RBFOX)family genes, related to a progression in TGCT. However, these data were mostly derived from a limited number of tumor samples and used exon microarray analyses; in contrast, there have been no comprehensive or systematic assessments of the AS landscape in TGCT.[27],[28],[29]

In the present study, we examined high-throughput RNA-seq data from 150 TGCT patients. We identified a total of 422 415 AS events in 10 332 genes encompassing all seven types of AS events. Of these, 300 events significantly associated with progression were identified, and we further analyzed the top 20 events in each AS type. Among the identified genes, several are known cancer drivers, such as tripartite motif containing 6 (TRIM6), TIR domain-containing adaptor protein(TIRAP), StAR related lipid transfer domain-containing 10(STARD10), and zinc finger protein 175(ZNF175).[30],[31],[32] One PFS-related AS event identified in our cohort was POLD1-51194-AA. Interestingly, Bonache et al.[33] demonstrated that downregulation of POLD1 expression was involved in the modulation of the cell cycle and post-transcriptional modifications in testicular samples. Hirvonen-Santti et al.[34] also showed that downregulation ofestrogen receptor beta(ER-β) and SNRPN upstream reading frame/ring finger protein 4 (SNURF/RNF4) complexes might play a role in testicular tumorigenesis; notably, RNF4-68572-ES was found to be significantly associated with PFS in our cohort. We also found that many of the AS events significantly associated with TGCT progression occurred in E3 ubiquitin ligase genes, including X-linked inhibitor of apoptosis(XIAP), ring finger protein 170(RNF170), Wolf-Hirschhorn syndrome candidate 1 (WHSC1), mouse double minute 2 homolog(MDM2), and pleckstrin homology domain-containing A5 (PLEKHA5). This finding suggests the possible involvement of aberrant protein ubiquitination in the development and/or progression of TGCT. Our functional analysis also uncovered enrichment of AS-related genes associated with DNA repair, microtubule-based process, regulation of GTPase activity, and DNA modification. A recent study by AlDubayan et al.[35] highlighted a deficiency in DNA repair processes as a prominent mechanism driving susceptibility to TGCT, which provided new insight into potential management strategies for individuals at high risk for TGCT progression. We speculate that dysfunction of some of the DNA repair-related splice variants identified here may correlate with tumor progression. However, only a few of the PFS-related AS events here involved DNA repair genes, and further studies must be performed to validate the robustness of our results.

Using the LASSO method, we obtained an ASS signature based on 15 key AS events (PEX1-80440-ES, RDX-18638-ES, NPLOC4-44135-ES, MBD1-45510-AA, CACNA2D2-65058-AA, ZNF669-10512-AD, AKAP2-87182-ES, TCEB1-84211-AD, SELENBP1-7618-ES, STARD10-17644-AP, RPL34-70298-AT, PPP4R1L-59958-ES, TGM2-59374-ES, SEC16A-88181-AA, and LIMK2-61838-ES). The ASS had good predictive value for 5-year progression and accurately stratified TGCT patients into high- and low-risk groups. Importantly, higher ASS values correlated significantly with poorer outcomes in this cohort. A comparison of the predictive value of ASS and a combination of traditional variables (age, pathological stage, and T stage) revealed that ASS had superior predictive power. These results suggest that the ASS identified here might have utility as a potential biomarker for predicting TGCT progression. Finally, we examined the correlation between AS events and the expression of SFs in TGCT, and we found that the majority of worse PFS-related AS events were associated with the expression levels of splicing factors positively, yet favorable prognosis AS events were in the opposite manner. This comprehensive AS-SF network could provide a better understanding of splicing patterns and their relationships with SFs in TGCT.

One of the strengths of our study is the genome-wide identification of PFS-related AS events in TGCT and the successful demonstration of a strong predictive model, for risk of disease progression. However, several limitations also exist. First, the number of samples was relatively small, perhaps reflecting the relative rarity of this disease. Additional samples will need to be analyzed to provide external validation of the ASS model. Second, the 15 significant AS events identified here will need to be further analyzed to improve our understanding of the underlying mechanisms in tumor progression. Finally, whether integrating ASS and other clinical variables into a combined model could further improve the predictive accuracy remains unclear, but it would certainly have potential translational significance.


  Conclusion Top


The results of this genome-wide profiling of AS events and association with SFs in TGCT add to our growing understanding of how aberrant AS affects cancer development and progression. Our results also provide new insights into the underlying mechanisms of TGCT progression and may help in the development of improved predictive biomarkers and therapeutic strategies for TGCT.


  Author Contributions Top


CJZ and ZTL analyzed the data and drafted the manuscript. KJS and YG helped analyze the data. CJZ, ZTL, and KJS prepared all figures. ZTL, KJS, and YG edited all tables. LC and DFX conceived the idea and designed the study. All authors have read and approved the final manuscript and agreed with the order of presentation of the authors.


  Competing Interests Top


All authors declared no competing interests.


  Acknowledgments Top


This work was supported by grants from the Youth Program of the National Natural Science Foundation of China (No. 81602238) and the Guangci Youth excellence program of Ruijin Hospital.

Supplementary Information is linked to the online version of the paper on the Asian Journal of Andrology website.



 
  References Top

1.
Miller KD, Nogueira L, Mariotto AB, Rowland JH, Yabroff KR, et al. Cancer treatment and survivorship statistics, 2019. CA Cancer J Clin 2019; 69: 363–85.  Back to cited text no. 1
    
2.
Litchfield K, Levy M, Orlando G, Loveday C, Law P, et al. Identification of 19 new risk loci and potential regulatory mechanisms influencing susceptibility to testicular germ cell tumor. Nat Genet 2017; 49: 1133–40.  Back to cited text no. 2
    
3.
Hamilton R, Nayan M, Anson-Cartwright L, Atenafu E, Bedard P, et al. Treatment of relapse of clinical stage I nonseminomatous germ cell tumors on surveillance. J Clin Oncol 2019; 37: 1919–26.  Back to cited text no. 3
    
4.
Rajpert-De Meyts E, McGlynn K, Okamoto K, Jewett M, et al. Testicular germ cell tumours. Lancet 2016; 387: 1762–74.  Back to cited text no. 4
    
5.
Boissier R, Hevia V, Bruins H, Budde K, Figueiredo A, et al. The risk of tumour recurrence in patients undergoing renal transplantation for end-stage renal disease after previous treatment for a urological cancer: a systematic review. Eur Urol 2018; 73: 94–108.  Back to cited text no. 5
    
6.
Mannisto S, Vähämurto P, Pollari M, Clausen M, Jyrkkiö S, et al. Intravenous but not intrathecal central nervous system-directed chemotherapy improves survival in patients with testicular diffuse large B-cell lymphoma. Eur J Cancer 2019; 115: 27–36.  Back to cited text no. 6
    
7.
Kerns S, Fung C, Monahan P, Ardeshir-Rouhani-Fard S, Abu Zaid M, et al. Cumulative burden of morbidity among testicular cancer survivors after standard cisplatin-based chemotherapy: a multi-institutional study. J Clin Oncol 2018; 36: 1505–12.  Back to cited text no. 7
    
8.
Killian J, Dorssers L, Trabert B, Gillis A, Cook M, et al. Imprints and DPPA3 are bypassed during pluripotency- and differentiation-coupled methylation reprogramming in testicular germ cell tumors. Genome Res 2016; 26: 1490–504.  Back to cited text no. 8
    
9.
Slansky J, Spellman P. Alternative splicing in tumors – a path to immunogenicity? N Engl J Med 2019; 380: 877–80.  Back to cited text no. 9
    
10.
Baralle F, Giudice J. Alternative splicing as a regulator of development and tissue identity. Nat Rev Mol Cell Biol 2017; 18: 437–51.  Back to cited text no. 10
    
11.
Kahles A, Lehmann K, Toussaint N, Hüser M, Stark S, et al. Comprehensive analysis of alternative splicing across tumors from 8,705 patients. Cancer Cell 2018; 34: 211–24.e6.  Back to cited text no. 11
    
12.
Xing S, Li Z, Ma W, He X, Shen S, et al. DIS3L2 promotes progression of hepatocellular carcinoma via hnRNP U-mediated alternative splicing. Cancer Res 2019; 79: 4923–36.  Back to cited text no. 12
    
13.
Li Y, Sun N, Lu Z, Sun S, Huang J, et al. PFS-related alternative mRNA splicing signature in non-small cell lung cancer. Cancer Lett 2017; 393: 40–51.  Back to cited text no. 13
    
14.
Zhu J, Chen Z, Yong L. Systematic profiling of alternative splicing signature reveals PFS-related predictor for ovarian cancer. Gynecol Oncol 2018; 148: 368–74.  Back to cited text no. 14
    
15.
Zhu GQ, Zhou YJ, Qiu LX, Wang B, Yang Y, et al. Prognostic alternative mRNA splicing signature in hepatocellular carcinoma: a study based on large-scale sequencing data. Carcinogenesis 2019; 40: 1077–85.  Back to cited text no. 15
    
16.
Song J, Liu YD, Su J, Yuan D, Sun F, et al. Systematic analysis of alternative splicing signature unveils prognostic predictor for kidney renal clear cell carcinoma. J Cell Physiol 2019; 234: 22753–64.  Back to cited text no. 16
    
17.
Ryan M, Cleland J, Kim R, Wong W, Weinstein J. SpliceSeq: a resource for analysis and visualization of RNA-Seq data on alternative splicing and its functional impacts. Bioinformatics 2012; 28: 2385–7.  Back to cited text no. 17
    
18.
Conway J, Lex A, Gehlenborg N. UpSetR: an R package for the visualization of intersecting sets and their properties. Bioinformatics 2017; 33: 2938–40.  Back to cited text no. 18
    
19.
Julienne H, Shi H, Pasaniuc B, Aschard H. RAISS: robust and accurate imputation from summary statistics. Bioinformatics 2019; 35: 4837–9.  Back to cited text no. 19
    
20.
Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi A, et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun 2019; 10: 1523.  Back to cited text no. 20
    
21.
Piva F, Giulietti M, Burini A, Principato G. SpliceAid 2: a database of human splicing factors expression data and RNA target motifs. Hum Mutat 2012; 33: 81–5.  Back to cited text no. 21
    
22.
Chovanec M, Hanna N, Cary K, Einhorn L, Albany C. Management of stage I testicular germ cell tumours. Nat Rev Urol 2016; 13: 663–73.  Back to cited text no. 22
    
23.
Pierpont T, Lyndaker A, Anderson C, Jin Q, Moore E, et al. Chemotherapy-induced depletion of OCT4-positive cancer stem cells in a mouse model of malignant testicular cancer. Cell Rep 2017; 21: 1896–909.  Back to cited text no. 23
    
24.
Pagliaro L. Role of high-dose chemotherapy with autologous stem-cell rescue in men with previously treated germ cell tumors. J Clin Oncol 2017; 35: 1036–40.  Back to cited text no. 24
    
25.
Terbuch A, Adiprasito J, Stiegelbauer V, Seles M, Klec C, et al. MiR-371a-3p serum levels are increased in recurrence of testicular germ cell tumor patients. Int J Mol Sci 2018; 19: 3130.  Back to cited text no. 25
    
26.
Wright P, Dyson H. Intrinsically disordered proteins in cellular signalling and regulation. Nat Rev Mol Cell Biol 2015; 16: 18–29.  Back to cited text no. 26
    
27.
Novikov L, Park J, Chen H, Klerman H, Jalloh A, et al. QKI-mediated alternative splicing of the histone variant MacroH2A1 regulates cancer cell proliferation. Mol Cell Biol 2011; 31: 4244–55.  Back to cited text no. 27
    
28.
Zhu R, Heaney J, Nadeau J, Ali S, Matin A. Deficiency of splicing factor 1 suppresses the occurrence of testicular germ cell tumors. Cancer Res 2010; 70: 7264–72.  Back to cited text no. 28
    
29.
Wang Y, Gray D, Robbins A, Crowgey E, Chanock S, et al. Subphenotype meta-analysis of testicular cancer genome-wide association study data suggests a role for RBFOX family genes in cryptorchidism susceptibility. Hum Reprod 2018; 33: 967–77.  Back to cited text no. 29
    
30.
Larabi A, Barnich N, Nguyen HT. New insights into the interplay between autophagy, gut microbiota and inflammatory responses in IBD. Autophagy 2020; 16: 38–51.  Back to cited text no. 30
    
31.
Rajsbaum R, Versteeg G, Schmid S, Maestre A, Belicha-Villanueva A, et al. Unanchored K48-linked polyubiquitin synthesized by the E3-ubiquitin ligase TRIM6 stimulates the interferon-IKKε kinase-mediated antiviral response. Immunity 2014; 40: 880–95.  Back to cited text no. 31
    
32.
Carrat G, Hu M, Nguyen-Tu M, Chabosseau P, Gaulton K, et al. Decreased STARD10 expression is associated with defective insulin secretion in humans and mice. Am J Hum Genet 2017; 100: 238–56.  Back to cited text no. 32
    
33.
Bonache S, Algaba F, Franco E, Bassas L, Larriba S. Altered gene expression signature of early stages of the germ line supports the pre-meiotic origin of human spermatogenic failure. Andrology 2014; 2: 596–606.  Back to cited text no. 33
    
34.
Hirvonen-Santti SJ, Sriraman V, Anttonen M, Savolainen S, Palvimo JJ, et al. Small nuclear RING finger protein expression during gonad development: regulation by gonadotropins and estrogen in the postnatal ovary. Endocrinology 2004; 145: 2433–44.  Back to cited text no. 34
    
35.
AlDubayan SH, Pyle LC, Gamulin M, Kulis T, Moore ND, et al. Association of inherited pathogenic variants in checkpoint kinase 2 (CHEK2) with susceptibility to testicular germ cell tumors. JAMA Oncol 2019; 5: 514–22.  Back to cited text no. 35
    


    Figures

  [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]



 

 
Top
 
 
  Search
 
 Search Pubmed for
 
    -  Zhang CJ
    -  Li ZT
    -  Shen KJ
    -  Chen L
    -  Xu DF
    -  Gao Y
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)  

 
  In this article
Abstract
Introduction
Materials and Me...
Results
Discussion
Conclusion
Author Contributions
Competing Interests
Acknowledgments
References
Article Figures

 Article Access Statistics
    Viewed263    
    PDF Downloaded14    

Recommend this journal