Gene abnormalities can lead to cell cycle disorders, uncontrolled proliferation, and thus tumors. The genomic defects of bladder cancer are complicated, ranging from single DNA mutations to gene polymorphisms to whole or partial deletions of chromosomes.

The study has found that all human malignancies have somatic gene mutations (9). Activation of oncogenes and suppression of tumor suppressor genes are very common. Abnormal activation of oncogenes such as HER2 can facilitate tumor metastasis. Inactivation of p53 and p16 and other tumor suppressor genes involved in cell cycle regulation can inhibit apoptosis of tumor cells. Common mutations in bladder cancer include TP53, PIK3CA, TSC1, FGFR3, HRAS, and HER2. Moreover, common abnormal expression genes include EGFR, Ki67, PD-L1, ERCC1, and BRCA1. Shariat et al. found that the proportion of p53 and p16 abnormally expressed in bladder cancer patients were 56% and 54%, respectively. It was highly correlated with muscle-invasive bladder cancer which often indicates a poor prognosis (5). Soria et al. studied the expression of HER2 in 354 bladder cancer patients and found that HER2 is highly expressed in more than 32% of patients. Higher expression of HER2 indicates more aggressive tumor invasiveness (10). Hayashi et al. inhibited highly expressed HER2 in bladder cancer cell lines and animal models with HER2 inhibitor T-DM1 and found that T-DM1 has a more prominent inhibitory effect on bladder cancer cell growth than the trastuzumab. In addition, platinum-resistant bladder cancer cells with have higher levels of HER2 expression. Interestingly, this group of cells is more sensitive to T-DM1, and cells treated with T-DM1 are more prone to apoptosis (11). The above studies indicate that the oncogene HER2 plays an important role in the pathogenesis of bladder cancer. Inhibition of HER2 expression and induction of apoptosis can significantly suppress the bladder cancer cell growth. HER2 mutations are also present in bladder cancer. Tschui et al. studied both exon 19 and exon 20 of HER2 gene in bladder cancer patients and found that mutations occur in exon 19 (12).

The cancer genome atlas (TCGA) analyzed DNA data from 131 patients with muscle-invasive bladder cancer and identified 32 mutant genes with high frequency. These mutations are mainly occurred in the cell cycle, chromatin regulation and kinase signaling pathways (13). Another comprehensive analysis of the full TCGA cohort of 412 MIBC cases was conducted in 2017. 58 significantly mutated genes (SMGs) were identified in the expanded cohort, in which 34 mutant genes were not identified in previous analysis and 7 mutant genes were identified in more than 10% of samples: KMT2A (11%), SPTAN1 (12%), ERBB2 (12%), CREBBP (12%), FAT1 (12%), ATM (14%) and KMT2C (18%). Moreover, 158 genes that were epigenetically silenced were identified by analyzing DNA methylation and gene expression, for example, SPATC1L (silenced in 19%), nicotinate phosphoribosyltransferase (NAPRT) (13%), Poly(ADP-ribose) polymerase PARP6 (26%) and latexin (LXN) (27%). However, the DNA hypermethylation in the promoter of other tumor suppressor genes were not found, such as RB1, NF2, NF1, TSC2, TSC1, PTEN, and TP53 (14).

In addition to single DNA abnormality, bladder cancer genomic research also focuses on polymorphism sites. There are multiple gene polymorphisms in bladder cancer such as ERCC1, XRCC1, GSTP1, CDA, GSTM1, and GSTT. Different gene polymorphisms could predict chemotherapy sensitivity in bladder cancer. Xu et al. studied the efficacy of 41 patients at stage IV of muscle-invasive bladder cancer with different ERCC1 genotypes after 2 to 6 cycles of platinum-based chemotherapy. The results suggest there is a discrepancy of short-term response to chemotherapy and median overall survival in patients with different ERCC1 genotype. Compared with C/T and T/T genotypes, patients with C/C genotype had a better short-term response to chemotherapy and median overall survival (15).

Studies have shown that there are changes in the PI3K, MAPK, Hedgehog, and Wnt pathways in the bladder cancer genome (16–18). Among them, the PI3K pathway is the most studied signaling pathway in bladder cancer. ERBB receptor family is the upstream promoter of the PI3K pathway. For example, ERBB1 (EGFR) can activate the PI3K pathway by activating RAS. Overexpression of EGFR, ERBB2 or ERBB3 is associated with tumor grade, stage, and prognosis in bladder cancer. Moreover, mutations in ERBB2 or ERBB3 have been found in some muscle-invasive bladder cancers (19). PIK3CA encodes the catalytic subunit p110α of PI3K, and its mutation occurs in 25% of non-muscle invasive bladder cancer (20). The above studies suggest that the mechanism of bladder cancer development and progression may be related to the PI3K signaling pathway.

Chromosomes are aggregates of genetic material, and chromosomal abnormalities are an important part of the genetic defects of bladder cancer. Based on genetic techniques such as genomic hybridization and loss of heterozygosity, it is found that bladder cancer has complex chromosome number and structural variation. Bartoletti et al. have shown that the partial or complete loss of genetic material at chromosome 9 can lead to the loss of tumor suppressor genes such as p16 which often indicates the recurrence of low-grade bladder cancer. Abnormal numbers of chromosomes 3, 7, 13, and 17 have also occurred in bladder cancer (21). Among them, chromosome 3 polyploid and chromosome 7 aneuploidy may be related to the progression and malignancy of bladder cancer, and the aneuploidy of chromosome 17 often implies a high recurrence rate of bladder cancer (4, 22).

Telomerase is a ribonucleoprotein that is essential for the replication of most eukaryotic chromosome ends. In cancer cells, telomerase can be activated. In contrast, telomerase is epigenetically silenced and inactivated in normal cells (31). However, mutations of important genes can reverse telomerase silencings, such as tumor suppressor genes or mutations in the tumor suppressor pathway signaling molecules that may affect telomerase activity in human tumors (32). Telomerase consists of two subunits: telomerase RNA component (TERC) and telomerase reverse transcriptase tert (TERT) catalytic subunit. The TERT gene encodes a telomerase reverse transcriptase catalytic subunit and assembles into a ribonucleoprotein protease complex to maintain telomere length which plays an important role in the maintenance of tumor genome stability (33). In recent years, TERT promoter mutations have been found in malignant tumors such as melanoma (34) and glioma (35). Studies have shown that the TERT promoter mutation is highly correlated with the prognosis of transitional cell carcinoma and confirmed that the −124 C > T mutation (corresponding to C228T) is associated with high expression of TERT and telomerase activity (36).

Wu et al. found a coexistence relationship between the TERT promoter mutation point and the TC53/RB1 inert somatic mutation point. By determining the chromosomal instability index, it was found that the tumor chromosomal instability index of the TERT promoter mutation was significantly higher than that of the tumor without the TERT promoter mutation (26). TP53 is involved in the regulation of telomerase activity, maintenance of gene integrity, inert mutations, and its deletion is associated with chromosomal instability. These studies suggest that the coexistence of the two may have a synergistic effect on the chromosomal instability of bladder cancer (37).

Studies have found that the C228T mutation of the TERT promoter often occurs in BCSCs (27). Compared with non-BCSCs, C228T has a significantly higher mutation rate in BCSCs which is consistent with the phenomenon of high C228T expression in stem cells compared to normal somatic cells. Importantly, the TERT promoter C228T mutation can convert normal bladder stem cells (NBSCs) into tumor-initiating cells. This mutation is located at the promoter region and can lead to tumors formation which can be considered as tumor promoters. It can be seen that the high mutation rate of TERT promoter in BCSCs is a new feature of bladder cancer, but the transformation of NBSCs mutation into BCSCs is complex and further research is needed.

Telomerase activity is observed in almost all human tumor types, thus monitoring the tumor’s telomerase activity will greatly contribute to the diagnosis and screening of bladder cancer. Wu et al. screened for the presence of telomerase reverse transcriptase gene promoter somatic cell mutations by Sanger sequencing in 302 patients with different urological tumors. The result showed that 43% of genitourinary tumors had TERT promoter somatic mutations. This has determined high-frequency mutation hotspot of the telomerase reverse transcriptase gene promoter in clinical urogenital (26). In different types of urological organ tumors, the amplitude of somatic mutation of the TERT promoter was larger (0-63.7%). The urinary tract cancer has the highest mutation frequency among all, while the prostate cancer showed no mutation.

Chromatin remodeling is an important regulatory mechanism of epigenetics, which is characterized by changes in the nucleosome structure and its relative position to the DNA sequence, and alteration of the accessibility of gene promoter region sequences to further regulate gene expression. Chromatin remodeling works primarily through two pathways: ATP-dependent chromatin remodeling complex, the SWl/SNF complex; histone-modifying enzymes including histone methylation, acetylation, phosphorylation, and ubiquitination (38). Recent studies have found that chromatin remodeling related genes have high-frequency mutations in a variety of tumors including bladder cancer, kidney cancer, gastric cancer, ovarian clear cell carcinoma, breast cancer and glioblastoma (38). Mutations in chromatin remodeling-related genes can cause epigenetic changes in local histone modifications and chromatin conformation in cancer cells, leading to dysregulation of downstream signaling genes, changes in biological behavior such as cell proliferation and apoptosis, leading to tumor development (39).

Gui et al. performed full exon sequencing on 9 patients with transitional cell carcinoma and has found 8 frequently mutated chromatin remodeling genes in 59% of transitional cell bladder cancer: UTX, MLL-MLL3, CREBBP-EP300, NCOR1, ARID1A and CHD6 (40). UTX has a number of significantly enriched mutations including 11 nonsense mutations, 4 frameshift mutations, and 1 splice site change. One of these mutations was predicted to be able to truncate the JmjC domain, which is critical for the demethylase activity of the protein product. ARID1A has the same mutation pattern as UTX. In addition, mutations in the above genes have also been found in other malignant tumors such as ovarian cancer and renal cancer (41, 42). The study found that STAG2 is located on the X chromosome, encoding sister chromatid cohesion and segregation (SCCS) adhesion complex components, regulating the separation of sister chromosomes during cell division, and inactivation of STAG2 can cause weakening of chromosome binding and thus aneuploidy (43). Via gene sequencing and rigorous bioinformatics analysis, Guo et al. have found that STAG2, ESPL1 and NIPBL genes with frequent mutations in bladder cancer are involved in the SCCS process (44). The study further revealed that patients with mutant STAG2 had a higher frequency of chromosomal aneuploidy by detecting copy number changes in the chromosome arm, and significantly worse in the prognosis of STAG2 gene somatic cell mutation in bladder cancer patients compared with STAG2 bladder cancer patients without the mutation. Collectively, in SCCS, other types of tumors only report rare or low-frequency mutations in oncogenes. Bladder cancer is a tumor type that is first known to have a high-frequency genetic damage gene in the SCCS process in which the genetic mutation rate is approximately 32%. The genetic changes affect the SCCS process which may involve the development of bladder cancer. Despite the detailed mechanism of STAG2 leading to the pathogenesis of transitional cell carcinoma is still not known, the high-frequency recurrent mutant gene is indeed a new pathway associated with transitional cell carcinoma in the SCCS process of transitional cell bladder cancer (45).

Whole genome sequencing is the sequencing of the entire genome of tumor tissue with the DNA sequence of the germ cells from the same patient (tumor cell mutation is not present in the germ cells) as a control can effectively identify the changes in all regions including nucleotide replacement, structural rearrangement, and copy number changes (49). Therefore, whole genome sequencing is the most comprehensive depiction of the tumor genome, suitable for the study of tumor genome-wide association. However, whole genome sequencing requires the detection of a large number of sequences is redundant for studies that do not require the whole genome. In this case, shotgun sequencing based on the Sanger sequencing principle is more suitable for identifying somatic cell rearrangements and copy number changes in the genome.

Sanger sequencing is the gold standard for current gene sequencing. Wu et al. performed Sanger sequencing to sequence 302 different urinary tumor patients and found a high-frequency mutation hotspot of TERT promoter in urinary tumor patients wherein TERT promoter somatic mutation in 55.6% of bladder cancer (26). Guo et al. performed genome-wide sequencing of 99 tumors and corresponding peripheral blood samples from patients with transitional cell carcinoma by Sanger sequencing. SCCS-related gene mutations such as STAG2 and ESPL1 were confirmed from 11240 candidate somatic mutations. Moreover, mutant genes such as FGFR3, TP53, PIK3CA, and RB1 are ubiquitous in malignant tumors (44). Second-generation sequencing invention has received extensive attention at the beginning compared to Sanger sequencing. Andrea et al. recruited 50 patients with muscle-invasive bladder cancer and performed whole-exome sequencing on germline and pretreatment tumor DNA. Then these patients were treated with neoadjuvant cisplatin-based chemotherapy and underwent surgery to evaluate the pathologic response. They identified ERCC2, one nucleotide excision repair gene, was only significantly mutated in responders compared with nonresponders (50). David et al. investigated the correlation of ERCC2 with pathologic response to neoadjuvant cisplatin-based chemotherapy in an independent validation patient cohort, and they confirmed that ERCC2 was indeed associated with response to chemotherapy (51). Desai et al. used the second-generation sequencing technology to map the mutations of primary and recurrent transitional cell bladder cancer. By comparing the differential expression of tumor-associated DNA damage response genes, they also found that especially ERCC2 mutations indicate a better prognosis of chemoradiotherapy for transitional cell bladder cancer and a lower rate of recurrence and metastasis within 2 years (52).

Exon sequencing is a method of capturing whole genome exon DNA by using targeted sequence capture technology to enrich the construction of DNA library before performing high-throughput sequencing. It is characterized by sequencing of the open reading frame without sequencing the introns. Since exons account for only 2% of the entire genome, high-throughput sequencing of the targeted exon greatly reduces the amount of sequencing and achieves higher efficiency. Due to its deep sequencing of the coding region, it can be used to screen for mutations that Sanger sequencing could not. In addition, it is suitable for the study of single nucleotide polymorphism (SNP) and insertion-deletion (InDel) of the target gene. Longo et al. used targeted exon sequencing to detect 50 cases of patients with pTis-pT4b bladder cancer. Through the analysis of staged somatic mutations in patients with such high-risk bladder cancer, the mutated genes and mutation profiles were determined. In terms of single mutations, it was found that epigenetic and cell cycle-regulated genes have significantly higher mutation rates. PI3K/mTOR and cell cycle/DNA repair showed the highest mutation rate among the other assessed pathways. Moreover, RB1 and TP53 mutations were found to frequently coexist with NF1 and PIK3CA mutations (53).

However, exon sequencing also has certain limitations. It is unable to sequence non-coding genes and thus fail to conduct a comprehensive study of intron-regulated genes associated with the target gene, and also impossible to identify copies.

In the past, genome sequencing was the extraction of DNA from many cells or a piece of tissue. It is inevitable that sequencing results are obtained from many cell genomes. However, tumors are heterogeneous and the genomic information of different subtypes is more or less different. Investigation of the total DNA information extracted from different cell populations will inevitably cause researchers to ignore the differences between cells to varying degrees and further bias the study of the target cell genome (54, 55). Single cell sequencing is the sequencing of genomes from a single cell. The process involves the isolation of single cells and extraction of DNA followed by single-cell genomic amplification and sequencing analysis. This will prevent the influence of other cell genomes on the sequencing results of the target cells and thus accurately measure the copy number of a single nuclear gene (56).

Bladder cancer is a solid tumor with significant heterogeneity, and the process of research needs to be careful to prevent bias resulted by other cells. In order to study TERT promoter mutations in the bladder cancer, Li et al. had isolated the single cells of BCSCs, non-BCSCs, NBBCs, and non-NBBCs and extracted their genomes for single-cell gene amplification and sequencing. The high-frequency mutation sites of C228T were found in the TERT promoters of BCSCs and non-BCSCs, and the effect of C228T mutation on bladder cancer and BCSCs was investigated (27). In order to study the genetic basis and origin of bladder cancer stem cells, Yang et al. had performed single-cell sequencing on 59 single cells including bladder cancer stem cells (BCSCs), bladder cancer non-stem cells (BCNSCs), bladder epithelial stem cells (BESCs), and normal bladder epithelial cells. It was found that BCSCs showed clonal homogeneity, and phylogenetic analysis indicated that BCSCs were derived from BESCs or BCNSCs. This study has confirmed that 21 abnormal key genes in BSCSs wherein 6 of which were not reported in bladder cancer in the past, which fully demonstrated the accuracy of single-cell sequencing (57). Faridani et al. demonstrated that microRNAs as the potential markers of different cell types and cell status by performing single-cell sequencing of the small-RNA transcriptome on human embryonic cells and tumor cells (58).

The transcriptome sequencing is a technique to sequence the transcript RNA, which is a sequence of cDNA that is reverse transcribed from mRNA, total RNA or other RNAs (e.g., small RNAs) to obtain the sum of RNA in a state of the cell. It was found to be sensitive and effective for intron fusion including oncogene activation caused by in-frame fusion. Kekeeva et al. performed high-throughput transcriptome sequencing on bladder cancer patients and screened 4 fusion introns in bladder cancer from 819 suspected fusion introns such as SEPT9/CYHR, IGF1R/TTC23, SYT8/TNNI2 and CASZ1/DFFA (59). Transcriptome sequencing can analyze gene expression profiles to reveal differentially expressed genes in tumors. Based on the transcriptome sequencing data, Zhang et al. confirmed the differential gene expressions of several bladder cancers through the comparison of bladder cancer and normal tissue transcriptome, in which ELF3 and MYBL2 are the tumor suppressors while MEG3, APEX1, and EZH2 are the inducer of bladder cancer progression (60). In addition, transcriptome sequencing can also be applied to detect somatic mutations, but it is difficult to find a matching normal sample control which limits the application in this aspect. Liu et al. performed transcriptome sequencing of transitional cell carcinoma and identified 937 differentially expressed genes (61).

It is well known that miRNAs can form RNA silencing-inducing complexes to inhibit translation of target mRNAs which is an important mechanism for gene expression regulation. Studies have shown that miRNAs also play an important role in tumor pathogenesis (62). Therefore, transcriptome sequencing technology has broad application prospects in detecting the abnormal miRNA expression in tumors.