Mutational Landscape of Bladder Cancer in Mexican Patients: KMT2D Mutations and chr11q15.5 Amplifications Are Associated with Muscle Invasion

Bladder cancer (BC) is the most common neoplasm of the urinary tract, which originates in the epithelium that covers the inner surface of the bladder. The molecular BC profile has led to the development of different classifications of non-muscle invasive bladder cancer (NMIBC) and muscle-invasive bladder cancer (MIBC). However, the genomic BC landscape profile of the Mexican population, including NMIBC and MIBC, is unknown. In this study, we aimed to identify somatic single nucleotide variants (SNVs) and copy number variations (CNVs) in Mexican patients with BC and their associations with clinical and pathological characteristics. We retrospectively evaluated 37 patients treated between 2012 and 2021 at the National Cancer Institute—Mexico (INCan). DNA samples were obtained from paraffin-embedded tumor tissues and exome sequenced. Strelka2 and Lancet packages were used to identify SNVs and insertions or deletions. FACETS was used to determine CNVs. We found a high frequency of mutations in TP53 and KMT2D, gains in 11q15.5 and 19p13.11-q12, and losses in 7q11.23. STAG2 mutations and 1q11.23 deletions were also associated with NMIBC and low histologic grade.


Introduction
Bladder cancer (BC) is the most common neoplasm of the urinary tract. It generally originates from the epithelium that covers the inner surface of the bladder [1]. BC is the tenth most common cancer type, with an estimated 570,000 new cases worldwide in 2020 [2]. Smoking is the leading risk factor associated with BC incidence [3]. Within the total incidence rate, men tend to be diagnosed with bladder cancer 3 to 4 times more frequently than women. In addition, Non-Hispanic white patients are the most frequently affected, followed by Non-Hispanic blacks and Hispanics [4].
BC arises in two different pathways depending on the origin site. Non-muscular invading bladder cancer (NMIBC) or superficial BC represents about 60% of all BC cases. NMIBC is characterized by the loss of heterozygosity in chromosome 9 and mutations in FGFR3. In 10% to 20% of the cases, NMIBC can invade the muscular layer and evolve as muscular invasive bladder cancer (MIBC); this evolution is characterized by the loss of the tumor suppressor genes TP53 and RB1 [5]. MIBC is the most aggressive BC type, and it is characterized by genomic instability and a high mutational rate. Patients with MIBC tend to present with a 5-year survival rate of 60% when they have a localized tumor, but less than 10% when distant metastases are present [6].
Most NMIBCs are exophytic tumors; however, some lesions present an endophytic growth pattern affecting the lamina propria. Two main subepithelial growth patterns have been described: the endophytic growth pattern (EGP) and the von Brunn's nest involvement (VBNI). These growth patterns have been associated with high-grade tumors and higher stages of the disease; however, no positive correlations with bladder cancer recurrence have been associated with these growth patterns [7,8].
The molecular analysis of BC has led to the development of different classifications of both NMIBC and MIBC [9][10][11][12]. After an international consensus, MIBC was classified into six different molecular subtypes: Basal/Squamous, Neuroendocrine-like, Stroma-rich, Luminal-papillary, Luminal-unstable, and Luminal-non-specified [6]. Molecular analysis has also identified mutational signatures associated with the mutational processes for carcinogenesis in bladder cancer. One of the most common mutational signatures in bladder cancer is the single base substitution signature 13 (SBS13), which is based on six types of substitutions: C > A, C > G, C > T, T > A, T > C, and T > G, and is related to the cytidine deaminase activity of APOBEC. Other mutational signatures reported for bladder cancer included SBS01 and SBS05 associated with the aging process, SBS02 associated with APOBEC, SBS04 associated with smoking, and SBS22 associated with aristolochic acid exposure [13,14].
The improvements in molecular biology and our understanding of tumorigenesis opened the window to the personalized medicine era for patients with BC [1]. For example, the description of mutational processes affecting bladder cancer tumorigenesis. However, the molecular and genomic data found for BC had been obtained principally from European and European descendant patients; other ethnic groups, such as Hispanics, are underrepresented in these classifications. Here, we aim to identify the somatic single nucleotide variants (SNVs) and copy number variations (CNVs) present in Mexican patients with bladder cancer, exploring their association with clinical-pathological characteristics.

Clinical-Pathological Characteristics of the BC Patients
We included 37  When evaluating the overall survival of the patients, a median value for survival in this cohort was not reached (Figure 1a). When grouping patients according to their muscle invasion phenotype, the patients with MIBC showed a poor overall survival compared to patients with NMIBC (p-value = 0.001), with a median survival of 25 months (Figure 1b). In addition, we found a significant difference in the overall survival of the patients when sorting them according to their histological grade (p-value = 0.047); however, we did not observe any differences related to overall survival in the Kaplan-Meier curves related to smoking or sex (Supplementary Figure S1). Additionally, when performing a sensitivity analysis, we found no modification in the influence of muscular invasion on mortality (

Tumor Mutational Burden and Clinical-Pathological Characteristics
Whole exome sequencing (WES) was achieved successfully with an average sequencing depth on the target region of 183.90× and an average coverage of the target region of 99.65%. Samples had a mean of 94.83% Q30 quality in the sequenced bases. FASTQ files were obtained, and further bioinformatics analysis was performed to detect somatic and structural variations. Samples showed a median of 623 somatic variants (ranging from 136 to 2552 mutations per case), including SVNs and InDels (Insertion-or-Deletion). The total number of somatic variants, including driver and passenger mutations, present in the exome region, was used to identify the tumor mutational burden (TMB) of each tumor. The TMB values ranged from 2.27 to 42.53 mutations/Mb. A total of 43% tumors had TMB values greater than 10 mutations/Mb and were classified as having high TMB status ( Figure 2). The relationship between TMB and clinical features is shown in Supplementary Figure S2.

Somatic Variants and Clinical-Pathological Characteristics of the Patients
High-confidence somatic variants were classified as driver and passenger mutations. All the samples had driver somatic variants. The tumor suppressor gene TP53 was the most frequently affected in our study, which was mutated in 43% of the patients. Other genes which frequently mutated included the gene encoding for the lysine

Tumor Mutational Burden and Clinical-Pathological Characteristics
Whole exome sequencing (WES) was achieved successfully with an average sequencing depth on the target region of 183.90× and an average coverage of the target region of 99.65%. Samples had a mean of 94.83% Q30 quality in the sequenced bases. FASTQ files were obtained, and further bioinformatics analysis was performed to detect somatic and structural variations. Samples showed a median of 623 somatic variants (ranging from 136 to 2552 mutations per case), including SVNs and InDels (Insertion-or-Deletion). The total number of somatic variants, including driver and passenger mutations, present in the exome region, was used to identify the tumor mutational burden (TMB) of each tumor. The TMB values ranged from 2.27 to 42.53 mutations/Mb. A total of 43% tumors had TMB values greater than 10 mutations/Mb and were classified as having high TMB status ( Figure 2). The relationship between TMB and clinical features is shown in Supplementary Figure S2.

Tumor Mutational Burden and Clinical-Pathological Characteristics
Whole exome sequencing (WES) was achieved successfully with an average sequencing depth on the target region of 183.90× and an average coverage of the target region of 99.65%. Samples had a mean of 94.83% Q30 quality in the sequenced bases. FASTQ files were obtained, and further bioinformatics analysis was performed to detect somatic and structural variations. Samples showed a median of 623 somatic variants (ranging from 136 to 2552 mutations per case), including SVNs and InDels (Insertion-or-Deletion). The total number of somatic variants, including driver and passenger mutations, present in the exome region, was used to identify the tumor mutational burden

Somatic Variants and Clinical-Pathological Characteristics of the Patients
High-confidence somatic variants were classified as driver and passenger mutations. All the samples had driver somatic variants. The tumor suppressor gene TP53 was the most frequently affected in our study, which was mutated in 43% of the patients. Other genes which frequently mutated included the gene encoding for the lysine

Somatic Variants and Clinical-Pathological Characteristics of the Patients
High-confidence somatic variants were classified as driver and passenger mutations. All the samples had driver somatic variants. The tumor suppressor gene TP53 was the most frequently affected in our study, which was mutated in 43% of the patients. Other genes which frequently mutated included the gene encoding for the lysine methyltransferase 2C (KMT2C) (35%), the fibroblast growth factor receptor 3 (FGRFR3) (32%), the cell division cycle protein (CDC27) (27%), the lysine demethylase 6A (KDM6A) (27%), the SWI/SNF-Related protein ARID1A (27%), and lysine methyltransferase 2D KTM2D (27%). Missense SNVs were the most common type of somatic variant ( Figure 3).

Structural Variants and Clinical-Pathological Characteristics of the Patients
The structural variants were identified for the patients as CNV gains and CNV losses. As shown in Figure 5a, chromosomes 8 and 19 and were affected almost entirely by CNV gains, while chromosomes 9, 10, 17, 21, and 22 were affected mainly by CNV losses. Even though large chromosomic regions were affected by structural variants, our cohort also showed cytobands with a higher frequency of CNVs (Figure 5a). Cytoband chr5q31.3 was the most frequently affected, showing CNV losses in 98% of the samples. Therefore, the protocadherin gamma cluster genes (PCDHG), encoded in chr5q31.3, were the most frequently affected genes, with CNV losses in 95% of the patients. ZMAT2 was also present in this cytoband and was found to be affected in 73% of the patients. Chr1p36.21 was affected completely by CNV losses in 95% of the cases and chr19q13.42 was affected in 82% of the cases by CNV gains and losses. Within chr1p36.21 and ch19q13.42, the genes frequently affected by CNV included: RPS9 (78%), LILRB3 (73%), PRAMEF12 (54%), and PRAMEF1 (38%). Figure 5b shows an arrangement of the cytobands frequently affected by CNV gains and losses among the patients evaluated and according to their muscle invasive phenotype.
When grouping according to the muscle invasive phenotype of the patients (Figure  6a), those with MIBC tended to show a high frequency of CNV gains in cytobands chr11p15.5 (p-value = 0.010, FDR = 0.540) and chr19p13.11-q12 (p-value = 0.024, FDR = 0.540). MUC2 encoded in chr11p15.5 was found in all the patients affected by CNV gains, with log2 values ranging from 0.22 to 2.19. In the same direction, ZNF429 harbored in

Structural Variants and Clinical-Pathological Characteristics of the Patients
The structural variants were identified for the patients as CNV gains and CNV losses. As shown in Figure 5a, chromosomes 8 and 19 and were affected almost entirely by CNV gains, while chromosomes 9, 10, 17, 21, and 22 were affected mainly by CNV losses. Even though large chromosomic regions were affected by structural variants, our cohort also showed cytobands with a higher frequency of CNVs (Figure 5a). Cytoband chr5q31.3 was the most frequently affected, showing CNV losses in 98% of the samples. Therefore, the protocadherin gamma cluster genes (PCDHG), encoded in chr5q31.3, were the most frequently affected genes, with CNV losses in 95% of the patients. ZMAT2 was also present in this cytoband and was found to be affected in 73% of the patients. Chr1p36.21 was affected completely by CNV losses in 95% of the cases and chr19q13.42 was affected in 82% of the cases by CNV gains and losses. Within chr1p36.21 and ch19q13.42, the genes frequently affected by CNV included: RPS9 (78%), LILRB3 (73%), PRAMEF12 (54%), and PRAMEF1 (38%). Figure 5b shows an arrangement of the cytobands frequently affected by CNV gains and losses among the patients evaluated and according to their muscle invasive phenotype.
When grouping according to the muscle invasive phenotype of the patients (Figure 6a), those with MIBC tended to show a high frequency of CNV gains in cytobands chr11p15.5 (p-value = 0.010, FDR = 0.540) and chr19p13.11-q12 (p-value = 0.024, FDR = 0.540). MUC2 encoded in chr11p15.5 was found in all the patients affected by CNV gains, with log2 values ranging from 0.22 to 2.19. In the same direction, ZNF429 harbored in chr19p13.11-q12 was found in all the patients affected by CNV gains. CNV gains in chr11p15.5 also tended to be frequent in patients with a high histologic grade, but did not reach the threshold of statistical significance for the raw p-value (p-value = 0.057, FDR 0.729), whereas CNV losses in chr7q11.23 (p-value = 0.005, FDR = 0.729), chr18p11.21 (p-value = 0.019, FDR = 0.729), and chr7q35 (p-value = 0.021, FDR = 0.729) were frequently affected in patients with low histologic grade (Figure 6b). Regarding the smoking status of the patients, CNV deletions in cytoband chr15q13.1 (p-value = 0.046, FDR = 0.513) were found to be as significant in patients with a smoking history (Figure 6c). No significant associations between the sex of the patients and the affected cytobands were found (Figure 6d). chr19p13.11-q12 was found in all the patients affected by CNV gains. CNV gains in chr11p15.5 also tended to be frequent in patients with a high histologic grade, but did not reach the threshold of statistical significance for the raw p-value (p-value = 0.057, FDR 0.729), whereas CNV losses in chr7q11.23 (p-value = 0.005, FDR = 0.729), chr18p11.21 (pvalue = 0.019, FDR = 0.729), and chr7q35 (p-value = 0.021, FDR = 0.729) were frequently affected in patients with low histologic grade (Figure 6b). Regarding the smoking status of the patients, CNV deletions in cytoband chr15q13.1 (p-value = 0.046, FDR = 0.513) were found to be as significant in patients with a smoking history (Figure 6c). No significant associations between the sex of the patients and the affected cytobands were found ( Figure  6d). The red dot indicates CNV events with a log2 greater than 3 or lower than ₋ 3 per patient; (b) Bar plot presenting the ten cytobands more frequently affected by CNV gains and losses in the cohort. Green bars indicate CNV gains, and pink bars indicate CNV losses.   In addition, we found no modification in the influence of muscular invasion on mortality when performing a sensitivity analysis adjusting for the genomic alterations that were found to be associated with muscular invasion (HR = 13.11, 95%CI = 1.66, 103.60). A mild reduction in the influence was observed when adjusting for TP53 mutation (HR = 8.09, CI95% = 0.95, 68.78); however, no modifications were observed when adjusting for KMT2D mutation, chr11p15.5 amplification, and chr19p13.11-q12 amplification. (Supplementary  Table S2).

Mutational Processes and Clinical-Pathological Characteristics
The mutational processes implicated in bladder cancer were evaluated through mutational signature analysis, identifying the contribution of single-base substitution signatures (SBS). Signature SBS16, related to liver cancer and defective nucleotide excision repair pathway, was the most frequent mutational process, with a contribution greater than 0.2 and found in 89% of the patients, followed by SBS05 present in 51% and SBS13 in 19% of the patients (Figure 7a). When evaluating the association of the mutational processes, we observed that patients with smoking history were associated with greater contribution values of SBS05 (Supplementary Figure S3). High TMB status was also associated with SBS02 and SBS13, whereas a low TMB status was associated with SBS01 and SBS05 (Figure 7b). In addition, we found no modification in the influence of muscular invasion on mortality when performing a sensitivity analysis adjusting for the genomic alterations that were found to be associated with muscular invasion (HR = 13.11, 95%CI = 1.66, 103.60). A mild reduction in the influence was observed when adjusting for TP53 mutation (HR = 8.09, CI95% = 0.95, 68.78); however, no modifications were observed when adjusting for KMT2D mutation, chr11p15.5 amplification, and chr19p13.11-q12 amplification. (Supplementary Table S2).

Mutational Processes and Clinical-Pathological Characteristics
The mutational processes implicated in bladder cancer were evaluated through mutational signature analysis, identifying the contribution of single-base substitution signatures (SBS). Signature SBS16, related to liver cancer and defective nucleotide excision repair pathway, was the most frequent mutational process, with a contribution greater than 0.2 and found in 89% of the patients, followed by SBS05 present in 51% and SBS13 in 19% of the patients (Figure 7a). When evaluating the association of the mutational processes, we observed that patients with smoking history were associated with greater contribution values of SBS05 (Supplementary Figure S3). High TMB status was also associated with SBS02 and SBS13, whereas a low TMB status was associated with SBS01 and SBS05 (Figure 7b).  with a contribution more significant than cero were depicted. The gradient red color indicates the grade of the contribution of each signature; (b) Violin plot depicting the association between the tumoral mutational burden (TMB) status of the patients and SBS01, SBS02, SBS05, SBS13, and SBS16. We did not find any association between TMB and clinical features (Supplementary Figure S1).

Discussion
To our best knowledge, this is the first report of the somatic mutational landscape of bladder cancer in Mexican patients. Using WES, we identified a differential mutational pattern between patients with MIBC and NMIBC. Patients with MIBC were associated with a higher frequency of mutations in TP53 and KMT2D and CNV gains in chr11q15.5 and chr19p13.11-q12. In contrast, mutations in STAG2 and CNV deletions in chr7q11.23 were associated with patients with NMIBC and a low histologic grade.
The patients included in our study showed similar characteristics to other previously published Mexican bladder cancer studies. A descriptive epidemiological study of 20 years of experience with bladder cancer in the INCan identified a median age at diagnosis of 54 years with a predominance of male patients and smoking as the main risk factor. A high rate of patients were diagnosed in late disease stages, with 70% of the cases with metastasis [15]. Another study performed in a general hospital in Mexico reported a median age at diagnosis of 62.5 years with a predominance of male patients. Within this study, a prevalence of 36% was reported for patients with MIBC [16].
Overall, non-Hispanic whites are the ethnic group with the highest incidence of bladder cancer, followed by Blacks and Hispanics [8]. These differences are also seen to be associated with the overall survival of the patients, with non-Hispanic whites as the ethnic group with the best rates [8]. Differences in the tumor biology and somatic mutations could be influenced by the ethnicity of the patients; a higher frequency of TP53, ARID1A, ERBB3, and CDKN1A mutations have been reported for white patients compared to nonwhite patients [17]. TP53 was the gene most frequently mutated in our cohort. Similar to reports in other populations [18], we identified an association between TP53 mutation and muscular invasion. In addition, mutations in TP53 have been associated with poor overall survival [18]. Within the multivariate analysis performed in this study, we observed a positive influence of TP53 on the mortality of the patients, suggesting that mutations in TP53 affect Mexican patients in the same way that affects other populations.
This group of patients evaluated included a similar proportion of patients with MIBC and NMIBC, with a greater proportion of cases with high histologic grades. We identified KMT2D as a gene frequently mutated in patients with MIBC and high histologic grade. KMT2D encodes the lysine methyltransferase 2D; the loss of this gene has been associated with abnormal epigenetic reprogramming of different molecular pathways [19]. KMT2D is mutated in different types of cancer, including bladder cancer [20]. In bladder cancer cell lines, KMT2D overexpression was associated with tumor suppressor effects, as it promoted the expression of PTEN and TP53, as well as the repression of STAG2 [21]. In lung cancer, it has been observed that KMT2D mutations increase the glycolysis of tumor cells. For this reason, it has been proposed to study the pharmacological inhibition of glycolysis in tumors deficient in KMT2D [22]. The expression of KMT2D has also been proposed as a prognostic biomarker for bladder cancer in European patients, observing a trend towards greater survival in patients with higher expression of this protein [23]. However, it is unknown whether this effect was also present in Mexican patients.
We also observed that patients with MIBC frequently showed copy number gains of the chr11p15.5 cytoband. MUC2 was seen to be affected in all patients with these gains. This protein has been seen to contribute to the development of colorectal cancer [24]. Similarly, there is evidence suggesting that the expression of MUC2 is associated with invasion and metastasis in various malignant tumors, including gastric cancer, prostate cancer, and colorectal cancer [25]. In the case of bladder cancer, the presence of MUC2 has been associated with the non-invasive proliferation of tumors or with a favorable outcome for patients [26]. It has also been seen that the expression of MUC2 is observed more frequently in low-grade urothelial carcinomas and correlates with a low pathological stage [26].
Within our cohort, nearly half of the patients were associated with TMB values greater than 10 mutations/Mb. TMB has been used as a biomarker for the selection of different immune checkpoint inhibitors presenting a wider range of therapeutic strategies for the treatment of bladder cancer patients [27,28]. To our best knowledge, there are no studies exposing the clinical uses of TMB in Mexican patients with bladder cancer. We found no associations between High-TMB and clinical characteristics, which could mean a greater number of patients that could benefit from TMB as a biomarker for treatment selection.
We are aware that the limited number of patients included in this study could hinder the statistical power of the analysis. Nonetheless, this study provides new information about the clinical and pathological characteristics of Mexican patients regarding bladder cancer, which could guide us in the design of further studies that feature a larger cohort to test some of the identified variants as biomarkers of prognosis for patients treated.

Population
A total of 37 patients treated between 2012 and 2021 at the Instituto Nacional de Cancerología (INCan), located in Mexico City, were selected randomly for this study from the total of patients that met the inclusion criteria. Inclusion criteria involved patients with formalin-fixed paraffin-embedded (FFPE) tumor tissue blocks stored at the pathology department of the INCan, who were older than 18 years old and had a positive diagnosis of urothelial bladder cancer. FFPE samples were collected by transurethral resection of the bladder tumor (TURBT) before any kind of treatment. Patients with a history of other cancer or any type of treatment prior to TURBT were not included in this study. Tumor samples were selected from formalin-fixed paraffin-embedded tissue (FFPE) blocks for an experienced pathologist. Selected tumor samples had at least 70% of tumor cellularity. This project was approved by the institutional ethics board (017/034/IBI) and the institutional ethics in research committee (CEI/1175/17) of the INCan. Due to the minimal risk and the nature of this study for the patients and following the guidelines of the institutional ethics board, the informed consent was waived.

Clinical Data Collection
Clinical data, including age at diagnosis, body mass index (BMI), sex, education level, smoking status, symptoms at diagnosis, muscular invasion, subepithelial invasion, and histological grade, were collected from the electronic clinical records of each patient. The follow-up time was calculated from the clinical records as the period between the date of diagnosis and the date of death or loss of follow-up.

DNA Extraction and Quality Control
Samples were homogenized using QIAshredder (QIAGEN, 79654), and DNA was extracted using the QIAamp ® DNA FFPE Tissue Kit (QIAGEN, 56404), following the protocol recommendations. Purity of DNA was evaluated with a Thermo Fisher Scientific NanoDrop 2000. Quantity and DNA fragmentation status were evaluated with the Agilent 2200 TapeStation System using the genomic DNA ScreenTape assay (Agilent, 5067-5365) to obtain a DNA integrity number [29]. Samples with DIN from 6 to 10 and a minimum concentration of 20 ng/µL were selected for WES.

Library Preparation, Hybridation Capture, and WES
Library preparation, hybridization capture, and WES procedures were performed by the New York Genome Center. TruSeq DNA PCR-Free libraries were prepared from FFPE tissues using 1µg of input DNA according to the manufacturer's instructions (Ilumina, San Diego, CA, USA). Sequencing was performed on HiSeq2500 (Ilumina, San Diego, CA, USA).

Bioinformatics Pipeline
Sequencing reads for the tumor samples were first trimmed for adapters using Trim-Galore (v0.4.0). The trimmed reads were then aligned to the reference genome using BWA-MEM (v0.7.15) [30], GATK (v4.1.0) [31] FixMateInformation was run to verify and fix mate-pair information, followed by Novosort (v1.03.01) markDuplicates to merge indi-vidual lane BAM files into a single BAM file per sample. Duplicates were then sorted and marked, and GATK's base quality score recalibration (BQSR) was performed to obtain a coordinated sorted BAM file for each sample.
A matched normal sample was not available. In its place, we used the HapMap sample NA12878, which was prepared and sequenced using the same protocol as the tumor sample. This normal sample was used to remove some of the false positives that were due to library preparation and sequencing (that would manifest in the same way in the tumor and NA12878), as well as some germline variants that were common to the tumor sample and NA12878.
For CNVs, segments with log2 > 0.2 were categorized as amplifications, and segments with log2 < −0.235 were categorized as deletions (corresponding to a single copy change at 30% purity in a diploid genome, or a 15% Variant Allele Fraction). CNVs of a size less than 20 Mb were denoted as focal and the rest were considered large-scale. Only focal CNV were selected for subsequent analysis. We used Bedtools [45] for annotating CNVs. All predicted CNVs were annotated with germline variants by overlapping with known variants in 1000 Genomes and Database of Genomic Variants (DGV) [46].

Mutational Signature Analysis
Mutational signatures analysis of single-base substitutions (SBS) was performed to identify the contribution of the different mutational processes to carcinogenesis according to COSMIC database [47,48]. Mutational signatures were calculated for each sample using the R package deconstuctSigs (v2) [49]. COSMIC signatures were set as reference signatures, and the count method was set as default. Mutational signature weight ≥0.2 was considered a positive contribution to the mutational process. The "tri.counts.method" parameter in deconstructSigs was set to "exome2genome" and a custom exome trinucleotide counts file based on the target interval was provided.

Statistical Analysis
Associations between clinical characteristics such as muscular invasion, histologic grade smoking history, and sex were assessed with the presence of somatic variations and CNVs. Patients were grouped according to their clinical characteristics, then the frequency of the variants between the groups was compared using Fisher's exact test. p-values were adjusted using the Benjamini-Hochberg procedure. The Kaplan-Meier curve was analyses by Log rank test. The mutational signatures contribution was compared with the clinical characteristics of the patients using the Mann-Whitney U test. Statistical significance was set at p-value < 0.05. All statistical analyses were performed using R software (https://cran.r-project.org, accessed on 23 March 2022).

Conclusions
This is the first work in a Mexican population in which the mutational panorama of BC is characterized for both MIBC and NMIBC. Patients with MIBC showed a higher frequency of mutations in TP53 and KMT2D, gains in chr11q15.5 and chr19p13.11-q12, and losses in chr7q11. 23

Informed Consent Statement:
Due to the retrospective design of this study, informed consent was not obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available upon request to the corresponding author if you want to partner with or contribute to the project. The data are not publicly available due to Institutional Review Board policy.