FGFR3 Mutations in Urothelial Carcinoma: A Single-Center Study Using Next-Generation Sequencing

Background: Mutations of fibroblast growth factor receptor 3 (FGFR3) are associated with urothelial carcinoma (UC) oncogenesis and are considered an important therapeutic target. Therefore, we evaluated the FGFR3 mutation rate and its clinical significance in urothelial carcinoma (UC) using next-generation sequencing. Methods: A total of 123 patients with UC who were treated at Chonnam National University Hospital (Gwang-ju, Korea) from January 2018 to December 2020 were enrolled. We performed NGS using the Oncomine panel with tumor specimens and blood samples corresponding to each specimen. We analyzed the FGFR3 mutation results according to the type of UC and the effects on early recurrence and progression. Results: The mean age of the patients was 71.39 ± 9.33 years, and 103 patients (83.7%) were male. Overall, the FGFR3 mutation rate was 30.1% (37 patients). The FGFR3 mutation rate was the highest in the non-muscle-invasive bladder cancer (NMIBC) group (45.1%), followed by the muscle-invasive bladder cancer (22.7%) and upper tract UC (UTUC) (14.3%) groups. Patients with FGFR3 mutations had a significantly lower disease stage (p = 0.019) but a high-risk of NMIBC (p < 0.001). Conclusions: Our results revealed that FGFR3 mutations were more prevalent in patients with NMIBC and lower stage UC and associated with a high-risk of NMIBC. Large multicenter studies are needed to clarify the clinical significance of FGFR3 mutations in UC.


Introduction
Urothelial carcinoma (UC) can be found in the upper and lower urinary tracts.The most important risk factor for developing this type of cancer is tobacco smoking, which accounts for 50% of cases [1].The majority of UCs are bladder cancer-the 10th most commonly diagnosed cancer worldwide [2]-which can be divided into two main categories, i.e., non-muscle-invasive bladder cancer (NMIBC) and muscle-invasive bladder cancer (MIBC).Approximately 75% of patients with bladder cancer are diagnosed as having NMIBC, with a high cancer-specific survival rate, whereas MIBC, which is more likely to spread to lymph nodes or other organs, has a poor prognosis [3,4].Upper tract UC (UTUC), which is uncommon and accounts for only 5-10% of cases of UC, usually exhibits an aggressive behavior [5].With recent developments in medical oncology, although platinum-based chemotherapy has been the cornerstone of therapeutic strategies for patients with advanced or metastatic UC, the survival benefits remain dismal, with a median overall survival of approximately 15 months [6].
In recent years, the therapeutic scenario has evolved with the development and approval of several immune checkpoint inhibitors [7].In addition, the rapid development of clinical genetic testing techniques such as next-generation sequencing (NGS) has facilitated a better understanding of the molecular landscape, which enables disease risk assessment and informative biomarker identification and leads to the development of novel selective genomic-targeting therapeutics for the treatment of UC patients.Therefore, the use of NGS is rapidly gaining popularity in the clinical practice of UC, similar to non-small cell lung, breast and colorectal cancers [8].
Among potentially targetable genomic alterations, mutations of fibroblast growth factor receptor 3 (FGFR3), which are one of the most common somatic mutations in UC, are associated with UC oncogenesis and are considered an important therapeutic target [6,9].As mentioned above, NGS has facilitated the identification of FGFR3 mutations in a wide range of cancers as well as UC [8].As a result, FGFR3 mutation tests are increasingly performed, leading to the development and commercialization of FGFR3-targeting therapeutics (e.g., erdafitinib) in clinical practice [9].FGFR3 mutations are generally associated with low-grade and low-stage UC and favorable disease-specific survival [10,11].However, there is also evidence suggesting that FGFR3 mutations are associated with a less favorable prognosis in advanced UC [12].Furthermore, compared with Western countries, there are limited data on FGFR3 mutations in Asian countries, especially in South Korea.Hence, in the present study, we aimed to investigate the FGFR3 mutation rate and its clinical significance in UC using NGS.

Study Population and Data Collection
The present study screened patients who visited Chonnam National University Hospital (Gwang-ju, Korea) for the treatment of UC from January 2018 to December 2020.All patients were diagnosed by histologic confirmation through transurethral resection of bladder tumor (TURBT), radical nephroureterectomy (RNU) or tumor biopsy (only for metastatic diseases).Urothelial tumors were staged according to the 2017 TNM classification of UC [13].The histologic grading of urothelial tumors was performed using the 2004/2016 World Health Organization (WHO) grading system [14].In addition, patients diagnosed with NMIBC were stratified into 4 groups according to the European Association of Urology prognostic factor risk groups [3].UC specimens for NGS analysis were obtained at the initial TURBT, RNU or tumor biopsy, and blood samples corresponding to each specimen were collected when the patients visited the outpatient department as scheduled.Patients were excluded for any of the following reasons: UC specimens not suitable for NGS analysis, histologic diagnosis other than pure transitional cell carcinoma, inability to collect blood samples or obtain informed consent or follow-up loss.A total of 123 patients were included in the final analysis based on these exclusion criteria.
A detailed medical history, including age, body mass index (BMI), sex, diabetes mellitus, hypertension, clinical stage, histologic grade, risk stratification (only if NMIBC), bladder tumor recurrence and disease progression were obtained from the patients' medical records.Patients were monitored by cystoscopy regularly every 3 to 6 months and computed tomography (CT) every 3 to 6 months according to the stage of the disease.Disease recurrence was defined as the cystoscopic detection of a new bladder tumor with histologic confirmation.Disease progression was defined as the worsening of clinical status with or without radiologic progression, consequently leading to a change in treatment.

NGS Analysis
An NGS analysis of FGFR3 mutations in UC was performed on UC specimens (formalin-fixed paraffin-embedded tissue) obtained from the initial TURBT, RNU or tumor biopsy and blood samples corresponding to each specimen collected from patients.The slides of all bladder cancer specimens were reviewed by a histopathologic specialist.Genomic DNA was extracted from 123 tumor samples and their corresponding blood samples.Genomic DNA was extracted using the QIAGEN GeneRead DNA FFPE Kit (QIAGEN, Hilden, Germany) following the manufacturer's instructions.DNA concentration was quantified using the Qubit™ ds DNA High-Sensitive Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) on the Qubit fluorometer.
Sequencing was performed with the Oncomine Comprehensive Assay v3 (OCAv3) panel (Thermo Fisher Scientific, Waltham, MA, USA).A library was prepared using the OCAv3 kit (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's manual.Multiplex PCR was performed using 20 ng of DNA.The completed library was quantified with a High Sensitivity DNA Kit (Cat.5067-4626) on a 4200 TapeStation system (Agilent, Santa Clara, CA, USA) and diluted to a final concentration of 14 pM.Diluted samples were subjected to template prep with Ion Chef XL equipment (Thermo Fisher Scientific, Waltham, MA, USA).Template-prepped samples were loaded into the Ion 530™ Chip Kit (Cat.A45850) and analyzed on the Ion S5 XL sequencing equipment.Alignment was performed using t-map (v5.10.1).Hg19 was used as the reference genome [15].
The sequenced data were processed using a series of steps.We aligned the sequenced files (FASTQ file) to the reference genome (human reference genome g1k v37) using BBmap (38.96), and sorting and indexing were performed using Samtools (samtools-1.3.1).Next, filtered alignments were further processed to improve the alignment quality, including local realignment around indels and base quality score recalibration using the Genome Analysis Toolkit (gatk-4.2.6.1).Base quality score recalibration was carried out to recalculate base quality scores for all sequenced reads based on known polymorphisms.The base and mapping quality scores were used to filter reads during variant calling, and the fine-tuning that occurs in this step is important to ensure only high-confidence variants are called [15].
Variant calling was performed in GATK-Mutect2.Mutect is a method developed for detecting the most likely somatic point mutations in NGS data using a Bayesian classifier approach.We used FilterMutectCalls of the Mutect2 pipeline for a variant filter that filters based on the probability of a somatic variant and optimizes the threshold of the "F score" by considering the average of sensitivity and precision.By performing this variant calling with tumor (bladder tissue) and normal (whole blood) samples ("tumor with matched normal" mode) on each sample, germ-line mutations were excluded, and we analyzed oncogene mutations among pure somatic mutations [15].Among them, we extracted the FGFR3 mutation data, and analyzed the FGFR3 mutation results according to the type of UC and the effects on disease recurrence and progression within 1 year.

Statistical Analysis
Statistical analysis was performed using STATA version 16.1 (StataCorp, College Station, TX, USA).Descriptive analysis was performed to assess patient demographics.Continuous variables are presented as means and standard deviations, and categorical variables are presented as frequencies (%).Student's t-test, Wilcoxon rank test, one-way ANOVA and Kruskal-Wallis rank test for continuous variables and Pearson chi-square test and Fisher's exact test for categorical variables were used to compare clinical characteristics according to the type of UC and presence of FGFR3 mutations.Logistic regression was performed to evaluate the effects on disease recurrence and progression within 1 year.A p value of <0.05 was considered indicative of statistical significance.

Ethics Statement
The study protocol was reviewed and approved by the institutional review board of Chonnam National University Hospital (IRB-approved protocol: No. CNUH-2023-254).The study was performed in accordance with the principles of the Declaration of Helsinki and the Ethical Guidelines for Clinical Studies.

Discussion
UC is a heterogeneous malignancy and has a different clinical course depending on its histopathology and location [3,4,16].In order to improve the assessment of the disease risk and prediction of the treatment response and prognosis, physicians have been focusing on advanced technologies such as genomic evaluation or artificial intelligence and gaining insights into a comprehensive cancer landscape [17].Recently, the rapid development of NGS technology has allowed researchers to obtain comprehensive genetic information on cancer by leveraging genomic data based on NGS analysis [8].Among the genomic biomarkers of UC, FGFR3 is a potential biomarker for clinical decision making in disease diagnosis and management [9,10,12,18].Therefore, in the present study, we evaluated the FGFR3 mutation rate and its clinical significance in UC using NGS, and we found that FGFR3 mutations were more prevalent in NMIBC and lower stage UC and associated with a high-risk of NMIBC.In addition, our findings suggest the potential of FGFR3 as a biomarker for UC.
FGFRs are highly conserved tyrosine kinase receptors and play essential roles in different cellular processes such as regulation of cell growth, proliferation, differentiation and death [19].The FGFR3 gene, which encodes one of the members of the FGFR family, has recently been found to be associated with UC oncogenesis, invasiveness and prognosis [20].In addition, FGFR3 mutations are one of the most common somatic mutations in UC, occurring in approximately 40-60% of cases of primary bladder and upper tract UC [10,21,22].In the present study, 30.1% of 123 UC tumors had FGFR3 mutations, which is relatively low compared with the percentage in other studies.This difference may be attributed to the varying location and stage of UC in the present study, unlike other studies.In particular, in the NMIBC group, the FGFR3 mutation rate was 45.1%.
As mentioned above, the FGFR3 mutation rate may be different according to the location and stage of UC, which is generally associated with lower disease grade and stage [10].In several studies, FGFR3 mutation rates were 49-84% in NMIBC compared with the rates (15-20%) in MIBC [10,11,21,23,24].Most NMIBCs are characterized by papillary tumors, activating FGFR3 mutations and genomic stability, in which FGFR3 mutations play a role in inducing constitutive receptor activation that functions to promote proliferation via downstream activation of the extracellular signal-regulated kinases (ERKs) [25,26].On the other hand, most MIBCs are characterized by non-papillary tumors, TP53 and RB1 inactivation and genomic instability [23,25].In addition, FGFR3 mutations are known to be mutually exclusive to some bladder cancer oncogenes, such as TP53 and RB1, which are generally associated with MIBC [9].Therefore, activating FGFR3 mutations are less common in MIBC than in NMIBC.In the present study, the FGFR3 mutation rates of the NMIBC and MIBC groups were 45.1% (23/51) and 22.7% (10/44), respectively, consistent with the findings of previous studies.
In comparison with bladder cancer, UTUC is rare and accounts for only 5-10% of all UCs [5].Consequently, studies on FGFR3 mutations in UTUC have been uncommon.Nevertheless, several studies have reported FGFR3 mutation rates in the range of 10-40% in UTUC [22,[27][28][29][30].In addition, FGFR3 mutations in UTUC may be differentiated according to the disease stage, similar to bladder cancer.In particular, Sfakianos et al. and Lyle et al. reported FGFR3 mutation rates of 30.4% and 40%, respectively, which were predominantly associated with non-muscle-invasive tumors [22,27].In contrast, Springer et al. and Lee et al. reported FGFR3 mutation rates of 21% and 13% in UTUC tissues, respectively, and the majority of UTUCs were categorized as muscle-invasive [29,30].All UTUCs included in the present study were muscle-invasive, and the FGFR3 mutation rate in the UTUC group was 14.3%, which is consistent with previous reports on FGFR3 mutations in muscle-invasive UTUC.
As a prognostic biomarker of UC, FGFR3 mutations are generally associated with a less aggressive tumor and favorable prognosis [10,11,21,24,31,32].In previous studies on NMIBC, the FGFR3 mutation rate was significantly higher in lower grade disease than in higher grade disease, which was associated with favorable outcomes [10,21,24].In a study on the role of FGFR3 mutations in primary T1 tumors, better progression-free survival was observed among patients with bladder cancer harboring FGFR3 mutations [31].In addition, studies on the molecular subtypes of bladder cancer revealed that FGFR3 mutations may be associated with the luminal-papillary subtype, which itself is associated with a less aggressive phenotype and significantly longer overall survival compared with other subtypes [9,32].Even in cases of MIBC, FGFR3 mutations were associated with favorable pathologic features and longer disease-specific survival [11].However, despite reports on the association of FGFR3 mutations with favorable characteristics, there is also evidence suggesting that FGFR3 mutations may be associated with less favorable outcomes in UC.Particularly in the context of tumor recurrence, FGFR3 mutations may be strongly associated with a high-risk of recurrence in stage Ta tumors [21].Although the underlying mechanism is unclear, bladder cancer development is thought to involve a tumoral epithelial-to-mesenchymal transition [33].In the present study, FGFR3 mutations were associated with lower stage UC but a high-risk of NMIBC.Taken together, the findings suggest that FGFR3 may be one of the critical biomarkers of UC, and further studies will be needed to evaluate its clinical significance and apply it to actual clinical practice.
The present study is not without its limitations.First, this small-scale study only included patients from a single institution.In addition, the follow-up period of the present study is relatively short; we evaluated the effects on disease recurrence and progression within 1 year.Second, all UTUCs included in the present study were muscle-invasive, which may have a selection bias.Therefore, further studies with a longer follow-up period using more UC samples from multiple institutions are needed to clarify the precise role of FGFR3 mutations.Third, our targeted sequencing approach for cancer sample tissues could be limited by biopsy bias and make it difficult to evaluate precise evolutionary relationships over time.Lastly, we did not consider the effect of adjuvant treatment.As adjuvant treatment can result in temporal changes in the genomic environment of UC, further studies on recurrent or progressed UC samples are needed to account for this aspect.
Nevertheless, our results indicated that FGFR3 may be an important biomarker of UC.Considering the results of this study, current strategies for the diagnosis of UC via conventional imaging and urine cytology may have a poor detection rate; thus, FGFR3 mutation analysis should be considered for diagnosis and disease monitoring.In addition, although the mainstay of therapeutic approaches for the treatments of advanced or metastatic UC has so far been platinum-based chemotherapies, the development of FGFR3-targeting therapeutics (e.g., erdafitinib) will represent a milestone for the treatment of UC patients with FGFR mutation.Moreover, uncovering the complete mechanisms of FGFR3-targeting agent could provide advanced therapeutic strategies to increase efficacy, such as the combination with other immune checkpoint inhibitors or anti-neoplastic drugs.Lastly, although FGFR3-targeted therapy is performed in clinical practice for patients with

Table 1 .
Demographic characteristics of patients.

Table 2 .
Comparisons of clinical features according to the type of UC.

Table 3 .
Comparison of clinical features according to the presence of FGFR3 mutations.

Table 4 .
Clinical factors associated with the recurrence of NMIBC within 1 year.

Table 5 .
Clinical factors associated with the progression of MIBC and UTUC within 1 year.

Table 6 .
Clinical factors associated with bladder tumor recurrence in UTUC within 1 year.