Whole-genome Mutational Analysis for Tumor-informed Detection of Circulating Tumor DNA in Patients with Urothelial Carcinoma.

BACKGROUND AND OBJECTIVE: Circulating tumor DNA (ctDNA) can be used for sensitive detection of minimal residual disease (MRD). However, the probability of detecting ctDNA in settings of low tumor burden is limited by the number of mutations analyzed and the plasma volume available. We used a whole-genome sequencing (WGS) approach for ctDNA detection in patients with urothelial carcinoma. METHODS: We used a tumor-informed WGS approach for ctDNA-based detection of MRD and evaluation of treatment responses. We analyzed 916 longitudinally collected plasma samples from 112 patients with localized muscle-invasive bladder cancer who received neoadjuvant chemotherapy (NAC) before radical cystectomy. Recurrence-free survival (primary endpoint), overall survival, and ctDNA dynamics during NAC were assessed. KEY FINDINGS AND LIMITATIONS: We found that WGS-based ctDNA detection is prognostic for patient outcomes with a median lead time of 131 d over radiographic imaging. WGS-based ctDNA assessment after radical cystectomy identified recurrence with sensitivity of 91% and specificity of 92%. In addition, genomic characterization of post-treatment plasma samples with a high ctDNA level revealed acquisition of platinum therapy-associated mutational signatures and copy number variations not present in the primary tumors. The sequencing depth is a limitation for studying tumor evolution. CONCLUSIONS AND CLINICAL IMPLICATIONS: Our results support the use of WGS for ultrasensitive ctDNA detection and highlight the possibility of plasma-based tracking of tumor evolution. WGS-based ctDNA detection represents a promising option for clinical use owing to the low volume of plasma needed and the ease of performing WGS, eliminating the need for personalized assay design. PATIENT SUMMARY: Detection of tumor DNA in blood samples from patients with cancer of the urinary tract is associated with poorer outcomes. Disease recurrence after surgery can be identified by the presence of tumor DNA in blood before it can be detected on radiography scans.

Field Cancerization Is Associated with Tumor Development, T-cell Exhaustion, and Clinical Outcomes in Bladder Cancer.

BACKGROUND: Field cancerization is characterized by areas of normal tissue affected by mutated clones. Bladder field cancerization may explain the development and recurrence of bladder cancer and may be associated with treatment outcomes. OBJECTIVE: To investigate the predictive and prognostic roles of field cancerization in patients with high-risk non-muscle-invasive bladder cancer (NMIBC) treated with bacillus Calmette-Guérin (BCG). DESIGN, SETTING, AND PARTICIPANTS: We conducted comprehensive genomic and proteomic analyses for 751 bladder biopsies and 234 urine samples from 136 patients with NMIBC. The samples were collected at multiple time points during the disease course. Field cancerization in normal-appearing bladder biopsies was measured using deep-targeted sequencing and error correction models. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Endpoints included the rates of recurrence and progression. Cox regression and Wilcoxon rank-sum and Fisher's exact tests were used. RESULTS AND LIMITATIONS: A high level of field cancerization was associated with high tumor mutational burden (p = 0.007), high tumor neoantigen load (p = 0.029), and high tumor-associated CD8 T-cell exhaustion (p = 0.017). In addition, high field cancerization was associated with worse short-term outcomes (p = 0.029). Nonsynonymous mutations in bladder cancer-associated genes such as KDM6A, ARID1A, and TP53 were identified as early disease drivers already found in normal-appearing bladder biopsies. Urinary tumor DNA (utDNA) levels reflected the bladder tumor burden and originated from tumors and field cancerization. High levels of utDNA after BCG were associated with worse clinical outcomes (p = 0.027) and with disease progression (p = 0.003). High field cancerization resulted in high urinary levels of proteins associated with angiogenesis and proliferation. Limitations include variation in the number of biopsies and time points analyzed. CONCLUSIONS: Field cancerization levels are associated with tumor development, immune responses, and clinical outcomes. utDNA measurements can be used to monitor disease status and treatment response. PATIENT SUMMARY: Molecular changes in the tissue lining the bladder result in tumor recurrence. Urinary measurements may be used to monitor bladder cancer status and treatment responses.

Elevated T-cell Exhaustion and Urinary Tumor DNA Levels Are Associated with Bacillus Calmette-Guérin Failure in Patients with Non-muscle-invasive Bladder Cancer.

BACKGROUND: The functional status of immune cells in the tumor microenvironment and tumor characteristics may explain bacillus Calmette-Guérin (BCG) failure in high-risk non-muscle-invasive bladder cancer (NMIBC). OBJECTIVE: To characterize molecular correlates of post-BCG high-grade (HG) recurrence using multiomics analysis. DESIGN, SETTING, AND PARTICIPANTS: Patients with BCG-treated NMIBC (n = 156) were included in the study. Metachronous tumors were analyzed using RNA sequencing (n = 170) and whole-exome sequencing (n = 195). Urine samples were analyzed for immuno-oncology-related proteins (n = 190) and tumor-derived DNA (tdDNA; n = 187). OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: The primary endpoint was post-BCG HG recurrence. Cox regression and Wilcoxon rank-sum, t, and Fisher's exact tests were used for analyses. RESULTS AND LIMITATIONS: BCG induced activation of the immune system regardless of clinical response; however, immunoinhibitory proteins were observed in the urine of patients with post-BCG HG recurrence (CD70, PD1, CD5). Post-BCG HG recurrence was associated with post-BCG T-cell exhaustion (p = 0.002). Pre-BCG tumors from patients with post-BCG T-cell exhaustion had high expression of genes related to cell division and immune function. A high predicted post-BCG exhaustion score for pre-BCG tumors was associated with worse post-BCG HG recurrence-free survival (HGRFS; p = 0.002). This was validated in independent cohorts. Pre-BCG class 2a and 2b tumors (UROMOL2021 scheme) were associated with worse post-BCG HGRFS (p = 0.015). Post-BCG exhaustion was observed in patients with high pre-BCG neoantigen load (p = 0.017) and MUC4 mutations (p = 0.002). Finally, the absence of post-BCG tdDNA clearance identified patients at high risk of recurrence (p = 0.018). The retrospective design and partial overlap for analyses are study limitations. CONCLUSIONS: Post-BCG HG recurrence may be caused by T-cell exhaustion. Tumor subtype and pre-BCG tumor characteristics may identify patients at high risk of post-BCG HG recurrence. Urinary measurements have potential for real-time assessment of treatment response. PATIENT SUMMARY: A dysfunctional immune response to bacillus Calmette-Guérin (BCG) therapy may explain high-grade recurrences of bladder cancer.

Genome-wide Meta-analysis Identifies Novel Genes Associated with Recurrence and Progression in Non-muscle-invasive Bladder Cancer.

BACKGROUND: Non-muscle-invasive bladder cancer (NMIBC) is characterized by frequent recurrences and a risk of progression in stage and grade. Increased knowledge of underlying biological mechanisms is needed. OBJECTIVE: To identify single nucleotide polymorphisms (SNPs) associated with recurrence-free (RFS) and progression-free (PFS) survival in NMIBC. DESIGN, SETTING, AND PARTICIPANTS: We analyzed outcome data from 3400 newly diagnosed NMIBC patients from the Netherlands, the UK, Canada, and Spain. We generated genome-wide germline SNP data using Illumina OmniExpress and Infinium Global Screening Array in combination with genotype imputation. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Cohort-specific genome-wide association studies (GWASs) for RFS and PFS were performed using a Cox proportional hazard model. Results were combined in a fixed-effect inverse-variance weighted meta-analysis. Candidate genes for the identified SNP associations were prioritized using functional annotation, gene-based analysis, expression quantitative trait locus analysis, and transcription factor binding site databases. Tumor expression levels of prioritized genes were tested for association with RFS and PFS in an independent NMIBC cohort. RESULTS AND LIMITATIONS: This meta-analysis revealed a genome-wide significant locus for RFS on chromosome 14 (lead SNP rs12885353, hazard ratio [HR] C vs T allele 1.55, 95% confidence interval [CI] 1.33-1.82, p = 4.0 × 10-8), containing genes G2E3 and SCFD1. Higher expression of SCFD1 was associated with increased RFS (HR 0.70, 95% CI 0.59-0.84, pFDR = 0.003). Twelve other loci were suggestively associated with RFS (p < 10-5), pointing toward 18 additional candidate genes. For PFS, ten loci showed suggestive evidence of association, indicating 36 candidate genes. Expression levels of ten of these genes were statistically significantly associated with PFS, of which four (IFT140, UBE2I, FAHD1, and NME3) showed directional consistency with our meta-analysis results and published literature. CONCLUSIONS: In this first prognostic GWAS in NMIBC, we identified several novel candidate loci and five genes that showed convincing associations with recurrence or progression. PATIENT SUMMARY: In this study, we searched for inherited DNA changes that affect the outcome of non-muscle-invasive bladder cancer (NMIBC). We identified several genes that are associated with disease recurrence and progression. The roles and mechanisms of these genes in NMIBC prognosis should be investigated in future studies.

SPTAN1, APC, and FGFR3 Mutation Status and APOBEC Mutation Signatures are Predictive of Mitomycin C Response in Non-muscle-invasive Bladder Cancer.

BACKGROUND: Currently, no biomarkers of response to mitomycin C have been identified in non-muscle-invasive bladder cancer patients. Predictive biomarkers could improve the treatment outcome and eliminate adverse events from unnecessary treatment. OBJECTIVE: To identify and validate predictive biomarkers of chemoresection with mitomycin C. DESIGN SETTING AND PARTICIPANTS: The intervention group of a randomised controlled trial was identified for analyses. The study was conducted between January 2018 and June 2019 in two major urological departments in Denmark. Patients had a history of Ta low-grade/high-grade disease and were included upon recurrence. The intervention group (58 patients) received chemoresection with mitomycin C. Tumour and reference germline DNA from prior tumours were analysed by whole exome sequencing. Predictive biomarkers were validated in the context of Ta low-grade tumours from the UROMOL study. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Response to chemotherapy (intervention group from the randomised controlled trial) and recurrence-free survival (UROMOL cohort) were measured. Groups were compared using Fisher's exact test and Wilcoxon rank sum test. RESULTS AND LIMITATIONS: Chemoresponse was associated with the mutation status of SPTAN1, APC, and FGFR3, and the level of APOBEC signature contribution (p = 0.035, p = 0.034, p = 0.055, and p = 0.035, respectively). The main limitations include no biopsy for biomarker discovery immediately prior to chemoresection and the unmatched validation cohort. CONCLUSIONS: Mutation status of APC, SPTAN1, and FGFR3 and the level of mutational contribution from APOBEC-related signatures were identified as potential predictive biomarkers for chemoresection with mitomycin C in non-muscle-invasive bladder cancer patients. A prospective validation study is however needed. PATIENT SUMMARY: We investigated DNA from noninvasive bladder tumours in order to predict treatment response to chemotherapy. Four biomarkers showed promising results, which should be tested in future studies.

An integrated multi-omics analysis identifies prognostic molecular subtypes of non-muscle-invasive bladder cancer.

The molecular landscape in non-muscle-invasive bladder cancer (NMIBC) is characterized by large biological heterogeneity with variable clinical outcomes. Here, we perform an integrative multi-omics analysis of patients diagnosed with NMIBC (n = 834). Transcriptomic analysis identifies four classes (1, 2a, 2b and 3) reflecting tumor biology and disease aggressiveness. Both transcriptome-based subtyping and the level of chromosomal instability provide independent prognostic value beyond established prognostic clinicopathological parameters. High chromosomal instability, p53-pathway disruption and APOBEC-related mutations are significantly associated with transcriptomic class 2a and poor outcome. RNA-derived immune cell infiltration is associated with chromosomally unstable tumors and enriched in class 2b. Spatial proteomics analysis confirms the higher infiltration of class 2b tumors and demonstrates an association between higher immune cell infiltration and lower recurrence rates. Finally, the independent prognostic value of the transcriptomic classes is documented in 1228 validation samples using a single sample classification tool. The classifier provides a framework for biomarker discovery and for optimizing treatment and surveillance in next-generation clinical trials.

Molecular correlates of cisplatin-based chemotherapy response in muscle invasive bladder cancer by integrated multi-omics analysis.

Overtreatment with cisplatin-based chemotherapy is a major issue in the management of muscle-invasive bladder cancer (MIBC), and currently none of the reported biomarkers for predicting response have been implemented in the clinic. Here we perform a comprehensive multi-omics analysis (genomics, transcriptomics, epigenomics and proteomics) of 300 MIBC patients treated with chemotherapy (neoadjuvant or first-line) to identify molecular changes associated with treatment response. DNA-based associations with response converge on genomic instability driven by a high number of chromosomal alterations, indels, signature 5 mutations and/or BRCA2 mutations. Expression data identifies the basal/squamous gene expression subtype to be associated with poor response. Immune cell infiltration and high PD-1 protein expression are associated with treatment response. Through integration of genomic and transcriptomic data, we demonstrate patient stratification to groups of low and high likelihood of cisplatin-based response. This could pave the way for future patient selection following validation in prospective clinical trials.

Identification of Differential Tumor Subtypes of T1 Bladder Cancer.

Stage T1 bladder cancers have the highest progression and recurrence rates of all non-muscle-invasive bladder cancers (NMIBCs). Most T1 cancers are treated with bacillus Calmette-Guérin (BCG), but many will progress or recur, and some T1 patients will die from bladder cancer. Particularly aggressive tumors could be treated with early cystectomy. To better understand the molecular heterogeneity of T1 cancers, we performed transcriptome profiling and unsupervised clustering, and identified five consensus subtypes of T1 tumors treated with repeat transurethral resection (reTUR) and induction and maintenance BCG. The T1-LumGU subtype was associated with carcinoma in situ (CIS; six/13, 46% of all CIS), had high E2F1 and EZH2 expression, and was enriched in E2F target and G2M checkpoint hallmarks. The T1-Inflam subtype was inflamed and infiltrated with immune cells. While most T1 tumors were classified as luminal papillary, the T1-TLum subtype had the highest median luminal papillary score and FGFR3 expression, no recurrence events, and the fewest copy number gains. T1-Myc and T1-Early subtypes had the most recurrences (14/30 within 24 mo), the highest median MYC expression, and, when combined, had significantly worse recurrence-free survival than the other three subtypes. T1-Early had five (38%) recurrences within the first 6 mo of BCG, and repressed IFN-α and IFN-γ hallmarks and inflammation. We developed a single-patient T1 classifier and validated our subtype biology in a second cohort of T1 tumors. Future research will be necessary to validate the proposed T1 subtypes and to determine if therapies can be individualized for each subtype. PATIENT SUMMARY: We identified and characterized expression subtypes of high-grade stage T1 bladder cancer that are biologically heterogeneous and have variable responses to bacillus Calmette-Guérin treatment. We validated the subtypes and describe a single-patient classifier.

Early Detection of Metastatic Relapse and Monitoring of Therapeutic Efficacy by Ultra-Deep Sequencing of Plasma Cell-Free DNA in Patients With Urothelial Bladder Carcinoma.

PURPOSE: Novel sensitive methods for early detection of relapse and for monitoring therapeutic efficacy may have a huge impact on risk stratification, treatment, and ultimately outcome for patients with bladder cancer. We addressed the prognostic and predictive impact of ultra-deep sequencing of cell-free DNA in patients before and after cystectomy and during chemotherapy. PATIENTS AND METHODS: We included 68 patients with localized advanced bladder cancer. Patient-specific somatic mutations, identified by whole-exome sequencing, were used to assess circulating tumor DNA (ctDNA) by ultra-deep sequencing (median, 105,000×) of plasma DNA. Plasma samples (n = 656) were procured at diagnosis, during chemotherapy, before cystectomy, and during surveillance. Expression profiling was performed for tumor subtype and immune signature analyses. RESULTS: Presence of ctDNA was highly prognostic at diagnosis before chemotherapy (hazard ratio, 29.1; P = .001). After cystectomy, ctDNA analysis correctly identified all patients with metastatic relapse during disease monitoring (100% sensitivity, 98% specificity). A median lead time over radiographic imaging of 96 days was observed. In addition, for high-risk patients (ctDNA positive before or during treatment), the dynamics of ctDNA during chemotherapy was associated with disease recurrence (P = .023), whereas pathologic downstaging was not. Analysis of tumor-centric biomarkers showed that mutational processes (signature 5) were associated with pathologic downstaging (P = .024); however, no significant correlation for tumor subtypes, DNA damage response mutations, and other biomarkers was observed. Our results suggest that ctDNA analysis is better associated with treatment efficacy compared with other available methods. CONCLUSION: ctDNA assessment for early risk stratification, therapy monitoring, and early relapse detection in bladder cancer is feasible and provides a basis for clinical studies that evaluate early therapeutic interventions.