The molecular landscape in non-muscle-invasive bladder cancer (NMIBC) is characterized by large biological heterogeneity with variable clinical outcomes. Here, we performed a large integrative multi-omics analysis of patients diagnosed with NMIBC (n=834). Transcriptomic analysis identified four classes (1, 2a, 2b and 3) reflecting tumor biology and disease aggressiveness. Both transcriptome-based subtyping and the level of chromosomal instability provided independent prognostic value beyond established prognostic clinicopathological parameters. High chromosomal instability, p53-pathway disruption and APOBEC-related mutations were significantly associated with transcriptomic class 2a and poor outcome. RNA-derived immune cell infiltration was associated with chromosomally unstable tumors and enriched in class 2b. Spatial proteomics analysis confirmed the higher infiltration of class 2b tumors and demonstrated an association between higher immune cell infiltration and lower recurrence rates. Finally, a single-sample classification tool was built and the independent prognostic value of the transcriptomic classes was documented in 1306 validation samples. The classifier provides a framework for novel biomarker discovery and for optimizing treatment and surveillance in next-generation clinical trials.

455 Background: The POUT trial (CRUK/11/027; NCT01993979) previously reported (with median follow-up 30.3 months) that adjuvant chemotherapy improves disease free survival (DFS) for patients (pts) with histologically confirmed pT2-T4 N0-3 M0 UTUC. Here we present results of a pre-planned analysis updating the primary endpoint and reporting key secondary endpoints including overall survival. Methods: 261 pts with UTUC were enrolled following nephro-ureterectomy and randomised (1:1) to 4 cycles of gemcitabine-cisplatin (gemcitabine-carboplatin if GFR 30-49ml/min) or surveillance with subsequent chemotherapy if required. Pts had 6 monthly imaging and cystoscopy for 2 years, then annually to 5 years. Toxicity was assessed by CTCAE v4. Primary endpoint was DFS. Secondary endpoints included metastasis free survival (MFS), overall survival (OS), toxicity and patient reported quality of life (QoL). The trial closed to recruitment early on advice of the independent data monitoring committee due to evidence of efficacy. Time-to-event endpoints are analysed (intention-to-treat) by Cox proportional hazard models. Unadjusted and adjusted (by nodal status, planned chemotherapy type, microscopic margin status, pathological stage) hazard ratios (HR, < 1 favouring chemotherapy) are reported. Results: From May 2012 to Nov 2017, 261 pts were recruited (129 surveillance; 132 chemotherapy) at 56 UK centres. One participant withdrew consent for data usage and was excluded from analyses. Pts had median age 69 years (range 37-88), 28% pT2, 66% pT3; 91% pN0. To 09/09/2020, median follow up was 48.1 months (IQR: 36.0-60.1). The unadjusted/adjusted HR for DFS was 0.48 (95% CI: 0.33-0.71; p = 0.0003) / 0.50 (95%CI: 0.34-0.75; p = 0.001), and for MFS was 0.52 (95% CI: 0.35-0.77; p = 0.001) / 0.54 (95% CI: 0.36-0.81; p = 0.002). 93/260 (35.8%) pts have died (52/129 [40.3%] surveillance and 41/131 [31.3%] chemotherapy). Chemotherapy conferred a non-statistically significant 28% reduction in relative risk of death (HR = 0.72, 95% CI: 0.47-1.08; p = 0.11; adjusted HR = 0.79, 95% CI: 0.52-1.19; p = 0.26). 3 year OS was surveillance: 67% (95% CI: 58-74%; chemotherapy: 79% (71%-85%). There was no evidence of long-term toxicity associated with chemotherapy (Wilcoxon rank-sum test p-value for worst grade post-6 months = 0.32). Most common grade 2+ adverse events were hypertension (25/240 [10.4%]), lethargy (25/240 [10.4%]) and hearing loss (13/240 [5.4%]). There was no evidence of statistically or clinically relevant differences in QoL. 12 months after treatment (EORTC Q30 global health status mean difference 4.1 and 4.8 at 12 and 24 months respectively in favour of chemotherapy). Conclusions: With additional follow-up, the previously reported DFS benefit for chemotherapy was maintained with no detrimental long-term toxicity. No statistically significant improvement in OS was observed. Clinical trial information: NCT01993979.

IntroductionNon-muscle-invasive bladder cancer (NMIBC) is a common and heterogeneous disease; many patients develop recurrent or progress to muscle-invasive disease. Intravesical drug therapy is a pillar in the current management of NMIBC; notwithstanding, Mitomycin C (MMC) and Bacillus Calmette-Guérin (BCG) have numerous limitations including international supply issues, and local and systemic toxicity. Here we review novel intravesical therapeutic options and drug delivery devices with potential for clinical use in the treatment of NMIBC.MethodsPubMed, ClinicalTrials.gov and Cochrane Library searches were undertaken. Systematic reviews, meta-analyses, randomised controlled trials, single-arm clinical trials and national/international conference proceedings were included.ResultsNovel intravesical drugs, including chemotherapeutic agents, immune checkpoint inhibitors, monoclonal antibodies and gene therapies, have demonstrated varying efficacy in the treatment of NMIBC. Current evidence for the majority of treatments is mostly limited to single-arm trials in patients with recurrent NMIBC. Various novel methods of drug delivery have also been investigated, with encouraging preliminary results supporting the intravesical delivery of hyperthermic MMC and MMC hydrogel formulations.ConclusionsNovel therapeutic agents and drug delivery systems will be important in the future intravesical management of NMIBC. As our understanding of the molecular diversity of NMIBC develops, molecular subtyping will become fundamental in the personalisation of intravesical treatments. Further randomised studies are urgently required to investigate the efficacy of novel intravesical treatments and novel regimens, in comparison to current standards-of-care, particularly in the context of international BCG shortages.

446 Background: The diagnostic pathways for bladder cancer are largely unchanged for 30 years. The BladderPath trial objectives are to: improve staging, accelerate treatment, and reduce iatrogenic tumour spread by avoiding TURBT. We hypothesise that substituting TURBT with mpMRI will avoid unnecessary surgery, accelerate correct MIBC treatment and improve outcomes. Methods: A randomised phase 2/3 trial of standard of care (Pathway 1) versus risk-stratified mpMRI-directed care (Pathway 2) in visually-diagnosed incident bladder cancer. Patients with bladder lesions suspicious for malignancy at flexible cystoscopy are stratified by a 5-point Likert scale: strongly agree (1) or agree (2) that the lesion is NMIBC, equivocal (3) NMIBC or MIBC, and agree (4) or strongly agree (5) the lesion is MIBC. All patients are randomised; those with probable NMIBC (Likert 1 & 2) undergo TURBT in both pathways, and those with possible MIBC (Likert 3-5) undergo TURBT (Pathway 1) or mpMRI (Pathway 2). Endpoints: feasibility, time to correct MIBC therapy and clinical progression-free survival. We report preliminary feasibility data. Results: To date, 218 potentially eligible patients have been registered; on cystoscopy, 151 patients had no bladder lesions and 5 were excluded for other reasons. Of 62 patients randomised, we report results from 45 with preliminary data. Of the 21 patients in Pathway 1, 11 patients classified as probable NMIBC underwent TURBT (pathology: 11/11 NMIBC); 9/10 patients classified as possible MIBC underwent TURBT (pathology: 4/9 NMIBC & 5/9 MIBC) and 1 patient incorrectly underwent mpMRI (staging: MIBC). Of the 24 patients in Pathway 2, 11/12 patients classified as probable NMIBC underwent TURBT (pathology: 9/11 NMIBC & 2/11 MIBC) and one patient incorrectly underwent mpMRI (staging: NMIBC); 12 patients classified as possible MIBC underwent mpMRI (staging: 6/12 NMIBC & 6/12 MIBC). Conclusions: A 5-point Likert scale accurately identifies patients with a low risk of MIBC (Likert 1 & 2). It is feasible to randomise possible MIBC patients to TURBT or mpMRI for staging. The study is ongoing to investigate the intermediate outcome of time to correct therapy for MIBC and NMIBC and the final outcome of clinical progression-free survival. Clinical trial information: 35296862.

Although there has been an overall reduction in the incidence of bladder cancer, the mortality rate has remained unchanged, hence survival is now worse. Most patients have non‐muscle‐invasive bladder cancer at presentation, which is not immediately life threatening, so efforts must be made to improve the early diagnosis and expeditious treatment of this disease.