679 Background: In view of the efficacy of FGFR targeting in early and advanced bladder cancer, as has been shown for erdafitinib in the THOR and NORSE trial, molecular testing of FGFR mutations and fusions will soon become clinical routine worldwide. Therascreen FGFR testing has been approved as companion diagnostic to select patients for erdafitinib. However, technical problems and shortages as well as a limited number of detectable mutations indicate the need of FGFR test systems that can be performed easily & fast worldwide. The objective was to evaluate concordance of UroTyper FGFR with Therascreen FGFR mutation testing and unravel molecular characteristics of FGFR mutated tumors to assess potential co-targeting options for erdafitinib to justify subsequent adopted clinical trials with the Bladder BRIDGister framework. Methods: FFPE tumor tissues from 100 TURB samples were prospectively collected as part of the Bladder BRIDGister. RNA from FFPE tissues were extracted by commercial kits to assess Therascreen FGFR & UroTyper FGFR MODAPLEX (BIOTYPE GmbH, Dresden) in conjunction with relative gene expression of subtyping markers (i.a.. PPARG), ADC targets (TROP2, NECTIN4) and CPI targets by RT-qPCR systems. Hierarchical clustering, Spearman correlation tests were done by JMP 9.0.0 (SAS software). Results: After exclusion of FGFR fusion samples, a total of 94 samples was available for comparative mutation analysis. Both PCR based FGFR tests concordantly identified S249C (n=15), R248C (n=2) and G370C (n=1) mutations, while the results for Y373C were in part discrepant with Therascreen FGFR detecting 9 mutations, while UroTyper FGFR did detect 6 mutations. Further examination of the 3 discrepant Y373C cases by SNaPshot and Uromonitor FGFR test revealed that the Therascreen FGFR test resulted in 3 false positive results. Moreover, one sample was invalid from Therascreen FGFR test. Overall, the concordance was very good and reached a kappa value of 0.92. Interestingly, FGFR3 mutations by UroTyper FGFR test were positively associated not only with FGFR3 mRNA overexpression (r=0.6699, p<0.0001), but also with TROP2 and NECTIN4 (r=0.5026 and r=0.4406, p<0.0001) indicating, that combinations of FGFR targeting with ADC targeting of TROP2 or NECTIN4 could have strong synergistic effects. Hierarchical clustering revealed a homogenous FGFR3 mutated group of tumors simultaneously co-expressing the target genes TROP2, NECTIN4 and PPARG. Conclusions: PCR-based FGFR assessment by Therascreen and UroTyper is highly concordant and enables fast, local FGFR assessment within few hours. FGFR3 mutations are associated with increased TROP2 & NECTIN4 expression indicating potential synergistic options which warrants further exploration as part of molecularly stratified clinical trials.

550 Background: Patients with muscle invasive urothelial carcinoma achieving pathological complete response (pCR) upon neoadjuvant chemotherapy (NACT) have improved prognosis. Previously we did show that luminal tumors respond better to NACT (Ecke et al 2022), while the radioligand targets CXCR4 and FAP are found in chemoresistant, stroma-associated tumors. The objective of this study was to exploit the expression of CXCR4 & FAP for theranostic imaging & treatment in selected patients by radioligand instillation into the bladder to justify subsequent adopted clinical trials within the Bladder BRIDGister framework. Methods: FFPE tumor tissues from 511 TURB samples were prospectively collected as part of the Bladder BRIDGister. RNA was extracted by commercial kits, relative gene expression of subtyping markers and radioligand target genes were analyzed by RT-qPCR. Hierarchical clustering, Kruskal-Wallis and contingency tests were done by JMP 9.0.0 (SAS software). PET/CT Imaging by CXCR4 and/or FAP instillation was performed in selected patients followed by systemic radioligand application after completion of cisplatinum based NACT to enable improved staging. Results: The prospective molecular registry cohort consisted of 511 bladder cancer patients included into the Bladder BRIDGister (median age: 75, male 74% vs. female 26%) with 5% Tis, 54% Ta, 16% T1 and 15% T2 stage and concomitant CIS being in 7%. CXCR4 expression was negatively associated with ERBB2 & FGFR3 mRNA (r=-0,3021 and r=-0,3326, p<0,0001). Inverse relation of FAP expression with ERBB2 & FGFR3 was even more pronounced (r=-0,4883 and r=-0,5177, p<0,0001). Clustering validated that CXCR4 and FAP are elevated in non-luminal tumors not responding to NACT. Exemplarily one stromal rich pT2 G3 MIBC patient with elevated FAP expression, who was predicted to be unresponsive to NACT, was selected for FDG and FAP PET/CT imaging after two cycles of NACT. Indeed metastatic lesion were detected in the liver and pancreatic tissue. FAP imaging revealed to be more sensitive than conventional FDG imaging. In a next step 68Ga-CXCR4 was instilled into the bladder of a patient with NACT-resistant MIBC rejecting cystectomy to establish a non-systemic approach for radioligand therapy. Theranostic instillation imaging revealed strong uptake into the invading MIBC and subsequent instillation therapy with 177Lu-FAP led to partial remission and increased immune cell infiltration indicating initial effectivity. Conclusions: Expression of the radioligand targets CXCR4 and FAP are associated with aggressive stromal and basal like tumors being resistant to NACT. This could be validated by selecting patients for FAP & CXCR4 PET/CT imaging. Instillation of radioligands into the bladder revealed to be safe with first signs of effectivity justifying clinical approaches as part of phase 1 / 2 clinical trials.

545 Background: Patients with muscle invasive urothelial carcinoma achieving pathological complete response (pCR) upon neoadjuvant chemotherapy (NACT) have improved prognosis. Molecular subtypes of bladder cancer differ markedly with regard to sensitivity to cisplatinum based chemotherapy. Previously we did show that luminal tumors respond better to NACT, while FGFR1 expression is associated with chemo resistance (Ecke et al. 2022). The objective of this study was to determine which patients may benefit from Antibody Drug Conjugate (ADC) treatment in addition to NACT to justify subsequent prospective analysis within the "Bladder BRIDGister". Methods: Formalin fixed paraffin embedded (FFPE) tissues from transurethral resections (TUR) before chemotherapy and cystectomy samples after NACT of 36 patients were retrospectively collected. RNA from FFPE tissues were extracted by commercial kits, relative gene expression of subtyping markers (KRT5, KRT20, FGFR1) and radioligand target genes (NECTIN4, TROP2) were analyzed by standardized RT-qPCR systems (STRATIFYER Molecular Pathology GmbH, Cologne). Spearman correlation, hierarchical clustering, Kruskal-Wallis, chi square and contingency tests were done by JMP 9.0.0 (SAS software). Results: The neoadjuvant cohort consisted of 36 patients (median age: 69, male 83% vs. female 17%) with 92% of patients being pathohistologically node negative. When comparing pretreatment with post treatment samples the median expression of KRT20 dropped 128fold, while FGFR1 expression increased 6.8 fold. Interestingly, TROP2 and NECTIN4 mRNA expression also dropped significantly upon NACT by 5.7 fold and 7.1 fold, respectively. TROP2 and NECTIN4 were positively associated with the response marker KRT20 in therapy naïve TUR biopsies (r=0.5562 p=0.0004; r=0.5833 p=0.0002), but negatively associated with the resistance marker FGFR1 (r=-0.2903 p=0,0858; r=-0.3396 p=0,0427). However, TROP2 and NECTIN4 were not associated with pCR in spearman analysis with minor trend for TROP2 (r=0,2139 p=0,2103). Cluster analysis revealed a subgroup of KRT20 positive and FGFR1 negative tumors expression TROP2 and NECTIN4, which achieved 80% pCR. In addition elevated TROP2 and NECTIN4 expression was found in KRT20 positive tumors coexpressing FGFR1 and being resistant to NACT. Conclusions: Expression of the ADC targets TROP2 and NECTIN4 is associated with KRT20 positive, luminal tumors being highly sensitive to neoadjuvant chemotherapy alone. KRT5 positive, basal tumors do exhibit only very low expression of TROP2 and NECTIN4 mRNA. In view of toxicities the addition of TROP2 and NECTIN4 treatment to NACT might be considered only in luminal tumors exhibiting elevated FGFR1 expression as resistance mechanism and therefore do not respond to NACT.

Introduction and objective: BTA stat (R) , NMP22 (R) BladderChek (R) , UBC (R) Rapid Test, and CancerCheck (R) UBC (R) rapid VISUAL are uri-nary-based rapid tests. This multicenter study is the first study comparing all available rapid tests on a large cohort of bladder cancer patients and healthy controls in one setting. Methods: In total 732 urine samples (second morning urine) in a real-world assessment have been analyzed. We evaluated clinical sam-ples from 464 patients with histologically confirmed urothelial tumors of the urinary bladder (17 solitary CIS, 189 low-grade, 187 high-grade nonmuscle invasive, 71 high-grade muscle invasive), 77 patients with No Evidence of Disease (NED), and from 191 healthy controls. Urine samples were analyzed by the BTA stat (R), NMP22 (R) BladderChek (R), UBC (R) Rapid Test point-of-care (POC) system using the concile Omega 100 POC reader, and CancerCheck (R) UBC (R) rapid VISUAL. Sensitivities and specificities were calculated by contingency analyses. Results: All investigated urinary markers detected more pathological concentrations in urine of bladder cancer patients compared to tumor-free patients. The calculated diagnostic sensitivities for BTA stat (R), NMP22 (R) BladderChek (R), UBC (R) Rapid Test, CancerCheck (R) UBC (R) rapid VISUAL, and cytology were 62.4%, 13.4%, 58.2%, 28.6%, 36.2% for low-grade, 83.4%, 49.5%, 84.5%, 63.1%, 71.2% for high-grade nonmuscle invasive, and 95.8%, 35.2%, 76.1%, 50.7%, 67.7% for high-grade muscle-invasive bladder cancer. The specificity was 67.9%, 95.5%, 79.4%, 94.4%, and 83.7%, respectively. The area under the curve (AUC) after receiver operating characteristics (ROC) analysis for high-grade non-muscle-invasive tumors was 0.757, 0.725, 0.819, 0.787, and 0.774, respectively. Conclusions: The analysis of more than 700 urine samples offers an objective view on urine-based rapid diagnostics. Elevated patholog-ical concentrations of markers in urine of bladder cancer patients were detected in all investigated tests. The highest sensitivities for high-grade non-muscle-invasive tumors were calculated for BTA stat (R) and UBC (R) Rapid Test, whereas NMP22 (R) BladderChek (R), and cytology showed the highest specificities. BTA stat (R) and UBC (R) Rapid Test have the potential to be used as a clinical valuable urinary protein bio-marker for the detection of high-grade non-muscle-invasive bladder cancer patients and could be included in the management of these tumors. (c) 2023 Elsevier Inc. All rights reserved.

e16603 Background: Patients with muscle invasive urothelial carcinoma achieving pathological complete response upon neoadjuvant chemotherapy (NACT) have improved prognosis. Previously we did show that luminal tumors respond better to NACT, while FGFR1 expression is associated with NACT resistance (Ecke et al. 2022). Interestingly the expression of the radioligand targets CXCR4 and FAP is found in chemoresistant, stroma-associated tumors. The objective of this study was to prospectively validate the predictive value of molecular target typing from TUR biopsy samples and compare PET CT imaging by FDG and FAP radioligand imaging in selected patients after two cycles of NACT to justify subsequent adaptive trial concepts within the "Bladder BRIDGister." Methods: Formalin fixed paraffin embedded tissues (FFPE) from transurethral resections (TUR) before chemotherapy and cystectomy samples after NACT of 36 patients were retrospectively collected and 650 TURB samples were prospectively collected as part of the Bladder BRIDGister. RNA from FFPE tissues were extracted by commercial kits, relative gene expression of subtyping markers (KRT5, KRT20, FGFR1) and radioligand target genes (CXCR4, FAP) were analyzed by standardized RT-qPCR systems (STRATIFYER Molecular Pathology GmbH, Cologne). PET CT Imaging by FDG and FAP was performed after two cycles cisplatinum based NACT. Results: The neoadjuvant cohort consisted of 36 patients (median age 69, male 83%, female 17%) with 92% of patients being node negative. Hierarchical clustering revealed CXCR4 and FAP to be elevated in stromal rich, KRT5 & KRT20 negative tumors not responding to NACT. Elevated FAP mRNA expression was sig. associated with resistance to NACT (chi 2 4.314 p=0,0378). Combining elevated FAP and CXCR4 mRNA expression did identify 1/3 of the patients to be at high risk of NACT resistance (90%). Exemplarily one pT3 G3 patient was selected for PET CT imaging after two cycles that was predicted to be unresponsive to NACT by molecular subtyping. FDG PET CT revealed a hepatic metastatic lesion. In contrast FAP PET CT indicated multiple hepatic and a pancreatic metastatic lesion indicating tumor progression under NACT. Therapy was switched to MVAC with persistent non response. Then pembrolizumab monotherapy was administered due to PD-L1 positivity of the initial TURB (10% TPS /CPC 15). However there still was fulminant progression of the liver metastasis so that ICI therapy had to be stopped. Conclusions: Expression of the radioligand targets CXCR4 and FAP has been shown to be associated with aggressive stromal associated tumors and being resistant to NACT. This could be validated by selecting patients after two cycles of NACT for PET/CT imaging. Stratified FAP PET/CT turned out to be more sensitive than conventional FDG PET CT and may enable future theranostic treatment combinations in patients unresponsive to standard chemo- or immunotherapies.

Patients with muscle-invasive urothelial carcinoma achieving pathological complete response (pCR) upon neoadjuvant chemotherapy (NAC) have improved prognosis. Molecular subtypes of bladder cancer differ markedly regarding sensitivity to cisplatin-based chemotherapy and harbor FGFR treatment targets to various content. The objective of the present study was to evaluate whether preoperative assessment of molecular subtype as well as FGFR target gene expression is predictive for therapeutic outcome—rate of ypT0 status—to justify subsequent prospective validation within the “BladderBRIDGister”. Formalin-fixed paraffin-embedded (FFPE) tissue specimens from transurethral bladder tumor resections (TUR) prior to neoadjuvant chemotherapy and corresponding radical cystectomy samples after chemotherapy of 36 patients were retrospectively collected. RNA from FFPE tissues were extracted by commercial kits, Relative gene expression of subtyping markers (e.g., KRT5, KRT20) and target genes (FGFR1, FGFR3) was analyzed by standardized RT-qPCR systems (STRATIFYER Molecular Pathology GmbH, Cologne). Spearman correlation, Kruskal–Wallis, Mann–Whitney and sensitivity/specificity tests were performed by JMP 9.0.0 (SAS software). The neoadjuvant cohort consisted of 36 patients (median age: 69, male 83% vs. female 17%) with 92% of patients being node-negative during radical cystectomy after 1 to 4 cycles of NAC. When comparing pretreatment with post-treatment samples, the median expression of KRT20 dropped most significantly from DCT 37.38 to 30.65, which compares with a 128-fold decrease. The reduction in gene expression was modest for other luminal marker genes (GATA3 6.8-fold, ERBB2 6.3-fold). In contrast, FGFR1 mRNA expression increased from 33.28 to 35.88 (~6.8-fold increase). Spearman correlation revealed positive association of pretreatment KRT20 mRNA levels with achieving pCR (r = 0.3072: p = 0.0684), whereas pretreatment FGFR1 mRNA was associated with resistance to chemotherapy (r = −0.6418: p < 0.0001). Hierarchical clustering identified luminal tumors of high KRT20 mRNA expression being associated with high pCR rate (10/16; 63%), while the double-negative subgroup with high FGFR1 expression did not respond with pCR (0/9; 0%). Molecular subtyping distinguishes patients with high probability of response from tumors as resistant to neoadjuvant chemotherapy. Targeting FGFR1 in less-differentiated bladder cancer subgroups may sensitize tumors for adopted treatments or subsequent chemotherapy.

Introduction: The aim of the study is to compare length of hospital stay, transfusion rates, and re-intervention rates during hospitalization for transurethral resection of the prostate (TUR-P), open prostatectomy (OP), and laser therapy (LT) for surgical treatment of benign prostatic obstruction (BPO). Methods: URO-Cert is an organization, in which clinical data of prostatic diseases from 2 university, 19 public, and 3 private hospitals and 270 office-based urologists are collected in order to document treatment quality. Data on diagnostics, therapy, and course of disease are recorded web based. The analysis includes datasets from 2005 to 2017. Results: Of 10,420 patients, 8,389 were treated with TUR-P, 1,334 with OP, and 697 with LT. Median length of hospital stay was 6 days (IQR: 4–7) for TUR-P, 9 days (IQR: 7–11) for OP, and 5 days (IQR: 4–6) for LT (p < 0.001). Risk for a hospital stay ≥7 days was higher for OP versus TUR-P (OR: 7.25; 95% CI = 6.27–8.36; p < 0.001) and LT (OR: 17.89; 95% CI = 14.12–22.65; p < 0.001) and higher for TUR-P versus LT (OR: 2.47; 95% CI = 2.03–3.01; p < 0.001). OP had a significantly higher risk for transfusions than TUR-P (OR: 2.44; 95% CI = 1.74–3.41; p < 0.001) and LT (OR: 3.32; 95% CI = 1.56–7.01; p < 0.001). Transfusion rates were not significantly different between TUR-P and LT (OR: 1.36; 95% CI = 0.66–2.79; p = 0.51). Risk of re-intervention was not different between all 3 approaches. Conclusion: OP was associated with higher transfusion rates and longer hospital stay than TUR-P and LT. Risk of transfusion was not different between TUR-P and LT, but TUR-P was inferior to LT concerning length of hospital stay. Re-intervention rates during hospitalization did not differ between the groups.