Abstract PR04: Tumor immune microenvironment drives prognostic relevance correlating with bladder cancer subtypes

Background: Muscle-invasive bladder cancer (MIBC) represents approximately two thirds of invasive urothelial bladder cancers (UBC) and has high morbidity and mortality. Despite intensive efforts to improve patient treatment and outcome, two thirds of patients with UBC will have a recurrence or disease progression within 5 years. We conducted this study to gain further insights in the immunologic tumor microenvironment (TIME).

Material and Methods: Stromal tumor-infiltrating lymphocytes (sTILs) were scored continuously on HE slides in a cohort of 135 patients with MIBC treated by radical cystectomy (adjuvant chemotherapy n= 34) according to current recommendations (Salgado et al., 2015). In parallel, we assessed intrinsic subtypes by 21-gene Nanostring signature adapted from the MDACC-subtyping approach. Tertiary lymph structures were assessed by whole slide immunohistochemistry of CD3, CD8, CD68, and CD79a. Spatial immune profiling was carried out on regionally (tumor center, invasive margin) designed TMAs by CD3, CD8, CD56 (NK-cells), CD68, PD-1, and PD-L1 and revealed spatial organized immune phenotypes. Results were validated in 407 MIBC of the TCGA cohort by hierarchical clustering analysis, immune cell population analysis via CIBERSORT, and sTIL-scoring on digitalized HE-slides. Furthermore, tumor mutational burden, neoantigen load, and mutational patterns as well as mutational signatures were correlated with immune phenotypes in the TCGA cohort.

Results: We demonstrate that quantity and spatial distribution of sTILs within the tumor immune microenvironment (TIME) predict stages of tumor inflammation, subtypes, and patient survival and correlate with expression of immune checkpoints in an analysis of 542 MIBC. High sTILs indicate an inflamed subtype with 80% 5-year disease-specific survival. A lack of immune infiltrates identifies an uninflamed subtype with a survival rate of less than 25%. A separate immune-evading phenotype with upregulated immune checkpoints was associated with poor survival. Within the TIME are tertiary lymph node structures (TLS), which can mediate antitumor activity via active immune cells. High TLS amounts and close tumor distance correlated significantly with an inflamed phenotype and favorable survival. The uninflamed and evasion phenotypes showed lowest TLS numbers and farthest tumor distances and shortest survival. High inflammation also correlated with increased neoantigen load, high TMB, and specific mutational patterns (TCGA-MSig1, TCGA-MSig3/4). Patients treated with adjuvant chemotherapy showed a favorable prognosis dependent on high sTILs.

Conclusion: Determination of sTILs and tumor subtypes may stratify therapy success and patient survival. Considering sTILs can easily be quantified using simple morphologic parameters such as hematoxylin-eosin, sTILs can be implemented for predicting patient survival and outcome after adjuvant platinum-containing chemotherapy in a routine manner.

This abstract is also being presented as Poster A03.

Citation Format: Markus Eckstein, Carolin Pfannstiel, Katherine B. Chiappinelli, Danijel Sikic, Sven Wach, Ralph M. Wirtz, Adrian Wullweber, Helge Taubert, Johannes Breyer, Wolfgang Otto, Thomas Worst, Maximilian Burger, Bernd Wullich, Christian Bolenz, Nicole Fuhrich, Carol Geppert, Veronika Weyerer, Robert Stoehr, Simone Bertz, Bastian Keck, Franziska Erlmeier, Philipp Erben, Arndt Hartmann, Pamela Strissel, Reiner Strick. Tumor immune microenvironment drives prognostic relevance correlating with bladder cancer subtypes [abstract]. In: Proceedings of the AACR Special Conference on Bladder Cancer: Transforming the Field; 2019 May 18-21; Denver, CO. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(15_Suppl):Abstract nr PR04.

Abstract A15: Uninflamed immunologic microenvironment of muscle-invasive bladder cancer associates with activating FGFR3 gene alterations

Background and Objective: Muscle-invasive bladder cancer (MIBC) is a complex disease for which perioperative chemotherapy followed by radical cystectomy is a standard treatment. In recent years new therapy options for advanced urothelial tumors have emerged, representing a personalized medicine approach for patients. For example, targeting fibroblast growth factor receptors (FGFR) as well as restoring antitumor activity using PD-1/PD-L1 inhibitors are currently being evaluated in clinical trials and show promising results. Our present goal in this study was to assess the relationship of FGFR3-gene alterations, immune cell types, gene expression of checkpoint inhibitors like PD-L1/PD-1, and immune-related genes in MIBC to gain insight into significant differential expression.

Materials and Methods: We analyzed the TCGA cohort (n=407). Tumors were identified with activating FGFR3-gene alterations (FGFR3-altered: mutation, amplification, fusion, gain). A comparison was performed between tumors with FGFR3-altered or wild-type tumors and tumor mutational burden (TMB), neoantigen load, immune checkpoint, and immune-related gene expression as well as specific immune cell populations using the CIBERSORT-algorithm.

Results: Activating FGFR3-altered tumors showed significantly less TMB and neoantigen load compared to wild-type cases (p=0.037 and p=0.04). Tumors with high expression of PD-1, PD-L1, CTLA4, and IDO1 were predominately not FGFR3-altered (p<0.0001). Additionally, expression of immune-related genes like CD3Z, CD8A, FOXP3 and amounts of specific immune cell populations were significantly inversely associated with an FGFR3-altered status.

Conclusion: Strikingly, our analysis demonstrates that activating FGFR3 gene alterations in MIBC show low immune checkpoint and immune-related gene expression as well as low amounts of immune cells, demonstrating association with an uninflamed tumor microenvironment. Therefore, combination therapies targeting FGFR3 and, e.g., PD-1/PD-L1 might not be effective. It will be essential to validate these results in another MIBC cohort as well as unravel the role of aberrant signaling of FGFR3.

Citation Format: Veronika Weyerer, Robert Stoehr, Pamela Strissel, Reiner Strick, Christian Bolenz, Arndt Hartmann, Philipp Erben, Markus Eckstein. Uninflamed immunologic microenvironment of muscle-invasive bladder cancer associates with activating FGFR3 gene alterations [abstract]. In: Proceedings of the AACR Special Conference on Bladder Cancer: Transforming the Field; 2019 May 18-21; Denver, CO. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(15_Suppl):Abstract nr A15.

FGFR3 Mutation Status and FGFR3 Expression in a Large Bladder Cancer Cohort Treated by Radical Cystectomy: Implications for Anti-FGFR3 Treatment?†

Fibroblast growth factor receptor 3 (FGFR3) is an actionable target in bladder cancer (BC). FGFR3 mutations are common in noninvasive BC and associated with favorable BC prognosis. Overexpression was reported in up to 40% of FGFR3 wild-type muscle-invasive BC. We analyzed FGFR3 mutations, FGFR3, and p53 protein expression and assessed their prognostic value in a cohort of 1000 chemotherapy-naive radical cystectomy specimens. FGFR3 mutations were found in 11%, FGFR3 overexpression was found in 28%, and p53 overexpression was found in 69% of tumors. Among FGFR3 mutant tumors, 73% had FGFR3 overexpression versus 22% among FGFR3 wild-type tumors. FGFR3 mutations were significantly associated with lower pT stage, tumor grade, absence of carcinoma in situ, pN0, low-level p53, and longer disease-specific survival (DSS). FGFR3 overexpression was associated only with lower pT stage and tumor grade. Moreover, FGFR3 overexpression was not associated with DSS in patients with FGFR3 wild-type tumors. In conclusion, FGFR3 mutations identified patients with favorable BC at cystectomy. Our results suggest that FGFR3 mutations have a driver role and are functionally distinct from FGFR3 overexpression. Hence, patients with FGFR3 mutations would be more likely to benefit from anti-FGFR3 therapy. Ideally, further research is needed to test this hypothesis. PATIENT SUMMARY: Oncogenic fibroblast growth factor receptor 3 (FGFR3) mutations are very common in bladder cancer. In this report, we found that these FGFR3 mutations were associated with favorable features and prognosis of bladder cancer compared with patients with FGFR3 overexpressed tumors only. As a consequence, patients with FGFR3 mutant tumors would be more likely to benefit from anti-FGFR3 therapy than patients with FGFR3 protein overexpression only.

Variant morphology and random chromosomal integration of BK polyomavirus in posttransplant urothelial carcinomas

BK polyomavirus (BKPyV) causes major complications in solid organ transplant recipients but little is known about its role in the development of urothelial carcinoma (UC) during immunosuppression. Immunohistochemistry (IHC) screening for polyomavirus large T antigen (LTag) was performed in 94 micropapillary UC (MPUC), 480 unselected UC, 199 muscle invasive UC (including 83 UC with variant differentiation), 76 cases of plasmocytoid, nested and large nested UC and 15 posttransplant UC. LTag expressing UC were reevaluated regarding their histomorphological features and characterized by IHC for p53 and HER2, chromogenic in situ hybridization for HER2 and SNaPshot analysis of the TERT promoter and HRAS. Real-time PCR and next generation sequencing (NGS) were performed to search for BKPyV-DNA and for variants in the tumor and viral genomes. We detected five LTag expressing UC which were diagnosed between 2 and 18 years after kidney (n = 4) or heart (n = 1) transplantation. 89 MPUC without history of organ transplantation and overall 755 UC (including cases with variant histology) were LTag negative. Of the five LTag expressing UC, three were MPUC, one showed extensive divergent differentiation with Mullerian type clear cell carcinoma, and one displayed focal villoglandular differentiation. All five tumors had aberrant nuclear p53 expression, 2/5 were HER2-amplified, and 3/5 had TERT promoter mutations. Within the 50 most common cancer related genes altered in UC we detected very few alterations and no TP53 mutations. BKPyV-DNA was present in 5/5 UC, chromosomal integration of the BKPyV genome was detectable in 4/5 UC. Two UC with BKPyV integration showed small deletions in the BKPyV noncoding control region (NCCR). The only UC without detectable BKPyV integration had a high viral load of human herpesvirus 6 (HHV-6). Our results suggest that LTag expression of integrated BKPyV genomes and resulting p53 inactivation lead to aggressive high-grade UC with unusual, often micropapillary morphology.

A multicenter phase II trial of the combination cisplatin/ docetaxel/durvalumab/tremelimumab as single-cycle induction treatment in locally advanced HNSCC (CheckRad-CD8 trial)

6519

Background: PD-1/PD-L1 inhibitors are efficient in head and neck squamous cell cancer (HNSCC). Combination with anti-CTLA4 agents may enhance anti-tumor activity compared to anti-PD-1/PD-L1 monotherapy in different tumor types. In the CheckRad-CD8 trial the typical induction treatment consisting of Cisplatin/Docetaxel was combined with Durvalumab/Tremelimumab. Patients with pathological complete response (pCR) in the re-biopsy after induction treatment or at least 20% increase of intratumoral CD8 density in the re-biopsy compared to baseline entered radioimmunotherapy with concomitant Durvalumab/Tremelimumab. Methods: In this prospective multicenter phase II trial, patients with HNSCC stage III-IVB received a single cycle of Cisplatin 30mg/m² d1-3, Docetaxel 75mg/m² d1, Durvalumab 1500mg fix dose d5 and Tremelimumab 75mg fix dose d5. Objectives of this interim analysis were to quantify the effect of the induction treatment on intratumoral CD8 density and the pCR rate and to generate safety data. Results: Between Sep 2018 and Dec 2019, 57 patients were enrolled. Median age was 59 years, 22 patients (37%) were current smokers, 27 patients (47%) had oropharyngeal tumors (52% p16 positive). The median pre-treatment intratumoral CD8 density was 335 CD8+ cells/mm². After induction treatment 27 patients (47%) had a pCR in the re-biopsy and further 25 patients (44%) had a relevant increase of intratumoral CD8+ cells (median increase by factor 3.0). Response according to RECIST criteria was CR in 1 (2%), PR in 19 (33%) and SD in 20 patients (35%) (17 patients not evaluable). Adverse events (AE) grade 3-4 appeared in 39 patients (68%) and mainly consisted of leucopenia (43%) and infections (28%). 6 patients (11%) developed grade 3-4 immune-related AEs. In multivariable analysis the intratumoral CD8 density was the only independently significant predictor of pCR (odds ratio 1.0013 per cell/mm², 95%-CI 1.00023-1.0023, p=0.017). 42 patients (74%) continued with Durvalumab/ Tremelimumab concomitant to radiotherapy. Conclusions: Single cycle induction treatment with Cisplatin/Docetaxel/Durvalumab/Tremelimumab is feasible and achieves a high pCR rate. CD8 density may have a predictive role for further treatment planning in locally advanced HNSCC. Clinical trial information: NCT03426657 .

Use of real-world clinical lab data to reveal Asian NTRK fusion test availability patterns

e15574

Background: The development and launch of first-generation tropomyosin receptor kinase (TRK) inhibitors, such as larotrectinib and entrectinib, has brought targeted treatment options to neurotrophic tropomyosin-related kinase (NTRK) fusion–positive adult and pediatric cancer patients who may have been out of treatment options. Widespread adoption of NTRK fusion testing is needed to support successful selection of patients for these TRK inhibitors as well as other targeted therapies. The objective of this study was to investigate global availability of NTRK fusion testing, with a specific focus on Asia. Methods: For this study, real-world clinical pathology laboratory data, derived from Diaceutics’ proprietary Global Diagnostic Index (GDI) of more than 2500 labs, were investigated to assess the current footprint of NTRK fusion testing in Asia. Availability of clinical routine testing was analyzed in selected cohorts of laboratories across different regions, comparing Asia to the US and Europe. Laboratories were included based on their large volume of testing or on their importance (e.g., large academic and large commercial laboratories). Results: Data on availability of NTRK fusion testing by clinical laboratories reveal different patterns of NTRK fusion test adoption in Asia, depending on the region under study. In China and similarly in the US, Italy, and France clinical NTRK fusion testing is already being performed by the majority of laboratories from the selected cohort. In Japan, physicians have limited access to testing, with only a few laboratories from the selected cohort performing testing for NTRK fusions. Conclusions: NTRK fusion testing is essential for identifying patients eligible for first-generation TRK-inhibitors. Various laboratories across selected cohorts of labs in China and Japan lack the capability to initiate clinical NTRK fusion testing in-house, denying some patients the potential benefit from these breakthrough therapies. Consequently, patient outcomes will be suboptimal due to lack of testing infrastructure. The readiness of Asian markets to support patient selection successfully can affect the rate of testing and the impact of these innovative therapies on patients’ lives. [Table: see text]

Hrd in ovarian cancer: Defined today, evolving for the future

e18052

Background: Homologous Recombination Repair (HRR) gene mutations result in Homologous Recombination Deficiency (HRD) associated with increased risk of high grade serous ovarian (HGOC) cancer and subsequent response to PARP inhibitors (PARPi). Traditionally, HRD has been determined by testing for germline and/or somatic BRCA1/2 mutations. Today, a growing number of HRR gene mutations are known to result in HRD and genomic instability, thus being a suitable target for PARPi. Therapy response to PARPi is highest in BRCA-mutant followed by HRD+/non-BRCA-mutant HGOC. Today, no standard HRD testing methods exist, causing confusion for physicians, and leading to poor outcomes for missed PARPi eligible patients. Thus, there is need to understand HRD testing utilization and methods in HGOC to inform best practices and optimize HRD testing in the clinic. Methods: We assessed the testing landscape for determining HRD status in ovarian cancer using a data set of 8,400 newly diagnosed and metastatic ovarian cancer patients in the US from Q3-2018 through Q2-2019 identified from Diaceutics’ proprietary Global Diagnostic Index (GDI). Analysis of real-world BRCA1/2 and NGS associated testing data and laboratory profile mapping exercise of 82 US labs was carried out using Diaceutics proprietary methods and data sources to evaluate BRCA1/2 and/or HRD germline/somatic testing rates, test availability, and test panel HRR gene composition. Results: Overall, germline mutation testing rates were 3x greater than somatic testing rates. Excluding BRCA1/2, 67 labs offered comprehensive solid tumor NGS panels capable of measuring HRD with varied HRR gene target composition. Across 34 labs, 5 HRR genes were commonly found on panels: PALB2, ATM, BARD1, BRIP1 and CHEK2. 3 labs currently offering panels explicitly intended for HRD determination only include BRCA1/2 and at least one genomic instability marker (loss of heterozygosity, large-scale state transitions or telomeric allelic imbalance). Conclusions: Lack of standardized HRD panels and low testing rate identifying patients with somatic mutations in BRCA1/2 and other HRR genes is leading to poorer outcomes for missed patients eligible for PARPi’s. As clinical evidence linking HRD status with PARPi efficacy grows in ovarian as well as prostate and pancreatic cancer, Diaceutics recommends organizations such as ASCO, CAP or AMP establish defined universal HRD testing panels including relevant somatic/germline HRR genes and BRCA1/2 as well as genomic instability markers and educate stake holders aiding harmonization and ultimately, better treatment outcomes.

Collective forces of tumor spheroids in three-dimensional biopolymer networks

We describe a method for quantifying the contractile forces that tumor spheroids collectively exert on highly nonlinear three-dimensional collagen networks. While three-dimensional traction force microscopy for single cells in a nonlinear matrix is computationally complex due to the variable cell shape, here we exploit the spherical symmetry of tumor spheroids to derive a scale-invariant relationship between spheroid contractility and the surrounding matrix deformations. This relationship allows us to directly translate the magnitude of matrix deformations to the total contractility of arbitrarily sized spheroids. We show that our method is accurate up to strains of 50% and remains valid even for irregularly shaped tissue samples when considering only the deformations in the far field. Finally, we demonstrate that collective forces of tumor spheroids reflect the contractility of individual cells for up to 1 hr after seeding, while collective forces on longer timescales are guided by mechanical feedback from the extracellular matrix.

Deep Learning Predicts Molecular Subtype of Muscle-invasive Bladder Cancer from Conventional Histopathological Slides

BACKGROUND

Muscle-invasive bladder cancer (MIBC) is the second most common genitourinary malignancy, and is associated with high morbidity and mortality. Recently, molecular subtypes of MIBC have been identified, which have important clinical implications.

OBJECTIVE

In the current study, we tried to predict the molecular subtype of MIBC samples from conventional histomorphology alone using deep learning.

DESIGN, SETTING, AND PARTICIPANTS

Two cohorts of patients with MIBC were used: (1) The Cancer Genome Atlas Urothelial Bladder Carcinoma dataset including 407 patients and (2) our own cohort including 16 patients with treatment-naïve, primary resected MIBC. This resulted in a total of 423 digital whole slide images of tumor tissue to train, validate, and test the deep learning algorithm to predict the molecular subtype.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Various accuracy measurements including the area under the receiver operating characteristic curves were used to evaluate the deep learning model. A sliding window approach to visualize classification was used. Class activation maps were used to identify image features that are most relevant to call a specific class.

RESULTS AND LIMITATIONS

The deep learning model showed great performance in the prediction of the molecular subtype of MIBC patients from hematoxylin and eosin (HE) slides alone-similar to or better than pathology experts. Using different visualization techniques, we identified new histopathological features that were most relevant to our model.

CONCLUSIONS

Deep learning can be used to predict important molecular features in MIBC patients from HE slides alone, potentially improving the clinical management of this disease significantly.

PATIENT SUMMARY

In patients with bladder cancer, a computer program found changes in the appearance of tumor tissue under the microscope and used these to predict genetic alterations. This could potentially benefit patients.

FOXA1 Gene Expression for Defining Molecular Subtypes of Muscle-Invasive Bladder Cancer after Radical Cystectomy

It remains unclear how to implement the recently revealed basal and luminal subtypes of muscle-invasive bladder cancer (MIBC) into daily clinical routine and whether molecular marker panels can be reduced. The mRNA expression of basal (KRT5) and luminal (FOXA1, GATA3, KRT20) markers was measured by reverse transcription quantitative real-time polymerase chain reaction (RT-qPCR) and correlated to clinicopathological features, recurrence-free survival (RFS), disease-free survival (DFS), and overall survival (OS) in 80 patients with MIBC who underwent radical cystectomy. Additionally, the correlation of single markers with the basal and non-basal subtypes defined by a 36-gene panel was examined and then validated in the TCGA (The Cancer Genome Atlas) cohort. High expression of FOXA1 (p = 0.0048) and KRT20 (p = 0.0317) was associated with reduced RFS. In the multivariable analysis, only FOXA1 remained an independent prognostic marker for DFS (p = 0.0333) and RFS (p = 0.0310). FOXA1 expression (AUC = 0.79; p = 0.0007) was closest to the combined marker expression (AUC = 0.79; p = 0.0015) in resembling the non-basal subtype defined by the 36-gene panel. FOXA1 in combination with KRT5 may be used to distinguish the basal and non-basal subtypes of MIBC.

Pure Large Nested Variant of Urothelial Carcinoma (LNUC) Is the Prototype of an FGFR3 Mutated Aggressive Urothelial Carcinoma with Luminal-Papillary Phenotype

Since 2016, large nested urothelial carcinoma (LNUC) has been included within the WHO classification of urothelial tumors. Limited reports with mainly small case series have confirmed the malignant behavior of LNUC despite its bland morphological appearance. We evaluated, for the first time, markers for new immunooncological or targeted therapies including FGFR3 mutational status and PD-L1 status, the frequency of TERT-promoter mutations and the molecular subtype in a cohort of 25 LNUC using SNaPshot analysis and immunohistochemistry. Of the 25 cases, 17 were pure LNUC, with 13 showing an additional exophytic papillary/papillary-like component. Seven mixed LNUCs presented areas of classical nested variant urothelial carcinoma (NVUC) and one showed a component of conventional urothelial carcinoma. Of the 17 evaluable pure LNUCs, 16 were FGFR3-mutated with identical mutations in their concomitant papillary/papillary-like components. An FGFR3 mutation was found in 1/7 evaluable mixed LNUCs combined with NVUC. TERT-promoter mutations were detected in 86.7% pure and 83.3% mixed tumors. Immunohistochemistry revealed a luminal phenotype; PD-L1 was negative in the majority of tumor cells and tumor-associated immune cells. Pure LNUC is a prime example of a luminal, FGFR3-mutated, mostly PD-L1-negative tumor. In contrast, FGFR3 mutations seem to be rare in mixed LNUC, which may indicate a different pathway of tumor development.

Therapeutic implications of PD-L1 expression in bladder cancer with squamous differentiation

Background

Immune checkpoint inhibitors (ICI) are an integral part of bladder cancer therapy, however, the relevance of ICI treatment for mixed and pure squamous cell carcinoma of the bladder remains poorly studied. Therefore, we analysed the expression of programmed death-ligand 1 (PD-L1) in urothelial carcinomas with squamous differentiation (UC/SCC) and pure squamous cell carcinoma (SCC) of the bladder and studied a UC/SCC patient with ICI therapy.

Methods

Tissue microarrays of 45 UC/SCC and 63 SCC samples were immunohistochemically stained with four anti-PD-L1 antibodies (28–8, 22C3, SP142 and SP263). PD-L1 expression was determined for tumour cells (TP-Score), immune cells (IC-Score) and combined (CPS, combined positive score). In addition, we present clinical and histological data of an UC/SCC patient with nivolumab therapy.

Results

Overall, positive PD-L1 staining ranged between 4.8 and 61.9% for IC and 0 and 51.2% for TC depending on the used antibody. There were no significant differences between UC/SCC and SCC. According to current FDA guidelines for example for first line therapy of urothelial cancer with pembrolizumab (CPS ≥ 10), a subset of SCC patients up to 20% would be eligible. Finally, our UC/SCC index patient revealed excellent therapy response regarding his lung metastasis.

Conclusions

Our data reveal a PD-L1 expression in squamous differentiated carcinomas comparable with current data shown for urothelial tumours. In accordance with the encouraging clinical data of the index patient we suggest ICI treatment also for mixed and pure SCC of the urinary bladder.

EMT transcription factor ZEB1 alters the epigenetic landscape of colorectal cancer cells

Epigenetic deregulation remarkably triggers mechanisms associated with tumor aggressiveness like epithelial-mesenchymal transition (EMT). Since EMT is a highly complex, but also reversible event, epigenetic processes such as DNA methylation or chromatin alterations must be involved in its regulation. It was recently described that loss of the cell cycle regulator p21 was associated with a gain in EMT characteristics and an upregulation of the master EMT transcription factor ZEB1. In this study, in silico analysis was performed in combination with different in vitro and in vivo techniques to identify and verify novel epigenetic targets of ZEB1, and to proof the direct transcriptional regulation of SETD1B by ZEB1. The chorioallantoic-membrane assay served as an in vivo model to analyze the ZEB1/SETD1B interaction. Bioinformatical analysis of CRC patient data was used to examine the ZEB1/SETD1B network under clinical conditions and the ZEB1/SETD1B network was modeled under physiological and pathological conditions. Thus, we identified a self-reinforcing loop for ZEB1 expression and found that the SETD1B associated active chromatin mark H3K4me3 was enriched at the ZEB1 promoter in EMT cells. Moreover, clinical evaluation of CRC patient data showed that the simultaneous high expression of ZEB1 and SETD1B was correlated with the worst prognosis. Here we report that the expression of chromatin modifiers is remarkably dysregulated in EMT cells. SETD1B was identified as a new ZEB1 target in vitro and in vivo. Our study demonstrates a novel example of an activator role of ZEB1 for the epigenetic landscape in colorectal tumor cells.

Results from BLASST-1 (Bladder Cancer Signal Seeking Trial) of nivolumab, gemcitabine, and cisplatin in muscle invasive bladder cancer (MIBC) undergoing cystectomy

439

Background: Cisplatin-based neoadjuvant chemotherapy (NAC) in MIBC improves survival which correlates with pathologic response (PaR) at radical cystectomy (RC). The combination of immunotherapy and NAC may improve PaR and outcomes in MIBC. We tested the efficacy and safety of nivolumab (N) with gemcitabine-cisplatin (GC) as neoadjuvant therapy for MIBC in our phase II trial (NCT03294304). Methods: Eligible pts with MIBC (cT2-T4a, N≤1, M0) who were candidates for RC were enrolled. Pts received C (70mg/m2) IV on D1, G (1000mg/m2) on D1,D8 and N (360 mg) IV on D8 every 21 days for 4 cycles followed by RC within 8 weeks. The primary endpoint was PaR (≤pT1,N0). Secondary objectives were safety of GC+N and PFS at 2 years. The correlative objectives based on pre-treatment biopsies were correlation of PaR with 1) WGS 2) molecular subtypes of BC; 3) PD-L1 expression; 4) baseline TILs, CD3, CD8 and CD56.. Evaluable pts. should have received at least 1 dose of N. PaR will be summarized by the PaR rate as estimated by the sample proportion with exact 95% confidence intervals. We specified a null PaR of 0.35 and an alternative hypothesis of 0.55; we will reject the null hypothesis if at least 20 of 41 pts. have a PaR. Results: Between Feb 2018 and June 2019, 41 pts. were enrolled (cT2N0 90%, cT3N0 7%, cT4N1 3%); 2 patients refused surgery but were included in ITT population. PaR was observed in 27/41 pts. (65.8%), including pts with N1 disease. The combination was safe with manageable toxicities and no deaths from treatment. Majority of AEs were from GC; the overall rates of grade 3-4 AEs was 20%, majority being neutropenia, thrombocytopenia and renal insufficiency. Immune related AEs were seen in 3 patients, 2 had "adenitis" which wasymptomatic,1 pt developed Guillian Barre Syndrome after surgery, which resolved with IVIG; and none of them required steroids. There was no delay in time to RC and no unexpected surgical complications from treatment. Patients are being followed for progression and survival. Correlative work is ongoing. Conclusions: Neoadjuvant N+GC is safe and effective in MIBC with significant pathologic downstaging rates and no added toxicities or delay to surgery. Clinical trial information: NCT03294304.

Association of KRT20 and KRT5 with response to neoadjuvant chemotherapy in patients diagnosed with muscle invasive bladder cancer

562

Background: Patients with muscle-invasive bladder cancer (MIBC) that underwent neoadjuvant chemotherapy (NAC) prior to radical cystectomy (RC) show improved overall survival. Those with a pathological complete response (pCR) usually have the best prognosis. In the literature, improved response to NAC has been associated with basal tumor characteristics in MIBC so far. The aim of the present study was to examine the association of luminal (KRT20) and basal (KRT5) mRNA expression patterns at transurethral resection (TUR) with pCR at RC after NAC in a contemporary cohort of consecutive MIBC patients. Methods: Clinical Data and formalin fixed paraffin embedded tumor tissue samples from TUR and RC of 49 patients with MIBC were retrospectively analyzed. Using RT-PCR KRT20 and KRT5 mRNA expression were measured in 40-∆Ct values and normalized against the control gene CALM2. Statistical analyses comprised nonparametric and chi2 testing, partition models and spearman correlation analyses. Results: The study cohort had a median age of 63 years and consisted of 38/49 (78%) males. After NAC, 17/49 (35%) patients had cPR. Using partition models, we found that patients with high-KRT20 (≥39.5 ∆Ct) had a higher chance of pCR (57% vs. 26%, p=0.04). Using a cutoff for KRT5 at <38.1 ∆Ct within the subgroup of patients with low-KRT20 (<39.5 ∆Ct, n=35), we found poorest response among low-KRT20/low-KRT5 compared to low-KRT20/high-KRT5 and high-KRT20 (13% vs. 37% vs. 57%, p=0.29), respectively. For low-KRT20/low-KRT5, low-KRT20/high-KRT5 and high-KRT20 median KRT5 was 34.8 vs. 39.5 vs. 34.1 ∆Ct ( p=0.001) and median KRT20 was 37.9 vs. 32.9 vs. 40.1 ∆Ct,( p=0.001), respectively. Conclusions: Patients with MIBC showing high expression of KRT20 were more likely to show pCR at RC after NAC. Moreover, we were able to identify a high risk group of patients with lowKRT20/lowKRT5 that was less likely to achieve pCR at RC after NAC. Our findings are contradicting previous studies and need further verification in larger cohorts. However, our results might be useful for treatment stratification in MIBC patients.

Predictive value of fibroblast growth factor receptor (FGFR) alterations on anti-PD-(L)1 treatment outcomes in patients (Pts) with advanced urothelial cancer (UC): Pooled analysis of real-world data

493

Background: The tumor microenvironment in UC harboring FGFR gene alterations is characterized by decreased T-cell infiltration and low immune marker expression, potentially implicating suboptimal response to immune checkpoint inhibitors. The association between FGFR gene mutations/fusions and anti-PD-(L)1 treatment outcomes in advanced UC was assessed using real-world pt data. Methods: A pooled dataset of matched clinical and genomic data for advanced UC pts treated with anti-PD-(L)1 in any line from the Bladder Cancer Research Initiative for Drug Targets in Germany (BRIDGE) Consortium and UNC-CH was assessed. FGFR status was defined by a prespecified panel of FGFR2/3 mutations and fusions. Overall survival (OS) was analyzed using Kaplan-Meier estimates and Cox proportional hazards models. Multivariate analyses were performed using potential prognostic covariates (sex, age, baseline tumor stage, urothelial histology, smoking history, primary tumor location, and ECOG) in a Cox regression model for OS to assess their impact on the effect of FGFR alterations. Results: Median OS for FGFR+ pts (n=28) who received any line of anti-PD-(L)1 therapy was 9.5 mo vs 7.5 mo for FGFR− pts (n=139) (HR: 1.03, 95% CI: 0.60-1.76, p=0.93). Median OS for pts treated with first-line anti-PD-(L)1 was 5.42 mo in FGFR+ pts (n=10) and was not reached for FGFR− pts (n=31) (HR: 2.06, 95% CI: 0.68-6.24, p=0.19); median OS in second-line anti-PD-(L)1 was 6.5 mo (FGFR+; n=14) vs 5.7 mo (FGFR−; n=86) (HR: 0.89, 95% CI: 0.44-1.81, p=0.74). The multivariate analyses showed a significant trend of poorer OS in FGFR+ pts with first-line anti-PD-(L)1 (HR: 10.42, 95% CI: 1.45-74.97, p=0.02); wide CI may be attributed to small sample size for some categories in several covariates. Conclusions: Treatment with first-line anti-PD-(L)1 in FGFR+ pts may be associated with poorer OS outcomes in FGFR+ pts; however, this trend was not observed in FGFR+ pts treated with any line and second-line anti-PD-(L)1. Investigation of the predictive value of FGFR alterations to immunotherapy outcomes in larger real-world pt datasets is warranted.

Smoking status and PD-L1 mRNA-expression as a predictor of response to neoadjuvant chemotherapy in patients diagnosed with muscle invasive bladder cancer

530

Background: Patients with muscle-invasive bladder cancer (MIBC) that underwent neoadjuvant chemotherapy (NAC) prior to radical cystectomy (RC) show improved overall survival. Patients with a pathological complete response (pCR) usually have the best prognosis. In the literature, impaired response to immune checkpoint therapy has been reported in active smokers. The aim of our study was to examine the association of smoking status with pCR at RC after NAC. Moreover, we investigated the interaction of smoking status and Programmed Death Ligand 1 (PD-L1) mRNA expression at transurethral resection (TUR) and pCR prediction at RC after NAC. Methods: Clinical Data and formalin fixed paraffin embedded tumor tissue samples from TUR and RC of 49 patients with MIBC were retrospectively analyzed. Using RT-PCR PD-L1 mRNA expression was measured in 40-∆Ct values and normalized against the control gene CALM2. Smoking status was defined as never, former and active. After NAC, RC was performed and the specimens were evaluated for pCR, defined as ypT0N0M0. Statistical analyses comprised nonparametric and chi2 testing, partition models and spearman correlation analyses. Results: The study cohort had a median age of 63 years and consisted of 38/49 (78%) males. Regarding smoking status 11/49 (22%) were never, 17/49 (35%) were former and 21/49 (43%) were active smokers. After NAC, 17/49 patients (35%) had a pCR. Never/former smokers did not show a higher rate of pCR compared to active smokers (43%vs.24%, p=0.16). Comparing smoking status (never/former vs. active smokers) within the subgroup showing high PD-L1 expression (≥32.1∆Ct), a higher rate of pCR was found in never/former smokers (58% vs. 25%, p=0.047). Conclusions: Never and former smokers with MIBC that show high PD-L1 mRNA expression patterns are more likely to show pCR at RC after NAC. Smoking cessation is important for the management of MIBC patients undergoing NAC and RC.

Same same but different: A Web-based deep learning application revealed classifying features for the histopathologic distinction of cortical malformations

OBJECTIVE

The microscopic review of hematoxylin-eosin-stained images of focal cortical dysplasia type IIb and cortical tuber of tuberous sclerosis complex remains challenging. Both entities are distinct subtypes of human malformations of cortical development that share histopathological features consisting of neuronal dyslamination with dysmorphic neurons and balloon cells. We trained a convolutional neural network (CNN) to classify both entities and visualize the results. Additionally, we propose a new Web-based deep learning application as proof of concept of how deep learning could enter the pathologic routine.

METHODS

A digital processing pipeline was developed for a series of 56 cases of focal cortical dysplasia type IIb and cortical tuber of tuberous sclerosis complex to obtain 4000 regions of interest and 200 000 subsamples with different zoom and rotation angles to train a neural network. Guided gradient-weighted class activation maps (Guided Grad-CAMs) were generated to visualize morphological features used by the CNN to distinguish both entities.

RESULTS

Our best-performing network achieved 91% accuracy and 0.88 area under the receiver operating characteristic curve at the tile level for an unseen test set. Novel histopathologic patterns were found through the visualized Guided Grad-CAMs. These patterns were assembled into a classification score to augment decision-making in routine histopathology workup. This score was successfully validated by 11 expert neuropathologists and 12 nonexperts, boosting nonexperts to expert level performance.

SIGNIFICANCE

Our newly developed Web application combines the visualization of whole slide images with the possibility of deep learning-aided classification between focal cortical dysplasia IIb and tuberous sclerosis complex. This approach will help to introduce deep learning applications and visualization for the histopathologic diagnosis of rare and difficult-to-classify brain lesions.

ANLN and TLE2 in Muscle Invasive Bladder Cancer: A Functional and Clinical Evaluation Based on In Silico and In Vitro Data

Anilin actin binding protein (ANLN) and transducing-like enhancer protein 2 (TLE2) are associated with cancer patient survival and progression. The impact of their gene expression on progression-free survival (PFS) of patients with muscle invasive bladder cancer (MIBC) treated with radical cystectomy (RC) and subtype association has not yet been investigated. qRT-PCR was used to measure the transcript levels of ANLN and TLE2 in the Mannheim cohort, and validated in silico by The Cancer Genome Atlas (TCGA) cohort. Uni- and multivariate Cox regression analyses identified predictors for disease-specific survival (DSS) and overall survival (OS). In the Mannheim cohort, tumors with high ANLN expression were associated with lower OS and DSS, while high TLE2 expression was associated with a favorable OS. The TCGA cohort confirmed that high ANLN and low TLE2 expression was associated with shorter OS and disease-free survival (DFS). In both cohorts, multivariate analyses showed ANLN and TLE2 expression as independent outcome predictors. Furthermore, ANLN was more highly expressed in cell lines and patients with the basal subtype, while TLE2 expression was higher in cell lines and patients with the luminal subtype. ANLN and TLE2 are promising biomarkers for individualized bladder cancer therapy including cancer subclassification and informed MIBC prognosis.

PD-L1 assessment in urothelial carcinoma: a practical approach

Five programmed death-1 (PD-1)/programmed death-ligand 1 (PD-L1) inhibitors are currently approved for treatment of locally advanced or metastatic urothelial carcinoma of the bladder and the upper urinary tract. Due to restrictions by the FDA and EMA first-line treatment with Atezolizumab and Pembrolizumab in platinum-ineligible patients requires immunohistochemical PD-L1 testing. In the second-line setting all drugs are approved without PD-L1 testing. Used PD-L1 assays in clinical trials include the 28-8 pharmDx (Nivolumab), the 22C3 pharmDx (Pembrolizumab), Ventana SP142 (Atezolizumab), and the Ventana PD-L1 SP263 assays (Durvalumab). Differences in antibodies, needed platforms and testing algorithms have raised questions about interchangeability and comparability among these assays and their diagnostic use. We provide a practical review about the current recommendations, used assays and algorithms of PD-L1 testing in urothelial carcinoma to help oncologists, urologists and pathologists to understand analytical features, differences in antibody assays, differences in scoring algorithms and comparability of various PD-L1 assays. We reviewed and summarized published studies from the last four years (2016-2019) on PD-L1 testing in bladder cancer and present a condensed practical guideline including pre-analytical, analytical and test-specific issues.