Tyson, Mark D.

Abstract IntroductionCG0070 is a serotype 5 adenovirus engineered to express GM-CSF and replicate in cells with mutated or deficient RB, with response rates (RR) of approximately 45% observed in patients with recurrent NMIBC after BCG (J Urol. 2012;188:2391-7; Urol Oncol. 2018;36(10):440-7). This single arm phase 3 study (NCT04452591) was launched to confirm the clinical activity of CG0070 in patients with BCG Unresponsive NMIBC.Methods110 patients with BCG-unresponsive CIS with or without concurrent Ta or T1 disease will be treated with intravesical (IVe) CG0070 at a dose of 1x1012 vp. CG0070 will be administered as follows: induction weekly x 6 followed by weekly x 3 maintenance instillations at months 3, 6, 9, 12, and 18. Patients with persistent CIS or HG Ta at 3 months (m) may receive re-induction with weekly x 6 CG0070. Assessment of response will include q 3 m cystoscopy with biopsy of areas suspicious for disease, urine cytology, CTU/MRU, and mandatory bladder mapping at 12 m. Detection of high grade disease within the bladder will be enumerated as recurrence or non-response.ResultsThe primary endpoint of the study is CR at any time on study as assessed by biopsy (directed to cystoscopic abnormalities and mandatory mapping at 12 m), urine cytology, and radiography, as above. Secondary endpoints include CR at 12 m, duration of response, progression free survival, cystectomy free survival and safety. Correlative assessments include changes in the tumor immune microenvironment, systemic immune induction as reflected in the peripheral blood and urine, as well as viral replication and transgene expression. Baseline expression of coxsackie adenovirus receptor, E2F transcription factor as well as anti-adenovirus antibody titer will be correlated with tumor response.ConclusionThe study an ongoing trial, currently enrolling in the United States, Japan, South Korea, Australia, and Taiwan.Refs. Burke JM, Lamm DL, Meng MV, Nemunaitis JJ, Stephenson JJ, Arseneau JC, Aimi J, Lerner S, Yeung AW, Kazarian T, Maslyar DJ, McKiernan JM. A first in human phase 1 study of CG0070, a GM-CSF expressing oncolytic adenovirus, for the treatment of nonmuscle invasive bladder cancer. Citation Format: Ed Uchio, Donald Lamm, Neil Shore, Trinity Bivalacqua, Ashish Kamat, Mark Tyson, Paul Anderson, Paola Grandi, James Burke. A phase 3, single-arm study of CG0070 in subjects with non-muscle invasive bladder cancer (NMIBC) unresponsive to Bacillus Calmette-Guerin (BCG) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr CT233.

e16550 Background: Improved computational power and modern algorithms have generated significant interest in radiomics for cancer diagnosis and staging. Here we assess the performance of deep learning (DL) models as a means for feature extraction in combination with supervised machine learning (ML) algorithms for accurate staging and chemotherapy response assessment of bladder cancer. Methods: Deidentified grayscale CT images from bladder cancer patients scheduled to undergo radical cystectomy were included in this retrospective study. These images were manually annotated with two regional masks (normal region and cancer region). Five DL models- namely, AlexNet, GoogleNet, InceptionV3, ResNet-50, and XceptionNet pre-trained on the ImageNet dataset, a public dataset, were then fine-tuned on our bladder CT scan data to extract features. Through feature selection process, the subset of the features was used to build ML classifiers for classification. The classification was performed using five different ML classifiers, namely k-Nearest Neighbor (KNN), Naïve-Bayes (NB), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), and Decision Tree (DT). The classification task was performed with 10-fold cross-validation, and each of the experiments contained a different but not mutually exclusive subset of samples. The evaluation metrics include accuracy, sensitivity, specificity, precision, and F1-score. Results: A total of 200 deidentified grayscale CT images of 100 patients with histologically proven bladder cancer, were included in this study. For experiment (1) normal vs. cancer, the LDA classifier on XceptionNet based features provides the best performance with an accuracy of 86.07%, sensitivity of 96.75%, specificity of 69.65%, precision of 83.07%, and F1-score of 89.39%. For experiment (2) non-muscle invasive Cancer (NMIBC) vs. muscle invasive bladder cancer (MIBC), the LDA classifier on XceptionNet based features provided the best performance with an accuracy of 79.72%, sensitivity of 66.62%, specificity of 87.39%, precision of 75.58%, and F1-score of 70.81%. For experiment (3) T0 lesion vs. MIBC, the LDA classifier on XceptionNet based features provides the best performance with an accuracy of 74.96%, sensitivity of 80.51%, specificity of 70.22%, precision of 69.78%, and F1-score of 74.73%. Conclusions: Our proposed model has shown good results in differentiating normal vs cancer and promising performance in differentiating T0 vs MIBC after chemotherapy treatment. We are expanding our dataset to further improve the performance in differentiating T0 vs MIBC. In addition, we will investigate the applicability of GAN for data augmentation to address data limit. We believe the hybrid DL and ML framework may facilitates radiologists' decisions and clinical decision-making in patients with bladder cancer.

TPS598 Background: CG0070 is a serotype 5 adenovirus engineered to express GM-CSF and replicate in cells with mutated or deficient RB, with response rates (RR) of approximately 45% observed in patients with recurrent NMIBC after BCG (J Urol. 2012;188:2391–7; Urol Oncol. 2018;36(10):440–7). This single arm phase 3 study (NCT04452591) was launched to confirm the clinical activity of CG0070 in patients with BCG Unresponsive NMIBC. Methods: 110 patients with BCG-unresponsive CIS with or without concurrent Ta or T1 disease will be treated with intravesical (IVe) CG0070 at a dose of 1x1012 vp. CG0070 will be administered as follows: induction weekly x 6 followed by weekly x 3 maintenance instillations at months 3, 6, 9, 12, and 18. Patients with persistent CIS or HG Ta at 3 months (m) may receive re-induction with weekly x 6 CG0070. Assessment of response will include q 3 m cystoscopy with biopsy of areas suspicious for disease, urine cytology, CTU/MRU, and mandatory bladder mapping at 12 m. Detection of high grade disease within the bladder will be enumerated as recurrence or non-response. The primary endpoint of the study is CR at any time on study as assessed by biopsy (directed to cystoscopic abnormalities and mandatory mapping at 12 m), urine cytology, and radiography, as above. Secondary endpoints include CR at 12 m, duration of response, progression free survival, cystectomy free survival and safety. Correlative assessments include changes in the tumor immune microenvironment, systemic immune induction as reflected in the peripheral blood and urine, as well as viral replication and transgene expression. Baseline expression of coxsackie adenovirus receptor, E2F transcription factor as well as anti-adenovirus antibody titer will be correlated with tumor response. Study enrollment globally is ongoing including in North America, Taiwan, Japan, and South Korea. Refs. Burke JM, Lamm DL, Meng MV, Nemunaitis JJ, Stephenson JJ, Arseneau JC, Aimi J, Lerner S, Yeung AW, Kazarian T, Maslyar DJ, McKiernan JM. A first in human phase 1 study of CG0070, a GM-CSF expressing oncolytic adenovirus, for the treatment of nonmuscle invasive bladder cancer. Clinical trial information: NCT04452591.

482 Background: Prostate radiotherapy is associated with worse oncologic outcomes in patients with bladder cancer. This is thought to be secondary to microvascular damage resulting in propagation of inflammation. There is sparse data on the effect of prior radiotherapy on the efficacy of intravesical Bacillus Calmette-Guerin (BCG) in patients with non-muscle invasive bladder cancer (NMIBC). We sought to evaluate BCG response and outcomes in NMIBC patients who have previously undergone prostate radiotherapy compared to patients with no prior history of pelvic radiotherapy. Methods: This is a retrospective cohort study that includes patients who received intravesical BCG for NMIBC at our institution from 2001 to 2019. Data were collected on patient and pathologic characteristics. Patients were stratified into three cohorts: prior radiotherapy (RT), radical prostatectomy (RP), and no prostate cancer (No PCa). Wilcoxon rank sum test and Chi-squared test were used for comparison between groups. The primary endpoints included recurrence at one-year, progression to MIBC and progression to metastatic disease. Results: In 199 total patients who underwent BCG for NMIBC, 23 had a prior history of radiotherapy treatment for prostate cancer, while 17 underwent prior radical prostatectomy. Overall, 41.2% of patients had recurrence at one year. There was no difference in recurrence at one year between the cohorts: 42.8% No PCa, 29.4% RP and 39.1% RT; p = 0.56. There was also no difference in progression to MIBC (6.9% No PCA, 0% RP, and 8.7% RT) or progression to metastatic disease (6.9% No PCA, 5.9% RP and 4.3% RT) with p = 0.50 and 0.89 respectively. Conclusions: The efficacy of intravesical BCG does not appear to be modified by prior radiation treatment for prostate cancer in patients with NMIBC.[Table: see text]

Objectives To evaluate the role of blue‐light cystoscopy (BLC) in detecting invasive tumours that were not visible on white‐light cystoscopy (WLC). Patients and Methods Using the multi‐institutional Cysview registry database, patients who had at least one white‐light negative (WL–)/blue‐light positive (BL+) lesion with invasive pathology (≥T1) as highest stage tumour were identified. All WL–/BL+ lesions and all invasive tumours in the database were used as denominators. Relevant baseline and outcome data were collected. Results Of the 3514 lesions (1257 unique patients), 818 (23.2%) lesions were WL−/BL+, of those, 55 (7%) lesions were invasive (48 T1, seven T2; 47 unique patients) including 28/55 (51%) de novo invasive lesions (26 unique patients). In all, 21/47 (45%) patients had WL−/BL+ concommitant carcinoma in situ and/or another T1 lesions. Of 22 patients with a WL−/BL+ lesion who underwent radical cystectomy (RC), high‐risk pathological features leading to RC was only visible on BLC in 18 (82%) patients. At time of RC, 11/22 (50%) patients had pathological upstaging including four (18%) with node‐positive disease. Conclusions A considerable proportion of invasive lesions are only detectable by BLC and the rate of pathological upstaging is significant. Our present findings suggest an additional benefit of BLC in the detection of invasive bladder tumours that has implications for treatment approach.