Moore, Cassandra N.

Abstract Background: Globally, Black men suffer the greatest prostate cancer (CaP) burden. Blacks are diagnosed more at advanced stages of CaP, greater morbidity and mortality and poorer survivorship outcomes. Blacks are exposed to adverse, discriminatory societal determinants. How these determinants impact the CaP disparities, in particular survivorship and health related quality of life inequities are understudied. Studies examining patient-reported outcomes of Black CaP survivors reveal heightened negative HRQOL sequelae, and depression, family, work and living situation instability due to cancer treatment and societal determinants of health (SDOH). SDOH is used mostly in Public Health to predict population health risk and outcomes, but rarely integrated into HRQOL research and practice. This project is novel by incorporating SDOH into the assessment and relief of HRQOL threats. Despite the Black CaP disparities in survivorship, there remains unacceptable lack of focus and prioritization to provide comprehensive relief. Methods: In response to this noticeable scientific gap, our requisite multidisciplinary investigatory team including Survivor-Advocate Investigators joins forces to achieve the study goal. We are employing community engaged research practice to create a CaP survivorship care plan (SCP) template targeted to Blacks. Our SCP will provide a comprehensive best-practice roadmap to document medical information with treatment history and status, along with relevant resources and health advisories to provide relief for unfavorable sequalae due to cancer and its treatments as well as improve quality of life for CaP survivors. We are building upon the Science of Survivorship, and Contextual Socioecological and the Behavioral Precision Medicine Models with full survivor-advocate partnership to create the SCP. This project to develop a SCP template employs an informative consensus panel to inform the initial SCP. Results: Based on advocate-survivor input and guidance, the SCP will include resources and tools focused on SDOH, treatment adherence, treatment side effects and symptom relief, co-occurring chronic conditions, cardio-protective strategies, and physical, emotional and social wellbeing towards improving in patient outcomes and HRQOL in Black CaP survivors. Data from the Engagement Core and Survivor-Advocate Community Advisory Board, and the preliminary SCP CaP for Black men will be presented at the meeting. Conclusion: Our Consortium builds upon the work of the multidisciplinary PIs, and provide team science approach with robust scientific methods to better understand and address the HRQOL needs of Black CaP survivors. Importantly, this research will make available a patient-centered SCP for CaP survivors, focused on Black men. The iCCaRE Consortium will use this SCP_CaP to inform the development of Artificial Intelligence interventions addressing medical, follow-up care, surveillance, social and emotional support, SDOH and health advisories that will be deployed on popular mobile platforms for greater reach and on-demand use. Citation Format: Kimlin T. Ashing, Folakemi T. Odedina, Cassandra N. Moore, Che Ngufor, Getachew A. Dagne, Fornati Bedell, Diana Londoño, John McCall, Arnold Merriweather, JoAnne S. Oliver, Rotimi Rotimi Oladapo. The iCCARE Consortium for Prostate Cancer in Black men: Creating a survivorship care plan for Black prostate cancer survivors [abstract]. In: Proceedings of the 15th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2022 Sep 16-19; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2022;31(1 Suppl):Abstract nr A033.

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.

Abstract Background Radium (Ra)‐223 is an established treatment option for patients with metastatic castrate‐resistant prostate cancer (mCRPC) who have symptomatic bone metastases without soft tissue disease. Studies have indicated genetic aberrations that regulate DNA damage response (DDR) in prostate cancer can increase susceptibility to treatments such as poly ADP‐ribose polymerase inhibitors and platinum‐based therapies. This study aims to evaluate mCRPC response to Ra‐223 stratified by tumor genomics. Methods This is a retrospective study of mCRPC patients who received Ra‐223 and genetic testing within the Mayo Clinic database (Arizona, Florida, and Minnesota) and Tulane Cancer Center. Patient demographics, genetic aberrations, treatment responses in terms of alkaline phosphatase (ALP) and prostate‐specific antigen (PSA), and survival were assessed. Primary end points were ALP and PSA response. Secondary end points were progression‐free survival (PFS) and overall survival (OS) from time of first radium treatment. Results One hundred and twenty‐seven mCRPC patients treated with Ra‐223 had germline and/or somatic genetic sequencing. The median age at time of diagnosis and Ra‐223 treatment was 61.0 and 68.6 years, respectively. Seventy‐nine (62.2%) had Gleason score ≥ 8 at time of diagnosis. 50.4% received prior docetaxel, and 12.6% received prior cabazitaxel. Notable alterations include TP53 (51.7%), BRCA 1/2 (15.0%), PTEN (13.4%), ATM (11.7%), TMPRSS2‐ERG (8.2%), RB deletion (3.4%), and CDK12 (1.9%). There was no significant difference in ALP or PSA response among the different genetic aberrations. Patients with a TMPRSS2‐ERG mutation exhibited a trend toward lower OS 15.4 months (95% confidence interval [CI] 10.0−NR) versus 26.8 months (95% CI 20.9−35.1). Patients with an RB deletion had a lower PFS 6.0 months (95% CI 1.28−NR) versus 9.0 months (95% CI 7.3−11.1) and a lower OS 13.9 months (95% CI 5.2−NR) versus 26.5 months (95% CI 19.8−33.8). Conclusions Among mCRPC patients treated with Ra‐223 at Mayo Clinic and Tulane Cancer Center, we did not find any clear negative predictors of biochemical response or survival to treatment. TMPRSS2‐ERG and RB mutations were associated with a worse OS. Prospective studies and larger sample sizes are needed to determine the impact of genetic aberrations in response to Ra‐223.

e17587 Background: Radium (Ra)-223 has become the cornerstone treatment in patients with metastatic castrate-resistant prostate cancer (mCRPC) with bone metastases. Ra-223 has been shown to suppress abnormal bone formation by inducing DNA double-strand breaks in tumor cells. This can reduce skeletal-related events, as well as improve survival and quality of life. Studies have indicated mutations that regulate DNA damage response in prostate cancer as susceptible to PARP inhibitors and platinum-based therapies. This study aims to evaluate mCRPC response to Ra-223 stratified by tumor genomics. Methods: This is a retrospective study of mCRPC patients who received Ra-223 and genetic testing within the Mayo Clinic database (Arizona, Florida, and Minnesota). Patient demographics, genetic mutations, treatment responses in terms of alkaline phosphatase (ALP) and prostate-specific antigen (PSA), and survival were assessed. Primary end points were progression-free survival (PFS) and overall survival (OS) from time of first radium treatment. Results: 239 mCRPC patients treated with Ra-223 were identified. Germline and/or somatic genetic sequencing was available in 50 patients, who had a median age of 61 years at time of diagnosis and 68 years at time of Ra-223 treatment. Median Gleason score at time of diagnosis was 8.0. 100% of patients received prior androgen deprivation therapy. 72% received prior docetaxel, and 18% received prior cabazitaxel. Notable mutations included TP53 (45.2%), TMPRSS2-ERG (28.6%), PTEN (26.2%), BRCA1/2 (14.3%), ATM (9.3%), CDK12 (7.1%), and Rb (4.9%). TMPRSS2-ERG was the second most commonly found mutation, with median OS 12.7 mo versus 15.4 mo in patients without TMPRSS2-ERG mutations (p = 0.099). In patients with CDK12 mutations (2 of 28 tested positive), PFS was 10.4 mo versus 5.6 mo in patients without CDK12 mutations (p = 0.157). The assessment of ALP and PSA response to radium treatment stratified by genetic mutation did not reach statistical significance. Conclusions: Among mCRPC patients treated with Ra-223 at Mayo Clinic, TMPRSS2-ERG was the second most commonly found genetic mutation and had observed worse survival outcomes that approached statistical significance. Prospective studies and increased sample sizes are needed to determine the impact of genetic mutations in response to Ra-223.