UROFAIR Virtual 2020

© The Canadian Journal of Urology TM : International Supplement, July 2020 Percutaneous Nephrolitotomy for Calyx Inferior Stones: Stone Burdenwill Affect? Stone Burden as Predisposition Factor in Calyx Inferior Stones by Supine Percutaneous Nephrolitotomy Jufriady Ismy (1) (1) Medical Faculty Syiah Kuala University Introduction and Objectives: To compare the stone free rates of inferior calyceal stones with different stone burden on supine percutaneous nephrolithotomy patients in Zainoel Abidin Hospital, Aceh, Indonesia. Methods: The data was collected retrospectively from medical records in Zainoel Abidin Hospital, Aceh between January 2014 until December 2019. Patients were followed-up with plain abdominal radiography (BNO) or renal ultrasonography (USG). Stone free status was defined as no residual fragments on radiography or USG. Results: 97 patients with inferior calyceal stones who underwent PCNLwere included. 53 cases had stone burden < 20 mm, 34 cases with stone burden 21-30 mm, and 10 cases with stone burden > 30 mm. Overall, 91 (94%) cases were defined as stone free. On group < 20 mm, 21-30 mm, and > 30 mm; 51 (96%), 32 (94%), and 8 (80%) cases defined as stone free respectively (p = 0.485). Conclusions: Supine PCNLwas still superior inmanagement of calyx inferior stones with different stone burden. The stone free rate of these three groups showed no statistically significant difference. Safety and Outcomes of Combined External Beam Radiation Therapy (EBRT) with High Dose Rate (HDR) Brachytherapy for Localized Prostate Cancer Yufei Qiao (1) (1) Singapore General Hospital Introduction andObjectives: Combined external beam radiotherapy (EBRT) and high dose rate brachytherapy boost (HDR-BT) has been proven a safe and effective option to deliver high dose radiation in selected patients with intermediate-risk to high-risk prostate cancer. This study aims to report the clinical outcomes of Asian population treated with this combined approach in a tertiary academic institution. Methods: Medical records of 37 patients who underwent EBRT with HDR- BT boost 2015-2019 retrospectively reviewed. Mid-term safety and efficacy data particularly gastrointestinal (GI) and genitourinary (GU) toxicities evaluated. Short to mid-term outcomes including biochemical recurrence, distal metastasis, progression-free survival, and overall survival were assessed. Adverse effects assessed using CTCAE version 4.0. Biochemical failure defined by the Phoenix consensus definition of PSA level 2.0 ng/mL above the nadir value. Results: Total 37 patients received 15 Gy in one fraction. Two defaulted, one was lost to follow up. For rest 34 patients, median follow-up 28.5 months. Patients stratified by Damico risk classification: 21% patients (n=7) intermediate risk and 79% patients (n=27) high risk. All patients received androgen deprivation therapy. Combined median dose to 2 cc (D2 cc) of rectumand bladder was of 63.0Gy and 71.4Gy respectively. One patient (2.9%) developed grade 3 GU toxicity requiring admission for washout; no grade 3-4 GI toxicity. One patient (2.9%) developed biochemical recurrence 2 years post treatment, and distal metastasis 15months later. The overall progression-free survival rate was 97.1%, and overall survival rate was 100%. Conclusions: EBRT with HDR-BT is a safe and effective approach for intermediate and high-risk prostate cancer, with excellent biochemical control and long-term toxicities. C-10090 C-10086 Increasing Rate of Bladder Cancer within Australia Brielle Wood (1) (1) Queensland Health Introduction and Objectives: Bladder cancer is the seventh most common cancer andmakes up 2% of all new cases diagnosed inAustralia. Occurs most commonly in males and within the fifth and seventh decades of life. The aim of the study was to determine if bladder cancer diagnosis is increasing within Australia over a 15-year period. Methods: This was an observational study of theAustralian rates of bladder cancer from the year 2000 until 2015. The study used publicly available Australian Institute of Health and Welfare (AIHW) data, and included both sexes and all ages. Results: Over the study period a total of 37,630 people were diagnosed with bladder cancer withinAustralia. The average number of new diagnosis increased by 10.3% per annum of the study period. Bladder cancer was most common in the male population with 75% (28,271) of diagnosis being males and 25%were females with bladder cancer. The rate of bladder cancer increasedwith increasing age, with the age group of 75-79 years (6,666) having the highest incidence of bladder cancer. The total incidence of bladder cancer over the time period was highest in NSW (11,827), followed by VIC (8,875), QLD (7,737), WA (3,310), SA (3,264), TAS (1,099), ACT (399) and lastly NT (138) with the lowest incidence. Conclusions: The incidence of bladder cancer has continued to increase within Australia since the year 2000. The increase in bladder cancer incidence is likely multifactorial, and may indicate: increasing diagnosis and investigation; increasing ageing population; or increasing risk factors. CanMachine-Learning (ML) AugmentedAudio-UroflowmetryDistinguish Between Normal and Abnormal Flows from Voiding Sounds? Han Jie Lee (1) (1) Singapore General Hospital Introduction and Objectives: Urinary flows predicted by an artificial intelligence (AI)-assisted system analysing voiding sounds have correlated well with standard uroflowmetry measurements. Presently, we evaluate its accuracy in distinguishing normal and abnormal urinary flows based on voiding sounds. Methods: 233 male participants aged 21-80 years with or without lower urinary tract symptoms (LUTS), were enrolled between 01 December 2017 and 30 June 2018. Participants voided into a gravimetric uroflowmetry machine while voiding sounds were recorded on a smartphone.Audio recordings were digitally processed, pairedwith the corresponding uroflowmetry parameters and split into two groups, for training and testing theAI system. The training dataset was dichotomised into normal and abnormal flows by two urologists, and used to train two ML algorithms - Gaussian Mixture ModelUniversal Background (GMM-UBM) and Long-Short-Term Memory (LSTM) model. Both trained algorithms were asked to categorise audio recordings from the testing dataset into normal/abnormal groups, and this was referenced against the urologists diagnoses.Accuracy and area under ROC (AROC) curves were calculated for each ML algorithm. Results: From 233 paired audio recordings/uroflow measurements, 131 (63 normal and 68 abnormal flow patterns) were used to train the ML algorithm. On testing, the GMM-UBM model had an accuracy of 89.2% and AROC of 0.93, correctly classifying 34/40 normal and 57/62 abnormal flows. The LSTM model had an accuracy of 91.1% andAROC of 0.92, correctly classifying 39/40 normal and 54/62 abnormal flows. Conclusions: Audio-uroflow predictions from voiding sounds using AI- assisted algorithms can distinguish normal and abnormal flows with good accuracy, but further validation is required. C-10087 C-10085 8