Search

PURPOSE There is ongoing clinical need to improve estimates of disease outcome in prostate cancer. Machine learning (ML) approaches to pathologic diagnosis and prognosis are a promising and increasingly used strategy. In this study, we use an ML algorithm for prediction of adverse outcomes at radical prostatectomy (RP) using whole-slide images (WSIs) of prostate biopsies with Grade Group (GG) 2 or 3 disease. METHODS We performed a retrospective review of prostate biopsies collected at our institution which had corresponding RP, GG 2 or 3 disease one or more cores, and no biopsies with higher than GG 3 disease. A hematoxylin and eosin–stained core needle biopsy from each site with GG 2 or 3 disease was scanned and used as the sole input for the algorithm. The ML pipeline had three phases: image preprocessing, feature extraction, and adverse outcome prediction. First, patches were extracted from each biopsy scan. Subsequently, the pre-trained Visual Geometry Group-16 convolutional neural network was used for feature extraction. A representative feature vector was then used as input to an Extreme Gradient Boosting classifier for predicting the binary adverse outcome. We subsequently assessed patient clinical risk using CAPRA score for comparison with the ML pipeline results. RESULTS The data set included 361 WSIs from 107 patients (56 with adverse pathology at RP). The area under the receiver operating characteristic curves for the ML classification were 0.72 (95% CI, 0.62 to 0.81), 0.65 (95% CI, 0.53 to 0.79) and 0.89 (95% CI, 0.79 to 1.00) for the entire cohort, and GG 2 and GG 3 patients, respectively, similar to the performance of the CAPRA clinical risk assessment. CONCLUSION We provide evidence for the potential of ML algorithms to use WSIs of needle core prostate biopsies to estimate clinically relevant prostate cancer outcomes.

283 Background: The widely-used National Comprehensive Cancer Network (NCCN) risk criteria for prostate cancer are impactful for patient and clinicians, particularly when assessing suitability for initial active surveillance. However, in the era of MRI-guided biopsy, fewer patients may meet stringent “very-low risk” criteria due to improved sampling, which may lead to over-treatment of non-lethal cancers. We evaluated the distribution of prostate cancer risk classification obtained using MRI-ultrasound fusion versus systematic 12 core biopsy. Methods: We performed a retrospective study of patients who underwent prostate MRI-ultrasound fusion biopsy at a single institution from January 2017 to July 2021. The primary study objective was to characterize the proportion of patients meeting NCCN “very-low risk” designation within a contemporary biopsy cohort. We calculated NCCN risk classifications (very low, low, favorable intermediate, unfavorable intermediate, high, and very high) based on constituent components (PSA, PSA density, number of biopsy cores positive, greatest extent of core positivity). We conducted comparisons within patients based on findings obtained through MRI-ultrasound fusion biopsy versus systematic biopsy only. Results: We identified 1,132 patients, including 1,014 (89.5%) and 118 (10.5%) who underwent first-time biopsy and surveillance biopsies. The median PSA was 7.15 (interquartile range: 5.2-10.8). Using data obtained through systematic biopsy alone, 123 (10.9%) and 242 (21.4%) were classified as very-low and low NCCN risk, respectively. Incorporating data from MRI-ultrasound fusion biopsy, 86 (7.6%) and 255 (22.5%) were classified as very-low and low NCCN risk, respectively. Among patients initially electing active surveillance as management, 7.4%, 60.2%, 26.9% met very-low, low, and favorable-intermediate risk criteria, respectively. Conclusions: A small proportion of patients undergoing MRI-ultrasound fusion biopsy will be classified as ‘very-low risk’ by the NCCN definition, and the majority of patients managed initially with active surveillance fall outside of this criteria. These findings imply a need to broaden support for active surveillance outside of “very-low risk” criteria to promote greater uptake.[Table: see text]

s S3 an increased trend in positive HPV in smears which have >3000 cells/PT. 4. HPV test results may be unreliable in PT with low cellularity ( 3000 cells/PT in women 45 years, especially if they are in the low risk category may be appropriate. Table Cell groups with HPV correlations 0-10 cells/ HPF (NZ356) 11-20 cells/ HPF (NZ181) 21-30 cells/ HPF (NZ100) 31 cells/ HPF (NZ149) Total (NZ786) HPV (-) 79 49 31 39 198 HPV (+) 5 2 2 4 13 HPV (+) % 5.95% 3.92% 6.06% 9.30% 6.16% Figure HPV testing clinical follow-up. Table 1 Prior HPV Test Results in Patients with Invasive Cervical Cancers Patient # HPV Test# Positive# %