The Brazilian national prospective active surveillance (AS) cohort of patients with low-risk prostate cancer in the public health system: vigiaSUS study protocol.

BACKGROUND: Prostate cancer exhibits a very diverse behaviour, with some patients dying from the disease and others never needing treatment. Active surveillance (AS) consists of periodic PSA assessment (prostate-specific antigen), DRE (digital rectal examination) and periodic prostate biopsies. According to the main guidelines, AS is the preferred strategy for low-risk patients, to avoid or delay definitive treatment. However, concerns remain regarding its applicability in certain patient subgroups, such as African American men, who were underrepresented in the main cohorts. Brazil has a very racially diverse population, with 56.1% self-reporting as brown or black. The aim of this study is to evaluate and validate the AS strategy in low-risk prostate cancer patients following an AS protocol in the Brazilian public health system. METHODS: This is a multicentre AS prospective cohort study that will include 200 patients from all regions of Brazil in the public health system. Patients with prostate adenocarcinoma and low-risk criteria, defined as clinical staging T1-T2a, Gleason score ≤ 6, and PSA < 10 ng/ml, will be enrolled. Archival prostate cancer tissue will be centrally reviewed. Patients enrolled in the study will follow the AS strategy, which involves PSA and physical examination every 6 months as well as multiparametric MRI (mpMRI) every two years and prostate biopsy at month 12 and then every two years. The primary objective is to evaluate the reclassification rate at 12 months, and secondary objectives include determining the treatment-free survival rate, metastasis-free survival, and specific and overall survival. Exploratory objectives include the evaluation of quality of life and anxiety, the impact of PTEN loss and the economic impact of AS on the Brazilian public health system. DISCUSSION: This is the first Brazilian prospective study of patients with low-risk prostate cancer under AS. To our knowledge, this is one of the largest AS study cohort with a majority of nonwhite patients. We believe that this study is an opportunity to better understand the outcomes of AS in populations underrepresented in studies. Based on these data, an AS national clinical guideline will be developed, which may have a beneficial impact on the quality of life of patients and on public health. TRIAL REGISTRATION: Clinicaltrials registration is NCT05343936.

The value of artificial intelligence for detection and grading of prostate cancer in human prostatectomy specimens: a validation study.

BACKGROUND: The Gleason grading system is an important clinical practice for diagnosing prostate cancer in pathology images. However, this analysis results in significant variability among pathologists, hence creating possible negative clinical impacts. Artificial intelligence methods can be an important support for the pathologist, improving Gleason grade classifications. Consequently, our purpose is to construct and evaluate the potential of a Convolutional Neural Network (CNN) to classify Gleason patterns. METHODS: The methodology included 6982 image patches with cancer, extracted from radical prostatectomy specimens previously analyzed by an expert uropathologist. A CNN was constructed to accurately classify the corresponding Gleason. The evaluation was carried out by computing the corresponding 3 classes confusion matrix; thus, calculating the percentage of precision, sensitivity, and specificity, as well as the overall accuracy. Additionally, k-fold three-way cross-validation was performed to enhance evaluation, allowing better interpretation and avoiding possible bias. RESULTS: The overall accuracy reached 98% for the training and validation stage, and 94% for the test phase. Considering the test samples, the true positive ratio between pathologist and computer method was 85%, 93%, and 96% for specific Gleason patterns. Finally, precision, sensitivity, and specificity reached values up to 97%. CONCLUSION: The CNN model presented and evaluated has shown high accuracy for specifically pattern neighbors and critical Gleason patterns. The outcomes are in line and complement others in the literature. The promising results surpassed current inter-pathologist congruence in classical reports, evidencing the potential of this novel technology in daily clinical aspects.

Detecting and grading prostate cancer in radical prostatectomy specimens through deep learning techniques.

OBJECTIVES: This study aims to evaluate the ability of deep learning algorithms to detect and grade prostate cancer (PCa) in radical prostatectomy specimens. METHODS: We selected 12 whole-slide images of radical prostatectomy specimens. These images were divided into patches, and then, analyzed and annotated. The annotated areas were categorized as follows: stroma, normal glands, and Gleason patterns 3, 4, and 5. Two analyses were performed: i) a categorical image classification method that labels each image as benign or as Gleason 3, Gleason 4, or Gleason 5, and ii) a scanning method in which distinct areas representative of benign and different Gleason patterns are delineated and labeled separately by a pathologist. The Inception v3 Convolutional Neural Network architecture was used in categorical model training, and a Mask Region-based Convolutional Neural Network was used to train the scanning method. After training, we selected three new whole-slide images that were not used during the training to evaluate the model as our test dataset. The analysis results of the images using deep learning algorithms were compared with those obtained by the pathologists. RESULTS: In the categorical classification method, the trained model obtained a validation accuracy of 94.1% during training; however, the concordance with our expert uropathologists in the test dataset was only 44%. With the image-scanning method, our model demonstrated a validation accuracy of 91.2%. When the test images were used, the concordance between the deep learning method and uropathologists was 89%. CONCLUSION: Deep learning algorithms have a high potential for use in the diagnosis and grading of PCa. Scanning methods are likely to be superior to simple classification methods.

Detecting and grading prostate cancer in radical prostatectomy specimens through deep learning techniques

OBJECTIVES: This study aims to evaluate the ability of deep learning algorithms to detect and grade prostate cancer (PCa) in radical prostatectomy specimens. METHODS: We selected 12 whole-slide images of radical prostatectomy specimens. These images were divided into patches, and then, analyzed and annotated. The annotated areas were categorized as follows: stroma, normal glands, and Gleason patterns 3, 4, and 5. Two analyses were performed: i) a categorical image classification method that labels each image as benign or as Gleason 3, Gleason 4, or Gleason 5, and ii) a scanning method in which distinct areas representative of benign and different Gleason patterns are delineated and labeled separately by a pathologist. The Inception v3 Convolutional Neural Network architecture was used in categorical model training, and a Mask Region-based Convolutional Neural Network was used to train the scanning method. After training, we selected three new whole-slide images that were not used during the training to evaluate the model as our test dataset. The analysis results of the images using deep learning algorithms were compared with those obtained by the pathologists. RESULTS: In the categorical classification method, the trained model obtained a validation accuracy of 94.1% during training; however, the concordance with our expert uropathologists in the test dataset was only 44%. With the image-scanning method, our model demonstrated a validation accuracy of 91.2%. When the test images were used, the concordance between the deep learning method and uropathologists was 89%. CONCLUSION: Deep learning algorithms have a high potential for use in the diagnosis and grading of PCa. Scanning methods are likely to be superior to simple classification methods.

HPV Genotype Prevalence and Success of Vaccination to Prevent Cervical Cancer.

BACKGROUND: Nearly 500,000 new cases of cervical cancer are estimated annually worldwide. Three vaccines are currently licensed to prevent cervical cancer. The success of vaccination depends mainly on the prevalence of HPV genotypes, and many cases of HPV infection have been diagnosed after vaccination. Our aim was to search for HPV genotyping in cervical samples to verify the proportion of women that remain susceptible to infection even after vaccination. METHODS: 21,017 liquid-based cervical (LBC) specimens were received for cytology and HPV detection from 2015 to 2018. Before slide preparations for cytology, a 1,000-µL aliquot was taken from the LBC fixative and subjected to automated DNA extraction and multiplex PCR followed by capillary electrophoresis to detect and classify HPV. RESULTS: HPV was detected in 895 (4.3%) specimens. The most prevalent genotype was HPV-16, followed by HPV-58 and HPV-66. A total of 258 (28.8%) cases were positive for high-risk (HR)-HPV types (66, 59, 39, 56, 30, 35, 53, 51, 68, 82, and 70) that are not covered by the HPV vaccines. CONCLUSION: A significant proportion of HPV types detected in cytological specimens are representative of HR-HPV not covered by the available vaccines. The health system should be aware of the considerable percentage of women who are not being immunized and will continue to need cervical cancer screening.