Active surveillance (prostate cancer)
Prostate cancer ranks as the most common cancer among Norwegian men, with about 5000 new diagnoses yearly. The high incidence of overdiagnosis and overtreatment—impacting nearly two-thirds of cases—underscores the critical need for a more accurate risk stratification system. Our project aims to address this challenge by developing a precise risk assessment tool utilizing artificial intelligence (AI) biomarkers applicable to biopsy specimens.
The conventional treatments of radical prostatectomy or radiation therapy, while effective, often lead to significant adverse effects, impacting patients' quality of life. Active surveillance (AS) emerges as a promising alternative for patients with low- and intermediate-risk disease. Under this approach, patients undergo regular monitoring and receive treatment when there are signs of disease progression. However, the current criteria for selecting patients for AS and determining when to transition to treatment in AS protocols primarily rely on biopsies with standard clinicopathologic characteristics. Unfortunately, these characteristics do not adequately distinguish between patients who would benefit from immediate treatment and those who can safely postpone or completely avoid treatment.
Our project focuses on refining AS protocols by incorporating robust AI-driven markers to improve efficacy of AS. These markers encompass various factors such as mitotic figure detection, stroma fraction, DNA ploidy, histotyping, and protein expression. We will develop these markers using advanced image analysis and deep learning techniques. Additionally, our project incorporates an innovative method that enables multiple staining techniques (H&E, Feulgen, and immunohistochemistry) on a single tissue section. This approach allows for the sequential assessment of numerous biomarkers in the same tissue section, enabling a comprehensive cell-by-cell comparison and optimizing the utilization of limited biopsy tissue.
The project comprises two phases: marker development and protocol integration. We will start by working on markers initially established in radical prostatectomy specimens, adapting and transitioning these techniques to prostate cancer biopsies. With access to seven datasets comprising over 2000 patients, our AI-driven methods will undergo rigorous training, testing, and validation. To assess their effectiveness, we'll use metrics such as biochemical recurrence for radical prostatectomy datasets and transfer to treatment for AS datasets. Upon successful validation, these AI-driven markers will be incorporated into an updated AS protocol. To ensure its effectiveness in various clinical settings, the updated AS protocol will undergo evaluation across collaborating hospitals, including Vestfold Hospital Trust, Telemark Hospital Trust, and Vestre Viken Hospital Trust.