PTEN and DNA Ploidy Status by Machine Learning in Prostate Cancer.

Abstract

Simple SummaryMolecular tissue-based prognostic biomarkers are anticipated to complement the current risk stratification systems in prostate cancer, but their manual assessment is subjective and time-consuming. Objective assessment of such biomarkers by machine learning-based methods could advance their adoption in a clinical workflow. PTEN and DNA ploidy status are well-studied biomarkers, which can provide clinically relevant information in prostate cancer at a low cost. Using a cohort of 253 patients who received radical prostatectomy, we developed a novel, fully-automated PTEN scoring in immunohistochemically-stained tissue slides, which could be used to assess PTEN status in a reliable and reproducible manner. In an independent validation cohort of 259 patients, automatically assessed PTEN status was significantly associated with time to biochemical recurrence after radical prostatectomy, and the combination of PTEN and DNA ploidy status further improved risk stratification. These results demonstrate the utility of machine learning in biomarker assessment.Machine learning (ML) is expected to improve biomarker assessment. Using convolution neural networks, we developed a fully-automated method for assessing PTEN protein status in immunohistochemically-stained slides using a radical prostatectomy (RP) cohort (n = 253). It was validated according to a predefined protocol in an independent RP cohort (n = 259), alone and by measuring its prognostic value in combination with DNA ploidy status determined by ML-based image cytometry. In the primary analysis, automatically assessed dichotomized PTEN status was associated with time to biochemical recurrence (TTBCR) (hazard ratio (HR) = 3.32, 95% CI 2.05 to 5.38). Patients with both non-diploid tumors and PTEN-low had an HR of 4.63 (95% CI 2.50 to 8.57), while patients with one of these characteristics had an HR of 1.94 (95% CI 1.15 to 3.30), compared to patients with diploid tumors and PTEN-high, in univariable analysis of TTBCR in the validation cohort. Automatic PTEN scoring was strongly predictive of the PTEN status assessed by human experts (area under the curve 0.987 (95% CI 0.968 to 0.994)). This suggests that PTEN status can be accurately assessed using ML, and that the combined marker of automatically assessed PTEN and DNA ploidy status may provide an objective supplement to the existing risk stratification factors in prostate cancer.