Abstract
Existing tools for post-radical prostatectomy (RP) prostate cancer biochemical recurrence (BCR) prognosis rely on human pathologist-derived parameters such as tumor grade, with the resulting inter-reviewer variability. Genomic companion diagnostic tests such as Decipher tend to be tissue destructive, expensive, and not routinely available in most centers. We present a tissue non-destructive method for automated BCR prognosis, termed "Histotyping", that employs computational image analysis of morphologic patterns of prostate tissue from a single, routinely acquired hematoxylin and eosin slide. Patients from two institutions (n = 214) were used to train Histotyping for identifying high-risk patients based on six features of glandular morphology extracted from RP specimens. Histotyping was validated for post-RP BCR prognosis on a separate set of n = 675 patients from five institutions and compared against Decipher on n = 167 patients. Histotyping was prognostic of BCR in the validation set (p < 0.001, univariable hazard ratio [HR] = 2.83, 95% confidence interval [CI]: 2.03–3.93, concordance index [c-index] = 0.68, median years-to-BCR: 1.7). Histotyping was also prognostic in clinically stratified subsets, such as patients with Gleason grade group 3 (HR = 4.09) and negative surgical margins (HR = 3.26). Histotyping was prognostic independent of grade group, margin status, pathological stage, and preoperative prostate-specific antigen (PSA) (multivariable p < 0.001, HR = 2.09, 95% CI: 1.40–3.10, n = 648). The combination of Histotyping, grade group, and preoperative PSA outperformed Decipher (c-index = 0.75 vs. 0.70, n = 167). These results suggest that a prognostic classifier for prostate cancer based on digital images could serve as an alternative or complement to molecular-based companion diagnostic tests.
Highlights
Tumor morphology is associated with cancer aggressiveness in prostate cancer (PCa)
Histotyping was significantly prognostic of biochemical recurrence (BCR) in the training (p < 0.001, hazard ratio (HR) = 2.64, 95% confidence interval [CI]: 1.56–4.44, c-index = 0.63) and validation (p < 0.001, HR = 2.83, 95% CI: 2.03–3.93, cindex = 0.68) sets
While there was a separation between Histotyping lowrisk and high-risk patients in all five sites, this separation was not significant in the University of Turku (UTurku) and Mount Sinai (MS) cohorts, a result potentially influenced by the small number of patients in these sets, with just 48 and 22 patients, respectively
Summary
Tumor morphology is associated with cancer aggressiveness in prostate cancer (PCa). Gleason grading, used by pathologists to score the loss of glandular structure and organization in tissue[1], is strongly correlated with patient outcome[2]. We present a QH method for BCR prognosis using continuous score (p = 0.002, HR = 1.17, 95% CI: 1.06–1.28) and as a automated analysis of an H&E slide from the dominant tumor categorical low/high-risk grouping The Cox model selected six features associated with highrisk disease and used the weighted sum of these features to estimate the BCR risk for each patient. This model, termed Histotyping, was validated on n = 675 patients. Though the 95% CIs of Histotyping+ and Decipher overlapped, Histotyping+ had the higher c-index in 81% of bootstrap
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
