Background Improved prognostication is needed to minimize overtreatment in grade group (GG) 2–4 prostate cancer. Our aim was to determine, at messenger RNA (mRNA) level, the performance of the genes in the commercial panels Decipher, Oncotype DX, Prolaris, and mutational panel MSK‐IMPACT to predict metastasis‐free and prostate cancer‐specific death (PCSD) in patients with GG 2–4 prostate cancer at radical prostatectomy. Methods The retrospective cohort consisted of GG 2–4 patients treated with radical prostatectomy (median follow‐up 10.4 years). Seventy‐six cases with postoperative metastasis or PCSD and 84 controls with similar clinical baseline risk, but without progression, were analyzed. Index lesion mRNA transcripts were analyzed using NanoString technology. Random forest models were trained using panel gene sets to predict clinical endpoints and area under the curve (AUC), sensitivity, specificity, Youden index, and number needed to diagnose (NND) was measured. Survival probability was assessed with Kaplan–Meier estimator. Results All gene sets outperformed clinical parameters and predicted metastasis‐free and prostate cancer‐specific survival. However, there were significant differences between the panels. In metastasis prediction, the genes in Oncotype DX had inferior performance (area under the curve [AUC] = 0.65) compared to other panels (AUC = 0.73–0.74). Decipher, MSK‐IMPACT and Prolaris showed similar NND (2.83–3.12) with Oncotype DX having highest NND (4.79). In PCSD prediction, the Prolaris gene set performed worse (AUC = 0.66) than MSK‐IMPACT or Decipher (AUC = 0.72). Oncotype DX performed similarly to other panels (AUC = 0.69, p > .05). Oncotype DX demonstrated lowest NND (2.79) compared to other panels (4.22–5.66). Conclusion Transcript analysis of genes included in commercial panels is feasible in survival prediction of GG 2‐4 patients after radical prostatectomy and may aid in clinical decision making. There were significant differences between the panels, and overall stronger predictive gene sets are needed. Prospective investigation is warranted in biopsy materials.

No abstract

A pathologist’s histological diagnosis is the gold standard of prostate cancer diagnostic measures. In addition to cancer grade, the stage of cancer determines the follow-up and possible adjuvant therapies after surgery. Not only glandular pattern recognition but also the assessment of three-dimensional extent and size of cancer, in relation to the whole organ, are parts of the subjective diagnosis, which may vary among pathologists. Furthermore, evaluation of lymph nodes for any possible metastases is time-consuming and missing cancer in lymph nodes can lead to undertreatment of the patient. For the abovementioned reasons, our team is developing an algorithm that calculates the amount of cancer tissue objectively, assisting the pathologist in the diagnostic procedure. In addition, we are creating an assistant for cancer detection in lymph nodes, which are removed for histological evaluation. All in all, we aim to create a tool that will help pathologists to a better, faster, more secure and more accurate diagnosis. Our material consists of full sets of scanned whole slide images from 302 prostates that were retrieved by radical prostatectomy. After supervised learning procedure, convolutional neural networks were employed for the classification of cancerous and non-cancerous regions in the images. A tiling based approach was used in which a slide was divided into square shaped small tiles. Millions of cancerous and non-cancerous tiles were sampled from the dataset for training and validation. During the convolutional neural network training, several different tile sizes were used, i.e., 256x256, 512x512, 1024x1024 pixels. Four different types of architectures were fine-tuned and trained for the task of tile-wise binary classification, namely InceptionV3, Xception, ResNet50 and a custom convolutional neural network architecture. In our preliminary assays for cancer detection, in both pixel-wise and tile-wise evaluation, InceptionV3 performed outstandingly well with an AUC score of 0.97 and 0.951, respectively. In conclusion, our algorithm has developed very well thus far with an accuracy in cancer detection of 97%. It is not only a versatile assisting tool, aiding pathologists to a more objective, standardized and accurate diagnosis, but also serves as a second opinion in difficult and challenging diagnostic cases. Citation Format: Carolin Stürenberg, Umair Khan, Kevin Sandeman, Oguzhan Gencoglu, Adrian Malen, Andrew Erickson, Timo Heikkinen, Antti Rannikko, Tuomas Mirtti. Detection and local histological staging of prostate cancer foci in H&E whole slide images using convolutional neural networks [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1396.

No abstract