AimsAntibodies targeting the checkpoint molecules programmed cell death 1 (PD-1) and its ligand PD-L1 are emerging cancer therapeutics. We systematically investigated PD-1 and PD-L1 expression patterns in the poor-prognosis tumor entity high-grade serous ovarian carcinoma.MethodsPD-1 and PD-L1 protein expression was determined by immunohistochemistry on tissue microarrays from 215 primary cancers both in cancer cells and in tumor-infiltrating lymphocytes (TILs). mRNA expression was measured by quantitative reverse transcription PCR. An in silico validation of mRNA data was performed in The Cancer Genome Atlas (TCGA) dataset.ResultsPD-1 and PD-L1 expression in cancer cells, CD3+, PD-1+, and PD-L1+ TILs densities as well as PD-1 and PD-L1 mRNA levels were positive prognostic factors for progression-free (PFS) and overall survival (OS), with all factors being significant for PFS (p < 0.035 each), and most being significant for OS. Most factors also had prognostic value that was independent from age, stage, and residual tumor. Moreover, high PD-1+ TILs as well as PD-L1+ TILs densities added prognostic value to CD3+TILs (PD-1+: p = 0.002,; PD-L1+: p = 0.002). The significant positive prognostic impact of PD-1 and PD-L1 mRNA expression could be reproduced in the TCGA gene expression datasets (p = 0.02 and p < 0.0001, respectively).ConclusionsDespite their reported immune-modulatory function, high PD-1 and PD-L1 levels are indicators of a favorable prognosis in ovarian cancer. Our data indicate that PD-1 and PD-L1 molecules are biologically relevant regulators of the immune response in high-grade serous ovarian carcinoma, which is an argument for the evaluation of immune checkpoint inhibiting drugs in this tumor entity.

The majority of patients with solid malignancies die from metastatic burden. However, our current understanding of the mechanisms and resulting patterns of dissemination is limited. Here, we analyzed patterns of metastatic progression across 16 major cancer types in a cohort of 1008 patients with metastatic cancer autopsied between 2000 and 2013 to assess cancer specific progression patterns of disease and related risk predictions. The frequency and location of metastases were evaluated in and across 1) 16 major cancers, 2) smoking- and non-smoking-related cancers and 3) adeno- and squamous cell carcinoma. Associations between primary and secondary sites were analyzed by the fractional and the relative risk methods. We detected significantly different cancer specific patterns of metastatic progression with specific relative risk profiles for secondary site involvement. Histology and smoking etiology influenced these patterns. Backward analysis showed that metastatic patterns help to predict unknown primary sites. Solid malignancies maintain a unique and recurrent organ tropism to specific secondary sites which does not appear to be strongly influenced by advances in cancer medicine as shown by comparison with previous data sets. The delineated landscape of metastatic progression patterns is a comprehensive data resource to both clinical and basic scientists which aids fostering new hypotheses for cancer research and cancer therapies.

Deep learning has recently gained popularity in digital pathology due to its high prediction quality. However, the medical domain requires explanation and insight for a better understanding beyond standard quantitative performance evaluation. Recently, explanation methods have emerged, which are so far still rarely used in medicine. This work shows their application to generate heatmaps that allow to resolve common challenges encountered in deep learning-based digital histopathology analyses. These challenges comprise biases typically inherent to histopathology data. We study binary classification tasks of tumor tissue discrimination in publicly available haematoxylin and eosin slides of various tumor entities and investigate three types of biases:(1) biases which affect the entire dataset, (2) biases which are by chance correlated with class labels and (3) sampling biases. While standard analyses focus on patch-level evaluation, we advocate pixel-wise heatmaps, which offer a more precise and versatile diagnostic instrument and furthermore help to reveal biases in the data. This insight is shown to not only detect but also to be helpful to remove the effects of common hidden biases, which improves generalization within and across datasets. For example, we could see a trend of improved area under the receiver operating characteristic curve by 5% when reducing a labeling bias. Explanation techniques are thus demonstrated to be a helpful and highly relevant tool for the development and the deployment phases within the life cycle of real-world applications in digital pathology. Related work Models in digital pathologySimilar to the recent trend in computer vision, where end-to-end training with deep learning clearly prevails classification of handcrafted features, an increased use of deep learning, e.g convolutional neural networks (CNN) is also noticeable in digital pathology. Nonetheless, there are some works on combining support vector machines with image feature algorithms 21, 23-25 . Meanwhile, while some works propose custom-designed networks 26, 27 , e. g. spatially constrained, locality sensitive CNN for the detection of nuclei in histopathological images 26 , most often standard deep learning architectures (e. g. AlexNet 28 , GoogLeNet 3 , ResNet 29 ) as well as hybrids are used for digital pathology 22, 30-33 . According to 6 , currently the most common architecture is the GoogLeNet Inception-V3 model. Interpretability in computational pathologyAs discussed above, more and more developments have emerged that introduce the possibility of explanation (e. g. [11][12][13][14][15][16][17][18] ; for a summary of implementations see 34 ); few of which have been applied in digital pathology 21,22,30,[35][36][37][38][39][40] . The visualization of a support vector machine's decision on Bag-of-Visual-Words features in a histopathological discrimination task is explored in 21 . The authors present an explanatory approach for evidence of tumor and lymphocytes in H&E images as well as for molecular properties which-unli...