New insights in the interaction of FGF/FGFR and steroid receptor signaling in breast cancer

Luminal breast cancer (BrCa) has a favorable prognosis compared to other tumor subtypes. However, with time tumors may evolve and lead to disease progression. Thus, there is a great interest in unraveling the mechanisms that drive tumor metastasis and endocrine resistance. In this review we focused in one of the many pathways that have been involved in tumor progression, the FGF/FGFR axis. We emphasized in data obtained from in vivo experimental models since we believe that in luminal BrCa, tumor growth relies in a crosstalk with the stromal tissue. We revisited the studies that illustrate the interaction between hormone receptors and FGFR. We also highlighted the most frequent alterations found in BrCa cell lines and we provide a short review on the trials that use FGFR inhibitors in combination with endocrine therapies. The analysis of this data suggests that there are many players involved in this pathway that might be also targeted to decrease FGF signaling in addition to specific FGFR inhibitors that may be exploited to increase their efficacy.

DeepMed: A unified, modular pipeline for end-to-end deep learning in computational pathology

The interpretation of digitized histopathology images has been transformed thanks to artificial intelligence (AI). End-to-end AI algorithms can infer high-level features directly from raw image data, extending the capabilities of human experts. In particular, AI can predict tumor subtypes, genetic mutations and gene expression directly from hematoxylin and eosin (H&E) stained pathology slides. However, existing end-to-end AI workflows are poorly standardized and not easily adaptable to new tasks. Here, we introduce DeepMed, a Python library for predicting any high-level attribute directly from histopathological whole slide images alone, or from images coupled with additional meta-data (https://github.com/KatherLab/deepmed). Unlike earlier computational pipelines, DeepMed is highly developer-friendly: its structure is modular and separates preprocessing, training, deployment, statistics, and visualization in such a way that any one of these processes can be altered without affecting the others. Also, DeepMed scales easily from local use on laptop computers to multi-GPU clusters in cloud computing services and therefore can be used for teaching, prototyping and for large-scale applications. Finally, DeepMed is user-friendly and allows researchers to easily test multiple hypotheses in a single dataset (via cross-validation) or in multiple datasets (via external validation). Here, we demonstrate and document DeepMed's abilities to predict molecular alterations, histopathological subtypes and molecular features from routine histopathology images, using a large benchmark dataset which we release publicly. In summary, DeepMed is a fully integrated and broadly applicable end-to-end AI pipeline for the biomedical research community.

Multi-path Convolutional Neural Network to identify Tumorous Sub-classes for Breast Tissue from Histopathological Images

Malignancy is one of the leading causes of death globally. It is on the rise in the developed and low-income countries with survival rates of less than 40%. However, early diagnosis may increase survival chances. Histopathology images acquired from the biopsy are a popular method for cancer diagnosis. In this article, we propose a deep convolutional neural network-based method that helps classify breast cancer tumor subtypes from histopathology images. The model is trained on the BreakHis dataset but is also tested on images from other datasets. The model is trained to recognized eight different tumor subtypes, and also to perform binary classification (malignant / non-malignant). The CNN model uses an encoder-decoder architecture as well as a parallel feed-forward network. The proposed model provides higher cumulative training accuracy and statistical scoring after five-fold cross-validation. Comparing with the other models, the accuracy of the proposed model is higher at different magnification and patient levels.

Delineation of colorectal cancer ligand-receptor interactions and their roles in the tumor microenvironment and prognosis

Background Immunotherapies targeting ligand-receptor interactions (LRIs) are advancing rapidly in the treatment of colorectal cancer (CRC), and LRIs also affect many aspects of CRC development. However, the pattern of LRIs in CRC and their effect on tumor microenvironment and clinical value are still unclear. Methods We delineated the pattern of LRIs in 55,539 single-cell RNA sequencing (scRNA-seq) samples from 29 patients with CRC and three bulk RNA-seq datasets containing data from 1411 CRC patients. Then the influence of tumor microenvironment, immunotherapy and prognosis of CRC patients were comprehensively investigated. Results We calculated the strength of 1893 ligand-receptor pairs between 25 cell types to reconstruct the spatial structure of CRC. We identified tumor subtypes based on LRIs, revealed the relationship between the subtypes and immunotherapy efficacy and explored the ligand-receptor pairs and specific targets affecting the abundance of tumor-infiltrating lymphocytes. Finally, a prognostic model based on ligand-receptor pairs was constructed and validated. Conclusion Overall, through the comprehensive and in-depth investigation of the existing ligand-receptor pairs, this study provides new ideas for CRC subtype classification, a new risk screening tool for CRC patients, and potential ligand-receptor pair targets and pathways for CRC therapy.

A Network-Centric Framework for the Evaluation of Mutual Exclusivity Tests on Cancer Drivers

One of the key concepts employed in cancer driver gene identification is that of mutual exclusivity (ME); a driver mutation is less likely to occur in case of an earlier mutation that has common functionality in the same molecular pathway. Several ME tests have been proposed recently, however the current protocols to evaluate ME tests have two main limitations. Firstly the evaluations are mostly with respect to simulated data and secondly the evaluation metrics lack a network-centric view. The latter is especially crucial as the notion of common functionality can be achieved through searching for interaction patterns in relevant networks. We propose a network-centric framework to evaluate the pairwise significances found by statistical ME tests. It has three main components. The first component consists of metrics employed in the network-centric ME evaluations. Such metrics are designed so that network knowledge and the reference set of known cancer genes are incorporated in ME evaluations under a careful definition of proper control groups. The other two components are designed as further mechanisms to avoid confounders inherent in ME detection on top of the network-centric view. To this end, our second objective is to dissect the side effects caused by mutation load artifacts where mutations driving tumor subtypes with low mutation load might be incorrectly diagnosed as mutually exclusive. Finally, as part of the third main component, the confounding issue stemming from the use of nonspecific interaction networks generated as combinations of interactions from different tissues is resolved through the creation and use of tissue-specific networks in the proposed framework. The data, the source code and useful scripts are available at: https://github.com/abu-compbio/NetCentric.

Coexistence of Craniopharyngioma and Cranial Fibrous Dysplasia: A Clinicopathological Study of 5 Cases and Literature Review

Background: Craniopharyngioma (CP) and cranial fibrous dysplasia (CFD) are both rare embryonic benign cranial diseases that most commonly present during childhood or adolescence. Co-existence of CP and CFD is extremely rare, which has never been reported before.Methods: We retrospectively reviewed the data of 5 patients coexisted with CP and CFD in Beijing Tiantan Hospital from January 2003 to January 2021. Their clinicopathological features, treatment modalities, and outcomes were summarized. Moreover, a comprehensive literature review was conducted, and in order to explore the potential connection leading to this coexistence, the CFD characteristic GNAS gene and corresponding Gsα protein were tested in the CPResults: There were 4 males and 1 female (median age, 39 years) in the present series. The symptoms mainly included headache, dizziness, fatigue, polyuria/polydipsia, hypogonadism and blurred vision. Sphenoidal bone is the most common involved bone by CFD (n =4). Four patients had undergone surgery to remove the CP (1 transsphnoid and 3 transcranial). Complete and subtotal resection were achieved in 2 cases respectively. The tumor subtypes were 3 adamantinomatous, 1 unknown subtype. The common postoperative complications are pan-hypopituitarism, diabetes insipidus, and hypothyroidism. The mean follow-up time was 57.2 months. Postoperative hormone replacement was required in 2 patients. 3 patients underwent a genetic study of tumor specimens. GNAS mutations were not detected, but they were positive for Gsα protein. Conclusions: Though the definite causative relationship has not been proved, the coexistence of CP and CFD should not completely be excluded potential interplay or atypical FD course for the uncommon manifestations of CPs. Prompt diagnosis and appropriate treatment are more challenging than for solitary CPs for the deformations of skull base, as of now, management strategies are aimed at surgical treating the CP and regularly monitoring the CFD.

Exploiting convergent evolution to derive a pan-cancer cisplatin sensitivity gene expression signature

Precision medicine offers remarkable potential for the treatment of cancer, but is largely focused on tumors that harbor actionable mutations. Gene expression signatures can expand the scope of precision medicine by predicting response to traditional(cytotoxic) chemotherapy agents without relying on changes in mutational status. We present a novel signature extraction method, inspired by the principle of convergent evolution, which states that tumors with disparate genetic backgrounds may evolve similar phenotypes independently. This evolutionary-informed method can be utilized to produce signatures predictive of response to over 200 chemotherapeutic drugs found in the Genomics of Drug Sensitivity in Cancer Database. Here, we demonstrate its use by extracting the Cisplatin Response Signature, CisSig, for use in predicting a common trait (sensitivity to cisplatin) across disparate tumor subtypes (epithelial-origin tumors). CisSig is predictive of cisplatin response within the cell lines and clinical trends in independent datasets of tumor samples. This novel methodology can be used to produce robust signatures for the prediction of traditional chemotherapeutic response, dramatically increasing the reach of personalized medicine in cancer.

CPTAC Pancancer Phosphoproteomics Kinase Enrichment Analysis with ProKAP Provides Insights into Immunogenic Signaling Pathways

The National Cancer Institute (NCI) Clinical Proteomic Tumor Analysis Consortium (CPTAC) initiative has generated extensive phosphoproteomics and proteomics data for tumor and tumoradjacent normal tissue across multiple cancer types. This dataset provides an unprecedented opportunity to systematically characterize pan-cancer kinase activities, which is essential for coupling tumor subtypes with kinase inhibitors as potential treatment. In this work, we performed Kinase Enrichment Analysis (KEA) using a CPTAC phosphoproteomics dataset to identify putative differences in kinase state between tumor and normal tissues within and across five types of cancer. We then implemented an interactive web-portal, the ProTrack Kinase Activity Portal (ProKAP), for querying, visualizing, and downloading the derived pan-cancer kinase activity scores together with the corresponding sample metadata, and protein and phosphoprotein expression profiles. To illustrate the usage of this digital resource, we analyzed the association between kinase activity scores and immune subtypes of clear cell renal cell carcinoma (ccRCC) derived from the CPTAC ccRCC study. We found multiple kinases, whose inhibition has been suggested to have therapeutic effect in other tumor types, are highly active in CD8+-enriched ccRCC tumors. The ProTrack Kinase Activity Portal (ProKAP) is available at: https://pancan-kea3.cptac-data-view.org.

CBIO-22. PARP INHIBITION SYNERGIZES WITH DNA DAMAGING DRUGS IN PEDIATRIC CNS TUMORS

BACKGROUND Central nervous system (CNS) tumors are the second most common childhood cancer. Despite innovations in surgery and chemo-/radiotherapy, CNS tumors remain the major cause of cancer-related death in children. Previous sequencing analyses in a pediatric cancer cohort identified BRCA and DSB repair signatures as potentially targetable events. Based on these findings, we propose the use of PARP inhibitors (PARPi) for aggressive CNS tumor subtypes, including high-grade glioma (HGG), medulloblastoma (MB) and ependymoma (EPN). METHODS We tested multiple PARPi in tumor cell lines (n=8) as well as primary patient-derived models (n=11) of pediatric HGG, MB, EPN and atypical teratoid/rhabdoid tumors (ATRTs). Based on PARPi sensitivity, selected models were further exposed to a combination of PARPi and DNA-damaging/modifying agents. The mode of action was investigated using Western blot and flow cytometry. RESULTS We show that a fraction of pediatric MB, EPN and ATRT demonstrate sensitivity towards PARP inhibition, which is paralleled by susceptibility to the DNA damaging drugs cisplatin and irinotecan. Interestingly, talazoparib, the most potent PARPi, showed synergistic cytotoxicity with DNA-damaging/modifying drugs. In addition, cell cycle blockade and increased DNA damage combined with reduced DNA repair signaling, such as activation of the ATR/Chk1 pathway were observed. Corroboratively, talazoparib exhibited a synergistic anti-cancer effect in combination with inhibitors of ATR, a major regulator of DNA damage response. CONCLUSION/OUTLOOK To sum up, we demonstrate that PARP inhibition synergizes with DNA damaging anti-cancer compounds or DNA repair inhibitors and, thus, represents a promising therapeutic strategy for a defined subgroup of pediatric high-risk CNS tumors patients. More in depth characterization of the underlying molecular events will most likely allow the identification of predictive biomarkers for most efficient implementation of this strategy into clinical application.