MuTILs: explainable, multiresolution computational scoring of Tumor-Infiltrating Lymphocytes in breast carcinomas using clinical guidelines

Tumor-Infiltrating Lymphocytes (TILs) have strong prognostic and predictive value in breast cancer, but their visual assessment is subjective. We present MuTILs, a convolutional neural network architecture specifically optimized for the assessment of TILs in whole-slide image scans in accordance with clinical scoring recommendations. MuTILs is a concept bottleneck model, designed to be explainable and to encourage sensible predictions at multiple resolutions. Our computational scores match visual scores and have independent prognostic value in invasive breast cancers from the TCGA dataset.

Path to Clonal Theranostics in Luminal Breast Cancers

Integrating tumor heterogeneity in the drug discovery process is a key challenge to tackle breast cancer resistance. Identifying protein targets for functionally distinct tumor clones is particularly important to tailor therapy to the heterogeneous tumor subpopulations and achieve clonal theranostics. For this purpose, we performed an unsupervised, label-free, spatially resolved shotgun proteomics guided by MALDI mass spectrometry imaging (MSI) on 124 selected tumor clonal areas from early luminal breast cancers, tumor stroma, and breast cancer metastases. 2868 proteins were identified. The main protein classes found in the clonal proteome dataset were enzymes, cytoskeletal proteins, membrane-traffic, translational or scaffold proteins, or transporters. As a comparison, gene-specific transcriptional regulators, chromatin related proteins or transmembrane signal receptor were more abundant in the TCGA dataset. Moreover, 26 mutated proteins have been identified. Similarly, expanding the search to alternative proteins databases retrieved 126 alternative proteins in the clonal proteome dataset. Most of these alternative proteins were coded mainly from non-coding RNA. To fully understand the molecular information brought by our approach and its relevance to drug target discovery, the clonal proteomic dataset was further compared to the TCGA breast cancer database and two transcriptomic panels, BC360 (nanoString®) and CDx (Foundation One®). We retrieved 139 pathways in the clonal proteome dataset. Only 55% of these pathways were also present in the TCGA dataset, 68% in BC360 and 50% in CDx. Seven of these pathways have been suggested as candidate for drug targeting, 22 have been associated with breast cancer in experimental or clinical reports, the remaining 19 pathways have been understudied in breast cancer. Among the anticancer drugs, 35 drugs matched uniquely with the clonal proteome dataset, with only 7 of them already approved in breast cancer. The number of target and drug interactions with non-anticancer drugs (such as agents targeting the cardiovascular system, metabolism, the musculoskeletal or the nervous systems) was higher in the clonal proteome dataset (540 interactions) compared to TCGA (83 interactions), BC360 (419 interactions), or CDx (172 interactions). Many of the protein targets identified and drugs screened were clinically relevant to breast cancer and are in clinical trials. Thus, we described the non-redundant knowledge brought by this clone-tailored approach compared to TCGA or transcriptomic panels, the targetable proteins identified in the clonal proteome dataset, and the potential of this approach for drug discovery and repurposing through drug interactions with antineoplastic agents and non-anticancer drugs.

Pan-Cancer Analysis Shows Enrichment of Macrophages, Overexpression of Checkpoint Molecules, Inhibitory Cytokines, and Immune Exhaustion Signatures in EMT-High Tumors

Epithelial–mesenchymal transition (EMT) is a highly dynamic process that occurs under normal circumstances; however, EMT is also known to play a central role in tumor progression and metastasis. Furthermore, role of tumor immune microenvironment (TIME) in shaping anticancer immunity and inducing the EMT is also well recognized. Understanding the key features of EMT is critical for the development of effective therapeutic interventions. Given the central role of EMT in immune escape and cancer progression and treatment, we have carried out a pan-cancer TIME analysis of The Cancer Genome Atlas (TCGA) dataset in context to EMT. We have analyzed infiltration of various immune cells, expression of multiple checkpoint molecules and cytokines, and inflammatory and immune exhaustion gene signatures in 22 cancer types from TCGA dataset. A total of 16 cancer types showed a significantly increased (p < 0.001) infiltration of macrophages in EMT-high tumors (mesenchymal samples). Furthermore, out of the 17 checkpoint molecules we analyzed, 11 showed a significant overexpression (p < 0.001) in EMT-high samples of at least 10 cancer types. Analysis of cytokines showed significant enrichment of immunosuppressive cytokines—TGFB1 and IL10—in the EMT-high group of almost all cancer types. Analysis of various gene signatures showed enrichment of inflammation, exhausted CD8+ T cells, and activated stroma signatures in EMT-high tumors. In summary, our pan-cancer EMT analysis of TCGA dataset shows that the TIME of EMT-high tumors is highly immunosuppressive compared to the EMT-low (epithelial) tumors. The distinctive features of EMT-high tumors are as follows: (i) the enrichment of tumor-associated macrophages, (ii) overexpression of immune checkpoint molecules, (iii) upregulation of immune inhibitory cytokines TGFB1 and IL10, and (iv) enrichment of inflammatory and exhausted CD8+ T-cell signatures. Our study shows that TIMEs of different EMT groups differ significantly, and this would pave the way for future studies analyzing and targeting the TIME regulators for anticancer immunotherapy.

Pervasive conditional selection of driver mutations and modular epistasis networks in cancer

Cancer driver mutations often display mutual exclusion or co-occurrence, underscoring the key role of epistasis in carcinogenesis. However, estimating the magnitude of epistatic interactions and their quantitative effect on tumor evolution remains a challenge. We developed a method to quantify COnditional SELection on the Excess of Nonsynonymous Substitutions (Coselens) in cancer genes. Coselens infers the number of drivers per gene in different partitions of a cancer genomics dataset using covariance-based mutation models and determines whether coding mutations in a gene affect selection for drivers in any other gene. Using Coselens, we identified 296 conditionally selected gene pairs across 16 cancer types in the TCGA dataset. Conditional selection accounts for 25-50% of driver substitutions in tumors with >2 drivers. Conditionally co-selected genes form modular networks, whose structures challenge the traditional interpretation of within-pathway mutual exclusivity and across-pathway synergy, suggesting a more complex scenario, where gene-specific across-pathway interactions shape differentiated cancer subtypes.

Network models of prostate cancer immune microenvironments identify ROMO1 as heterogeneity and prognostic marker

Prostate cancer (PCa) is the fifth leading cause of death from cancer in men worldwide. Its treatment remains challenging due to the heterogeneity of the tumor, mainly because of the lack of effective and targeted prognostic markers at the system biology level. First, the data were retrieved from TCGA dataset, and valid samples were obtained by consistent clustering and principal component analysis; next, key genes were analyzed for prognosis of PCa using WGCNA, MEGENA, and LASSO Cox regression model analysis, while key genes were screened based on disease-free survival significance. Finally, TIMER data were selected to explore the relationship between genes and tumor immune infiltration, and GSCAlite was used to explore the small-molecule targeted drugs that act with them. Here, we used tumor subtype analysis and an energetic co-expression network algorithm of WGCNA and MEGENA to identify a signal dominated by the ROMO1 to predict PCa prognosis. Cox regression analysis of ROMO1 was an independent influence, and the prognostic value of this biomarker was validated in the training set, the validated data itself, and external data, respectively. This biomarker correlates with tumor immune infiltration and has a high degree of infiltration, poor prognosis, and strong correlation with CD8+T cells. Gene function annotation and other analyses also implied a potential molecular mechanism for ROMO1. In conclusion, we putative ROMO1 as a portal key prognostic gene for the diagnosis and prognosis of PCa, which provides new insights into the diagnosis and treatment of PCa.

Identification of Key Pathways and Genes Related to Immunotherapy Resistance of LUAD Based on WGCNA Analysis

Immunotherapy resistance is a major barrier in the application of immune checkpoint inhibitors (ICI) in lung adenocarcinoma (LUAD) patients. Although recent studies have found several mechanisms and potential genes responsible for immunotherapy resistance, ways to solve this problem are still lacking. Tumor immune dysfunction and exclusion (TIDE) algorithm is a newly developed method to calculate potential regulators and indicators of ICI resistance. In this article, we combined TIDE and weighted gene co-expression network analysis (WGCNA) to screen potential modules and hub genes that are highly associated with immunotherapy resistance using the Cancer Genome Atlas (TCGA) dataset of LUAD patients. We identified 45 gene co-expression modules, and the pink module was most correlated with TIDE score and other immunosuppressive features. After considering the potential factors in immunotherapy resistance, we found that the pink module was also highly related to cancer stemness. Further analysis showed enriched immunosuppressive cells in the extracellular matrix (ECM), immunotherapy resistance indicators, and common cancer-related signaling pathways in the pink module. Seven hub genes in the pink module were shown to be significantly upregulated in tumor tissues compared with normal lung tissue, and were related to poor survival of LUAD patients. Among them, THY1 was the gene most associated with TIDE score, a gene highly related to suppressive immune states, and was shown to be strongly expressed in late-stage patients. Immunohistochemistry (IHC) results demonstrated that THY1 level was higher in the progressive disease (PD) group of LUAD patients receiving a PD-1 monoclonal antibody (mAb) and positively correlated with SOX9. Collectively, we identified that THY1 could be a critical biomarker in predicting ICI efficiency and a potential target for avoiding tumor immunotherapy resistance.

Phosphoserine phosphatase as a prognostic biomarker in patients with gastric cancer and its potential association with immune cells

Background Because of dismal prognosis in gastric cancer, identifying relevant prognostic factors is necessary. Phosphoserine phosphatase (PSPH) exhibits different expression patterns in many cancers and has been reported to affect the prognosis of patients with cancer. In this study, we examined the prognostic role of metabolic gene PSPH in gastric cancer based on the TCGA dataset and our hospital–based cohort cases. Methods We collected and analysed RNA-seq data of Pan-cancer and gastric cancer in the TCGA dataset and PSPH expression data obtained from immunohistochemical analysis of 243 patients with gastric cancer from Sun Yat-sen University cancer center. Further, Kaplan–Meier survival analysis and Cox analysis were used to assess the effect of PSPH on prognosis. The ESTIMATE and Cibersort algorithms were used to elucidate the relationship between PSPH and the abundance of immune cells using the TCGA dataset. Results We observed that PSPH expression displayed considerably high in gastric cancer and it was significantly associated with inferior prognosis (P = 0.043). Surprisingly, there was a significant relationship between lower immune scores and high expression of PSPH (P < 0.05). Furthermore, patients with a low amount of immune cells exhibited poor prognosis (P = 0.046). The expression of PSPH significantly increased in activated memory CD4 T cells, resting NK cells and M0 macrophages (P = 0.037, < 0.001, and 0.005, respectively). Conclusions This study highlighted that PSPH influences the prognosis of patients with gastric cancer, and this is associated with the infiltration of tumour immune cells, indicating that PSPH may be a new immune-related target for treating gastric cancer.

Prognostic Significance of SATB1, SMAD3, Ezrin and β-Catenin in Patients with Pancreatic Adenocarcinoma

The present study aimed to explore the role of SATB1, SMAD3, Ezrin and β-catenin as individual and combined biomarkers for the survival prediction in pancreatic adenocarcinoma (PAC). Notably, this study distinguished for the first time a potential prognostic value of SATB1 corresponding to its subcellular localization in PAC. Immunohistochemical staining on tissue macroarrays, as well as RNA-seq data from public sources, were investigated, and the results correlated with overall survival (OS) and clinicopathological features. The connectivity between the examined factors, as well as their common signaling pathways, were demonstrated by the functional enrichment analysis. Herein, the prognostic ability of cytoplasmic SATB1 in OS analysis was even superior to nuclear SATB1. Both staining patterns tended to have opposite roles in the prognosis of PAC: SATB1c was an independent prognostic factor for poor OS, whereas SATB1n expression reached no statistical significance, but Kaplan–Meier curves separated patients with low expression and adverse prognosis from patients with high expression and favorable prognosis. High levels of SATB1 mRNA appeared as an independent prognostic indicator for better OS. Furthermore, individual expression of SMAD3 or Ezrin, as well as combined expression of SATB1/SMAD3/Ezrin/β-catenin, were associated with OS independently of conventional risk factors, both in our cohort and TCGA dataset. In our series, patients with tumors harboring combined expression of SATB1n-high/SMAD3low/Ezrinlow/β-cateninlow experienced the highest survival rates, while those with SATB1c-present/SMAD3high/Ezrinhigh/β-cateninhigh had the worst survival. In conclusion, protein and/or mRNA expression levels of SATB1, SMAD3, Ezrin and β-catenin may serve as potential prognostic biomarkers for PAC, both as single predictors and even better when combined.

β-Tubulin Isotype, TUBB4B, Regulates The Maintenance of Cancer Stem Cells

Recent advancements in cancer research have shown that cancer stem cell (CSC) niche is a crucial factor modulating tumor progression and treatment outcomes. It sustains CSCs by orchestrated regulation of several cytokines, growth factors, and signaling pathways. Although the features defining adult stem cell niches are well-explored, the CSC niche is poorly characterized. Since membrane trafficking proteins have been shown to be essential for the localization of critical proteins supporting CSCs, we investigated the role of TUBB4B, a probable membrane trafficking protein that was found to be overexpressed in the membranes of stem cell enriched cultures, in sustaining CSCs in oral cancer. Here, we show that the knockdown of TUBB4B downregulates the expression of pluripotency markers, depletes ALDH1A1+ population, decreases in vitro sphere formation, and diminishes the tumor initiation potential in vivo. As TUBB4B is not known to have any role in transcriptional regulation nor cell signaling, we suspected that its membrane trafficking function plays a role in constituting a CSC niche. The pattern of its expression in tissue sections, forming a gradient in and around the CSCs, reinforced the notion. Later, we explored its possible cooperation with a signaling protein, Ephrin-B1, the abrogation of which reduces the self-renewal of oral cancer stem cells. Expression and survival analyses based on the TCGA dataset of head and neck squamous cell carcinoma (HNSCC) samples indicated that the functional cooperation of TUBB4 and EFNB1 results in a poor prognosis. We also show that TUBB4B and Ephrin-B1 cohabit in the CSC niche. Moreover, depletion of TUBB4B downregulates the membrane expression of Ephrin-B1 and reduces the CSC population. Our results imply that the dynamics of TUBB4B is decisive for the surface localization of proteins, like Ephrin-B1, that sustain CSCs by their concerted signaling.

GPR87 Promotes Metastasis through the AKT-eNOS-NO Axis in Lung Adenocarcinoma

Lung adenocarcinoma is one of the leading causes of cancer-related deaths. Despite the availability of advanced anticancer drugs for lung cancer treatment, the prognosis of patients still remains poor. There is a need to explore novel oncogenic mechanisms to overcome these therapeutic limitations. The functional experiments in vitro and in vivo were performed to evaluate the role of GPR87 expression on lung adenocarcinoma metastasis. The public lung adenocarcinoma TCGA dataset was used to determine the clinical relevance of GPR87 expression in patients with lung adenocarcinoma. GPR87 is upregulated in various cancer; however, the biological function of GPR87 has not yet been established in lung adenocarcinoma. In this study, we found that GPR87 expression is upregulated in lung adenocarcinoma and is associated with poor patient prognosis. Additionally, we showed that GPR87 overexpression promotes invasiveness and metastasis of lung adenocarcinoma cells. Furthermore, we demonstrated that AKT-eNOS-NO signaling is a novel downstream pathway of GPR87 in lung adenocarcinoma. Conversely, we confirmed that silencing of GPR87 expression suppressed these phenotypes. Our results reveal the oncogenic function of GPR87 in cancer progression and metastasis through the activation of eNOS as a key mediator. Therefore, we propose that targeting eNOS could be a novel therapeutic strategy to improve the clinical treatment of lung adenocarcinoma.