Interferons can trigger robust anti-tumor responses following immune checkpoint blockade (ICB). However, chronic interferon signaling can also reshape tumor cell phenotypes, selecting for immune evasion mechanisms that drive resistance. In a cohort of 108 metastatic melanoma samples collected prior to first-line ICB, immune infiltration and bulk interferon gamma (IFNG) expression correlate with initial but not durable responses. Using a regression-based approach to deconvolve cancer cell signals from bulk tumors, we find that tumors from patients who initially respond but later acquire resistance exhibit heightened cancer cell-dependent IFN signaling at baseline, accompanied by MYC downregulation, dedifferentiation, and impaired major histocompatibility complex (MHC) class II induction. Genetically, acquired resistance also associates with reduced T cell interferon signaling, lower memory T cell activity, high tumor mutational burden, clonal diversification, and elevated subclonal neoantigen burden. These findings reveal baseline transcriptomic and genetic features distinguishing acquired resistance from durable ICB response, identifying candidate targets to prevent relapse and refine stratification.

BACKGROUND AND PURPOSE:Hidden cardiotoxicity is defined as drug-induced cardiotoxicity that becomes obvious only in the presence of comorbidities. However, the molecular mechanisms of hidden cardiotoxicity are not always known. Therefore, unbiased multi-omics approaches could assist in revealing regulatory pathways. The most notable representative of hidden cardiotoxic drugs is the cyclooxygenase-2-inhibitor, rofecoxib. We previously reported increased mortality in rats because of proarrhythmic effects of rofecoxib in ischaemic hearts. Here, we aimed to identify molecular mechanisms of hidden cardiotoxicity exemplified by rofecoxib that present prior to comorbidities. EXPERIMENTAL APPROACH:Rats were treated with rofecoxib or its vehicle for 4 weeks. RNA sequencing and proteomic datasets of heart samples were used for differential expression and pathway reconstruction analyses. KEY RESULTS:In this model, mechanisms of hidden cardiotoxicity could not be revealed by transcriptomic analyses. However, mass-spectrometry-based proteomics showed conspicuous changes, revealing 132 proteins that were dysregulated in expression or on phosphorylation sites. Importantly, the phospho-proteomics allowed us to identify two kinases that may mediate cardiotoxicity. Finally, pathway reconstruction maps a complex molecular machinery whose clustered proteins regulate processes involving cytoskeleton binding, mRNA processing, proteolysis, translation, citrate acid cycle and calcium ion signalling. CONCLUSION AND IMPLICATIONS:This is the first demonstration that multi-omics characterisation can reveal underlying regulatory pathways of hidden cardiotoxicity. Importantly, our study shows that transcriptomics gives limited information on the hidden cardiotoxic effects of rofecoxib, which are mainly mediated by changes in posttranslational modifications and protein expression. These changes, among other mechanisms, may disturb the cardiac calcium handling, which could explain the fatal arrhythmias following ischaemia/reperfusion observed with rofecoxib.

High-risk neuroblastoma is one of the most lethal childhood cancers. Half of these tumors are driven by MYCN gene amplification (MNA). Despite intensive chemo- and radiotherapy, only 40% of patients survive, and they often suffer from long-term side effects of these genotoxic treatments. Thus, less toxic and more efficacious therapies are needed. Here, we identified diphenyleneiodonium chloride (DPI) as tool compound that preferentially targeted MNA neuroblastoma. Using proteomic and metabolomic assays we investigated the DPI mode of action, finding that DPI could target alterations induced by MNA. These included cell cycle progression, DNA repair pathways, and profound changes in the expression of proteins participating in the mitochondrial electron transport chain. In addition, DPI also downregulated MYCN protein levels preferentially in MNA cells. Metabolic and biological assays suggested that alterations in mitochondrial function and the associated production of reactive oxygen species are critical DPI targets. DPI reduced the proliferation, survival, and malignant transformation of MNA neuroblastoma cells. DPI also shrank tumors and prevented metastatic spread in zebrafish models of neuroblastoma. These findings suggest that processes impacted by DPI could be valuable new targets for the development of non-genotoxic drugs against MNA neuroblastoma. ### Competing Interest Statement The authors have declared no competing interest.

Patients with high-grade serous ovarian cancer (HGSOC) typically present with widespread metastasis, obscuring a temporal understanding of tumor-immune dynamics. To address this, we perform multi-site global proteomics alongside matched immunohistochemistry (IHC) for CD4⁺ and CD8⁺ tumor-infiltrating lymphocytes (TILs) in patient samples. We order the protein expression profiles using an unbiased pseudotime analysis, recapitulating clinical observations of metastatic progression, and providing a framework to explore tumor-immune dynamics from localized to metastatic disease. Metastatic progression correlates with immune cell infiltration, the recruitment of regulatory T cells (Tregs) to counterbalance γδ T cell abundance, and an increased abundance of exhausted CD8⁺ T cells. The accumulation of Tregs at metastatic sites correlates with SNX8 expression, a critical regulator of the STING pathway. In early-stage tumors, keratin-expressing cancer cells recruit Tregs via MHC class II, fostering an inflammatory phenotype with limited IFNγ production and non-clonally expanded T cells. Together, our findings reveal novel mechanisms of immune escape associated with both localized disease and metastatic progression in HGSOC.

High risk - neuroblastoma (HR-NB) is a pediatric solid tumor with high lethality. Half of HR-NB are driven by MYCN gene amplification (MNA). These HR-NBs require high dosage chemotherapy and often relapse. Moreover, current therapies can cause severe long-term side effects and new therapies are urgently needed. This study investigates a novel therapeutic approach targeting the metabolic vulnerabilities of MNA NB cells. We discovered that Diphenyleneiodonium chloride (DPI), an inhibitor of flavoprotein enzymes and mitochondrial complex I, synergizes with mitoquinone mesylate (MitoQ), a mitochondria-targeted antioxidant in 2D and 3D in vitro models of NB. Similarly to DPI, MitoQ appears to have a greater effect on cells with higher MYCN levels. Furthermore, low nanomolar concentrations of MitoQ significantly decrease MYCN protein expression and induce differentiation of MNA cells. The DPI and MitoQ combination further synergizes with vincristine, a chemotherapeutic agent used in NB treatment. Phosphoproteomics and proteomics analysis suggests that the drug combination induces MNA NB cell death by arresting the cell cycle and inhibiting oxidative phosphorylation (OXPHOS) in the mitochondria. Thus, interference with mitochondrial metabolism may represent an effective strategy to enhance the activity of chemotherapeutic drugs in MNA-NB.

T cell co-inhibitory immune checkpoints, such as PD-1 or BTLA, are bona fide targets in cancer therapy. We used a human T cell reporter line to measure transcriptomic changes mediated by PD-1- and BTLA-induced signaling. T cell receptor (TCR)-complex stimulation resulted in the upregulation of a large number of genes but also in repression of a similar number of genes. PD-1 and BTLA signals attenuated transcriptomic changes mediated by TCR-complex signaling: upregulated genes tended to be suppressed and the expression of a significant number of downregulated genes was higher during PD-1 or BTLA signaling. BTLA was a significantly stronger attenuator of TCR-complex-induced transcriptome changes than PD-1. A strong overlap between genes that were regulated indicated quantitative rather than qualitative differences between these receptors. In line with their function as attenuators of TCR-complex-mediated changes, we found strongly regulated genes to be prime targets of PD-1 and BTLA signaling.

Patients with high-grade serous ovarian cancer (HGSOC) typically present with widespread metastasis, obscuring a temporal understanding of tumor-immune dynamics. We performed multi-site global proteomics and matched immunohistochemistry (IHC) of CD4+ and CD8+ tumor infiltrating lymphocytes (TILs) in patient samples. The protein expression profiles were ordered by pseudotime, recapitulating metastatic progression observed in the clinic, and providing a framework to explore tumor-immune dynamics from localized to metastatic disease. Metastatic progression correlated with immune cell infiltration and the recruitment of Tregs to counterbalance the effect of γδ and CD4+ T cells. Whilst later-stage metastases recruited Tregs via chemokines induced by SNX8, early-stage tumors relied more on antigen presentation. The exclusion of CD4+ TILs from the epithelium was correlated with metastatic progression, whereas CD8+ TILs were not, likely due to a predominance of exhausted CD8+ T cells. In early-stage tumors, keratin-expressing cancer cells recruited Tregs via MHC class II, resulting in an inflammatory phenotype characterized by limited IFNγ production and non-clonally expanded T cells. Additionally, IFI44L expression on macrophages caused immune exclusion by downregulating CD53 on T cells. Our findings reveal novel mechanisms of immune escape associated with localized disease and metastatic progression in HGSOC, highlighting potential targets to improve the efficacy of immunotherapy. ### Competing Interest Statement Daniel C. Liebler is a stockholder and employee of Inotiv, Inc., which provides analytical services described. Salisha Hill and Ryan D. Morrison are employees of Inotiv. The remaining authors declare no competing interests.

Immune checkpoint inhibitors (ICIs) have rejuvenated therapeutic approaches in oncology. Although responses tend to be durable, response rates vary in many cancer types. Thus, the identification and validation of predictive biomarkers is a key clinical priority, the answer to which is likely to lie in the tumour microenvironment (TME). A wealth of data demonstrates the huge impact of the TME on ICI response and resistance. However, these data also reveal the complexity of the TME composition including the spatiotemporal interactions between different cell types and their dynamic changes in response to ICIs. Here, we briefly review some of the modalities that sculpt the TME, in particular the metabolic milieu, hypoxia and the role of cancer-associated fibroblasts. We then discuss recent approaches to dissect the TME with a focus on single-cell RNA sequencing, spatial transcriptomics and spatial proteomics. We also discuss some of the clinically relevant findings these multi-modal analyses have yielded.

Abstract Single-cell technologies have elucidated mechanisms responsible for immune checkpoint inhibitor (ICI) response, but are not amenable to a clinical diagnostic setting. In contrast, bulk RNA sequencing (RNA-seq) is now routine for research and clinical applications. Our workflow uses transcription factor (TF)–directed coexpression networks (regulons) inferred from single-cell RNA-seq data to deconvolute immune functional states from bulk RNA-seq data. Regulons preserve the phenotypic variation in CD45+ immune cells from metastatic melanoma samples (n = 19, discovery dataset) treated with ICIs, despite reducing dimensionality by >100-fold. Four cell states, termed exhausted T cells, monocyte lineage cells, memory T cells, and B cells were associated with therapy response, and were characterized by differentially active and cell state–specific regulons. Clustering of bulk RNA-seq melanoma samples from four independent studies (n = 209, validation dataset) according to regulon-inferred scores identified four groups with significantly different response outcomes (P < 0.001). An intercellular link was established between exhausted T cells and monocyte lineage cells, whereby their cell numbers were correlated, and exhausted T cells predicted prognosis as a function of monocyte lineage cell number. The ligand–receptor expression analysis suggested that monocyte lineage cells drive exhausted T cells into terminal exhaustion through programs that regulate antigen presentation, chronic inflammation, and negative costimulation. Together, our results demonstrate how regulon-based characterization of cell states provide robust and functionally informative markers that can deconvolve bulk RNA-seq data to identify ICI responders.

OBJECTIVE:Investigate the clinical and functional implications of elevated CRABP2 expression in endometrial cancer (EC) patients.METHODS:Patients were stratified into high and low CRABP2 expression groups using a decision tree classifier. Univariate and multivariate statistical analyses determined the prognostic and clinicopathological consequences of increased CRABP2 expression. A CRABP2 gene signature was generated using differential expression analysis, and analyzed using network-based approaches. The findings were validated in The Clinical Proteomic Tumor Analysis Consortium (CPTAC), a newly generated cohort of 120 endometrial tissues, and The Cancer Dependency Map (DepMap).RESULTS:60 (11%) patients in TCGA had high CRABP2 expression, whilst 468 (89%) had low expression. High expression was associated with serous EC, reduced overall survival, advanced stage and grade. Downstream retinoic acid receptors (RARG and RARA) were correlated with CRABP2 expression and were associated with worse prognosis in serous EC. The CRABP2 gene signature was enriched for Polycomb target gene sets, and was regulated by ELP3 and BMP7. BMP7 expression was increased in the CRABP2-high group, was associated with worse prognosis, and CRISPR-Cas9 screens revealed correlations in its cell-fitness score with CRABP2 following gene knockout. The opposite was true for ELP3, suggesting opposing effects from both master regulators.CONCLUSIONS:CRABP2 expression is associated with poor prognosis and advanced EC. The expression of RARA and RARG correlates with CRABP2 and are associated with worse prognosis in advanced histological subtypes. Polycomb target gene sets and two master regulators, ELP3 and BMP7, were identified as functionally relevant mechanisms driving aberrant CRABP2 expression.

The T-cell immunoglobulin and ITIM domain (TIGIT) is a new inhibitory receptor that represents a novel target for the development of immunotherapy strategies. Using an in-silico approach, we identified differentially expressed genes (DEGs) and enriched pathways associated with TIGIT mRNA expression, in high grade serous ovarian cancer (HGSOC) using the Cancer Genome Atlas (TCGA) and the Australian Ovarian Cancer Study (AOCS). Methods DEGs between patients with high and low TIGIT expression, stratified based on an unsupervised tree analysis were calculated using EdgeR. Enriched pathways with the DEG list were identified using Gene Set Enrichment Analysis (GSEA) using a False Discovery Rate (FDR) <0.25 as significant. Results Increased TIGIT mRNA expression was associated with improved survival in HGSOC (p=0.034). 975 DEGs were identified in the TIGIT high group, and GSEA identified enriched pathways involved in complement activation humoral immune response, suggesting that TIGIT expression may be associated with an immunologically 'hot' tumour. This was confirmed by the finding that increased TIGIT expression was associated with an increased lymphocytic infiltration score, CD8+ T cells and Interferon Gamma Response score. Finally TIGIT expression was reduced in AOCS samples from women with acquired platinum resistance compared to matched primary tumour samples (p=0.014) Conclusion TIGIT represents an important prognostic marker in HGSOC. Similar to PD-1/PD-L1, TIGIT is associated with increased tumour infiltrating lymphocytes and an improved prognosis. Platinum resistance is associated with a reduction in TIGIT expression and warrants further study in HGSOC.