@article{KotlyarKrebsSolimandoetal.2023, author = {Kotlyar, Mischa J. and Krebs, Markus and Solimando, Antonio Giovanni and Marquardt, Andr{\´e} and Burger, Maximilian and K{\"u}bler, Hubert and Bargou, Ralf and Kneitz, Susanne and Otto, Wolfgang and Breyer, Johannes and Vergho, Daniel C. and Kneitz, Burkhard and Kalogirou, Charis}, title = {Critical evaluation of a microRNA-based risk classifier predicting cancer-specific survival in renal cell carcinoma with tumor thrombus of the inferior vena cava}, series = {Cancers}, volume = {15}, journal = {Cancers}, number = {7}, issn = {2072-6694}, doi = {10.3390/cancers15071981}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-311040}, year = {2023}, abstract = {(1) Background: Clear cell renal cell carcinoma extending into the inferior vena cava (ccRCC\(^{IVC}\)) represents a clinical high-risk setting. However, there is substantial heterogeneity within this patient subgroup regarding survival outcomes. Previously, members of our group developed a microRNA(miR)-based risk classifier — containing miR-21-5p, miR-126-3p and miR-221-3p expression — which significantly predicted the cancer-specific survival (CSS) of ccRCC\(^{IVC}\) patients. (2) Methods: Examining a single-center cohort of tumor tissue from n = 56 patients with ccRCC\(^{IVC}\), we measured the expression levels of miR-21, miR-126, and miR-221 using qRT-PCR. The prognostic impact of clinicopathological parameters and miR expression were investigated via single-variable and multivariable Cox regression. Referring to the previously established risk classifier, we performed Kaplan-Meier analyses for single miR expression levels and the combined risk classifier. Cut-off values and weights within the risk classifier were taken from the previous study. (3) Results: miR-21 and miR-126 expression were significantly associated with lymphonodal status at the time of surgery, the development of metastasis during follow-up, and cancer-related death. In Kaplan-Meier analyses, miR-21 and miR-126 significantly impacted CSS in our cohort. Moreover, applying the miR-based risk classifier significantly stratified ccRCC\(^{IVC}\) according to CSS. (4) Conclusions: In our retrospective analysis, we successfully validated the miR-based risk classifier within an independent ccRCC\(^{IVC}\) cohort.}, language = {en} } @article{MarquardtHartrampfKollmannsbergeretal.2023, author = {Marquardt, Andr{\´e} and Hartrampf, Philipp and Kollmannsberger, Philip and Solimando, Antonio G. and Meierjohann, Svenja and K{\"u}bler, Hubert and Bargou, Ralf and Schilling, Bastian and Serfling, Sebastian E. and Buck, Andreas and Werner, Rudolf A. and Lapa, Constantin and Krebs, Markus}, title = {Predicting microenvironment in CXCR4- and FAP-positive solid tumors — a pan-cancer machine learning workflow for theranostic target structures}, series = {Cancers}, volume = {15}, journal = {Cancers}, number = {2}, issn = {2072-6694}, doi = {10.3390/cancers15020392}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-305036}, year = {2023}, abstract = {(1) Background: C-X-C Motif Chemokine Receptor 4 (CXCR4) and Fibroblast Activation Protein Alpha (FAP) are promising theranostic targets. However, it is unclear whether CXCR4 and FAP positivity mark distinct microenvironments, especially in solid tumors. (2) Methods: Using Random Forest (RF) analysis, we searched for entity-independent mRNA and microRNA signatures related to CXCR4 and FAP overexpression in our pan-cancer cohort from The Cancer Genome Atlas (TCGA) database — representing n = 9242 specimens from 29 tumor entities. CXCR4- and FAP-positive samples were assessed via StringDB cluster analysis, EnrichR, Metascape, and Gene Set Enrichment Analysis (GSEA). Findings were validated via correlation analyses in n = 1541 tumor samples. TIMER2.0 analyzed the association of CXCR4 / FAP expression and infiltration levels of immune-related cells. (3) Results: We identified entity-independent CXCR4 and FAP gene signatures representative for the majority of solid cancers. While CXCR4 positivity marked an immune-related microenvironment, FAP overexpression highlighted an angiogenesis-associated niche. TIMER2.0 analysis confirmed characteristic infiltration levels of CD8+ cells for CXCR4-positive tumors and endothelial cells for FAP-positive tumors. (4) Conclusions: CXCR4- and FAP-directed PET imaging could provide a non-invasive decision aid for entity-agnostic treatment of microenvironment in solid malignancies. Moreover, this machine learning workflow can easily be transferred towards other theranostic targets.}, language = {en} } @phdthesis{Marquardt2023, author = {Marquardt, Andr{\´e}}, title = {Machine-Learning-Based Identification of Tumor Entities, Tumor Subgroups, and Therapy Options}, doi = {10.25972/OPUS-32954}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-329548}, school = {Universit{\"a}t W{\"u}rzburg}, year = {2023}, abstract = {Molecular genetic analyses, such as mutation analyses, are becoming increasingly important in the tumor field, especially in the context of therapy stratification. The identification of the underlying tumor entity is crucial, but can sometimes be difficult, for example in the case of metastases or the so-called Cancer of Unknown Primary (CUP) syndrome. In recent years, methylome and transcriptome utilizing machine learning (ML) approaches have been developed to enable fast and reliable tumor and tumor subtype identification. However, so far only methylome analysis have become widely used in routine diagnostics. The present work addresses the utility of publicly available RNA-sequencing data to determine the underlying tumor entity, possible subgroups, and potential therapy options. Identification of these by ML - in particular random forest (RF) models - was the first task. The results with test accuracies of up to 99\% provided new, previously unknown insights into the trained models and the corresponding entity prediction. Reducing the input data to the top 100 mRNA transcripts resulted in a minimal loss of prediction quality and could potentially enable application in clinical or real-world settings. By introducing the ratios of these top 100 genes to each other as a new database for RF models, a novel method was developed enabling the use of trained RF models on data from other sources. Further analysis of the transcriptomic differences of metastatic samples by visual clustering showed that there were no differences specific for the site of metastasis. Similarly, no distinct clusters were detectable when investigating primary tumors and metastases of cutaneous skin melanoma (SKCM). Subsequently, more than half of the validation datasets had a prediction accuracy of at least 80\%, with many datasets even achieving a prediction accuracy of - or close to - 100\%. To investigate the applicability of the used methods for subgroup identification, the TCGA-KIPAN dataset, consisting of the three major kidney cancer subgroups, was used. The results revealed a new, previously unknown subgroup consisting of all histopathological groups with clinically relevant characteristics, such as significantly different survival. Based on significant differences in gene expression, potential therapeutic options of the identified subgroup could be proposed. Concludingly, in exploring the potential applicability of RNA-sequencing data as a basis for therapy prediction, it was shown that this type of data is suitable to predict entities as well as subgroups with high accuracy. Clinical relevance was also demonstrated for a novel subgroup in renal cell carcinoma. The reduction of the number of genes required for entity prediction to 100 genes, enables panel sequencing and thus demonstrates potential applicability in a real-life setting.}, subject = {Maschinelles Lernen}, language = {en} } @article{MarquardtKollmannsbergerKrebsetal.2022, author = {Marquardt, Andr{\´e} and Kollmannsberger, Philip and Krebs, Markus and Argentiero, Antonella and Knott, Markus and Solimando, Antonio Giovanni and Kerscher, Alexander Georg}, title = {Visual clustering of transcriptomic data from primary and metastatic tumors — dependencies and novel pitfalls}, series = {Genes}, volume = {13}, journal = {Genes}, number = {8}, issn = {2073-4425}, doi = {10.3390/genes13081335}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-281872}, year = {2022}, abstract = {Personalized oncology is a rapidly evolving area and offers cancer patients therapy options that are more specific than ever. However, there is still a lack of understanding regarding transcriptomic similarities or differences of metastases and corresponding primary sites. Applying two unsupervised dimension reduction methods (t-Distributed Stochastic Neighbor Embedding (t-SNE) and Uniform Manifold Approximation and Projection (UMAP)) on three datasets of metastases (n = 682 samples) with three different data transformations (unprocessed, log10 as well as log10 + 1 transformed values), we visualized potential underlying clusters. Additionally, we analyzed two datasets (n = 616 samples) containing metastases and primary tumors of one entity, to point out potential familiarities. Using these methods, no tight link between the site of resection and cluster formation outcome could be demonstrated, or for datasets consisting of solely metastasis or mixed datasets. Instead, dimension reduction methods and data transformation significantly impacted visual clustering results. Our findings strongly suggest data transformation to be considered as another key element in the interpretation of visual clustering approaches along with initialization and different parameters. Furthermore, the results highlight the need for a more thorough examination of parameters used in the analysis of clusters.}, language = {en} } @article{KressJessenMarquardtetal.2021, author = {Kreß, Julia Katharina Charlotte and Jessen, Christina and Marquardt, Andr{\´e} and Hufnagel, Anita and Meierjohann, Svenja}, title = {NRF2 enables EGFR signaling in melanoma cells}, series = {International Journal of Molecular Sciences}, volume = {22}, journal = {International Journal of Molecular Sciences}, number = {8}, issn = {1422-0067}, doi = {10.3390/ijms22083803}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-260222}, year = {2021}, abstract = {Receptor tyrosine kinases (RTK) are rarely mutated in cutaneous melanoma, but the expression and activation of several RTK family members are associated with a proinvasive phenotype and therapy resistance. Epidermal growth factor receptor (EGFR) is a member of the RTK family and is only expressed in a subgroup of melanomas with poor prognosis. The insight into regulators of EGFR expression and activation is important for the understanding of the development of this malignant melanoma phenotype. Here, we describe that the transcription factor NRF2, the master regulator of the oxidative and electrophilic stress response, mediates the expression and activation of EGFR in melanoma by elevating the levels of EGFR as well as its ligands EGF and TGFα. ChIP sequencing data show that NRF2 directly binds to the promoter of EGF, which contains a canonical antioxidant response element. Accordingly, EGF is induced by oxidative stress and is also increased in lung adenocarcinoma and head and neck carcinoma with mutationally activated NRF2. In contrast, regulation of EGFR and TGFA occurs by an indirect mechanism, which is enabled by the ability of NRF2 to block the activity of the melanocytic lineage factor MITF in melanoma. MITF effectively suppresses EGFR and TGFA expression and therefore serves as link between NRF2 and EGFR. As EGFR was previously described to stimulate NRF2 activity, the mutual activation of NRF2 and EGFR pathways was investigated. The presence of NRF2 was necessary for full EGFR pathway activation, as NRF2-knockout cells showed reduced AKT activation in response to EGF stimulation compared to controls. Conversely, EGF led to the nuclear localization and activation of NRF2, thereby demonstrating that NRF2 and EGFR are connected in a positive feedback loop in melanoma. In summary, our data show that the EGFR-positive melanoma phenotype is strongly supported by NRF2, thus revealing a novel maintenance mechanism for this clinically challenging melanoma subpopulation.}, language = {en} } @article{MarquardtSolimandoKerscheretal.2021, author = {Marquardt, Andr{\´e} and Solimando, Antonio Giovanni and Kerscher, Alexander and Bittrich, Max and Kalogirou, Charis and K{\"u}bler, Hubert and Rosenwald, Andreas and Bargou, Ralf and Kollmannsberger, Philip and Schilling, Bastian and Meierjohann, Svenja and Krebs, Markus}, title = {Subgroup-Independent Mapping of Renal Cell Carcinoma — Machine Learning Reveals Prognostic Mitochondrial Gene Signature Beyond Histopathologic Boundaries}, series = {Frontiers in Oncology}, volume = {11}, journal = {Frontiers in Oncology}, issn = {2234-943X}, doi = {10.3389/fonc.2021.621278}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-232107}, year = {2021}, abstract = {Background: Renal cell carcinoma (RCC) is divided into three major histopathologic groups—clear cell (ccRCC), papillary (pRCC) and chromophobe RCC (chRCC). We performed a comprehensive re-analysis of publicly available RCC datasets from the TCGA (The Cancer Genome Atlas) database, thereby combining samples from all three subgroups, for an exploratory transcriptome profiling of RCC subgroups. Materials and Methods: We used FPKM (fragments per kilobase per million) files derived from the ccRCC, pRCC and chRCC cohorts of the TCGA database, representing transcriptomic data of 891 patients. Using principal component analysis, we visualized datasets as t-SNE plot for cluster detection. Clusters were characterized by machine learning, resulting gene signatures were validated by correlation analyses in the TCGA dataset and three external datasets (ICGC RECA-EU, CPTAC-3-Kidney, and GSE157256). Results: Many RCC samples co-clustered according to histopathology. However, a substantial number of samples clustered independently from histopathologic origin (mixed subgroup)—demonstrating divergence between histopathology and transcriptomic data. Further analyses of mixed subgroup via machine learning revealed a predominant mitochondrial gene signature—a trait previously known for chRCC—across all histopathologic subgroups. Additionally, ccRCC samples from mixed subgroup presented an inverse correlation of mitochondrial and angiogenesis-related genes in the TCGA and in three external validation cohorts. Moreover, mixed subgroup affiliation was associated with a highly significant shorter overall survival for patients with ccRCC—and a highly significant longer overall survival for chRCC patients. Conclusions: Pan-RCC clustering according to RNA-sequencing data revealed a distinct histology-independent subgroup characterized by strengthened mitochondrial and weakened angiogenesis-related gene signatures. Moreover, affiliation to mixed subgroup went along with a significantly shorter overall survival for ccRCC and a longer overall survival for chRCC patients. Further research could offer a therapy stratification by specifically addressing the mitochondrial metabolism of such tumors and its microenvironment.}, language = {en} } @article{MarquardtLandwehrRonchietal.2021, author = {Marquardt, Andr{\´e} and Landwehr, Laura-Sophie and Ronchi, Cristina L. and di Dalmazi, Guido and Riester, Anna and Kollmannsberger, Philip and Altieri, Barbara and Fassnacht, Martin and Sbiera, Silviu}, title = {Identifying New Potential Biomarkers in Adrenocortical Tumors Based on mRNA Expression Data Using Machine Learning}, series = {Cancers}, volume = {13}, journal = {Cancers}, number = {18}, issn = {2072-6694}, doi = {10.3390/cancers13184671}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-246245}, year = {2021}, abstract = {Simple Summary Using a visual-based clustering method on the TCGA RNA sequencing data of a large adrenocortical carcinoma (ACC) cohort, we were able to classify these tumors in two distinct clusters largely overlapping with previously identified ones. As previously shown, the identified clusters also correlated with patient survival. Applying the visual clustering method to a second dataset also including benign adrenocortical samples additionally revealed that one of the ACC clusters is more closely located to the benign samples, providing a possible explanation for the better survival of this ACC cluster. Furthermore, the subsequent use of machine learning identified new possible biomarker genes with prognostic potential for this rare disease, that are significantly differentially expressed in the different survival clusters and should be further evaluated. Abstract Adrenocortical carcinoma (ACC) is a rare disease, associated with poor survival. Several "multiple-omics" studies characterizing ACC on a molecular level identified two different clusters correlating with patient survival (C1A and C1B). We here used the publicly available transcriptome data from the TCGA-ACC dataset (n = 79), applying machine learning (ML) methods to classify the ACC based on expression pattern in an unbiased manner. UMAP (uniform manifold approximation and projection)-based clustering resulted in two distinct groups, ACC-UMAP1 and ACC-UMAP2, that largely overlap with clusters C1B and C1A, respectively. However, subsequent use of random-forest-based learning revealed a set of new possible marker genes showing significant differential expression in the described clusters (e.g., SOAT1, EIF2A1). For validation purposes, we used a secondary dataset based on a previous study from our group, consisting of 4 normal adrenal glands and 52 benign and 7 malignant tumor samples. The results largely confirmed those obtained for the TCGA-ACC cohort. In addition, the ENSAT dataset showed a correlation between benign adrenocortical tumors and the good prognosis ACC cluster ACC-UMAP1/C1B. In conclusion, the use of ML approaches re-identified and redefined known prognostic ACC subgroups. On the other hand, the subsequent use of random-forest-based learning identified new possible prognostic marker genes for ACC.}, language = {en} } @article{JessenKressBaluapurietal.2020, author = {Jessen, Christina and Kreß, Julia K. C. and Baluapuri, Apoorva and Hufnagel, Anita and Schmitz, Werner and Kneitz, Susanne and Roth, Sabine and Marquardt, Andr{\´e} and Appenzeller, Silke and Ade, Casten P. and Glutsch, Valerie and Wobser, Marion and Friedmann-Angeli, Jos{\´e} Pedro and Mosteo, Laura and Goding, Colin R. and Schilling, Bastian and Geissinger, Eva and Wolf, Elmar and Meierjohann, Svenja}, title = {The transcription factor NRF2 enhances melanoma malignancy by blocking differentiation and inducing COX2 expression}, series = {Oncogene}, volume = {39}, journal = {Oncogene}, issn = {0950-9232}, doi = {10.1038/s41388-020-01477-8}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-235064}, pages = {6841-6855}, year = {2020}, abstract = {The transcription factor NRF2 is the major mediator of oxidative stress responses and is closely connected to therapy resistance in tumors harboring activating mutations in the NRF2 pathway. In melanoma, such mutations are rare, and it is unclear to what extent melanomas rely on NRF2. Here we show that NRF2 suppresses the activity of the melanocyte lineage marker MITF in melanoma, thereby reducing the expression of pigmentation markers. Intriguingly, we furthermore identified NRF2 as key regulator of immune-modulating genes, linking oxidative stress with the induction of cyclooxygenase 2 (COX2) in an ATF4-dependent manner. COX2 is critical for the secretion of prostaglandin E2 and was strongly induced by H\(_2\)O\(_2\) or TNFα only in presence of NRF2. Induction of MITF and depletion of COX2 and PGE2 were also observed in NRF2-deleted melanoma cells in vivo. Furthermore, genes corresponding to the innate immune response such as RSAD2 and IFIH1 were strongly elevated in absence of NRF2 and coincided with immune evasion parameters in human melanoma datasets. Even in vitro, NRF2 activation or prostaglandin E2 supplementation blunted the induction of the innate immune response in melanoma cells. Transcriptome analyses from lung adenocarcinomas indicate that the observed link between NRF2 and the innate immune response is not restricted to melanoma.}, language = {en} } @article{KrebsSolimandoKalogirouetal.2020, author = {Krebs, Markus and Solimando, Antonio Giovanni and Kalogirou, Charis and Marquardt, Andr{\´e} and Frank, Torsten and Sokolakis, Ioannis and Hatzichristodoulou, Georgios and Kneitz, Susanne and Bargou, Ralf and K{\"u}bler, Hubert and Schilling, Bastian and Spahn, Martin and Kneitz, Burkhard}, title = {miR-221-3p Regulates VEGFR2 Expression in High-Risk Prostate Cancer and Represents an Escape Mechanism from Sunitinib In Vitro}, series = {Journal of Clinical Medicine}, volume = {9}, journal = {Journal of Clinical Medicine}, number = {3}, issn = {2077-0383}, doi = {10.3390/jcm9030670}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-203168}, year = {2020}, abstract = {Downregulation of miR-221-3p expression in prostate cancer (PCa) predicted overall and cancer-specific survival of high-risk PCa patients. Apart from PCa, miR-221-3p expression levels predicted a response to tyrosine kinase inhibitors (TKI) in clear cell renal cell carcinoma (ccRCC) patients. Since this role of miR-221-3p was explained with a specific targeting of VEGFR2, we examined whether miR-221-3p regulated VEGFR2 in PCa. First, we confirmed VEGFR2/KDR as a target gene of miR-221-3p in PCa cells by applying Luciferase reporter assays and Western blotting experiments. Although VEGFR2 was mainly downregulated in the PCa cohort of the TCGA (The Cancer Genome Atlas) database, VEGFR2 was upregulated in our high-risk PCa cohort (n = 142) and predicted clinical progression. In vitro miR-221-3p acted as an escape mechanism from TKI in PC3 cells, as displayed by proliferation and apoptosis assays. Moreover, we confirmed that Sunitinib induced an interferon-related gene signature in PC3 cells by analyzing external microarray data and by demonstrating a significant upregulation of miR-221-3p/miR-222-3p after Sunitinib exposure. Our findings bear a clinical perspective for high-risk PCa patients with low miR-221-3p levels since this could predict a favorable TKI response. Apart from this therapeutic niche, we identified a partially oncogenic function of miR-221-3p as an escape mechanism from VEGFR2 inhibition.}, language = {en} }