Molecular subtyping of head and neck cancer - Clinical applicability and correlations with morphological characteristics.

AIM: We aimed to evaluate the applicability of a customized NanoString panel for molecular subtyping of recurrent or metastatic head and neck squamous cell carcinoma (R/M-HNSCC). Additionally, histological analyses were conducted, correlated with the molecular subtypes and tested for their prognostic value. MATERIAL AND METHODS: We conducted molecular subtyping of R/M-HNSCC according to the molecular subtypes defined by Keck et al. For molecular analyses a 231 gene customized NanoString panel (the most accurately subtype defining genes, based on previous analyses) was applied to tumor samples from R/M-HNSCC patients that were treated in the CeFCiD trial (AIO/IAG-KHT trial 1108). A total of 130 samples from 95 patients were available for sequencing, of which 80 samples from 67 patients passed quality controls and were included in histological analyses. H&E stained slides were evaluated regarding distinct morphological patterns (e.g. tumor budding, nuclear size, stroma content). RESULTS: Determination of molecular subtypes led to classification of tumor samples as basal (n = 46, 45 %), inflamed/mesenchymal (n = 31, 30 %) and classical (n = 26, 25 %). Expression levels of Amphiregulin (AREG) were significantly higher for the basal and classical subtypes compared to the mesenchymal subtype. While molecular subtypes did not have an impact on survival, high levels of tumor budding were associated with poor outcomes. No correlation was found between molecular subtypes and histological characteristics. CONCLUSIONS: Utilizing the 231-gene NanoString panel we were able to determine the molecular subtype of R/M-HNSCC samples by the use of FFPE material. The value to stratify for different treatment options remains to be explored in the future. The prognostic value of tumor budding was underscored in this clinically well annotated cohort.

Transcriptomic Deconvolution of Neuroendocrine Neoplasms Predicts Clinically Relevant Characteristics.

Pancreatic neuroendocrine neoplasms (panNENs) are a rare yet diverse type of neoplasia whose precise clinical-pathological classification is frequently challenging. Since incorrect classifications can affect treatment decisions, additional tools which support the diagnosis, such as machine learning (ML) techniques, are critically needed but generally unavailable due to the scarcity of suitable ML training data for rare panNENs. Here, we demonstrate that a multi-step ML framework predicts clinically relevant panNEN characteristics while being exclusively trained on widely available data of a healthy origin. The approach classifies panNENs by deconvolving their transcriptomes into cell type proportions based on shared gene expression profiles with healthy pancreatic cell types. The deconvolution results were found to provide a prognostic value with respect to the prediction of the overall patient survival time, neoplastic grading, and carcinoma versus tumor subclassification. The performance with which a proliferation rate agnostic deconvolution ML model could predict the clinical characteristics was found to be comparable to that of a comparative baseline model trained on the proliferation rate-informed MKI67 levels. The approach is novel in that it complements established proliferation rate-oriented classification schemes whose results can be reproduced and further refined by differentiating between identically graded subgroups. By including non-endocrine cell types, the deconvolution approach furthermore provides an in silico quantification of panNEN dedifferentiation, optimizing it for challenging clinical classification tasks in more aggressive panNEN subtypes.

Elevated Flt3L Predicts Long-Term Survival in Patients with High-Grade Gastroenteropancreatic Neuroendocrine Neoplasms.

BACKGROUND: The clinical management of high-grade gastroenteropancreatic neuroendocrine neoplasms (GEP-NEN) is challenging due to disease heterogeneity, illustrating the need for reliable biomarkers facilitating patient stratification and guiding treatment decisions. FMS-like tyrosine kinase 3 ligand (Flt3L) is emerging as a prognostic or predictive surrogate marker of host tumoral immune response and might enable the stratification of patients with otherwise comparable tumor features. METHODS: We evaluated Flt3L gene expression in tumor tissue as well as circulating Flt3L levels as potential biomarkers in a cohort of 54 patients with GEP-NEN. RESULTS: We detected a prominent induction of Flt3L gene expression in individual G2 and G3 NEN, but not in G1 neuroendocrine tumors (NET). Flt3L mRNA expression levels in tumor tissue predicted the disease-related survival of patients with highly proliferative G2 and G3 NEN more accurately than the conventional criteria of grading or NEC/NET differentiation. High level Flt3L mRNA expression was associated with the increased expression of genes related to immunogenic cell death, lymphocyte effector function and dendritic cell maturation, suggesting a less tolerogenic (more proinflammatory) phenotype of tumors with Flt3L induction. Importantly, circulating levels of Flt3L were also elevated in high grade NEN and correlated with patients' progression-free and disease-related survival, thereby reflecting the results observed in tumor tissue. CONCLUSIONS: We propose Flt3L as a prognostic biomarker for high grade GEP-NEN, harnessing its potential as a marker of an inflammatory tumor microenvironment. Flt3L measurements in serum, which can be easily be incorporated into clinical routine, should be further evaluated to guide patient stratification and treatment decisions.

Gene Set Enrichment Analysis Reveals Individual Variability in Host Responses in Tuberculosis Patients.

Group-aggregated responses to tuberculosis (TB) have been well characterized on a molecular level. However, human beings differ and individual responses to infection vary. We have combined a novel approach to individual gene set analysis (GSA) with the clustering of transcriptomic profiles of TB patients from seven datasets in order to identify individual molecular endotypes of transcriptomic responses to TB. We found that TB patients differ with respect to the intensity of their hallmark interferon (IFN) responses, but they also show variability in their complement system, metabolic responses and multiple other pathways. This variability cannot be sufficiently explained with covariates such as gender or age, and the molecular endotypes are found across studies and populations. Using datasets from a Cynomolgus macaque model of TB, we revealed that transcriptional signatures of different molecular TB endotypes did not depend on TB progression post-infection. Moreover, we provide evidence that patients with molecular endotypes characterized by high levels of IFN responses (IFN-rich), suffered from more severe lung pathology than those with lower levels of IFN responses (IFN-low). Harnessing machine learning (ML) models, we derived gene signatures classifying IFN-rich and IFN-low TB endotypes and revealed that the IFN-low signature allowed slightly more reliable overall classification of TB patients from non-TB patients than the IFN-rich one. Using the paradigm of molecular endotypes and the ML-based predictions allows more precisely tailored treatment regimens, predicting treatment-outcome with higher accuracy and therefore bridging the gap between conventional treatment and precision medicine.

Combination of copanlisib with cetuximab improves tumor response in cetuximab-resistant patient-derived xenografts of head and neck cancer.

Despite recent advances, the treatment of head and neck squamous cell carcinoma (HNSCC) remains an area of high unmet medical need. HNSCC is frequently associated with either amplification or mutational changes in the PI3K pathway, making PI3K an attractive target particularly in cetuximab-resistant tumors. Here, we explored the antitumor activity of the selective, pan-class I PI3K inhibitor copanlisib with predominant activity towards PI3Kα and δ in monotherapy and in combination with cetuximab using a mouse clinical trial set-up with 33 patient-derived xenograft (PDX) models with known HPV and PI3K mutational status and available data on cetuximab sensitivity. Treatment with copanlisib alone resulted in moderate antitumor activity with 12/33 PDX models showing either tumor stabilization or regression. Combination treatment with copanlisib and cetuximab was superior to either of the monotherapies alone in the majority of the models (21/33), and the effect was particularly pronounced in cetuximab-resistant tumors (14/16). While no correlation was observed between PI3K mutation status and response to either cetuximab or copanlisib, increased PI3K signaling activity evaluated through gene expression profiling showed a positive correlation with response to copanlisib. Together, these data support further investigation of PI3K inhibition in HNSCC and suggests gene expression patterns associated with PI3K signaling as a potential biomarker for predicting treatment responses.

Discovery and Validation of Novel Biomarkers for Detection of Epithelial Ovarian Cancer.

Detection of epithelial ovarian cancer (EOC) poses a critical medical challenge. However, novel biomarkers for diagnosis remain to be discovered. Therefore, innovative approaches are of the utmost importance for patient outcome. Here, we present a concept for blood-based biomarker discovery, investigating both epithelial and specifically stromal compartments, which have been neglected in search for novel candidates. We queried gene expression profiles of EOC including microdissected epithelium and adjacent stroma from benign and malignant tumours. Genes significantly differentially expressed within either the epithelial or the stromal compartments were retrieved. The expression of genes whose products are secreted yet absent in the blood of healthy donors were validated in tissue and blood from patients with pelvic mass by NanoString analysis. Results were confirmed by the comprehensive gene expression database, CSIOVDB (Ovarian cancer database of Cancer Science Institute Singapore). The top 25% of candidate genes were explored for their biomarker potential, and twelve were able to discriminate between benign and malignant tumours on transcript levels (p < 0.05). Among them T-cell differentiation protein myelin and lymphocyte (MAL), aurora kinase A (AURKA), stroma-derived candidates versican (VCAN), and syndecan-3 (SDC), which performed significantly better than the recently reported biomarker fibroblast growth factor 18 (FGF18) to discern malignant from benign conditions. Furthermore, elevated MAL and AURKA expression levels correlated significantly with a poor prognosis. We identified promising novel candidates and found the stroma of EOC to be a suitable compartment for biomarker discovery.

Robust in-silico identification of Cancer Cell Lines based on RNA and targeted DNA sequencing data.

Cancer cell lines (CCL) are an integral part of modern cancer research but are susceptible to misidentification. The increasing popularity of sequencing technologies motivates the in-silico identification of CCLs based on their mutational fingerprint, but care must be taken when identifying heterogeneous data. We recently developed the proof-of-concept Uniquorn 1 method which could reliably identify heterogeneous sequencing data from selected sequencing technologies. Here we present Uniquorn 2, a generic and robust in-silico identification method for CCLs with DNA/RNA-seq and panel-seq information. We benchmarked Uniquorn 2 by cross-identifying 1612 RNA and 3596 panel-sized NGS profiles derived from 1516 CCLs, five repositories, four technologies and three major cancer panel-designs. Our method achieves an accuracy of 96% for RNA-seq and 95% for mixed DNA-seq and RNA-seq identification. Even for a panel of only 94 cancer-related genes, accuracy remains at 82% but decreases when using smaller panels. Uniquorn 2 is freely available as R-Bioconductor-package 'Uniquorn'.

Basal subtype is predictive for response to cetuximab treatment in patient-derived xenografts of squamous cell head and neck cancer.

Cetuximab is the single targeted therapy approved for the treatment of head and neck cancer (HNSCC). Predictive biomarkers have not been established and patient stratification based on molecular tumor profiles has not been possible. Since EGFR pathway activation is pronounced in basal subtype, we hypothesized this activation could be a predictive signature for an EGFR directed treatment. From our patient-derived xenograft platform of HNSCC, 28 models were subjected to Affymetrix gene expression studies on HG U133+ 2.0. Based on the expression of 821 genes, the subtype of each of the 28 models was determined by integrating gene expression profiles through centroid-clustering with previously published gene expression data by Keck et al. The models were treated in groups of 5-6 animals with docetaxel, cetuximab, everolimus, cis- or carboplatin and 5-fluorouracil. Response was evaluated by comparing tumor volume at treatment initiation and after 3 weeks of treatment (RTV). Tumors distributed over the 3 signature-defined subtypes: 5 mesenchymal/inflamed phenotype (MS), 15 basal type (BA), 8 classical type (CL). Cluster analysis revealed a strong correlation between response to cetuximab and the basal subtype. RTV MS 3.32 vs. BA 0.78 (MS vs. BA, unpaired t-test, p 0.0002). Cetuximab responders were distributed as following: 1/5 in MS, 5/8 in CL and 13/15 in the BA group. Activity of classical chemotherapies did not differ between the subtypes. In conclusion basal subtype was associated with response to EGFR directed therapy in head and neck squamous cell cancer patient-derived xenografts.

Robust in-silico identification of cancer cell lines based on next generation sequencing.

Cancer cell lines (CCL) are important tools for cancer researchers world-wide. However, handling of cancer cell lines is error-prone, and critical errors such as misidentification and cross-contamination occur more often than acceptable. Based on the fact that CCL today very often are sequenced (partly or entirely) anyway as part of the studies performed, we developed Uniquorn, a computational method that reliably identifies CCL samples based on variant profiles derived from whole exome or whole genome sequencing. Notably, Uniquorn does neither require a particular sequencing technology nor downstream analysis pipeline but works robustly across different NGS platforms and analysis steps. We evaluated Uniquorn by comparing more than 1900 CCL profiles from three large CCL libraries, embracing 1585 duplicates, against each other. In this setting, our method achieves a sensitivity of 97% and specificity of 99%. Errors are strongly associated to low quality mutation profiles. The R-package Uniquorn is freely available as Bioconductor-package.

CancerResource--updated database of cancer-relevant proteins, mutations and interacting drugs.

Here, we present an updated version of CancerResource, freely available without registration at http://bioinformatics.charite.de/care. With upcoming information on target expression and mutations in patients' tumors, the need for systems supporting decisions on individual therapy is growing. This knowledge is based on numerous, experimentally validated drug-target interactions and supporting analyses such as measuring changes in gene expression using microarrays and HTS-efforts on cell lines. To enable a better overview about similar drug-target data and supporting information, a series of novel information connections are established and made available as described in the following. CancerResource contains about 91,000 drug-target relations, more than 2000 cancer cell lines and drug sensitivity data for about 50,000 drugs. CancerResource enables the capability of uploading external expression and mutation data and comparing them to the database's cell lines. Target genes and compounds are projected onto cancer-related pathways to get a better overview about how drug-target interactions benefit the treatment of cancer. Features like cellular fingerprints comprising of mutations, expression values and drug-sensitivity data can promote the understanding of genotype to drug sensitivity associations. Ultimately, these profiles can also be used to determine the most effective drug treatment for a cancer cell line most similar to a patient's tumor cells.