USP7 inhibitors suppress tumour neoangiogenesis and promote synergy with immune checkpoint inhibitors by downregulating fibroblast VEGF.

BACKGROUND: Understanding how to modulate the microenvironment of tumors that are resistant to immune checkpoint inhibitors represents a major challenge in oncology.Here we investigate the ability of USP7 inhibitors to reprogram the tumor microenvironment (TME) by inhibiting secretion of vascular endothelial growth factor (VEGF) from fibroblasts. METHODS: To understand the role played by USP7 in the TME, we systematically evaluated the effects of potent, selective USP7 inhibitors on co-cultures comprising components of the TME, using human primary cells. We also evaluated the effects of USP7 inhibition on tumor growth inhibition in syngeneic models when dosed in combination with immune checkpoint inhibitors (ICIs). RESULTS: Abrogation of VEGF secretion from fibroblasts in response to USP7 inhibition resulted in inhibition of tumor neoangiogenesis and increased tumor recruitment of CD8-positive T-lymphocytes, leading to significantly improved sensitivity to immune checkpoint inhibitors. In syngeneic models, treatment with USP7 inhibitors led to striking tumor responses resulting in significantly improved survival. CONCLUSIONS: USP7-mediated reprograming of the TME is not linked to its previously characterized role in modulating MDM2 but does require p53 and UHRF1 in addition to the well-characterized VEGF transcription factor, HIF-1α. This represents a function of USP7 that is unique to fibroblasts, and which is not observed in cancer cells or other components of the TME. Given the potential for USP7 inhibitors to transform "immune desert" tumors into "immune responsive" tumors, this paves the way for a novel therapeutic strategy combining USP7 inhibitors with immune checkpoint inhibitors (ICIs).

surviveR: a flexible shiny application for patient survival analysis.

Kaplan-Meier (KM) survival analyses based on complex patient categorization due to the burgeoning volumes of genomic, molecular and phenotypic data, are an increasingly important aspect of the biomedical researcher's toolkit. Commercial statistics and graphing packages for such analyses are functionally limited, whereas open-source tools have a high barrier-to-entry in terms of understanding of methodologies and computational expertise. We developed surviveR to address this unmet need for a survival analysis tool that can enable users with limited computational expertise to conduct routine but complex analyses. surviveR is a cloud-based Shiny application, that addresses our identified unmet need for an easy-to-use web-based tool that can plot and analyse survival based datasets. Integrated customization options allows a user with limited computational expertise to easily filter patients to enable custom cohort generation, automatically calculate log-rank test and Cox hazard ratios. Continuous datasets can be integrated, such as RNA or protein expression measurements which can be then used as categories for survival plotting. We further demonstrate the utility through exemplifying its application to a clinically relevant colorectal cancer patient dataset. surviveR is a cloud-based web application available at https://generatr.qub.ac.uk/app/surviveR , that can be used by non-experts users to perform complex custom survival analysis.

Successful Proof-of-Concept for Topical Delivery of Novel Peptide ALM201 with Potential Usefulness for Treating Neovascular Eye Disorders.

Purpose: To evaluate the therapeutic benefit of a novel peptide, ALM201, in ocular pathologic vascularization. Design: Experimental study in mouse, rat, and rabbit animal models. Participants: Ten-week-old Lister Hooded male rats, 8-week-old Brown Norway male rats, 9-day-old C57BL/6J mice, and 12-month-old New Zealand male rabbits. Methods: Corneal vascularization was scored for vessel density and vessel distance to suture in a rat corneal suture model. Ocular penetration and biodistribution were evaluated by matrix-assisted laser desorption/ionization mass spectrometry imaging after topical ALM201 application to rabbit eyes. A mouse choroidal sprouting assay, with aflibercept as positive control, was used to evaluate choroidal neovascularization (CNV) in the posterior segment tissue. Efficacy of topical ALM201 was assessed using a rat laser CNV model of neovascular age-related macular degeneration. Main Outcome Measures: Clinical scoring and histologic analysis of vascularized corneas, sprouting area, lesion size, and vessel leakiness in posterior segments. Results: Assessment of ALM201 treatment in the rat corneal suture model showed a significant decrease in vessel density (P = 0.0065) and vessel distance to suture (P = 0.021) compared with vehicle control (phosphate-buffered saline [PBS]). Infiltration of inflammatory cells into the corneal stroma also was reduced significantly compared with PBS (724.5 ± 122 cells/mm2 vs. 1837 ± 195.9 cells/mm2, respectively; P = 0.0029). Biodistribution in rabbit eyes confirmed ALM201 bioavailability in anterior and posterior ocular segments 1 hour after topical instillation. ALM201 treatment significantly suppressed choroid vessel sprouting when compared with PBS treatment (44.5 ± 14.31 pixels vs. 120.9 ± 33.37 pixels, respectively; P = 0.04) and was not inferior to aflibercept (65.63 ± 11.86 pixels; P = 0.7459). Furthermore, topical ALM201 significantly improved vessel leakiness (leakage scores: 2.1 ± 0.7 vs. 2.9 ± 0.1; P = 0.0274) and lesion size (144,729 ± 33,239 µm3 vs. 187,923 ± 28,575 µm3; P = 0.03) in the rat laser CNV model when compared with topical PBS vehicle. Conclusions: ALM201 is a promising novel molecule with anti-inflammatory and antivascularization activity and is a strong candidate to meet the clinical need of a new, topically delivered therapeutic agent for treating inflammation and pathologic vascularization in the anterior and posterior segments of the eye.

Identification and development of a subtype-selective allosteric AKT inhibitor suitable for clinical development.

The serine/threonine protein kinase AKT plays a pivotal role within the PI3K pathway in regulating cellular proliferation and apoptotic cellular functions, and AKT hyper-activation via gene amplification and/or mutation has been implicated in multiple human malignancies. There are 3 AKT isoenzymes (AKT1-3) which mediate critical, non-redundant functions. We present the discovery and development of ALM301, a novel, allosteric, sub-type selective inhibitor of AKT1/2. ALM301 binds in an allosteric pocket created by the combined movement of the PH domain and the catalytic domain, resulting in a DFG out conformation. ALM301 was shown to be highly selective against a panel of over 450 kinases and potently inhibited cellular proliferation. These effects were particularly pronounced in MCF-7 cells containing a PI3KCA mutation. Subsequent cellular downstream pathway analysis in this sensitive cell line revealed potent inhibition of pAKT signalling up to 48 h post dosing. ALM301 treatment was well tolerated in an MCF-7 xenograft model and led to a dose-dependent reduction in tumour growth. Enhanced efficacy was observed in combination with tamoxifen. In summary, ALM301 is a highly specific AKT 1/2 inhibitor with an excellent pharmacological profile suitable for further clinical development.

classifieR a flexible interactive cloud-application for functional annotation of cancer transcriptomes.

BACKGROUND: Transcriptionally informed predictions are increasingly important for sub-typing cancer patients, understanding underlying biology and to inform novel treatment strategies. For instance, colorectal cancers (CRCs) can be classified into four CRC consensus molecular subgroups (CMS) or five intrinsic (CRIS) sub-types that have prognostic and predictive value. Breast cancer (BRCA) has five PAM50 molecular subgroups with similar value, and the OncotypeDX test provides transcriptomic based clinically actionable treatment-risk stratification. However, assigning samples to these subtypes and other transcriptionally inferred predictions is time consuming and requires significant bioinformatics experience. There is no "universal" method of using data from diverse assay/sequencing platforms to provide subgroup classification using the established classifier sets of genes (CMS, CRIS, PAM50, OncotypeDX), nor one which in provides additional useful functional annotations such as cellular composition, single-sample Gene Set Enrichment Analysis, or prediction of transcription factor activity. RESULTS: To address this bottleneck, we developed classifieR, an easy-to-use R-Shiny based web application that supports flexible rapid single sample annotation of transcriptional profiles derived from cancer patient samples form diverse platforms. We demonstrate the utility of the " classifieR" framework to applications focused on the analysis of transcriptional profiles from colorectal (classifieRc) and breast (classifieRb). Samples are annotated with disease relevant transcriptional subgroups (CMS/CRIS sub-types in classifieRc and PAM50/inferred OncotypeDX in classifieRb), estimation of cellular composition using MCP-counter and xCell, single-sample Gene Set Enrichment Analysis (ssGSEA) and transcription factor activity predictions with Discriminant Regulon Expression Analysis (DoRothEA). CONCLUSIONS: classifieR provides a framework which enables labs without access to a dedicated bioinformation can get information on the molecular makeup of their samples, providing an insight into patient prognosis, druggability and also as a tool for analysis and discovery. Applications are hosted online at https://generatr.qub.ac.uk/app/classifieRc and https://generatr.qub.ac.uk/app/classifieRb after signing up for an account on https://generatr.qub.ac.uk .

Functional Genomic Identification of Predictors of Sensitivity and Mechanisms of Resistance to Multivalent Second-Generation TRAIL-R2 Agonists.

Multivalent second-generation TRAIL-R2 agonists are currently in late preclinical development and early clinical trials. Herein, we use a representative second-generation agent, MEDI3039, to address two major clinical challenges facing these agents: lack of predictive biomarkers to enable patient selection and emergence of resistance. Genome-wide CRISPR knockout screens were notable for the lack of resistance mechanisms beyond the canonical TRAIL-R2 pathway (caspase-8, FADD, BID) as well as p53 and BAX in TP53 wild-type models, whereas a CRISPR activatory screen identified cell death inhibitors MCL-1 and BCL-XL as mechanisms to suppress MEDI3039-induced cell death. High-throughput drug screening failed to identify genomic alterations associated with response to MEDI3039; however, transcriptomics analysis revealed striking association between MEDI3039 sensitivity and expression of core components of the extrinsic apoptotic pathway, most notably its main apoptotic effector caspase-8 in solid tumor cell lines. Further analyses of colorectal cell lines and patient-derived xenografts identified caspase-8 expression ratio to its endogenous regulator FLIP(L) as predictive of sensitivity to MEDI3039 in several major solid tumor types and a further subset indicated by caspase-8:MCL-1 ratio. Subsequent MEDI3039 combination screening of TRAIL-R2, caspase-8, FADD, and BID knockout models with 60 compounds with varying mechanisms of action identified two inhibitor of apoptosis proteins (IAP) that exhibited strong synergy with MEDI3039 that could reverse resistance only in BID-deleted models. In summary, we identify the ratios of caspase-8:FLIP(L) and caspase-8:MCL-1 as potential predictive biomarkers for second-generation TRAIL-R2 agonists and loss of key effectors such as FADD and caspase-8 as likely drivers of clinical resistance in solid tumors.

SynLeGG: analysis and visualization of multiomics data for discovery of cancer 'Achilles Heels' and gene function relationships.

Achilles' heel relationships arise when the status of one gene exposes a cell's vulnerability to perturbation of a second gene, such as chemical inhibition, providing therapeutic opportunities for precision oncology. SynLeGG (www.overton-lab.uk/synlegg) identifies and visualizes mutually exclusive loss signatures in 'omics data to enable discovery of genetic dependency relationships (GDRs) across 783 cancer cell lines and 30 tissues. While there is significant focus on genetic approaches, transcriptome data has advantages for investigation of GDRs and remains relatively underexplored. SynLeGG depends upon the MultiSEp algorithm for unsupervised assignment of cell lines into gene expression clusters, which provide the basis for analysis of CRISPR scores and mutational status in order to propose candidate GDRs. Benchmarking against SynLethDB demonstrates favourable performance for MultiSEp against competing approaches, finding significantly higher area under the Receiver Operator Characteristic curve and between 2.8-fold to 8.5-fold greater coverage. In addition to pan-cancer analysis, SynLeGG offers investigation of tissue-specific GDRs and recovers established relationships, including synthetic lethality for SMARCA2 with SMARCA4. Proteomics, Gene Ontology, protein-protein interactions and paralogue information are provided to assist interpretation and candidate drug target prioritization. SynLeGG predictions are significantly enriched in dependencies validated by a recently published CRISPR screen.

The pseudo-caspase FLIP(L) regulates cell fate following p53 activation.

p53 is the most frequently mutated, well-studied tumor-suppressor gene, yet the molecular basis of the switch from p53-induced cell-cycle arrest to apoptosis remains poorly understood. Using a combination of transcriptomics and functional genomics, we unexpectedly identified a nodal role for the caspase-8 paralog and only human pseudo-caspase, FLIP(L), in regulating this switch. Moreover, we identify FLIP(L) as a direct p53 transcriptional target gene that is rapidly up-regulated in response to Nutlin-3A, an MDM2 inhibitor that potently activates p53. Genetically or pharmacologically inhibiting expression of FLIP(L) using siRNA or entinostat (a clinically relevant class-I HDAC inhibitor) efficiently promoted apoptosis in colorectal cancer cells in response to Nutlin-3A, which otherwise predominantly induced cell-cycle arrest. Enhanced apoptosis was also observed when entinostat was combined with clinically relevant, p53-activating chemotherapy in vitro, and this translated into enhanced in vivo efficacy. Mechanistically, FLIP(L) inhibited p53-induced apoptosis by blocking activation of caspase-8 by the TRAIL-R2/DR5 death receptor; notably, this activation was not dependent on receptor engagement by its ligand, TRAIL. In the absence of caspase-8, another of its paralogs, caspase-10 (also transcriptionally up-regulated by p53), induced apoptosis in Nutlin-3A-treated, FLIP(L)-depleted cells, albeit to a lesser extent than in caspase-8-proficient cells. FLIP(L) depletion also modulated transcription of canonical p53 target genes, suppressing p53-induced expression of the cell-cycle regulator p21 and enhancing p53-induced up-regulation of proapoptotic PUMA. Thus, even in the absence of caspase-8/10, FLIP(L) silencing promoted p53-induced apoptosis by enhancing PUMA expression. Thus, we report unexpected, therapeutically relevant roles for FLIP(L) in determining cell fate following p53 activation.

Pevonedistat (MLN4924): mechanism of cell death induction and therapeutic potential in colorectal cancer.

Pevonedistat (MLN4924), a selective inhibitor of the NEDD8-activating enzyme E1 regulatory subunit (NAE1), has demonstrated significant therapeutic potential in several malignancies. Although multiple mechanisms-of-action have been identified, how MLN4924 induces cell death and its potential as a combinatorial agent with standard-of-care (SoC) chemotherapy in colorectal cancer (CRC) remains largely undefined. In an effort to understand MLN4924-induced cell death in CRC, we identified p53 as an important mediator of the apoptotic response to MLN4924. We also identified roles for the extrinsic (TRAIL-R2/caspase-8) and intrinsic (BAX/BAK) apoptotic pathways in mediating the apoptotic effects of MLN4924 in CRC cells, as well as a role for BID, which modulates a cross-talk between these pathways. Depletion of the anti-apoptotic protein FLIP, which we identify as a novel mediator of resistance to MLN4924, enhanced apoptosis in a p53-, TRAIL-R2/DR5-, and caspase-8-dependent manner. Notably, TRAIL-R2 was involved in potentiating the apoptotic response to MLN4924 in the absence of FLIP, in a ligand-independent manner. Moreoever, when paired with SoC chemotherapies, MLN4924 demonstrated synergy with the irinotecan metabolite SN38. The cell death induced by MLN4924/SN38 combination was dependent on activation of mitochondria through BAX/BAK, but in a p53-independent manner, an important observation given the high frequency of TP53 mutation(s) in advanced CRC. These results uncover mechanisms of cell death induced by MLN4924 and suggest that this second-generation proteostasis-disrupting agent may have its most widespread activity in CRC, in combination with irinotecan-containing treatment regimens.

Whole transcriptome profiling of patient-derived xenograft models as a tool to identify both tumor and stromal specific biomarkers.

The tumor microenvironment is emerging as a key regulator of cancer growth and progression, however the exact mechanisms of interaction with the tumor are poorly understood. Whilst the majority of genomic profiling efforts thus far have focused on the tumor, here we investigate RNA-Seq as a hypothesis-free tool to generate independent tumor and stromal biomarkers, and explore tumor-stroma interactions by exploiting the human-murine compartment specificity of patient-derived xenografts (PDX).Across a pan-cancer cohort of 79 PDX models, we determine that mouse stroma can be separated into distinct clusters, each corresponding to a specific stromal cell type. This implies heterogeneous recruitment of mouse stroma to the xenograft independent of tumor type. We then generate cross-species expression networks to recapitulate a known association between tumor epithelial cells and fibroblast activation, and propose a potentially novel relationship between two hypoxia-associated genes, human MIF and mouse Ddx6. Assessment of disease subtype also reveals MMP12 as a putative stromal marker of triple-negative breast cancer. Finally, we establish that our ability to dissect recruited stroma from trans-differentiated tumor cells is crucial to identifying stem-like poor-prognosis signatures in the tumor compartment.In conclusion, RNA-Seq is a powerful, cost-effective solution to global analysis of human tumor and mouse stroma simultaneously, providing new insights into mouse stromal heterogeneity and compartment-specific disease markers that are otherwise overlooked by alternative technologies. The study represents the first comprehensive analysis of its kind across multiple PDX models, and supports adoption of the approach in pre-clinical drug efficacy studies, and compartment-specific biomarker discovery.

Multi-omic measurement of mutually exclusive loss-of-function enriches for candidate synthetic lethal gene pairs.

BACKGROUND: Identification of synthetic lethal interactions in cancer cells could offer promising new therapeutic targets. Large-scale functional genomic screening presents an opportunity to test large numbers of cancer synthetic lethal hypotheses. Methods enriching for candidate synthetic lethal targets in molecularly defined cancer cell lines can steer effective design of screening efforts. Loss of one partner of a synthetic lethal gene pair creates a dependency on the other, thus synthetic lethal gene pairs should never show simultaneous loss-of-function. We have developed a computational approach to mine large multi-omic cancer data sets and identify gene pairs with mutually exclusive loss-of-function. Since loss-of-function may not always be genetic, we look for deleterious mutations, gene deletion and/or loss of mRNA expression by bimodality defined with a novel algorithm BiSEp. RESULTS: Applying this toolkit to both tumour cell line and patient data, we achieve statistically significant enrichment for experimentally validated tumour suppressor genes and synthetic lethal gene pairings. Notably non-reliance on genetic loss reveals a number of known synthetic lethal relationships otherwise missed, resulting in marked improvement over genetic-only predictions. We go on to establish biological rationale surrounding a number of novel candidate synthetic lethal gene pairs with demonstrated dependencies in published cancer cell line shRNA screens. CONCLUSIONS: This work introduces a multi-omic approach to define gene loss-of-function, and enrich for candidate synthetic lethal gene pairs in cell lines testable through functional screens. In doing so, we offer an additional resource to generate new cancer drug target and combination hypotheses. Algorithms discussed are freely available in the BiSEp CRAN package at http://cran.r-project.org/web/packages/BiSEp/index.html .

Leptin, BMI, and a Metabolic Gene Expression Signature Associated with Clinical Outcome to VEGF Inhibition in Colorectal Cancer.

VEGF (vascular endothelial growth factor) signaling inhibitors are widely used in different cancer types; however, patient selection remains a challenge. Analyses of samples from a phase III clinical trial in metastatic colorectal cancer testing chemotherapy versus chemotherapy with the small molecule VEGF receptors inhibitor cediranib identified circulating leptin levels, BMI, and a tumor metabolic and angiogenic gene expression signature associated with improved clinical outcome in patients treated with cediranib. Patients with a glycolytic and hypoxic/angiogenic profile were associated with increased benefit from cediranib, whereas patients with a high lipogenic, oxidative phosphorylation and serine biosynthesis signature did not gain benefit. These findings translated to pre-clinical tumor xenograft models where the same metabolic gene expression profiles were associated with in vivo sensitivity to cediranib as monotherapy. These findings suggest a link between patient physiology, tumor biology, and response to antiangiogenics, which may guide patient selection for VEGF therapy in the future.

Evaluating Robustness and Sensitivity of the NanoString Technologies nCounter Platform to Enable Multiplexed Gene Expression Analysis of Clinical Samples.

Analysis of clinical trial specimens such as formalin-fixed paraffin-embedded (FFPE) tissue for molecular mechanisms of disease progression or drug response is often challenging and limited to a few markers at a time. This has led to the increasing importance of highly multiplexed assays that enable profiling of many biomarkers within a single assay. Methods for gene expression analysis have undergone major advances in biomedical research, but obtaining a robust dataset from low-quality RNA samples, such as those isolated from FFPE tissue, remains a challenge. Here, we provide a detailed evaluation of the NanoString Technologies nCounter platform, which provides a direct digital readout of up to 800 mRNA targets simultaneously. We tested this system by examining a broad set of human clinical tissues for a range of technical variables, including sensitivity and limit of detection to varying RNA quantity and quality, reagent performance over time, variability between instruments, the impact of the number of fields of view sampled, and differences between probe sequence locations and overlapping genes across CodeSets. This study demonstrates that Nanostring offers several key advantages, including sensitivity, reproducibility, technical robustness, and utility for clinical application.

Reproducible, Quantitative, and Flexible Molecular Subtyping of Clinical DLBCL Samples Using the NanoString nCounter System.

PURPOSE: Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous disease with distinct molecular subtypes. The most established subtyping approach, the "Cell of Origin" (COO) algorithm, categorizes DLBCL into activated B-cell (ABC) and germinal center B-cell (GCB)-like subgroups through gene expression profiling. Recently developed immunohistochemical (IHC) techniques and other established methodologies can deliver discordant results and have various technical limitations. We evaluated the NanoString nCounter gene expression system to address issues with current platforms. EXPERIMENTAL DESIGN: We devised a scoring system using 145 genes from published datasets to categorize DLBCL samples. After cell line validation, clinical tissue segmentation was tested using commercially available diagnostic DLBCL samples. Finally, we profiled biopsies from patients with relapsed/refractory DLBCL enrolled in the fostamatinib phase IIb clinical trial using three independent RNA expression platforms: NanoString, Affymetrix, and qNPA. RESULTS: Diagnostic samples showed a typical spread of subtypes with consistent gene expression profiles across matched fresh, frozen, and formalin-fixed paraffin-embedded tissues. Results from biopsy samples across platforms were remarkably consistent, in contrast to published IHC data. Interestingly, COO segmentation of longitudinal fostamatinib biopsies taken at initial diagnosis and then again at primary relapse showed 88% concordance (15/17), suggesting that COO designation remains stable over the course of disease progression. CONCLUSIONS: DLBCL segmentation of patient tumor samples is possible using a number of expression platforms. However, we found that NanoString offers the most flexibility and fewest limitations in regards to robust clinical tissue subtype characterization. These subtype distinctions should help guide disease prognosis and treatment options within DLBCL clinical practice.

RNA-Seq Differentiates Tumour and Host mRNA Expression Changes Induced by Treatment of Human Tumour Xenografts with the VEGFR Tyrosine Kinase Inhibitor Cediranib.

Pre-clinical models of tumour biology often rely on propagating human tumour cells in a mouse. In order to gain insight into the alignment of these models to human disease segments or investigate the effects of different therapeutics, approaches such as PCR or array based expression profiling are often employed despite suffering from biased transcript coverage, and a requirement for specialist experimental protocols to separate tumour and host signals. Here, we describe a computational strategy to profile transcript expression in both the tumour and host compartments of pre-clinical xenograft models from the same RNA sample using RNA-Seq. Key to this strategy is a species-specific mapping approach that removes the need for manipulation of the RNA population, customised sequencing protocols, or prior knowledge of the species component ratio. The method demonstrates comparable performance to species-specific RT-qPCR and a standard microarray platform, and allowed us to quantify gene expression changes in both the tumour and host tissue following treatment with cediranib, a potent vascular endothelial growth factor receptor tyrosine kinase inhibitor, including the reduction of multiple murine transcripts associated with endothelium or vessels, and an increase in genes associated with the inflammatory response in response to cediranib. In the human compartment, we observed a robust induction of hypoxia genes and a reduction in cell cycle associated transcripts. In conclusion, the study establishes that RNA-Seq can be applied to pre-clinical models to gain deeper understanding of model characteristics and compound mechanism of action, and to identify both tumour and host biomarkers.

Microarray gene expression analysis of fixed archival tissue permits molecular classification and identification of potential therapeutic targets in diffuse large B-cell lymphoma.

Refractory/relapsed diffuse large B-cell lymphoma (DLBCL) has a poor prognosis. Novel drugs targeting the constitutively activated NF-κB pathway characteristic of ABC-DLBCL are promising, but evaluation depends on accurate activated B cell-like (ABC)/germinal center B cell-like (GCB) molecular classification. This is traditionally performed on gene microarray expression profiles of fresh biopsies, which are not routinely collected, or by immunohistochemistry on formalin-fixed, paraffin-embedded (FFPE) tissue, which lacks reproducibility and classification accuracy. We explored the possibility of using routine archival FFPE tissue for gene microarray applications. We examined Affymetrix HG U133 Plus 2.0 gene expression profiles from paired archival FFPE and fresh-frozen tissues of 40 ABC/GCB-classified DLBCL cases to compare classification accuracy and test the potential for this approach to aid the discovery of therapeutic targets and disease classifiers in DLBCL. Unsupervised hierarchical clustering of unselected present probe sets distinguished ABC/GCB in FFPE with remarkable accuracy, and a Bayesian classifier correctly assigned 32 of 36 cases with >90% probability. Enrichment for NF-κB genes was appropriately seen in ABC-DLBCL FFPE tissues. The top discriminatory genes expressed in FFPE separated cases with high statistical significance and contained novel biology with potential therapeutic insights, warranting further investigation. These results support a growing understanding that archival FFPE tissues can be used in microarray experiments aimed at molecular classification, prognostic biomarker discovery, and molecular exploration of rare diseases.