Loss of CD20 expression as a mechanism of resistance to mosunetuzumab in relapsed/refractory B-cell lymphomas.

ABSTRACT: CD20 is an established therapeutic target in B-cell malignancies. The CD20 × CD3 bispecific antibody mosunetuzumab has significant efficacy in B-cell non-Hodgkin lymphomas (NHLs). Because target antigen loss is a recognized mechanism of resistance, we evaluated CD20 expression relative to clinical response in patients with relapsed and/or refractory NHL in the phase 1/2 GO29781 trial investigating mosunetuzumab monotherapy. CD20 was studied using immunohistochemistry (IHC), RNA sequencing, and whole-exome sequencing performed centrally in biopsy specimens collected before treatment at predose, during treatment, or upon progression. Before treatment, most patients exhibited a high proportion of tumor cells expressing CD20; however, in 16 of 293 patients (5.5%) the proportion was <10%. Analyses of paired biopsy specimens from patients on treatment revealed that CD20 levels were maintained in 29 of 30 patients (97%) vs at progression, where CD20 loss was observed in 11 of 32 patients (34%). Reduced transcription or acquisition of truncating mutations explained most but not all cases of CD20 loss. In vitro modeling confirmed the effects of CD20 variants identified in clinical samples on reduction of CD20 expression and missense mutations in the extracellular domain that could block mosunetuzumab binding. This study expands the knowledge about the occurrence of target antigen loss after anti-CD20 therapeutics to include CD20-targeting bispecific antibodies and elucidates mechanisms of reduced CD20 expression at disease progression that may be generalizable to other anti-CD20 targeting agents. These results also confirm the utility of readily available IHC staining for CD20 as a tool to inform clinical decisions. This trial was registered at www.ClinicalTrials.gov as #NCT02500407.

Non-equivalence of Wnt and R-spondin ligands during Lgr5+ intestinal stem-cell self-renewal.

The canonical Wnt/ß-catenin signalling pathway governs diverse developmental, homeostatic and pathological processes. Palmitoylated Wnt ligands engage cell-surface frizzled (FZD) receptors and LRP5 and LRP6 co-receptors, enabling ß-catenin nuclear translocation and TCF/LEF-dependent gene transactivation. Mutations in Wnt downstream signalling components have revealed diverse functions thought to be carried out by Wnt ligands themselves. However, redundancy between the 19 mammalian Wnt proteins and 10 FZD receptors and Wnt hydrophobicity have made it difficult to attribute these functions directly to Wnt ligands. For example, individual mutations in Wnt ligands have not revealed homeostatic phenotypes in the intestinal epithelium-an archetypal canonical, Wnt pathway-dependent, rapidly self-renewing tissue, the regeneration of which is fueled by proliferative crypt Lgr5+ intestinal stem cells (ISCs). R-spondin ligands (RSPO1-RSPO4) engage distinct LGR4-LGR6, RNF43 and ZNRF3 receptor classes, markedly potentiate canonical Wnt/ß-catenin signalling, and induce intestinal organoid growth in vitro and Lgr5+ ISCs in vivo. However, the interchangeability, functional cooperation and relative contributions of Wnt versus RSPO ligands to in vivo canonical Wnt signalling and ISC biology remain unknown. Here we identify the functional roles of Wnt and RSPO ligands in the intestinal crypt stem-cell niche. We show that the default fate of Lgr5+ ISCs is to differentiate, unless both RSPO and Wnt ligands are present. However, gain-of-function studies using RSPO ligands and a new non-lipidated Wnt analogue reveal that these ligands have qualitatively distinct, non-interchangeable roles in ISCs. Wnt proteins are unable to induce Lgr5+ ISC self-renewal, but instead confer a basal competency by maintaining RSPO receptor expression that enables RSPO ligands to actively drive and specify the extent of stem-cell expansion. This functionally non-equivalent yet cooperative interaction between Wnt and RSPO ligands establishes a molecular precedent for regulation of mammalian stem cells by distinct priming and self-renewal factors, with broad implications for precise control of tissue regeneration.

Prognostic mutational subtyping in de novo diffuse large B-cell lymphoma.

BACKGROUND: Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous disease defined using a number of well-established molecular subsets. Application of non-negative matrix factorization (NMF) to whole exome sequence data has previously been used to identify six distinct molecular clusters in DLBCL with potential clinical relevance. In this study, we applied NMF-clustering to targeted sequencing data utilizing the FoundationOne Heme® panel from the Phase III GOYA (NCT01287741) and Phase Ib/II CAVALLI studies (NCT02055820) in de novo DLBCL. Biopsy samples, survival outcomes, RNA-Seq and targeted exome-sequencing data were available for 423 patients in GOYA (obinutuzumab [G]-cyclophosphamide, doxorubicin, vincristine, and prednisone [CHOP] vs rituximab [R]-CHOP) and 86 patients in CAVALLI (venetoclax+[G/R]-CHOP). RESULTS: When the NMF algorithm was applied to samples from the GOYA study analyzed using a comprehensive genomic profiling platform, four of the six groups previously reported were observed: MYD88/CD79B, BCL2/EZH2, NOTCH2/TNFAIP3, and no mutations. Mutation profiles, cell-of-origin subset distributions and clinical associations of MYD88/CD79B and BCL2/EZH2 groups were similar to those described in previous NMF studies. In contrast, application of NMF to the CAVALLI study yielded only three; MYD88/CD79B-, BCL2/EZH2-like clusters, and a no mutations group, and there was a trend towards improved outcomes for BCL2/EZH2 over MYD88/CD79B. CONCLUSIONS: This analysis supports the utility of NMF used in conjunction with targeted sequencing platforms for identifying patients with different prognostic subsets. The observed trend for improved overall survival in the BCL2/EZH2 group is consistent with the mechanism of action of venetoclax, suggesting that targeting sequencing and NMF has potential for identifying patients who are more likely to gain benefit from venetoclax therapy.

Integration of cell of origin into the clinical CNS International Prognostic Index improves CNS relapse prediction in DLBCL.

Central nervous system (CNS) relapse carries a poor prognosis in diffuse large B-cell lymphoma (DLBCL). Integrating biomarkers into the CNS-International Prognostic Index (CNS-IPI) risk model may improve identification of patients at high risk for developing secondary CNS disease. CNS relapse was analyzed in 1418 DLBCL patients treated with obinutuzumab or rituximab plus cyclophosphamide, doxorubicin, vincristine, prednisone chemotherapy in the phase 3 GOYA study. Cell of origin (COO) was assessed using gene-expression profiling. BCL2 and MYC protein expression was analyzed by immunohistochemistry. The impact of CNS-IPI, COO, and BCL2/MYC dual-expression status on CNS relapse was assessed using a multivariate Cox regression model (data available in n = 1418, n = 933, and n = 688, respectively). High CNS-IPI score (hazard ratio [HR], 4.0; 95% confidence interval [CI], 1.3-12.3; P = .02) and activated B-cell‒like (ABC) (HR, 5.2; 95% CI, 2.1-12.9; P = .0004) or unclassified COO subtypes (HR, 4.2; 95% CI, 1.5-11.7; P = .006) were independently associated with CNS relapse. BCL2/MYC dual-expression status did not impact CNS relapse risk. Three risk subgroups were identified based on the presence of high CNS-IPI score and/or ABC/unclassified COO (CNS-IPI-C model): low risk (no risk factors, n = 450 [48.2%]), intermediate risk (1 factor, n = 408 [43.7%]), and high risk (both factors, n = 75 [8.0%]). Two-year CNS relapse rates were 0.5%, 4.4%, and 15.2% in the respective risk subgroups. Combining high CNS-IPI and ABC/unclassified COO improved CNS relapse prediction and identified a patient subgroup at high risk for developing CNS relapse. The study was registered at www.clinicaltrials.gov as #NCT01287741.

Prediction and characterization of diffuse large B-cell lymphoma cell-of-origin subtypes using targeted sequencing.

The aim of the present study was to determine cell of origin (COO) from a platform using a DNA-based method, COO DNA classifier (COODC). A targeted exome-sequencing platform that applies the mutational profile of a sample was used to classify COO subtype. Two major mutational signatures associated with COO were identified: Catalogue of Somatic Mutations in Cancer (COSMIC) signature 23 enriched in activated B cell (ABC) and COSMIC signature 3, which suggested increased frequency in germinal center B cell (GCB). Differential mutation signatures linked oncogenesis to mutational processes during B-cell activation, confirming the putative origin of GCB and ABC subtypes. Integrating COO with comprehensive genomic profiling enabled identification of features associated with COO and demonstrated the feasibility of determining COO without RNA.

Lay abstract To determine subtypes of diffuse large B-cell lymphoma (DLBCL), a cancer with poor survival, we aimed to identify DLBCL subtypes using DNA mutation-based tools. A targeted gene-sequencing platform, which measures the number and types of DNA mutations in a sample, was used to categorize DLBCL subtypes. Two major patterns of mutations associated with subtypes were identified: Catalogue of Somatic Mutations in Cancer (COSMIC) signature 23 and COSMIC signature 3. Differences in how the subtypes developed suggest a link between tumor developments and B cells being activated normally, confirming where the DLBCL subtypes came from. Combining this information with comprehensive genomic profiling, which determines all of the genes that a person has, allowed identification of features that are associated with DLBCL subtypes and showed that a DLBCL subtype can be determined without using RNA.

Prognostic impact of somatic mutations in diffuse large B-cell lymphoma and relationship to cell-of-origin: data from the phase III GOYA study.

Diffuse large B-cell lymphoma represents a biologically and clinically heterogeneous diagnostic category with well-defined cell-of-origin subtypes. Using data from the GOYA study (NCT01287741), we characterized the mutational profile of diffuse large B-cell lymphoma and evaluated the prognostic impact of somatic mutations in relation to cell-of-origin. Targeted DNA next-generation sequencing was performed in 499 formalin-fixed paraffin-embedded tissue biopsies from previously untreated patients. Prevalence of genetic alterations/mutations was examined. Multivariate Cox regression was used to evaluate the prognostic effect of individual genomic alterations. Of 465 genes analyzed, 59 were identified with mutations occurring in at least 10 of 499 patients (≥2% prevalence); 334 additional genes had mutations occurring in ≥1 patient. Single nucleotide variants were the most common mutation type. On multivariate analysis, BCL2 alterations were most strongly associated with shorter progression-free survival (multivariate hazard ratio: 2.6; 95% confidence interval: 1.6 to 4.2). BCL2 alterations were detected in 102 of 499 patients; 92 had BCL2 translocations, 90% of whom had germinal center B-cell-like diffuse large B-cell lymphoma. BCL2 alterations were also significantly correlated with BCL2 gene and protein expression levels. Validation of published mutational subsets revealed consistent patterns of co-occurrence, but no consistent prognostic differences between subsets. Our data confirm the molecular heterogeneity of diffuse large B-cell lymphoma, with potential treatment targets occurring in distinct cell-of-origin subtypes. clinicaltrials.gov identifier: NCT01287741.

Gene set meta-analysis with Quantitative Set Analysis for Gene Expression (QuSAGE).

Small sample sizes combined with high person-to-person variability can make it difficult to detect significant gene expression changes from transcriptional profiling studies. Subtle, but coordinated, gene expression changes may be detected using gene set analysis approaches. Meta-analysis is another approach to increase the power to detect biologically relevant changes by integrating information from multiple studies. Here, we present a framework that combines both approaches and allows for meta-analysis of gene sets. QuSAGE meta-analysis extends our previously published QuSAGE framework, which offers several advantages for gene set analysis, including fully accounting for gene-gene correlations and quantifying gene set activity as a full probability density function. Application of QuSAGE meta-analysis to influenza vaccination response shows it can detect significant activity that is not apparent in individual studies.

Follicular Lymphoma Evaluation Index (FLEX): A new clinical prognostic model that is superior to existing risk scores for predicting progression-free survival and early treatment failure after frontline immunochemotherapy.

Patients with advanced-stage follicular lymphoma (FL) who progress early after receiving first-line therapy have poor overall survival (OS). Currently applied clinical prognostic models such as FL International Prognostic Index [FLIPI], FLIPI-2 and PRIMA-Prognostic Index [PRIMA-PI] have suboptimal sensitivity and specificity to predict this poor prognosis subgroup. The primary objective was to develop a novel prognostic model, the FL Evaluation Index (FLEX) score, to identify high-risk patients and compare its performance with FLIPI, FLIPI-2 and PRIMA-PI. Progression-free survival (PFS) after first-line immunochemotherapy was the key endpoint, while OS and progression of disease within 24 months (POD24) were also assessed. The model, which includes nine clinical variables, was developed using a cohort of patients with previously untreated advanced-stage FL from the phase 3 GALLIUM trial (NCT01332968). The performance of the model was validated using data from the SABRINA trial (NCT01200758). In GALLIUM (n = 1004; 127 with and 877 without POD24), FLEX increased the intergroup (low-risk/high-risk) difference in 2-year and 3-year PFS rates and demonstrated superior intergroup differences in 2-year and 3-year OS rates compared with FLIPI, FLIPI-2 and PRIMA-PI. Sensitivity for a high-risk score to predict POD24 was 60% using FLEX compared with 53% for FLIPI and FLIPI-2, and 69% for PRIMA-PI, while specificity was 68% for FLEX compared with 58% for FLIPI, 59% for FLIPI-2 and 48% for PRIMA-PI. The prognostic value of FLEX in SABRINA was similar to FLIPI. Therefore, FLEX appears to perform better than existing prognostic models in previously untreated FL, in particular for the newer treatment regimens.

Multicenter Systems Analysis of Human Blood Reveals Immature Neutrophils in Males and During Pregnancy.

Despite clear differences in immune system responses and in the prevalence of autoimmune diseases between males and females, there is little understanding of the processes involved. In this study, we identified a gene signature of immature-like neutrophils, characterized by the overexpression of genes encoding for several granule-containing proteins, which was found at higher levels (up to 3-fold) in young (20-30 y old) but not older (60 to >89 y old) males compared with females. Functional and phenotypic characterization of peripheral blood neutrophils revealed more mature and responsive neutrophils in young females, which also exhibited an elevated capacity in neutrophil extracellular trap formation at baseline and upon microbial or sterile autoimmune stimuli. The expression levels of the immature-like neutrophil signature increased linearly with pregnancy, an immune state of increased susceptibility to certain infections. Using mass cytometry, we also find increased frequencies of immature forms of neutrophils in the blood of women during late pregnancy. Thus, our findings show novel sex differences in innate immunity and identify a common neutrophil signature in males and in pregnant women.

Dysregulation of B Cell Repertoire Formation in Myasthenia Gravis Patients Revealed through Deep Sequencing.

Myasthenia gravis (MG) is a prototypical B cell-mediated autoimmune disease affecting 20-50 people per 100,000. The majority of patients fall into two clinically distinguishable types based on whether they produce autoantibodies targeting the acetylcholine receptor (AChR-MG) or muscle specific kinase (MuSK-MG). The autoantibodies are pathogenic, but whether their generation is associated with broader defects in the B cell repertoire is unknown. To address this question, we performed deep sequencing of the BCR repertoire of AChR-MG, MuSK-MG, and healthy subjects to generate ∼518,000 unique VH and VL sequences from sorted naive and memory B cell populations. AChR-MG and MuSK-MG subjects displayed distinct gene segment usage biases in both VH and VL sequences within the naive and memory compartments. The memory compartment of AChR-MG was further characterized by reduced positive selection of somatic mutations in the VH CDR and altered VH CDR3 physicochemical properties. The VL repertoire of MuSK-MG was specifically characterized by reduced V-J segment distance in recombined sequences, suggesting diminished VL receptor editing during B cell development. Our results identify large-scale abnormalities in both the naive and memory B cell repertoires. Particular abnormalities were unique to either AChR-MG or MuSK-MG, indicating that the repertoires reflect the distinct properties of the subtypes. These repertoire abnormalities are consistent with previously observed defects in B cell tolerance checkpoints in MG, thereby offering additional insight regarding the impact of tolerance defects on peripheral autoimmune repertoires. These collective findings point toward a deformed B cell repertoire as a fundamental component of MG.

The Repertoire Dissimilarity Index as a method to compare lymphocyte receptor repertoires.

BACKGROUND: The B and T cells of the human adaptive immune system leverage a highly diverse repertoire of antigen-specific receptors to protect the human body from pathogens. The sequencing and analysis of immune repertoires is emerging as an important tool to understand immune responses, whether beneficial or harmful (in the case of autoimmunity). However, methods for studying these repertoires, and for directly comparing different immune repertoires, are lacking. RESULTS: In this paper, we present a non-parametric method for directly comparing sequencing repertoires, with the goal of rigorously quantifying differences in V, D, and J gene segment utilization. This method, referred to as the Repertoire Dissimilarity Index (RDI), uses a bootstrapped subsampling approach to account for variance in sequencing depth, and, coupled with a data simulation approach, allows for direct quantification of the average variation between repertoires. We use the RDI method to recapitulate known differences in the formation of the CD4+ and CD8+ T cell repertoires, and further show that antigen-driven activation of naïve CD8+ T cells is more selective than in the CD4+ repertoire, resulting in a more specialized CD8+ memory repertoire. CONCLUSIONS: We prove that the RDI method is an accurate and versatile method for comparisons of immune repertoires. The RDI method has been implemented as an R package, and is available for download through Bitbucket.

Quantitative gene set analysis generalized for repeated measures, confounder adjustment, and continuous covariates.

BACKGROUND: Gene set analysis (GSA) of gene expression data can be highly powerful when the biological signal is weak compared to other sources of variability in the data. However, many gene set analysis approaches utilize permutation tests which are not appropriate for complex study designs. For example, the correlation of subjects is broken when comparing time points within a longitudinal study. Linear mixed models provide a method to analyze longitudinal studies as well as adjust for potential confounding factors and account for sources of variability that are not of primary interest. Currently, there are no known gene set analysis approaches that fully account for these study design and analysis aspects. In order to do so, we generalize the QuSAGE gene set analysis algorithm, denoted Q-Gen, and provide the necessary estimation adjustments to incorporate linear mixed model analyses. RESULTS: We assessed the performance of our generalized method in comparison to the original QuSAGE method in settings such as longitudinal repeated measures analysis and accounting for potential confounders. We demonstrate that the original QuSAGE method can not control for type-I error when these complexities exist. In addition to statistical appropriateness, analysis of a longitudinal influenza study suggests Q-Gen can allow for greater sensitivity when exploring a large number of gene sets. CONCLUSIONS: Q-Gen is an extension to the gene set analysis method of QuSAGE, and allows for linear mixed models to be applied appropriately within a gene set analysis framework. It provides GSA an added layer of flexibility that was not currently available. This flexibility allows for more appropriate statistical modeling of complex data structures that are inherent to many microarray study designs and can provide more sensitivity.

Dynamic expression profiling of type I and type III interferon-stimulated hepatocytes reveals a stable hierarchy of gene expression.

Despite activating similar signaling cascades, the type I and type III interferons (IFNs) differ in their ability to antagonize virus replication. However, it is not clear whether these cytokines induce unique antiviral states, particularly in the liver, where the clinically important hepatitis B and C viruses cause persistent infection. Here, clustering and promoter analyses of microarray-based gene expression profiling were combined with mechanistic studies of signaling pathways to dynamically characterize the transcriptional responses induced by these cytokines in Huh7 hepatoma cells and primary human hepatocytes. Type I and III IFNs differed greatly in their level of interferon-stimulated gene (ISG) induction with a clearly detectable hierarchy (IFN-ß > IFN-α > IFN-λ3 > IFN-λ1 > IFN-λ2). Notably, although the hierarchy identified varying numbers of differentially expressed genes when quantified using common statistical thresholds, further analysis of gene expression over multiple timepoints indicated that the individual IFNs do not in fact regulate unique sets of genes. The kinetic profiles of IFN-induced gene expression were also qualitatively similar with the important exception of IFN-α. While stimulation with either IFN-ß or IFN-λs resulted in a similar long-lasting ISG induction, IFN-α signaling peaked early after stimulation then declined due to a negative feedback mechanism. The quantitative expression hierarchy and unique kinetics of IFN-α reveal potential specific roles for individual IFNs in the immune response, and elucidate the mechanism behind previously observed differences in IFN antiviral activity. While current clinical trials are focused on IFN-λ1 as a potential antiviral therapy, the finding that IFN-λ3 invariably possesses the highest activity among type III IFNs suggests that this cytokine may have superior clinical activity.

Quantitative set analysis for gene expression: a method to quantify gene set differential expression including gene-gene correlations.

Enrichment analysis of gene sets is a popular approach that provides a functional interpretation of genome-wide expression data. Existing tests are affected by inter-gene correlations, resulting in a high Type I error. The most widely used test, Gene Set Enrichment Analysis, relies on computationally intensive permutations of sample labels to generate a null distribution that preserves gene-gene correlations. A more recent approach, CAMERA, attempts to correct for these correlations by estimating a variance inflation factor directly from the data. Although these methods generate P-values for detecting gene set activity, they are unable to produce confidence intervals or allow for post hoc comparisons. We have developed a new computational framework for Quantitative Set Analysis of Gene Expression (QuSAGE). QuSAGE accounts for inter-gene correlations, improves the estimation of the variance inflation factor and, rather than evaluating the deviation from a null hypothesis with a P-value, it quantifies gene-set activity with a complete probability density function. From this probability density function, P-values and confidence intervals can be extracted and post hoc analysis can be carried out while maintaining statistical traceability. Compared with Gene Set Enrichment Analysis and CAMERA, QuSAGE exhibits better sensitivity and specificity on real data profiling the response to interferon therapy (in chronic Hepatitis C virus patients) and Influenza A virus infection. QuSAGE is available as an R package, which includes the core functions for the method as well as functions to plot and visualize the results.

NGS-determined molecular markers and disease burden metrics from ctDNA correlate with PFS in previously untreated DLBCL.

Personalized risk stratification and treatment may help improve outcomes among patients with diffuse large B-cell lymphoma (DLBCL). We developed a next-generation sequencing (NGS)-based method to assess a range of potential prognostic indicators, and evaluated it using pretreatment plasma samples from 310 patients with previously untreated DLBCL from the GOYA trial (NCT01287741). Variant calls and DLBCL subtyping with the plasma-based method were concordant with corresponding tissue-based methods. Patients with a tumor burden greater than the median (p = .003) and non-germinal center B-cell-like (non-GCB) DLBCL (p = .049) had worse progression-free survival than patients with a tumor burden less than the median or GCB DLBCL. Multi-factor assessment combining orthogonal features from a single pretreatment plasma sample has promise as a prognostic indicator in this setting (p = .085). This minimally invasive plasma-based NGS assay could enable comprehensive prognostic assessment of patients in a clinical setting, with greater accessibility than current methods.

Pharmacodynamics and molecular correlates of response to glofitamab in relapsed/refractory non-Hodgkin lymphoma.

Glofitamab, a novel CD20xCD3, T-cell-engaging bispecific antibody, exhibited single-agent activity in Study NP30179, a first-in-human, phase 1 trial in relapsed/refractory B-cell non-Hodgkin lymphoma. Preclinical studies showed that glofitamab leads to T-cell activation, proliferation, and tumor cell killing upon binding to CD20 on malignant cells. Here, we provide evidence of glofitamab's clinical activity, including pharmacodynamic profile, mode of action, and factors associated with clinical response, by evaluating biomarkers in patient samples from the dose-escalation part of this trial. Patients enrolled in Study NP30179 received single-dose obinutuzumab pretreatment (1000 mg) 7 days before IV glofitamab (5 µg-25 mg). Glofitamab treatment lasted ≤12 cycles once every 2 or 3 weeks. Blood samples were collected at predefined time points per the clinical protocol; T-cell populations were evaluated centrally by flow cytometry, and cytokine profiles were analyzed. Immunohistochemical and genomic biomarker analyses were performed on tumor biopsy samples. Pharmacodynamic modulation was observed with glofitamab treatment, including dose-dependent induction of cytokines, and T-cell margination, proliferation, and activation in peripheral blood. Gene expression analysis of pretreatment tumor biopsy samples indicated that tumor cell intrinsic factors such as TP53 signaling are associated with resistance to glofitamab, but they may also be interlinked with a diminished effector T-cell profile in resistant tumors and thus represent a poor prognostic factor per se. This integrative biomarker data analysis provides clinical evidence regarding glofitamab's mode of action, supports optimal biological dose selection, and will further guide clinical development. This trial was registered at www.clinicaltrials.gov as #NCT03075696.

A prognostic model integrating PET-derived metrics and image texture analyses with clinical risk factors from GOYA.

Image texture analysis (radiomics) uses radiographic images to quantify characteristics that may identify tumour heterogeneity and associated patient outcomes. Using fluoro-deoxy-glucose positron emission tomography/computed tomography (FDG-PET/CT)-derived data, including quantitative metrics, image texture analysis and other clinical risk factors, we aimed to develop a prognostic model that predicts survival in patients with previously untreated diffuse large B-cell lymphoma (DLBCL) from GOYA (NCT01287741). Image texture features and clinical risk factors were combined into a random forest model and compared with the international prognostic index (IPI) for DLBCL based on progression-free survival (PFS) and overall survival (OS) predictions. Baseline FDG-PET scans were available for 1263 patients, 832 patients of these were cell-of-origin (COO)-evaluable. Patients were stratified by IPI or radiomics features plus clinical risk factors into low-, intermediate- and high-risk groups. The random forest model with COO subgroups identified a clearer high-risk population (45% 2-year PFS [95% confidence interval (CI) 40%-52%]; 65% 2-year OS [95% CI 59%-71%]) than the IPI (58% 2-year PFS [95% CI 50%-67%]; 69% 2-year OS [95% CI 62%-77%]). This study confirms that standard clinical risk factors can be combined with PET-derived image texture features to provide an improved prognostic model predicting survival in untreated DLBCL.

Cell subset prediction for blood genomic studies.

BACKGROUND: Genome-wide transcriptional profiling of patient blood samples offers a powerful tool to investigate underlying disease mechanisms and personalized treatment decisions. Most studies are based on analysis of total peripheral blood mononuclear cells (PBMCs), a mixed population. In this case, accuracy is inherently limited since cell subset-specific differential expression of gene signatures will be diluted by RNA from other cells. While using specific PBMC subsets for transcriptional profiling would improve our ability to extract knowledge from these data, it is rarely obvious which cell subset(s) will be the most informative. RESULTS: We have developed a computational method (Subset Prediction from Enrichment Correlation, SPEC) to predict the cellular source for a pre-defined list of genes (i.e. a gene signature) using only data from total PBMCs. SPEC does not rely on the occurrence of cell subset-specific genes in the signature, but rather takes advantage of correlations with subset-specific genes across a set of samples. Validation using multiple experimental datasets demonstrates that SPEC can accurately identify the source of a gene signature as myeloid or lymphoid, as well as differentiate between B cells, T cells, NK cells and monocytes. Using SPEC, we predict that myeloid cells are the source of the interferon-therapy response gene signature associated with HCV patients who are non-responsive to standard therapy. CONCLUSIONS: SPEC is a powerful technique for blood genomic studies. It can help identify specific cell subsets that are important for understanding disease and therapy response. SPEC is widely applicable since only gene expression profiles from total PBMCs are required, and thus it can easily be used to mine the massive amount of existing microarray or RNA-seq data.