Search

Summary: Background: Multi-gene panel sequencing streamlines treatment selection for advanced non-small cell lung cancer (NSCLC). Implementation continues to be uneven across jurisdictions, partly due to uncertain clinical and economic impacts. In British Columbia (BC), Canada, the public healthcare system reimbursed a multi-gene panel in September 2016. This study determined the population-level cost-effectiveness of publicly reimbursed multi-gene panel sequencing compared to single-gene testing for advanced NSCLC. Methods: Our population-based retrospective study design used patient-level linked administrative health databases. We considered adult BC residents with a panel-eligible lung cancer diagnosis between September 2016 and December 2018. Using a machine learning approach, we conducted 1:1 genetic algorithm matching of recipients receiving multi-gene panel sequencing to controls receiving single-gene testing, maximising balance on observed demographic and clinical characteristics. Following matching, we estimated mean three-year survival time and costs (public healthcare payer perspective; 2021 CAD) and calculated the incremental net monetary benefit (INMB) for life-years gained (LYG) at conventional willingness-to-pay thresholds using inverse probability of censoring weighted linear regression and nonparametric bootstrapping. Findings: We matched 858 panel-eligible advanced NSCLC patients to controls, achieving balance for the 16 included covariates. Average test turnaround times were 18.6 days for multi-gene panel sequencing and 7.0 days for single-gene testing. After matching, mean incremental costs were $3529 (95% CI: −$4268, $10,942) and mean incremental LYG were 0.08 (95% CI: −0.04, 0.18). Among the 1000 bootstrap samples, 14.5% had lower costs and increased survival and 78.6% had higher costs and increased survival. The INMB was $523 (95% CI: −$6256, $7023) at $50,000/LYG, with a 57.5% probability of being cost-effective, and $4575 (95% CI: −$5468, $14,064) at $100,000/LYG, with an 84.0% probability of being cost-effective. Interpretation: Using population-based real-world data, we found a moderate to high probability that panel-based testing to inform targeted treatment for NSCLC would be cost-effective at higher thresholds. Funding: This research was supported by Genome British Columbia/Genome Canada (G05CHS) and the Terry Fox Research Institute. more...

Abstract Protein synthesis is frequently deregulated during tumorigenesis. However, the precise contexts of selective translational control and the regulators of such mechanisms in cancer is poorly understood. Here, we uncovered CNOT3, a subunit of the CCR4-NOT complex, as an essential modulator of translation in myeloid leukemia. Elevated CNOT3 expression correlates with unfavorable outcomes in patients with acute myeloid leukemia (AML). CNOT3 depletion induces differentiation and apoptosis and delayed leukemogenesis. Transcriptomic and proteomic profiling uncovers c-MYC as a critical downstream target which is translationally regulated by CNOT3. Global analysis of mRNA features demonstrates that CNOT3 selectively influences expression of target genes in a codon usage dependent manner. Furthermore, CNOT3 associates with the protein network largely consisting of ribosomal proteins and translation elongation factors in leukemia cells. Overall, our work elicits the direct requirement for translation efficiency in tumorigenesis and propose targeting the post-transcriptional circuitry via CNOT3 as a therapeutic vulnerability in AML. more...

Abstract Analyzing different omics data types independently is often too restrictive to allow for detection of subtle, but consistent, variations that are coherently supported based upon different assays. Integrating multi-omics data in one model can increase statistical power. However, designing such a model is challenging because different omics are measured at different levels. We developed the iNETgrate package ( https://bioconductor.org/packages/iNETgrate/ ) that efficiently integrates transcriptome and DNA methylation data in a single gene network. Applying iNETgrate on five independent datasets improved prognostication compared to common clinical gold standards and a patient similarity network approach. more...

Abstract Hematopoietic stem and progenitor cells (HSPCs) originate from an endothelial-to-hematopoietic transition (EHT) during embryogenesis. Characterization of early hemogenic endothelial (HE) cells is required to understand what drives hemogenic specification and to accurately define cells capable of undergoing EHT. Using Cellular Indexing of Transcriptomes and Epitopes by Sequencing (CITE-seq), we define the early subpopulation of pre-HE cells based on both surface markers and transcriptomes. We identify the transcription factor Meis1 as an essential regulator of hemogenic cell specification in the embryo prior to Runx1 expression. Meis1 is expressed at the earliest stages of EHT and distinguishes pre-HE cells primed towards the hemogenic trajectory from the arterial endothelial cells that continue towards a vascular fate. Endothelial-specific deletion of Meis1 impairs the formation of functional Runx1-expressing HE which significantly impedes the emergence of pre-HSPC via EHT. Our findings implicate Meis1 in a critical fate-determining step for establishing EHT potential in endothelial cells. more...

Summary: Hematopoietic stem and progenitor cells depend on bone marrow (BM) stromal cells for survival. Here, we present a protocol for performing three consecutive BM transplants in mice to study the role of BM niche in supporting hematopoiesis. We describe steps for transplanting cells to condition the marrow of the recipient mice and transplanting wild-type cells to examine the effect of the conditioned marrow in supporting hematopoiesis. We then detail procedures for transplanting into wild-type recipients to measure bone marrow chimerism.For complete details on the use and execution of this protocol, please refer to Gopal et al. (2022).1 : Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics. more...

Follicular lymphoma can transform to a more aggressive histology. Here, the authors use bulk and single cell analysis to create a 26 marker panel which could be used to profile FL samples and predict the risk of transformation using flow cytometry. more...

Summary: Background: Artificial intelligence (AI) is rapidly fuelling a fundamental transformation in the practice of pathology. However, clinical integration remains challenging, with no AI algorithms to date in routine adoption within typical anatomic pathology (AP) laboratories. This survey gathered current expert perspectives and expectations regarding the role of AI in AP from those with first-hand computational pathology and AI experience. Methods: Perspectives were solicited using the Delphi method from 24 subject matter experts between December 2020 and February 2021 regarding the anticipated role of AI in pathology by the year 2030. The study consisted of three consecutive rounds: 1) an open-ended, free response questionnaire generating a list of survey items; 2) a Likert-scale survey scored by experts and analysed for consensus; and 3) a repeat survey of items not reaching consensus to obtain further expert consensus. Findings: Consensus opinions were reached on 141 of 180 survey items (78.3%). Experts agreed that AI would be routinely and impactfully used within AP laboratory and pathologist clinical workflows by 2030. High consensus was reached on 100 items across nine categories encompassing the impact of AI on (1) pathology key performance indicators (KPIs) and (2) the pathology workforce and specific tasks performed by (3) pathologists and (4) AP lab technicians, as well as (5) specific AI applications and their likelihood of routine use by 2030, (6) AI's role in integrated diagnostics, (7) pathology tasks likely to be fully automated using AI, and (8) regulatory/legal and (9) ethical aspects of AI integration in pathology. Interpretation: This systematic consensus study details the expected short-to-mid-term impact of AI on pathology practice. These findings provide timely and relevant information regarding future care delivery in pathology and raise key practical, ethical, and legal challenges that must be addressed prior to AI's successful clinical implementation. Funding: No specific funding was provided for this study. more...

Abstract Blocked cellular differentiation is a central pathologic feature of the myeloid malignancies, myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). Treatment regimens promoting differentiation have resulted in incredible cure rates in certain AML subtypes, such as acute promyelocytic leukemia. Over the past several years, we have seen many new therapies for MDS/AML enter clinical practice, including epigenetic therapies (e.g., 5-azacitidine), isocitrate dehydrogenase (IDH) inhibitors, fms-like kinase 3 (FLT3) inhibitors, and lenalidomide for deletion 5q (del5q) MDS. Despite not being developed with the intent of manipulating differentiation, induction of differentiation is a major mechanism by which several of these novel agents function. In this review, we examine the new therapeutic landscape for these diseases, focusing on the role of hematopoietic differentiation and the impact of inflammation and aging. We review how current therapies in MDS/AML promote differentiation as a part of their therapeutic effect, and the cellular mechanisms by which this occurs. We then outline potential novel avenues to achieve differentiation in the myeloid malignancies for therapeutic purposes. This emerging body of knowledge about the importance of relieving differentiation blockade with anti-neoplastic therapies is important to understand how current novel agents function and may open avenues to developing new treatments that explicitly target cellular differentiation. Moving beyond cytotoxic agents has the potential to open new and unexpected avenues in the treatment of myeloid malignancies, hopefully providing more efficacy with reduced toxicity. more...

Since our previous studies found a low carbohydrate (CHO) diet containing soy protein and fish oil (15%Amylose/Soy/FO) significantly reduced lung and breast cancer in mice we asked herein if this low CHO diet could also delay the onset of myeloid malignancies. To test this we employed a miR-146a knock-out (KO) mouse model and found the 15%Amylose/Soy/FO diet increased their median lifespan by 8.5 month, compared to these mice on a Western diet. This was associated with increased lymphocytes and reduced monocytes, granulocytes, blood glucose and insulin levels. Inflammatory cytokine/chemokine studies carried out with 6-month-old mice, before any signs of illness, revealed the 15%Amylose/Soy/FO diet significantly reduced pro-inflammatory cytokines. This low CHO diet also led to an increase in plasma β-hydroxybutyrate and in liver fatty acid synthase levels. This, together with higher liver carnitine palmitoyltransferase I levels suggested that the 15%Amylose/Soy/FO diet was causing a systemic metabolic shift from glucose to fatty acids as an energy source. Lastly, we found the 15%Amylose/Soy/FO diet resulted in significantly higher numbers of primitive hematopoietic stem cells (HSCs) in the bone marrow of 6-month-old mice than those fed a Western diet. Taken together, these results suggest a 15%Amylose/Soy/FO diet reduces chronic inflammation and increases fatty acid oxidation and that this, in turn, may prevent HSC proliferation and exhaustion, thereby delaying myeloid malignancy-induced death of miR-146a KO mice. We suggest a low CHO diet containing soy protein and fish oil could be beneficial in reducing the risk of myeloid malignancies in patients with low miR-146a levels. more...

Abstract Background Single‐arm trials are common in precision oncology. Owing to the lack of randomized counterfactual, resultant data are not amenable to comparative outcomes analyses. Difference‐in‐difference (DID) methods present an opportunity to generate causal estimates of time‐varying treatment outcomes. Using DID, our study estimates within‐cohort effects of genomics‐informed treatment versus standard care on clinical and cost outcomes. Methods We focus on adults with advanced cancers enrolled in the single‐arm BC Cancer Personalized OncoGenomics program between 2012 and 2017. All individuals had a minimum of 1‐year follow up. Logistic regression explored baseline differences across patients who received a genomics‐informed treatment versus a standard care treatment after genomic sequencing. DID estimated the incremental effects of genomics‐informed treatment on time to treatment discontinuation (TTD), time to next treatment (TTNT), and costs. TTD and TTNT correlate with improved response and survival. Results Our study cohort included 346 patients, of whom 140 (40%) received genomics‐informed treatment after sequencing and 206 (60%) received standard care treatment. No significant differences in baseline characteristics were detected across treatment groups. DID estimated that the incremental effect of genomics‐informed versus standard care treatment was 102 days (95% CI: 35, 167) on TTD, 91 days (95% CI: −9, 175) on TTNT, and CAD$91,098 (95% CI: $46,848, $176,598) on costs. Effects were most pronounced in gastrointestinal cancer patients. Conclusions Genomics‐informed treatment had a statistically significant effect on TTD compared to standard care treatment, but at increased treatment costs. Within‐cohort evidence generated through this single‐arm study informs the early‐stage comparative effectiveness of precision oncology. more...

Abstract Integration of orthogonal data could provide new opportunities to pinpoint the underlying molecular mechanisms of hematologic disorders. Using a novel gene network approach, we integrated DNA methylation data from The Cancer Genome Atlas (n = 194 cases) with the corresponding gene expression profile. Our integrated gene network analysis classified AML patients into low‐, intermediate‐, and high‐risk groups. The identified high‐risk group had significantly shorter overall survival compared to the low‐risk group (p‐value ≤10−11). Specifically, our approach identified a particular subgroup of nine high‐risk AML cases that died within 2 years after diagnosis. These high‐risk cases otherwise would be incorrectly classified as intermediate‐risk solely based on cytogenetics, mutation profiles, and common molecular characteristics of AML. We confirmed the prognostic value of our integrative gene network approach using two independent datasets, as well as through comparison with European LeukemiaNet and LSC17 criteria. Our approach could be useful in the prognostication of a subset of borderline AML cases. These cases would not be classified into appropriate risk groups by other approaches that use gene expression, but not DNA methylation data. Our findings highlight the significance of epigenomic data, and they indicate integrating DNA methylation data with gene coexpression networks can have a synergistic effect. more...

Several genomic features have been found for acute myeloid leukaemia (AML) but targeted clinical genetic testing fails to predict prognosis. Here, the authors generate an AML prognostic score from RNA-seq data of patients, which successfully stratifies AML patients and which may provide guidance for therapeutic strategies. more...

Abstract Understanding how patient‐reported quality of life (QoL) and socioeconomic status (SES) relate to survival of acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) may improve prognostic information sharing. This study explores associations among QoL, SES, and survival through administration of the Euro‐QoL 5‐Dimension, 3‐level and Functional Assessment of Cancer Therapy‐Leukemia and financial impact questionnaires to 138 adult participants with newly diagnosed AML or MDS in a longitudinal, pan‐Canadian study. Cox regression and lasso variable selection models were used to explore associations among QoL, SES, and established predictors of survival. Secondary outcomes were changes in QoL, performance of the QoL instruments, and lost income. We found that higher QoL and SES were positively associated with survival. The Lasso model selected the visual analog scale of the EQ‐5D‐3L as the most important predictor among all other variables (P = .03; 92% selection). Patients with AML report improved QoL after treatment, despite higher mean out‐of‐pocket expenditures compared with MDS (up to $599 CDN/month for AML vs $239 for MDS; P = .05), greater loss of productivity‐related income (reaching $1786/month for AML vs $709 for MDS; P more...

Tissue context can dictate why a gene can have seemingly opposing functions in different settings. ELF3 is tumor suppressive in many cancers of epithelial origin but in lung cancer, the authors describe an oncogenic role in the adenocarcinoma histology of non-small cell lung cancer. more...

Studies with patient derived xenografts are hampered by factors such as genetic variability and sample availability. Here, the authors generate a leukemia mouse model by lentiviral transduction of normal human cord blood and show an oncogenic role of HOXB genes. more...

The analysis of cell-free circulating tumor DNA (ctDNA) is potentially a less invasive, more dynamic assessment of cancer progression and treatment response than characterizing solid tumor biopsies. Standard isolation methods require separation of plasma by centrifugation, a time-consuming step that complicates automation. To address these limitations, we present an automatable magnetic bead-based ctDNA isolation method that eliminates centrifugation to purify ctDNA directly from peripheral blood (PB). To develop and test our method, ctDNA from cancer patients was purified from PB and plasma. We found that allelic fractions of somatic single-nucleotide variants from target gene capture libraries were comparable, indicating that the PB ctDNA purification method may be a suitable replacement for the plasma-based protocols currently in use. more...

Summary: Increasing evidence of functional and transcriptional heterogeneity in phenotypically similar cells examined individually has prompted interest in obtaining parallel methylome data. We describe the development and application of such a protocol to index-sorted murine and human hematopoietic cells that are highly enriched in their content of functionally defined stem cells. Utilizing an optimized single-cell bisulfite sequencing protocol, we obtained quantitative DNA methylation measurements of up to 5.7 million CpGs in single hematopoietic cells. In parallel, we developed an analytical strategy (PDclust) to define single-cell DNA methylation states through pairwise comparisons of single-CpG methylation measurements. PDclust revealed that a single-cell epigenetic state can be described by a small ( more...

Myelodysplastic syndrome (MDS) is characterized by altered hematopoietic stem cell (HSC) and hematopoietic progenitor cell (HPC) regulation and reduction of miR-143 and miR-145 in some subtypes. Here the authors show that miR-143/145 loss leads to HSC depletion, HPC expansion and malignancy through Dab2 -mediated TGFβ pathway activation. more...

Abstract Background Approximately 5% to 10% of pancreatic ductal adenocarcinoma (PDAC) has a hereditary basis. In most of these defined hereditary cancer syndromes, PDAC is not the predominant cancer type. Traditional criteria for publicly funded genetic testing typically require the presence of a set combination of the predominant syndrome-associated cancer types in the family history. We report the identification of a CDKN2A pathogenic variant in a PDAC-prone family without the cutaneous features of multiple moles or melanoma that are characteristic of the Familial Atypical Multiple Mole Melanoma (FAMMM) Syndrome identified in a universal testing algorithm for inherited mutations in pancreatic cancer patients. Case presentation We present the case of two brothers of English ancestry diagnosed with PDAC within the same 12 month period, at the respective ages of 63 and 64 years of age. Neither brother reported a personal history of multiple moles or melanoma. Family history was positive for two second-degree relatives diagnosed with PDAC but was negative for other cancers or multiple moles in first- and second-degree relatives. Due to the absence of melanoma, this family did not meet provincial criteria for publicly funded genetic testing. Clinical genetic testing offered through a research grant identified a pathogenic variant in the CDKN2A gene c.377 T > A (p.Val126Asp). This variant is a North American founder mutation believed to pre-date colonization. Conclusions This case reminds clinicians to consider the possibility of a germline CDKN2A mutation in families with a high prevalence of PDAC, even in the absence of moles or melanoma. This case supports recent guidelines published by the American College of Medical Genetics and Genomics (ACMG) that genetics referrals are indicated in families with three or more cases of PDAC regardless of other cancer types in the family. A multi-gene panel approach is of particular benefit in diagnosing inherited cancer susceptibility in PDAC-only families. more...

Summary: The sterile alpha motif (SAM) and SRC homology 3 (SH3) domain containing protein 1 (Sash1) acts as a scaffold in TLR4 signaling. We generated Sash1−/− mice, which die in the perinatal period due to respiratory distress. Constitutive or endothelial-restricted Sash1 loss leads to a delay in maturation of alveolar epithelial cells causing reduced surfactant-associated protein synthesis. We show that Sash1 interacts with β-arrestin 1 downstream of the TLR4 pathway to activate Akt and endothelial nitric oxide synthase (eNOS) in microvascular endothelial cells. Generation of nitric oxide downstream of Sash1 in endothelial cells affects alveolar epithelial cells in a cGMP-dependent manner, inducing maturation of alveolar type 1 and 2 cells. Thus, we identify a critical cell nonautonomous function for Sash1 in embryonic development in which endothelial Sash1 regulates alveolar epithelial cell maturation and promotes pulmonary surfactant production through nitric oxide signaling. Lung immaturity is a major cause of respiratory distress and mortality in preterm infants, and these findings identify the endothelium as a potential target for therapy. : Surfactant deficiency due to lung immaturity is a major cause of respiratory distress in premature newborns. Coulombe et al. show that endothelial SAM and SH3 domain containing protein 1 (Sash1) drives perinatal lung maturation via nitric oxide signaling to alveolar cells. Sash1 interacts with β-arrestin1 to activate Akt-eNOS and induce alveolar epithelial cell maturation and surfactant synthesis. Keywords: TLR4, Sash1, surfactant, endothelium, alveolar type 2 cells, respiratory distress, lung development, β-arrestin, nitric oxide more...

Introduction: The initial point in the diagnostic workup of solid tumors remains manual, with the assessment of hematoxylin and eosin (H&E)-stained tissue sections by microscopy. This is a labor-intensive step that requires attention to detail. In addition, diagnoses are influenced by an individual pathologist's knowledge and experience and may not always be reproducible between pathologists. Methods: We introduce a deep learning-based method in colorectal cancer detection and segmentation from digitized H&E-stained histology slides. Results: In this study, we demonstrate that this neural network approach produces median accuracy of 99.9% for normal slides and 94.8% for cancer slides compared to pathologist-based diagnosis on H&E-stained slides digitized from clinical samples. Conclusion: Given that our approach has very high accuracy on normal slides, use of neural network algorithms may provide a screening approach to save pathologist time in identifying tumor regions. We suggest that this new method may be a powerful assistant for colorectal cancer diagnostics. more...

Extensive molecular profiling of leukemias and preleukemic diseases has revealed that distinct clinical entities, like acute myeloid (AML) and T-lymphoblastic leukemia (T-ALL), share similar pathogenetic mutations. It is not well understood how the cell of origin, accompanying mutations, extracellular signals or structural differences in a mutated gene determine the phenotypic identity of leukemias. We dissected the functional aspects of different protein regions of the MN1 oncogene and their effect on the leukemic phenotype, building on the ability of MN1 to induce leukemia without accompanying mutations. We found that the most C-terminal region of MN1 was required to block myeloid differentiation at an early stage, and deletion of an extended C-terminal region resulted in loss of myeloid identity and cell differentiation along the T-cell lineage in vivo. Megakaryocytic/erythroid lineage differentiation was blocked by the N-terminal region. In addition, the N-terminus was required for proliferation and leukemogenesis in vitro and in vivo through upregulation of HoxA9, HoxA10 and Meis2. Our results provide evidence that a single oncogene can modulate cellular identity of leukemic cells based on its active gene regions. It is therefore likely that different mutations in the same oncogene may impact cell fate decisions and phenotypic appearance of malignant diseases. more...

We examined whether lenalidomide exposure up-regulates miRNAs and mRNAs, previously shown to play a role in the disease phenotype of del(5q) myelodysplastic syndrome, in pre-treatment CD34+ marrow cells. We hypothesized that increased expression would predict for clinical response. Changes in miR-143, miR-145, miR-146a, miR-146b, miR-378, miR-584, SPARC and RPS14 were examined in del(5q) (n=10) and non-del(5q) (n=18) myelodysplastic syndrome patient samples. Significantly increased expression of miR-143 (1.8-fold and 1.5-fold in del(5q) and non-del(5q), respectively), and miR-145 (1.9-fold and 1.6-fold in del(5q) and non-del(5q), respectively) was observed. In the del(5q) myelodysplastic syndrome cohort, transfusion independence correlated with a 1.3-fold or more increase in miR-145 expression and response over 12 months correlated with a 1.5-fold or more increase. Knockdown of miR-143 and miR-145 in cord blood CD34+ cells resulted in increased erythroid progenitor activity. Lenalidomide selectively abrogated progenitor activity in cells depleted of miR-143 and miR-145 supporting a key role for miR-143/145 in the sensitivity to lenalidomide of del(5q) myelodysplastic syndrome patients. more...

Malformations of the cardiovascular system are the most common type of birth defect in humans, frequently affecting the formation of valves and septa. During heart valve and septa formation, cells from the atrio-ventricular canal (AVC) and outflow tract (OFT) regions of the heart undergo an epithelial-to-mesenchymal transformation (EMT) and invade the underlying extracellular matrix to give rise to endocardial cushions. Subsequent maturation of newly formed mesenchyme cells leads to thin stress-resistant leaflets. TWIST1 is a basic helix-loop-helix transcription factor expressed in newly formed mesenchyme cells of the AVC and OFT that has been shown to play roles in cell survival, cell proliferation and differentiation. However, the downstream targets of TWIST1 during heart valve formation remain unclear. To identify genes important for heart valve development downstream of TWIST1, we performed global gene expression profiling of AVC, OFT, atria and ventricles of the embryonic day 10.5 mouse heart by tag-sequencing (Tag-seq). Using this resource we identified a novel set of 939 genes, including 123 regulators of transcription, enriched in the valve forming regions of the heart. We compared these genes to a Tag-seq library from the Twist1 null developing valves revealing significant gene expression changes. These changes were consistent with a role of TWIST1 in controlling differentiation of mesenchymal cells following their transformation from endothelium in the mouse. To study the role of TWIST1 at the DNA level we performed chromatin immunoprecipitation and identified novel direct targets of TWIST1 in the developing heart valves. Our findings support a role for TWIST1 in the differentiation of AVC mesenchyme post-EMT in the mouse, and suggest that TWIST1 can exert its function by direct DNA binding to activate valve specific gene expression. more...

Follicular lymphoma patients display heterogeneous overall survival and variable risk of transformation. Recent studies have highlighted the role of the microenvironment. The contribution of microvessel density to follicular lymphoma survival remains controversial. We used a quantitative tumor mapping approach to determine whether the degree of vascularization correlated with outcome in a uniformly treated cohort. Whole-tissue sections of diagnostic biopsies from 84 cases were stained for CD34 and tumor-to-vessel-distance that encompassed 90% of the tumor (TVD90) was determined using image analysis. Twenty-one cases with lower TVD90 showed inferior overall survival (P=0.0001) and high risk of transformation (P=0.01). These cases significantly correlated with increased Lymphoma-Associated Macrophages (χ2=0.025). In multivariate analysis macrophages content, IPI and TVD90 were independent predictors of overall survival (P=0.05, P=0.001 and P=0.01, respectively) and IPI and TVD90 predicted risk of transformation (P=0.008 and P=0.08, respectively). Increased angiogenesis is an independent marker of inferior survival and may promote transformation. more...