Search

One major goal of cancer genome sequencing is to identify key genes and pathways that drive tumor pathogenesis. Although many studies have identified candidate driver genes based on recurrence of mutations in individual genes, subsets of genes with nonrecurrent mutations may also be defined as putative drivers if they affect a single biological pathway. In this fashion, we previously identified Wnt signaling as significantly mutated through large-scale massively parallel DNA sequencing of chronic lymphocytic leukemia (CLL). Here, we use a novel method of biomolecule delivery, vertical silicon nanowires, to efficiently introduce small interfering RNAs into CLL cells, and interrogate the effects of 8 of 15 mutated Wnt pathway members identified across 91 CLLs. In HEK293T cells, mutations in 2 genes did not generate functional changes, 3 led to dysregulated pathway activation, and 3 led to further activation or loss of repression of pathway activation. Silencing 4 of 8 mutated genes in CLL samples harboring the mutated alleles resulted in reduced viability compared with leukemia samples with wild-type alleles. We demonstrate that somatic mutations in CLL can generate dependence on this pathway for survival. These findings support the notion that nonrecurrent mutations at different nodes of the Wnt pathway can contribute to leukemogenesis.

Increasing evidence across malignancies suggests that infiltrating T cells at the site of disease are crucial to tumor control. We hypothesized that marrow-infiltrating immune populations play a critical role in response to donor lymphocyte infusion (DLI), an established and potentially curative immune therapy whose precise mechanism remains unknown. We therefore analyzed marrow-infiltrating immune populations in 29 patients (22 responders, 7 nonresponders) with relapsed chronic myelogenous leukemia who received CD4(+) DLI in the pre-tyrosine kinase inhibitor era. Immunohistochemical analysis of pretreatment marrow revealed that the presence of >4% marrow-infiltrating CD8(+) (but not CD4(+)) T cells predicted DLI response, even in the setting of high leukemia burden. Furthermore, mRNA expression profiling of marrow-infiltrating T cells of a subset of responders compared with nonresponders revealed enrichment of T-cell exhaustion-specific genes in pretreatment T cells of DLI responders and significant downregulation of gene components in the same pathway in responders in conjunction with clinical response. Our data demonstrate that response to DLI is associated with quantity of preexisting marrow CD8(+) T cells and local reversal of T-cell exhaustion. Our studies implicate T-cell exhaustion as a therapeutic target of DLI and support the potential use of novel anti-PD1/PDL1 agents in lieu of DLI.

Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is an aggressive malignancy thought to result from transformation of plasmacytoid dendritic cells (pDCs). Clinical outcomes are poor and pathogenesis is unclear. To better understand BPDCN genomics and disease mechanisms, we performed whole exome- (12 BPDCNs), targeted DNA- (additional 12 BPDCNs), bulk whole transcriptome RNA- (12 BPDCNs and 6 BPDCN patient-derived xenografts [PDXs]), and single cell RNA-sequencing (scRNA-seq) compared to normal DCs. We observed RNA splicing factor mutations in 16/24 cases (7 ZRSR2, 6 SRSF2, 1 each SF3B1, U2AF1, SF3A2, SF3B4). Additional recurrent alterations were in genes known to be mutated in other blood cancers: TET2, ASXL1, TP53, GNB1, NRAS, IDH2, ETV6, DNMT3A, and RUNX1. From exome sequencing we also discovered recurrent mutations in CRIPAK (6/12 cases), NEFH (4/12), HNF1A (2/12), PAX3 (2/12), and SSC5D (2/12) that may be unique to BPDCN. ZRSR2 is notable among the recurrently mutated splicing factors in hematologic malignancies in that all mutations are loss-of-function (e.g., nonsense, frameshift). Of note, BPDCN is very male predominant, ZRSR2 is located on chrX and all mutations are in males. ZRSR2 plays a critical role in "minor" or U12-type intron splicing (only 0.3% of all introns). Thus, we hypothesized that mis-splicing, possibly of U12 genes, contributes to BPDCN pathogenesis. Using RNA-seq, we measured aberrant splicing in BPDCN. Intron retention was the most frequent abnormality in ZRSR2 mutant BPDCNs and PDXs compared to non-mutant cases. ZRSR2 mutant intron retention predominantly affected U12 introns (patients: 29.4% of retained introns, P Disclosures Seiler: H3 Biomedicine: Employment. Buonamici:H3 Biomedicine: Employment. Lane:Stemline Therapeutics: Research Funding; N-of-1: Consultancy.

The mechanisms regulating key fate decisions such as self-renewal and differentiation in hematopoietic stem and progenitor cells (HSPC) remain poorly understood. We report here a screening strategy developed to assess modulators of human hematopoiesis using a lentiviral short hairpin RNA (shRNA) library transduced into cord blood-derived stem/progenitor cells. To screen for modifiers of self-renewal/differentiation, we used the limited persistence of HSPCs under ex vivo culture conditions as a baseline for functional selection of shRNAs conferring enhanced maintenance or expansion of the stem/progenitor potential. This approach enables complex, pooled screens in large numbers of cells. Functional selection identified novel specific gene targets (exostoses 1) or shRNA constructs capable of altering human hematopoietic progenitor differentiation or stem cell expansion, respectively, thereby demonstrating the potential of this forward screening approach in primary human stem cell populations.

Leukemia stem cells (LSCs) are found in most aggressive myeloid diseases and contribute to therapeutic resistance. LSCs are characterized by their gain of a self-renewal program that is normally associated with hematopoietic stem cells (HSCs). Previously we have shown that the RNA binding protein, Msi2 contributes to both HSC and myeloid leukemia function. Elevated MSI2 expression predicts a poor prognosis in a variety of leukemias and shRNA-mediated depletion in human AML cell lines reduces proliferation, increases differentiation and induces apoptosis. Despite these in vitroand correlative studies, MSI2’s molecular mechanism is not known and its role in LSC function has not been assessed. To elucidate MSI2’s role in LSC function, we utilized the MLL-AF9 leukemia mouse model. Initially we found MSI2 was elevated in the LSC enriched compartment (c-KitHigh cells) compared to non-LSCs (c-KitLow cells) based on flow cytometric intracellular staining. Therefore, to establish a model to study Msi2 and its contribution to myeloid LSCs, we have utilized the Msi2 conditional knockout mice that we previously crossed (Msi2f/f) into an Mx1-Cre background to generate the Msi2Δ/Δallele (injection of polyinositol-polycytosine; pIpC). In order to test if Msi2 is critical for MLL-AF9 mediated initiation, we transduced control Msi2f/f and Msi2Δ/ΔLin- Sca1+ c-Kit+cells (LSKs) with MLL-AF9 expressing retroviruses co-expressing GFP. Msi2 deleted LSKs or granulocyte-monocyte progenitors (GMPs) transduced with MLL-AF9 demonstrated delayed leukemogenesis with dramatically reduced diseased burden. Msi2 deficient leukemias were found to have a 4-fold reduced phenotypic LSC population and were more differentiated based on cellular morphology. Msi2 deficient leukemias failed to transplant into secondary recipients demonstrating that Msi2 is required for maintaining LSCs. Deletion of Msi2after leukemia engraftment led to a delay in leukemogenesis indicating that Msi2 is also important for leukemia maintenance. Gene expression profiling of the Msi2 ablated LSCs resulted in a loss of the HSC/LSC program and an increase in differentiation gene sets. The gene signature from the Msi2 deleted murine LSCs (121 genes) was overlapped and subjected to unsupervised clustering with the gene expression profiles from 336 AML patients (ECOG1900 dataset). This analysis resulted in distinct clusters that had differential MSI2 expression and the MSI2“high” cluster predicted a worse clinical outcome when compared to the other clusters. Overlapping of the differential transcriptional analysis of the Msi2 deleted murine LSCs with our global MSI2 direct mRNA targets (HITS-CLIP) led us to identify that MSI2 binds to transcripts that are associated with the downstream MLL self-renewal program, including Myc and Ikzf2. Ikzf2 is a member of the Ikaros transcription factor family and is known to regulate lymphocyte development by controlling regulatory T-cell function and chromatin remodeling. Ikzf2 shRNA mediated depletion resulted in reduced colony formation, decreased proliferation and increased apoptosis. The MLL associated targets were also reduced, which included Bcl-2 and Hoxa9. In contrast to its tumor suppressor role in hypodiploid B-ALL, these results suggest that Ikzf2 contributes to MLL leukemia cell maintenance. Thus, we provide evidence that MSI2 maintains the oncogenic LSC epigenetic program and the rationale for clinically targeting MSI2 in myeloid leukemia. Disclosures No relevant conflicts of interest to declare.

CLL exhibits great variability in genetic alterations and clinical outcome across patients. Gene expression can serve as an intermediate measurement to provide insight into the cellular circuitry linking genotype to phenotype. To explore this approach, we developed methods to statistically assess the associations between somatic mutations, transcriptional programs, and clinical outcome. First, in a cohort of 229 CLL patients, we associated IGHV-mutation status as well as 7 recurrent (>5% of CLLs) somatic alterations with 44 modules of co-expressed genes that were derived by clustering 3719 variably expressed genes measured by Affymetrix arrays. The somatic mutation genotypes were identified by whole exome sequencing (N=130) and/or SNP6 arrays (N=229). By permutation testing, we determined that 21 modules were associated with IGHV mutation status (P Second, having identified transcriptional modules associated with distinct genotypes, we sought to understand the functions of these modules and to infer the regulators of these programs. To associate modules with potential functional phenotypes, we performed gene enrichment analysis, and found multiple modules associated with inflammatory signatures, DNA repair and MYC targets. We then populated the intermediary layer between genotype and module with candidate transcription factors (TFs) by integrating curated TF datasets with TF expression, motif analysis and module expression. For example, CREB and ATF were candidate TF regulators in modules associated with SF3B1 and ATM mutations, respectively; while MYC- and NFkB-related TFs were candidate regulators of modules associated with both trisomy 12 or MYD88 mutations. We also identified several candidates of cellular convergence, where multiple genotypes lead to activation of the same transcriptional program. For example, the transcription factor EBF1, an important B cell regulator, was nominated as a candidate regulator in 8 of 44 modules which were associated with differing genotypes, suggesting the importance of EBF1 in mediating the genotype-phenotype relationship. Third, to complete the map from genotype to phenotype, we linked module expression with a clinical outcome. Using elastic-net Cox regression, we identified 2 modules associated with longer and 6 with shorter time from sample acquisition to treatment or death. Many of these modules were associated with well-established prognostic indicators (6 with IGHV status, 1 with P53/del17p, and 1 with SF3B1 status P In summary, this analysis serves as a proof of principle for a ‘genotype-phenotype map’ for CLL linking somatic alterations, gene expression programs, and clinical outcome. Our inferred CLL networks generate testable hypotheses that explain how genotypes affect the cellular circuitry of CLL cells that are currently being tested through functional gain/loss of function experiments. Disclosures: Brown: Pharmacyclics, Genentech, Celgene, Emergent, Onyx, Sanofi Aventis, Vertex, Avila, Novartis: Consultancy; Genzyme, Celgene: Research Funding.

We recently demonstrated that genetic heterogeneity contributes to evolution and clinical outcome in chronic lymphocytic leukemia (CLL) (Landau et al., Cell 2013). As both genetic and epigenetic alterations contribute to the cellular phenotype and its fitness for selection, we hypothesized that heterogeneity in DNA methylation impacts leukemic evolution as well. We analyzed existing CLL methylation data (Kulis et al., Nat Gen 2012), and found higher heterogeneity in 450K arrays of 139 CLLs compared to 23 normal B cell samples, with a higher proportion of sites having intermediate levels of methylation (13.7% vs 10.7%, P To validate this observation, we applied reduced representation bisulfite sequencing (RRBS) to generate genome-scale information of methylation at critical sites (e.g., CG islands) of 108 genetically-characterized CLLs. As predicted, we found higher proportions of discordant reads in CLL than either samples from 30 normal B cells or from 35 diverse healthy human tissues (average of 24.7% vs. 17.6% or 16.3%, respectively, P Increased epigenetic ‘noise’ is expected to result in a more plastic evolutionary landscape that facilitates the emergence of fitness-enhancing genetic and epigenetic alterations. Indeed, the increase in discordant reads in CLL compared to normal samples was most prominent in genomic regions important to transcription (promoters, CG islands [P To observe the relationship between genetic and epigenetic evolution, we used RRBS to longitudinally study 14 CLLs with characterized patterns of genetic evolution (median time between samples 3.5 yrs; 6 unevolved, 8 evolved). While unevolved CLLs revealed largely unchanged epigenetic landscapes, evolved CLLs demonstrated decreased methylation over time in promoters enriched for stem cell and MYC targets gene sets (N=347, Q Increased heterogeneity was found to be associated with shorter retreatment-free survival in a cohort of 46 patients treated after sampling (P=0.03, HR of 3.13), in a regression model including established risk indicators (IGHV unmutated status, del(17p), del(11q) and treatment prior to sample). Our data suggest that heterogeneity in CLL DNA methylation results from stochastic drift that subsequently undergoes positive selection. Thus, we propose that locally disordered DNA methylation may have a similar role to genetic instability, enhancing the ability of cancer cells to search for superior evolutionary trajectories. Disclosures: No relevant conflicts of interest to declare.

Tumor neoantigens are a promising class of vaccine immunogens as they arise from gene alterations in tumor cells and are hence exquisitely tumor-specific. We recently reported the development of a pipeline that leverages massively parallel sequencing data with HLA-peptide binding predictions to identify candidate neoantigens. By applying this pipeline to cases of chronic lymphocytic leukemia (CLL) with known HLA typing, we described the prediction of personal tumor neoantigens against which long-lived memory T cell responses developed following remission-inducing therapy. Our pipeline thus provides a method for selecting neoantigens for developing future personalized tumor vaccines. In order to extend this approach beyond CLL, we sought to gain estimates of tumor neoantigen loads across cancers. We hypothesized that the numbers of neoantigens within cancers would be proportional to their mutation frequency. To examine this hypothesis, we turned to the extensive collections of whole-exome sequencing (WES) data that have been generated through recent large-scale cancer sequencing projects. In order to generate accurate estimates of personal tumor neoantigen loads, HLA typing information is required. While in theory this information should be directly extractable from WES, direct inference of HLA type from standard WES reads has not been previously possible due to suboptimal alignments against a standard reference genome arising from the highly polymorphic nature of the HLA region. We therefore developed a strategy to optimize alignment. Based on the IMGT database, we constructed a reference library of all known HLA alleles (6597 unique entries) and aligned WES reads containing one or more short sequence segments corresponding to any HLA allele against this reference using the Novoalign software. HLA alleles were then inferred through a model that enabled calculation of allele probabilities by taking into account the number and quality of reads aligned to each allele. Alleles with the highest probabilities were then identified as winners. We trained the algorithm on 8 CLL cases for which WES data and HLA typing (based on conventional molecular typing) were available, and established a performance accuracy of ∼94% (45 of 48 alleles). This was further validated using a set of 133 Hap Map samples with known HLA typing, in which 94.61% (755 of 798) alleles were identified correctly at protein level resolution. We applied the HLA typing algorithm together with the neoantigen discovery pipeline across WES from 2488 cases collected from publicly available datasets of 13 diverse cancers. Mutation rates in solid tumor malignancies were consistently higher, in some cases by more than an order of magnitude, than the blood malignancies. For example, the high mutation rate tumor melanoma displayed a median of 300 (range, 34-4276) missense mutations per case, while renal cell carcinoma (RCC) had 41 (range, 10-101) and CLL had 16 (range, 0-75). The number of frame-shifting events (indels and termination read-throughs) was generally 10-fold or more lower in each tumor type than missense mutations and did not correlate with the number of missense mutations. As expected, the rate of predicted HLA binding peptides mirrored the somatic mutation rate per tumor type. The median number of predicted class I HLA-binding neopeptides (with IC50 < 500 nM) per sample generated from missense and frameshift events for melanoma was 488 (range: 18-5811), for RCC, 80 (range: 6-407), and for CLL 24 (range 2-124). Overall, we found an average of 1.5 HLA-binding peptides (i.e. with IC50 By predicting tumor neoantigens in a variety of low and high mutation rate cancers, we established that dozens to hundreds of potential neoantigens are present in most tumors. In the process, we developed a highly accurate analytic approach that provides a solution for extracting HLA typing information from WES data but which could, in principle, be applied to other highly polymorphic regions of the genome. Ongoing studies focus on integrating estimates of tumor neoantigen load with understanding of HLA expression in order to optimize selection of antigen targets to build future personalized tumor vaccines. Disclosures: No relevant conflicts of interest to declare.

Abstract 559 One goal of cancer genome sequencing is to identify key genes and pathways that drive tumor pathogenesis. While many studies identify driver genes based on recurrence of mutations in individual genes, subsets of genes with non-recurrent mutations may also be defined as drivers if they affect components of a single biological pathway. By large-scale DNA sequencing, we recently identified the Wnt pathway as a significantly mutated pathway in chronic lymphocytic leukemia (CLL) (Wang et al NEJM 2011). This pathway is known to be critical for hematopoietic stem cell development and differentiation, and is highly dysregulated in CLL. However, the functional alterations generated by specific mutations in members of this pathway and their contributions to CLL pathogenesis have not been previously characterized. In our previously reported DNA sequencing of 91 CLL samples, we identified a broad spectrum of mutated Wnt pathway members. Fifteen non-synonymous coding mutations were identified in 13 CLL samples (1–2 mutations per sample) within core Wnt pathway members. These nonsilent coding mutations involved components of the entire Wnt pathway–ranging from extracellular factors (DKK2, WNT1, WNT10A, RSPO4, CSNK1E), to cell surface receptors (FZD5, RYK), cytoplasmic factors (CSNK1E, PRICKLE1) and nuclear transcriptional factors (CHD8, BRD7, CREBBP, BCL9). All of the identified Wnt pathway mutations localized to evolutionarily conserved sites, supporting a role for these mutations in perturbing Wnt pathway function. To assess the contribution of the Wnt pathway to CLL-B cell survival, we optimized a novel miniaturized transfection platform, silicon nanowires (SiNWs), to deliver siRNAs targeting core Wnt pathway members into normal and CLL-B cells. By exploring various nanowire sizes, lengths and densities, we achieved conditions to consistently deliver siRNAs into B cells (90%) without impacting cellular viability (>95%). Using SiNW-mediated delivery of gene-specific siRNAs, we established that silencing of the Wnt pathway members LEF1, CTTNB1, and DVL1 in CLL samples (n=3) resulted in greater loss of cell viability (by∼20–35%) compared to delivery of non-targeting controls (P=0.06, 0.01, and 0.005, respectively). These results demonstrate that perturbation of key nodes in the Wnt pathway reduce CLL-B cell survival, and support the importance of this pathway in CLL. We then explored whether mutated Wnt pathway members could functionally alter Wnt pathway signaling. We focused our studies on two mutated Wnt pathway members, DKK2-R197H and BCL9-G548S. We generated plasmids expressing either wildtype or mutant alleles and measured Wnt pathway activation in 293T cells using a TCF/LEF-dependent luciferase reporter system following coexpression of WNT1 with the wildtype or mutant gene or both. For both DKK2 and BCL9, mutated forms abolished their normal repressive effects on Wnt activation. Elimination of the repressive effects of the wildtype protein was also observed when the wildtype and mutant alleles were expressed together, suggesting a dominant activating effect of the mutated allele. We hypothesized that chronic pathway activation by these mutations could lead to dependence of patient CLL cells on Wnt signaling for their survival. We directly addressed this question by examining the survival of normal B cells (n=4) and patient CLL-B cells with or without (n=7) mutated DKK2 or BCL9 following SiNW-mediated delivery of gene-specific siRNAs. This revealed that a BCL9-mutated CLL sample was more dependent on BCL9 expression for survival than normal B cells (p=0.02) or CLL samples lacking Wnt pathway mutations (p=0.07). Similarly, silencing of DKK2 in a DKK2- mutated CLL sample led to greater cell death than gene silencing in normal B cells, or in CLL samples without Wnt pathway mutations (p=0.003). These results demonstrate that CLL samples harboring mutations in DKK2 or BCL9 exhibit greater dependency on Wnt pathway signaling than samples without Wnt pathway mutations, and support a driving role for these mutations in CLL. Taken together, our results demonstrate that somatic mutations in Wnt pathway components in CLL-B cells alter the activity of this pathway and increase its sensitivity to pathway inhibition. Our findings further support the notion that non-recurrent mutations at different nodes of the Wnt pathway contribute to CLL pathogenesis. Disclosures: Brown: Calistoga, Celgene, Genentech, Pharmacyclics, Novartis, Avila: Consultancy; Genzyme, Celgene: Research Funding.

Abstract 1903 Donor lymphocyte infusion (DLI) can provide curative treatment for relapsed hematologic malignancies following allogeneic hematopoietic stem cell transplant (HSCT). However, the precise mechanism by which DLI eliminates leukemia cells in vivo has not been established. We hypothesized that marrow-infiltrating immune populations play a critical role in DLI responses since marrow is the primary site of disease and a reservoir of high-avidity antigen-specific memory T cells that can recognize tumor antigens, therefore potentially mediating graft-versus-leukemia (GvL) responses. We performed immunohistochemical staining of immune cells in serial marrow biopsies collected before and after DLI from 29 patients with relapsed CML. Twenty-two patients achieved cytogenetic remission within twelve months (‘responders’) while 7 patients demonstrated persistent disease (‘non-responders’). While no significant changes in the numbers of circulating T cells were seen between patient groups following DLI, the median number of marrow-infiltrating CD8+ T cells increased 2-fold in responders but remained unchanged in non-responders (P=0.02), demonstrating that clinical response to DLI is associated with T cell responses at the site of disease that may not be apparent in the peripheral blood. To investigate whether immune cell infiltration of the marrow could predict DLI response, we compared pre-treatment samples from both patient groups. Responders exhibited significantly higher proportions of CD8+ T cells in pre-DLI marrow compared to non-responders (5% vs 2.5%; P=0.01). Because disease burden is a known risk factor for ineffectual DLI response, we evaluated the interaction between disease burden and pre-existing CD8+ T cell infiltrate through the clinical course of 8 patients with high (≥70%) pre-treatment marrow cellularity. Three of 8 had ≥5% CD8+ T cell marrow infiltrates, and all 3 subsequently achieved cytogenetic remission. In contrast, 5 of 8 patients had To identify candidate mechanisms underlying T cell responses to DLI, we performed mRNA expression profiling of CD3+ T cells isolated from matched pre- and post-treatment marrows of 4 responders and 2 non-responders (U133+ Affymetrix cDNA microarrays). We first compared global gene expression patterns between pre-treatment marrow-infiltrating T cells of both groups. Notably, 28% of significantly upregulated genes in responders play critical roles in T cell exhaustion. This finding was strengthened by unbiased gene set enrichment analysis (GSEA) using 880 sets from the Molecular Signatures Database spiked with 17 additional specifically curated T cell exhaustion sets. Four of 15 positively enriched sets were T cell exhaustion-specific. We next compared differential gene expression in marrow-infiltrating T cells before and after therapy in responders compared to non-responders and found 21% of significantly down-regulated genes to be components within T cell exhaustion pathways along with repression of multiple, distinct T cell exhaustion gene sets. The robust reversibility of T cell exhaustion signatures after DLI therapy underscores this gene module as a likely therapeutic target of DLI. In conclusion, CD8+ T cell marrow infiltration may serve as a novel predictive marker of response to DLI, including patients with high disease burden. In addition, these data implicate T cell exhaustion in distinguishing responders from non-responders and, provocatively, propose reversal of T cell exhaustion as a potential marker of DLI responsiveness in patients with relapsed CML after HSCT. These studies illustrate the importance of local immune responses at the site of disease and suggest a potential tool to predict DLI response in other hematologic malignancies. The clinical debut of newer agents that reverse T cell exhaustion suggests their use in lieu of DLI to promote GvL responses after allogeneic HSCT. Disclosures: No relevant conflicts of interest to declare.

Abstract 954 A major goal of cancer immunotherapy is to target tumor cells with high specificity and minimal toxicity. Tumor neoantigens are attractive targets for immunotherapy as they are generated from cancer-driven gene mutations and are expressed exclusively in tumor cells. Neoantigens have not previously been systematically investigated in human tumors due to technical challenges in identifying them. To address this, and to identify immunogenic epitopes personal to an individual tumor, we sought to exploit results from next-generation DNA sequencing to detect somatic mutations together with algorithms that predict peptide binding to HLA class I. We tested this approach in chronic lymphocytic leukemia (CLL), a B cell malignancy associated with a rich source of mutations (Wang et al, NEJM 2011). In a published study, we sequenced 91 CLL samples using whole-genome (n=3) or whole-exome (n=88; WES) DNA sequencing. To detect mutations, sequences in each tumor sample were compared with corresponding normal sequences, and the frequency of various mutation classes that lead to amino acid alterations were determined. On a per-tumor level, we detected a median of 17 (range: 2–72) missense mutations; 1 (range: 0–6) splice-site mutation; 1 (range: 0–1) read-through mutation, and 1 (range: 0–5) frameshift mutation. Frameshift mutations were predicted to generate potential neo-open reading frames with a median length of 40 amino acids (range: 0–180). Only peptides binding HLA molecules have the potential to elicit immune responses. We therefore focused on mutations generating autologous HLA binding peptides. Using a well-validated MHC class I prediction algorithm, pan-NetMHC, we systematically evaluated the binding potential of 9- and 10-mers tiled around each mutation. For 31 of 91 samples with available HLA typing, a median of 22 strongly binding peptides (range: 9–75) was predicted to be generated from a median of 26 (range: 7–87) mutations per sample. Using a competitive HLA binding assay, we experimentally validated the predicted high-binding capacity of 60 of 112 (53.5%) synthesized peptides, generated from 3 patients. To determine if tumor neoantigens are naturally recognized by cytotoxic T lymphocytes, we focused our analysis on CLL patients who had undergone successful hematopoietic stem cell transplantation (HSCT) followed by repeated vaccinations with irradiated autologous whole CLL cells, as normal donor T cell reconstitution following HSCT can overcome endogenous immune defects of the host and since donor T cells in this setting are already primed against host leukemia cells in vivo. Analysis of one patient identified 31 coding mutations, which generated 47 peptides predicted to bind autologous HLA. Of these, 24 were experimentally confirmed to bind autologous HLA alleles and thereafter were screened for inducing T cell reactivity. Using IFN-g ELISPOT reactivity following 2 ex vivo peptide stimulations, memory CD8+ T cell responses could be detected against a HLA A2-binding peptide derived from mutated FNDC3B (VVMSWAPPV). These cells were preferentially reactive to the mutated (but not wildtype) peptide. Results were confirmed with peptide-reactive T cell clones and staining with HLA-A2+/mutated FNDC3B peptide-specific tetramers. Mutated FNDC3B-reactive T cells also recognized HLA-A2+ expressing cells transfected with a minigene encompassing 300 base pairs surrounding the FNDC3B mutation, consistent with endogenous processing and presentation of the peptide. Ongoing studies focus on exploring the tumor cytolytic potential of mutation-specific T cells, and the kinetics of developing neoantigen-specific T cell reactivity in relation to immune reconstitution. Our studies provide proof-of-concept for systematic identification of tumor neoantigens by integrating information from next-generation sequencing of tumors with predictive HLA binding algorithms. These studies now set the stage for the implementation of clinical studies to explore the therapeutic potential of targeting tumor neoantigens. Disclosures: No relevant conflicts of interest to declare.

Abstract 463 The somatic genetic basis of chronic lymphocytic leukemia (CLL), a common and clinically heterogenous adult leukemia, remains poorly understood. Massively parallel sequencing technology provides a method for systematic discovery of genetic alterations that underlie disease, and for uncovering new therapeutic targets and biomarkers. We therefore performed DNA sequencing of 91 CLL samples with matched germline, consisting of 88 exomes and 3 genomes. Samples were collected from patients displaying a range of characteristics representing the broad clinical spectrum of CLL. Libraries were constructed and sequenced on an Illumina GA-II sequencer to deep coverage. Point mutations and indels were detected by comparing sequences of each tumor to its corresponding normal. We detected 1828 non-silent mutations in protein-coding sequences, corresponding to an average somatic mutation rate of 0.72/Mb (range 0.075–2.14). Nine cancer genes were identified, as they were mutated significantly more often than the background rate given their sequence composition across all 91 CLL/normal pairs. Four have been previously described in CLL (TP53, ATM, MYD88 and NOTCH1); the others were novel (SF3B1, ZMYM3, MAPK1, FBXW7 and DDX3X). Strikingly, SF3B1, which functions at the catalytic core of the spliceosome, was mutated the second most frequently (15%). All 14 identified mutations localized to a discrete region (exons 12–18), and 7 were recurrent (K700E). We additionally validated the SF3B1-K700E mutation through genotyping of 101 independent CLL samples. The 9 significantly mutated genes fell into 5 core signaling pathways, in which the genes play established roles: DNA damage repair and cell-cycle control (TP53, ATM), Notch signaling (FBXW7, NOTCH1), inflammatory pathways (MYD88, DDX3X, MAPK1) and RNA splicing/processing (SF3B1, DDX3X). The common cytogenetic aberrations in CLL carry strong prognostic significance, implying that they reflect distinct pathogenesis. Supporting this hypothesis, we discovered strong associations between different driver mutations and key FISH abnormalities. Most TP53 mutations were in samples with del(17p) (p To define the impact of the mutated cancer genes on disease outcome, we constructed a Cox multivariable regression model for factors contributing to earlier time to first therapy (TTFT) in the 91 CLLs. SF3B1 mutation was predictive of poor prognosis (HR 3.17, p=0.004), independent of other established markers (i.e. IGVH mutation; del(17p); ATM mutation). Consistent with this analysis, patients harboring the SF3B1 mutation alone had short TTFT, as did patients with del(11q) or del(11q)+SF3B1 mutation. All 3 groups demonstrated significantly shorter TTFT than patients without either SF3B1 mutation or del(11q). Because SF3B1 encodes a splicing factor, we looked for functional evidence of splicing alterations associated with SF3B1 mutation. Indeed, mRNA targets of the spliceosome exhibited greater intron retention in CLLs with mutated vs nonmutated SF3B1, confirming alteration in normal SF3B1 function. Understanding the mutational landscape of CLL provides a starting point for systematic analyses to address fundamental questions in CLL, including how mutated genes alter cellular networks and phenotypes, and thereby contribute to disease heterogeneity. Our discovery of striking associations between driver mutations and standard CLL prognostic markers (cytogenetic aberrations, IGVH mutational status) suggest synergistic interactions. In particular, our studies highlight pre-mRNA splicing as a critical but thus far unexplored cellular process contributing to CLL. Disclosures: No relevant conflicts of interest to declare.

Abstract 801 We have recently found that the Wnt/b-catenin signaling pathway plays a key role in chronic lymphocytic leukemia (CLL). We were, however, intrigued by the question of whether this aberrant pathway may function differently in independent leukemias, and contribute to disease heterogeneity. To assess differential activity of the Wnt pathway across patients, we tested the effects of blocking Wnt activation on CLL cell survival. We knocked down a key downstream gene, LEF1, which is the most differentially expressed gene in CLL compared to normal B cells (based on gene expression microarrays). Addressing this question requires genetic manipulation of primary normal and malignant human B cells, and yet these cells are notoriously difficult to transfect. We therefore focused on developing a method for introducing siRNAs into normal and malignant B cells. We adapted a novel delivery system consisting of vertical silicon nanowires (SiNWs, Shalek et al PNAS 2010) that penetrate the plasma membrane in a minimally invasive fashion and deliver biomolecular cargo directly into the cytoplasm. We achieved consistent and reliable delivery of fluorescently labeled siRNAs (at 50–200 pmol) into normal and CLL B cells. siRNA was delivered to >90% of cells with >85% cell viability remaining after 48 hours. We used this platform to knockdown LEF1 in 20 CLL-B and 5 normal CD19+ B cell samples, and examined cell survival 48 hours after siRNA delivery using an ATP-based CellTiter-Glo assay. Indeed, our studies revealed a heterogeneous response among CLL-B cells to LEF1 inhibition. As a group, CLL-B cells were significantly more sensitive to LEF1 knockdown with a survival rate of 77% (12% s.e.m) compared to 97% (13% s.e.m) in normal B cells. CLL B cells from different patients showed differential sensitivity to LEF1 knockdown, with 8 non-responders, 8 intermediate responders and 4 strong responders (i.e. significant death). Sensitivity to LEF1 inhibition did not correlate with known CLL cytogenetic prognostic factors. To determine if the differential response to LEF1 knockdown was associated with specific gene signatures, we examined gene expression data generated from CLL-B cells from 12 (4 strong, 3 intermediate, and 5 non-responders) of the 20 CLLs tested (using the Affymetrix U133 Plus 2 Array). To increase statistical power, we used each CLL's expression profile (using only genes that showed variability across samples) to create clusters of ∼19 CLLs that showed similar expression profiles (using microarray data from our compendium of 177 additional CLLs). We further reduced the number of genes to ∼4000 genes by retaining only those whose expression levels were significantly different in at least one associated cluster relative to normal CD19+ B cell controls (T-test, FDR Disclosures: No relevant conflicts of interest to declare.

Abstract 284 Tumor evolution is a complex process, and is the biologic underpinning of disease progression, resistance to therapy and relapse. Using whole-exome sequencing (WES) of sequential samples from patients with relapsed chronic lymphocytic leukemia (CLL) treated with conventional chemotherapy, we studied genetic tumor evolution of cancer relapse. We performed WES using paired-end reads on DNA from two peripheral blood-derived CLL tumor samples at least one year apart and on germline DNA for 20 patients. Here we report the analysis of tumor exomes from the first seven patients, of whom 6 had relapsed disease after chemotherapy and one untreated patient without intervening therapy between samples. All samples had a tumor purity that exceeded 90%. Sequencing coverage was >86% of target territory, with 132x depth obtained for all samples. In total, 187 coding region mutations (124 nonsynonymous, 63 synonymous) were identified (median: 21 somatic mutations/patient; range: 10–64), not including Ig gene mutations which were >80% clonal and remained clonally stable in our cohort. We measured the abundance of specific mutations in each patient tumor to assess clonality. An allelic frequency of 0.3–0.6 likely represents heterozygous mutations in most or all tumor cells (‘clonal') while a frequency of In Patient A, a subclone with three mutations appeared to expand to become the dominant clone, with a change in allelic fraction from an average of 0.17 (0.14–0.23) to an average of 0.43 (0.41–0.46) (p Only one sample demonstrated the appearance of novel mutations with relapse (Patient C), with 19 new mutations (13 non-silent, 3 Silent) of a total of 46 appearing at relapse. All however were subclonal, and thus less likely to have driven tumor relapse. A comparison of the 10 mutations that were selected by chemotherapy to all other mutations demonstrated an enrichment in mutations seen in the COSMIC database (p In summary, our analysis of serial exomes from seven patients provided important insights into the genetic evolution of CLL under the selective pressure of chemotherapy. We demonstrate a significant change in clonal dynamics in one half of treated patients, which suggests that relapsed disease following treatment is driven by expansion of subclones under the selective pressure of chemotherapy rather than by novel mutagenesis. This observation may have clinical implications, as it suggests that pre-treatment WES may allow not only for the delineation of current genetic abnormalities, but through investigation of subclonal mutations, may also predict genetic evolution in future relapse. Disclosures: No relevant conflicts of interest to declare.

Abstract 693 The molecular pathogenesis of chronic lymphocytic leukemia (CLL) has not been well-characterized. Next generation sequencing technologies provide an unprecedented opportunity to screen entire genomes for genetic changes associated with tumorigenesis. To examine the mutational spectrum in CLL, we completed Illumina sequencing of 2 genomes and 4 exomes from patients with advanced CLL in parallel with germline DNA from autologous skin fibroblasts (average coverage 66x/59x for CLL/normal). While the somatic mutation rates varied widely between individuals, ranging from 1–10 per megabase, we discovered that 2 of 6 tumors harbored novel nonsynonymous mutations in Wnt pathway members (DKK2 [R197H] from Patient 1 and BCL9 [G458S] from Patient 3), suggesting that perturbation of this pathway may be critical in CLL. By targeted quantitative RNA sequencing, we detected both mutated DKK2 and BCL9 as heterozygous mutations in tumor RNA from Patients 1 and 3, respectively. We confirmed that the Wnt pathway is in fact highly dysregulated in CLL based on gene expression analysis of 141 Wnt pathway members, in a dataset comprised of 155 CLL and 20 normal CD19+ B cell samples. In CLL cells, Wnt family members (Wnt1, Wnt2, Wnt2B, Wnt3, Wnt4, Wnt5B, Wnt9A, Wnt10A, and Wnt16) and Wnt receptors (FZD3, FZD8, FZD9, LRP6) were overexpressed. Additionally, we observed downregulation of Wnt pathway repressors (TLE1, SOX4 and NKD1) and upregulation of activators (SMAD2, DVL2, DVL3 and PYGO1). Expression of the downstream transcription factor LEF1 was strikingly elevated in CLL, and we measured consistent 3–4 fold elevation of LEF1 protein in CLL-B cells (n=10) compared to normal B cells (n=5) by intracellular staining (p= 0.002). Patient CLL cells demonstrated 5-fold greater beta-catenin mediated activation of the Wnt pathway in CLL compared to normal B cells (p=0.01), as measured by direct introduction of the LEF1/TCF1 luciferase reporter genes into primary cells. Silencing of LEF1 in CLL cells (n=5) led to increased cell death in vitro by 70–90%. Collectively, these results demonstrate that Wnt pathway activation and dysregulation contributes to CLL survival. To examine the possible effects of mutated DKK2 and BCL9 on Wnt pathway activity, we introduced plasmids encoding wild-type or mutated genes, together with those encoding Wnt1 protein and Wnt reporter genes, into 293T cells. DKK2 is a secreted factor that normally represses Wnt activation by competing with Wnt proteins for binding to its surface receptors. We observed that DKK2 [R197H] is mutated in its second cysteine-rich domain, which is necessary for its interaction with the Wnt receptor. BCL9 interacts with the LEF1/TCF transcriptional factor complex and has been described to have both repressive and activating effects. In 3 of 3 experiments, we observed that the mutated forms of both genes lost their repressive activity. The loss of repression generated for both mutated genes was most marked in the presence of low concentrations of Wnt protein. Moreover, both mutated genes act as dominant alleles, especially in the presence of low concentrations of Wnt protein. Since DKK2 is a secreted factor, we examined the effects of plasma of Patient 1 on Wnt pathway activity. In both normal donors (n=4) and CLL patients (n=4), DKK2 protein could be consistently detected at equivalent amounts in blood plasma following immunoprecipitation and immunoblotting with DKK2-specific antibodies Co-culture of 293T cells expressing the Wnt pathway luciferase reporters with Patient 1 plasma (2-10%) resulted in dramatic elevation of Wnt activation, which was abolished by prior depletion of DKK2. We observed similar Wnt activation when testing plasma from 22 of 93 CLL patients, but not from 5 normal volunteers. Further sequencing of 100 Wnt pathway members from 60 additional CLL patients is in progress to define the extent to which somatic mutation is a common mechanism by which this pathway is altered. Our studies suggest that Wnt pathway inhibitors represent a promising therapeutic approach for CLL. Disclosures: No relevant conflicts of interest to declare.