Search

Peripheral T-cell lymphoma (PTCL) makes up about 12 percent of non-Hodgkin lymphoma, comprising 18 diseases that are poorly understood and carry a generally worse prognosis than B-lymphomas. PTCL not otherwise specified (PTCL-NOS), a diagnosis of exclusion, is most common, making up 25-30 percent. Gene-expression studies suggest a heterogeneous origin of this diagnosis, with overlap to other PTCL types, but the genetic factors underlying its pathogenesis are undefined. Current therapy for PTCL-NOS is empiric and ultimately ineffective for most patients. Identification of specific therapeutic targets is therefore a high priority. We have sought better understanding of pathogenesis through next-generation sequencing of PTCL-NOS tumor DNA. Whole-exome sequencing revealed candidate genes but low availability of fresh-frozen samples limited our ability to draw conclusions by this method alone. We therefore sequenced the coding regions of 237 candidate genes in a collection formalin-fixed paraffin-embedded samples. We used Nimblegen Sequence Capture for PCR amplification of exons and Illumina hiSeq for raw sequence generation. Results were aligned to hg19 and compared to dbSNP and the 1,000 genomes data to exclude germline variants. Analysis, including comparison to the COSMIC database of cancer-specific mutations, revealed high-confidence mutations affecting more than 60 known cancer-related genes in 25 PTCL-NOS cases. Recurrent mutations pointed to frequent activation of three key signaling pathways: NF-kB (TNFAIP3), WNT/B-Catenin (APC, CHD8, CELSR2), and NOTCH (NOTCH1, FBXW7). Recurrent deregulation of epigenetic processes was indicated by mutations in genes affecting histone acetylation (EP300, CREBBP), histone methylation (MLL2, KDM6A), and DNA methylation (TET2, DNMT3A). In addition, components of core tumor suppressor pathways showed evidence of frequent inactivation (TP53, ATM, RB1, CUL9, PRKDC). In all, 22 of 25 cases had mutations in at least one of these 17 recurrently mutated genes. Multiple additional candidate disease mechanisms also were suggested by lower-confidence mutations but require confirmation studies, which are under way. In sum, analysis of the coding region of PTCL-NOS tumor DNA suggests a complex and heterogeneous pathogenesis, in line with gene-expression profiling. This work provides an opportunity to better sub-classify entities within the diagnosis of PTCL-NOS and identify specific therapeutic targets and their associated biomarkers. Disclosures: Horwitz: Seattle Genetics, Inc.: Consultancy, Research Funding; Millennium: Consultancy, Research Funding.

Hairy cell leukemia (HCL) is a chronic lymphoproliferative disorder recently found to be characterized by somatic BRAFV600E mutations. The malignant cell in HCL exhibits features consistent with a mature B-lymphocyte, including cell-surface expression of the pan-B-cell marker CD19 and monotypic surface immunoglobulins with clonal rearrangements of immunoglobulin heavy and light chains. Despite possessing these stereotypic features, the cell of origin of HCL has been long debated, and no cell type along the continuum of developing B-lymphocytes has been definitively identified as the normal counterpart of HCL cells. We hypothesized that HCL may originate from immature hematopoietic cells, and therefore investigated the hematopoietic-stem/progenitor cell (HSPC) compartment in HCL patients. We found that HCL patients exhibited a significantly increased frequency of immunophenotypically defined long-term hematopoietic stem cells (LT-HSCs; lineage-negative (Lin-neg) CD34+CD38-CD90+CD45RA- cells), pro-B cells (Lin-neg CD10+ cells), and CD34-CD38+ CD10+CD19+ hematogones, as well as a decreased frequency of granulocyte-macrophage progenitor cells (Lin-neg CD34+CD38+CD45RA+CD123+) relative to age-matched normal controls. Sequencing of cDNA from highly pure FACS-sorted cell populations from the bone marrow of HCL patients revealed the presence of the BRAFV600E allele in LT-HSCs and in pro-B cells (Figure). Transplantation of LT-HSCs from the pretreatment bone marrow of HCL patients into NOD/SCID/IL2r-gnull mice resulted in stable human grafts characterized by an expanded B-progenitor population and development of a clonal population of hCD19+hCD103+hCD25+ B cells characteristic of HCL 6 months after transplantation. Together, these data suggest that HCL arises from HSCs that then differentiate into committed B-cells which ultimately give rise to the characteristic clonal B-cell proliferation of HCL. Given the human HSC genetic and functional cell data, we conditionally expressed BRafV600E from its endogenous locus at different stages of hematopoiesis, including in HSPCs and committed B cells. Mice with conditional expression of BRafV600E in Mx1Cre+ BRafV600E knock-in mice died of a lethal hematopoietic malignancy characterized by features of human HCL including splenomegaly, anemia, thrombocytopenia, increased circulating sCD25, and increased clonogenic capacity of B-lineage cells (evidenced by infinite serial replating in the presence of IL-7) (Figure). This disorder was transplantable into lethally-irradiated recipient mice. In contrast, mice with expression of BRafV600E restricted to the B-cell lineage with Cd19 Cre manifested no overt malignant phenotype up to one year of age. Stimulation of these mice with alloantigen through injections of sheep red blood cells resulted in germinal center B-cell hyperplasia, but still did not result in development of a clonal B-cell proliferation. Recent case reports have noted that refractory HCL patients respond to mutant BRAF inhibition with vemurafenib. We investigated the effect of vemurafenib on HSPCs and hematopoiesis in patients treated on a phase II study of the mutant BRAF inhibitor vemurafenib for relapsed/refractory HCL as well as in our in vivo murine models. Flow cytometric analysis of bone marrow cells from vemurafenib treated HCL patients revealed normalization of HSPC frequencies within three months of starting therapy, concomitant with an improvement in peripheral blood counts. Consistent with this, evaluation of the in vitro clonogenic capacity of sorted LT-HSC's from the bone marrow of HCL patients revealed a significant increase in myeloid/erythroid colony formation in HCL patients treated for 3 months with vemurafenib compared to their pretreatment marrows. Likewise, treatment of wildtype mice transplanted with Mx1Cre+ BRafV600E mutant bone marrow cells revealed improvement in anemia and hepatosplenomegaly with in vivo therapy. Overall, these findings link the pathogenesis of HCL to a specific somatic genetic abnormality present in HSCs and provide evidence that mature B-cell malignancies can initiate in the HSC compartment. Moreover, these data suggest that the use of therapies targeting MAP kinase signaling in HCL may lead to durable remissions not only by eliminating the mature leukemic cells but also through targeted inhibition of signaling and survival in HCL initiating cells. Disclosures: No relevant conflicts of interest to declare.

Diffuse large B-cell lymphomas (DLBCL) display marked clinical, pathologic, and genetic heterogeneity. With current frontline immunotherapy (RCHOP), only about 40% of patients are cured, with most relapses occurring within the first 2-3 years. Patients are currently risk-stratified based primarily on clinical features where the inclusion of molecular biomarkers into risk assessment could impact the potential to identify those patients most likely to have refractory disease or have an early relapse. Various cytogenomic studies have revealed the prognostic significance of genomic gain/loss in DLBCL, but their lack of utility and reproducibility across datasets can be attributed to not only different patient populations, but also the use of disparate platforms and analytical methods. The goal of the present study was to use a common analytical approach across different clinical datasets, to identify genomic loci with robust prognostic value in DLBCL. To this end, GISTIC was applied to copy number data of newly-diagnosed DLBCL (GSE11318 [n=170], E-MEXP-3463 [n=49]) and together with another similarly analyzed published dataset (PMID: 22975378 n= [180]). Regions of known copy number variations were excluded from the analysis. A total of 35 overlapping minimal common regions (MCRs) and 21 overlapping peaks of genomic gain/loss with well-defined calling parameters were identified in at least 2 of the 3 datasets with a minimum occurrence of 5%. Overlapping peaks mapped to loci of genes with known involvement in DLBCL such as TP53, CDKN2A, and TNFAIP3 confirming the validity of the approach. Individual MCRs were tested for association with overall survival using the Kaplan-Meier method and log-rank statistic in three RCHOP datasets: E-MEXP-3463, GSE15127 [N=124], and an in-house dataset for which array-CGH was performed using a targeted oligonucleotide (Agilent Technologies) with DNA extracted from 1-3 CD20+ enriched cores of formalin–fixed paraffin-embedded de novo DLBCL; IH [n=46]. Nine MCRs were found to significantly associate (P ≤0.05) with poor outcome: gain of 3q, 9q, 19p, and loss of 1p, 8p, 9p, 13q, 15q, and 17p. Four remained significant after multiple testing correction: gain of chr3:187,651,865-196,853,350 (P Correlation of gene expression levels to copy number change was evaluated in the (GSE11318 [n=162]) dataset. A total of 397 differentially expressed genes (P ≤0.05, FDR) showing positive correlation with copy number status were identified within 16 MCRs of which, five mapped to three peak MCRs. Genes mapped to the peak MCRs were found to have roles in NFKB activation, iron transport, phosphatidylcholine biosynthesis and regulation of transcription all of which represent novel therapeutic targets in DLBCL. For the nine MCRs associated with poor outcome, 69 genes exhibited positive correlation with copy number and enrichment of the KEGG pathway using these genes in the DAVID 6.7 program identified the endocytosis, apoptosis, and glycosylphosphatidylinositol (GPI)-anchor biosynthesis pathways. In summary, using this platform-agnostic approach, common and novel loci of genomic imbalance in DLBCL were identified, of which some were found to have clinical significance and could be included, with additional validation, in patient risk assessment. This analysis also afforded the identification of novel genes with possible roles in lymphomagenesis, representing potential therapeutic targets in DLBCL. Disclosures: Dias: Cancer Genetics, Inc.: Employment. Thodima:Cancer Genetics, Inc.: Employment. Mendiratta:Cancer Genetics, Inc.: Employment. Asha:Cancer Genetics, Inc.: Employment. Feldstein:Memorial Slaon-Kettering Cancer Center: Employment. Chaganti:Cancer Genetics, Inc.: Consultancy, Equity Ownership, Patents & Royalties. Houldsworth:Cancer Genetics, Inc.: Employment, Equity Ownership.

Whole genome and exome sequencing studies have identified numerous genomic alterations in DLBCL, but these methods have limited applicability for the clinical care of lymphoma patients due to cost, specific tissue requirements, and laborious bioinformatic analysis. FoundationOne-Heme (FOH) is a novel next-generation sequencing platform designed to provide targeted assessment of the genomic landscape of hematologic malignancies, including identification of mutations within specific genes, copy number changes, and translocations. FOH can be performed on small quantities of formalin-fixed paraffin-embedded (FFPE) tissue, detect rare variants due to extensive depth of sequencing coverage, and rapidly provide results via streamlined bioinformatic interpretation. Here we report the first experience using this novel platform to evaluate the genetic landscape of DLBCL. Genomic DNA and total RNA were isolated from FFPE tissue on a cohort of 53 cases of DLBCL, including de novo (n=30), relapsed/refractory (n=12), and large cell transformation from low-grade lymphoma (n=11). The cohort included 25 cases with combined MYC and BCL2 overexpression by IHC (criteria for positivity: >40% MYC, >70% BCL2), of which only one had a known translocation involving MYC. Adaptor ligated sequencing libraries were captured by solution hybridization using two custom bait sets targeting 374 cancer-related genes and 24 genes frequently rearranged for DNA-seq, and 258 frequently-rearranged genes for RNA-seq. All captured libraries were sequenced to high depth (Illumina HiSeq), averaging >658x for DNA and >20,000,000 total pairs for RNA, to enable the sensitive and specific detection of genomic alterations. Significant non-synonomous variants were identified as mutations from the COSMIC database, amplifications of established oncogenes, or homozygous deletions and/or clear loss-of-function mutations of known tumor suppressors. The DNA sequencing component of FOH detected translocations in BCL2, BCL6, and MYC, while the RNA sequencing component detected fusion transcripts involving BLC6 and MYC, in agreement with independent cytogenetic analysis via karyotype and FISH where available. The assay detected copy number alterations of 44 different genes, most commonly amplification of REL (15%) or loss of CDKN2A/CDKN2B (17%). The most frequent alterations of known significance are detailed in Figure 1. The most commonly altered gene was CDKN2A, exhibiting either homozygous deletion or loss of function mutation in 28% of cases. Chromatin modifying factors (e.g. MLL2, CREBBP, EZH2) represented the most frequently altered biologic category with alterations occurring in >50% of cases. Recurrent alterations in components of the Notch pathway (NOTCH1/2/4, FBXW7, SPEN), each predicted to activate the pathway, were identified in 23% of cases. Cell-of-origin was determined as per the Hans model using IHC for CD10, BCL6, and IRF4/MUM1; CD79B mutations were detected exclusively in non-GCB and EZH2 mutations were found exclusively in GCB-phenotype cases. Furthermore, IHC MYC+/BCL2+ de novo DLBCL cases (n=11) exhibited more frequent hypermutation of PIM1 (46%) compared with the 19 cases of IHC MYC-/BCL2- de novo DLBCL (11%). When comparing the various clinical categories, we found that mutations in tumor suppressors were significantly more common in relapsed/refractory than de novo DLBCL (47% vs 75%, p=0.02). Alterations in TP53 were most frequently observed in transformed lymphoma (55%). Our results demonstrate the feasibility of using a targeted next-generation sequencing platform on FFPE clinical specimens from patients with DLBCL as a means of providing an integrated analysis of gene mutations, copy number alterations, and translocations. This streamlined approach combines multiple molecular and cytogenetic tests into a single platform and uses a small amount of tissue to perform a multifaceted assessment of genomic alterations with potential diagnostic, prognostic, and therapeutic implications. Future efforts will be directed at analyzing additional cases of DLBCL to better establish the biologic and clinical significance of the observed genetic alterations, and to prospectively incorporate this novel platform to select patients for mechanism-based targeted therapy. Disclosures: Intlekofer: Foundation Medicine, Inc: Consultancy. Levine:Foundation Medicine, Inc: Consultancy. Zelenetz:Foundation Medicine, Inc: Consultancy. Palomba:Foundation Medicine, Inc: Consultancy. van den Brink:Foundation Medicine, Inc: Consultancy. Brennan:Foundation Medicine, Inc: Employment. Young:Foundation Medicine, Inc: Employment. He:Foundation Medicine, Inc: Employment. Nahas:Foundation Medicine, Inc: Employment. Yelensky:Foundation Medicine, Inc: Employment. Otto:Foundation Medicine, Inc: Employment. Lipson:Foundation Medicine, Inc: Employment. Stephens:Foundation Medicine, Inc: Employment. Miller:Foundation Medicine, Inc: Employment. Younes:Foundation Medicine, Inc: Consultancy.

Background Although treatment with purine analogs is associated with a high response rate, hairy cell leukemia (HCL) remains incurable with a relapse rate of approximately 30%. For these patients (pts), novel therapies are needed. The recent major finding of the near universal and exclusive presence of BRAFV600E mutations in HCL has identified BRAF as a promising therapeutic target. BRAF inhibition in HCL may not only represent the first targeted therapeutic approach, but also a more effective strategy for treatment of HCL. Therefore, we have designed a phase II trial to determine the clinical efficacy of the BRAF inhibitor vemurafenib in pts with relapsed or refractory HCL. Patients and Methods Pts with HCL who are resistant to/intolerant of purine analogs, or who have achieved suboptimal response to purine analogs (i.e. relapse 22 years following completion of therapy, or who have ≥ 2 relapses) were enrolled to the trial. Pts must have an indication for treatment, as defined by ANC 21.0, HGB 210, PLT 2100K. Eligible pts received vemurafenib 960mg bid continuously in cycles of 4 weeks for 3 cycles. Bone marrow (BM) evaluations were performed after cycle 1 and 3. Pts with partial response (PR) or complete response (CR) with detectable minimal residual disease (MRD) were allowed to receive vemurafenib for up to 3 additional cycles at the discretion of treating physicians. Serial peripheral blood (PB) and/or BM samples were tested for BRAF IHC, quantitative BRAF mutant allele burden assessment by allele-specific qRT-PCR, serum cytokine analysis, detailed multiparameter flow cytometric analysis to identify the effect of vemurafenib on myeloid and lymphoid cell regeneration, and targeted next-generation sequencing analysis of a panel of 200 genes known to be mutated in cancer and/or associated with response to MAP kinase pathway inhibitors to detect potential predictors of response to vemurafenib and identify genes collaborating with BRAF mutations in HCL pathogenesis. Results To date, 9 pts have been enrolled and 8 pts received the treatment. The median age was 62 years (range 49-77 years), and the median number of treatments prior to the study was 3.5 (range 1-7), including 2 pts with splenectomy. 5 pts are evaluable for toxicity and disease response. The most common adverse events were rash (Grade 2: 4 pts), photosensitivity (Grade 1-3: 3 pts), arthralgia (Grade 2-3: 3 pts), hand-foot syndrome (Grade 2: 1 pt), febrile neutropenia (Grade 3: 1 pt) and tumor lysis syndrome (Grade 3: 1 pt). One pt developed new squamous cell carcinoma while on therapy and successfully underwent complete resection. 4 pts required dose reductions to 480mg bid due to side effects of arthralgia, symptomatic hand-foot syndrome and febrile neutropenia, but all were able to complete all intended treatment. All 5 pts achieved complete hematologic recovery. 2 pts achieved marrow CR (1 MRD positive and the other MRD negative), and 3 pts achieved marrow PR with very minimal disease. Responses were rapid and evident by PB flow cytometric analysis 24 hours after the initial dose (Figure 1). Plasma levels of sCD25 normalized within 2-3 weeks, and this correlated with a rapid decrease of BRAF+ hairy cells and dephosphorylation of ERK after one cycle (Figure 2). Several additional inflammatory cytokines correlated closely with the burden of leukemic cells identifying novel tumor markers in HCL including sTNF-R2, sIL-1R2, and sIL4R. Continued improvement in response rates was noted when assessed at the end of cycle 1 and 3, as documented by paired PB and BM multiparametric flow cytometric analysis. 2 pts with PR after cycle 1 converted to CR after cycle 3, and, in the remaining 3 pts who maintained PR, the amount of disease burden further decreased from cycle 1 to cycle 3. Genetic analysis is ongoing and has identified mutations in MLL2 and CREBBP, genes previously described to be mutated in other B cell malignancies but not previously described in HCL. Conclusions While longer follow-up is needed to assess the durability of the response, our results indicate that vemurafenib has potent antitumor activity in pts with relapsed/refractory BRAF mutant HCL and confirm the MAP kinase pathway as a rational promising therapeutic target in HCL. However, the optimal dose and duration of the drug in HCL remain unknown, and further studies are warranted. Enrollment is ongoing to this study, and updated clinical and genetic analysis data will be presented. Disclosures: No relevant conflicts of interest to declare.

A malignant lymphoma arising in the lung of a pediatric HIV-positive patient exhibited histologic and clinical features of low-grade B-cell lymphoma of mucosa-associated lymphoid tissue (MALT). Clinically, the neoplasm consisted of a 4-cm mass in the left-upper lobe of the lung of a 7-year-old girl. The lung mass was surgically resected. Monoclonal immunoglobulin heavy and light chain gene rearrangements were shown by Southern blot. Monoclonality of light chain expression was demonstrated by immunohistochemistry. Coexpression of Leu-22 (CD43) by the tumor cells supported the diagnosis of lymphoma. The remainder of the pulmonary parenchyma distal to the mass was associated with pulmonary lymphoid hyperplasia/lymphocytic interstitial pneumonitis, which may have been a predisposing factor. Gastric MALT lymphomas have recently been described in adult HIV-antibody-positive patients. Ours represents the first reported case of a pulmonary MALT lymphoma in a pediatric HIV-positive patient. In addition, at age 7, this is the youngest patient reported with a MALT lymphoma.

Abstract 509 Osteoblasts, the bone forming cells, are implicated in the fate of healthy and malignant stem cells. They affect self-renewal and expansion of hematopoietic stem cells (HSCs) and homing of tumor cells into the bone marrow. Here we show that constitutive activation of canonical Wnt signaling in osteoblast precursors disrupts hematopoiesis in mice by shifting the differentiation potential of HSC progenitors to the myeloid lineage which results in accumulation of granulocyte/monocyte progenitors and concomitant development of acute myeloid leukemia (AML). B-lymphopoiesis is also decreased. The AML phenotype is associated with clonal evolution at the cytogenetic level since clonal abnormalities could be detected in leukemic blasts from mice with constitutive activation of the canonical Wnt target β-catenin in osteoblast precursors (βcateninosb mice). Bone marrow transplantation experiments from βcateninosb mice to wild type lethally irradiated mice resulted in development of AML within 8 weeks following transplantation, demonstrating progression towards AML. At the molecular level, cell-specific gene inactivation mouse models demonstrate that β-catenin interacts with FoxO1 in osteoblasts to induce development of AML. Downstream signaling events that confer osteoblast signaling to normal HSCs and lead to their leukemogenic transformation will be presented. Importantly, malignancy-inducing osteoblasts, detected by nuclear accumulation of β-catenin in bone marrow biopsies, were identified in > 25% of patients with myelodysplasia (MDS), acute myeloid leukemia (AML) or AML arising from a prior MDS. Specifically, 15 out of 53 patients with MDS (n=17 patients), AML (n=20 patients), or MDS that had transformed to AML (n=16) chosen at random showed nuclear localization of β-catenin in osteoblasts. Of note, 12 of the 15 (80%) patients with nuclear localization of β-catenin in osteoblasts had abnormalities of chromosome 5 and/or 7, very common cytogenetic abnormalities in patients with MDS and AML. The same signaling pathways mediating AML development in βcateninosb mice were also found to be activated in osteoblasts and hematopoietic cells from the patients with nuclear accumulation of β-catenin in osteoblasts. These findings demonstrate that genetic alterations in osteoblast precursors (1) can induce AML in mice and (2) are associated with AML development in humans. They also provide a molecular basis for the leukemogenic transformation. Disclosures: No relevant conflicts of interest to declare.

Abstract 2706 Objective: Adult T-cell leukemia/lymphoma (ATLL) is a mature T-cell malignancy associated with human T-cell lymphotropic virus -1 (HTLV-1) infection. Four clinical subtypes are recognized including acute/lymphomatous (aggressive) or chronic/smouldering (indolent). For the aggressive subtypes the median overall survival is 6 months despite aggressive combination chemotherapy. Novel therapeutic targets against CD25, CD52, and NF μB have shown preliminary activity in ATLL. Recently, the chemokine receptor 4 (CCR4) antibody (KW-0761) was approved in Japan based on a phase II overall response rate of 50% and median overall survival of 13.7 months in 28 patients with relapsed/refractory ATLL. CD30 (Ki-1) is a marker of immune activation expressed on lymphocytes, and is an emerging target in lymphoma. The novel antibody-drug conjugate brentuximab vedotin targets CD30 and has shown efficacy in Hodgkin lymphoma and anaplastic large cell lymphoma (ALCL). While CD30 positive ATLL is infrequent in Japanese ATLL patients, ranging from 12.1% to 19.4%, the prevalence of CD30 positive ATLL outside of Japan has not been extensively studied. Patients and Methods: To investigate the prevalence of CD30 expression and its importance in ATLL outside of Japan we conducted a retrospective review of all patients who underwent treatment or consultation for ATLL at our institution between 1998 and 2012. Individual chart review was performed to report the clinical presentation, pathological features, and outcomes. Pathology review of CD30 expression status was conducted if specimens were available. The CD30 (Ber-H2) antibody from Ventana was used on automated platforms per manufacturer's instructions. Specimens were determined to be positive if the CD30 expression was >20% by an expert hematopathologist (AC). Results: We identified 50 patients with ATLL. Patient characteristics were: median age 54 (range 30–82); male:female 19:31; Caribbean-39, American-6, Western African-2, Peruvian-2, and Greek-1. Subtype of disease at diagnosis was: Acute/Lymphomatous—41 and Chronic/Smouldering—9. Thirty-six of 50 (72%) patients' samples had been tested at least once for CD30. Eighteen patients (50%) were reported positive. Of all cases reviewed (N=50), eight had pathology available for confirmatory review. Of those, five patients remained positive, 1 patient was reclassified as positive after initially being reported as negative, and two remained negative. The median overall survival of all patients was 13 months. In the subset who had CD30 status assessed the median overall survival for CD30 negative and positive ATLL was 18 and 45 months respectively (p=0.116). Conclusions: CD30 expression on normal lymphocytes is rare. Our retrospective data suggests that ATLL patients presenting outside of Japan may demonstrate CD30 positivity at a higher prevalence compared to Japanese reports. In pre-clinical studies of CD30 positive ATLL, SGN-35 has been shown in vitro and in vivo to demonstrate similar cytotoxicity compared to ALCL (Maeda et al. Cancer Sci 2010). Novel strategies are clearly needed to improve the outcome of ATLL despite recent advances. Therefore, CD30 testing may be considered in ATLL as a novel target for future research strategies. Disclosures: Matasar: GSK: Research Funding; Genentech: Consultancy. Straus:Seattle Genetics: Consultancy. Zelenetz:GSK: Consultancy, Research Funding; Genentech/Roche: Consultancy, Research Funding; Gilead: Consultancy; Sanofi Aventis: Consultancy; Cephalon: Consultancy; Celgene: Consultancy; Seattle Genetics: Consultancy; Amgen: Research Funding. Horwitz:Seattle Genetics: Consultancy, Research Funding; Allos: Consultancy, Research Funding; Celgene: Consultancy, Research Funding; Bristol-Myers Squibb: Consultancy; Genzyme: Consultancy; Kyowa Hakko Kirin Pharma: Consultancy; Johnson & Johnson: Consultancy; Infinity Pharmaceuticals, Inc.: Research Funding.

Abstract 1583 Background: Diffuse large B cell lymphoma (DLBCL) is the most common lymphoma with various subtypes. Recent data of Tumor Associated Macrophages (TAM) in DLBCL have shown association with poor overall survival. We were interested in assessing these markers, in particular, for practical daily clinical use at a high volume cancer center. We therefore performed a retrospective study to evaluate expression of macrophage markers (CD68 and CD163) in DLBCL and determine overall survival (OS) and event free survival (EFS). Design: A total of 128 DLBCL cases were reviewed. Tissue microarrays (TMA) were constructed with cores in triplicate. CD163 and CD68 expression was assessed by immunohistochemistry (IHC) using anti-CD68 (KP-1, Ventana) and anti-CD163 (Vector Laboratories) performed on the automated Ventana XT platform according to manufacturer protocols (Tucson, AZ). The assessment of intratumoral CD163 infiltration was performed at objective magnification 40X by three observers (OB, HC and JTF) and a consensus was obtained. The number of CD163 macrophages was counted in three cores. The mean was calculated and the results were scored as follows: 0 (0 to 25 positive cells); 1 (between 26 to 50 positive cells); 2 (51 to 75 positive cells); 3 (76 to 100 positive cells). CD68 expression was assessed by IHC and the degree of intratumoral infiltration was scored as follows: 0 (less than 5% positive cells); 1 (between 5% to less than 50% of positive cells); 2 (more than 50% positive cells). Of 128 patients, 13 had limited clinical data, 23 were biopsies at relapse, and 92 were evaluable for OS and EFS from initial therapy. EFS was defined as progression of disease, secondary malignancy, or death due to disease or any other reason. OS was calculated from date of death or last follow-up date to date of diagnostic biopsy. EFS was calculated from event date or date of last follow-up to end of treatment date. SPSS software was used to calculate Kaplan Meier survival curves and statistics. Results: There were 51 (55%) males and 41 (45%) females (M:F ratio 1.2). The median age was 60 years (range from 10 to 90). A high number of tumor infiltrating CD163 expression (scores 0 [n=19], 1 [n=6], 2 [n=9] vs 3 [n=58]) was associated with significantly poorer OS (p=0.028) and EFS (p=0.005). A statistically significant difference was observed in patients with tumors that had abundant tumor associated macrophages with CD163 (score 3 >75 positive cells) on IHC. CD68 expression did not correlate with OS and EFS. Conclusion: High CD163 expression is statistically significantly associated with decreased OS and EFS in DLBCL. Because of cleaner staining, scoring on a daily practical basis in a high volume center is useful. Our data supports initial few prior reports of this marker in DLBCL. Disclosures: No relevant conflicts of interest to declare.

Abstract 299 Angioimmunoblastic T-cell lymphoma (AITL) is a subtype of T-cell lymphoma that manifests with a broad array of neoplastic and paraneoplastic manifestations, including generalized lymphadenopathy, hepatosplenomegaly, systemic symptoms, rash and hypergammaglobulinemia. AITL accounts for approximately 20% of T-cell lymphomas and 1–2% of all non-Hodgkin lymphomas (NHL). Median survival among patients with AITL is less than 3 years, although the course of illness varies widely. Histologic examination of AITL characteristically demonstrates proliferation of high endothelial venules and a polymorphic infiltrate, malignant cells that resemble CD4+ follicular helper T-cells and occasional EBV-positive lymphoid cells. The genetics of AITL remain almost completely undefined, in part because of the low tumor cell fraction and the abundance of non-malignant bystander cells. We reasoned that targeted sequencing of selected genes of interest would increase coverage and thereby enable the detection of low frequency variants present with malignant AITL cells. To define a targeted mutational landscape of AITL, we performed exon capture and next-generation sequencing of all coding exons of 197 genes known to be recurrently altered in cases of NHL. Single nucleotide variants (SNVs) and insertions/deletions were confirmed by manual review of sequencing reads and a subset was validated by Sanger sequencing of tumor and germline tissue. Twenty-six cases collected from three sites in the USA had adequate tumor specimens available and resulted in high-quality reads (median depth of coverage∼300). Median age among the 26 cases was 66.5 years (range 30–89). Twelve (46%) patients were male. Among 21 cases with adequate clinical annotation, all 21 had advanced stage disease (11 stage 3, 10 stage 4), 13 (62%) had B-symptoms and 19 (90%) had an ECOG performance score of 0–2. Ten patients received CHOP-like regimens, 4 received non-anthracycline containing chemotherapy regimens, 5 received prednisone monotherapy, 2 received cyclosporine monotherapy, and one patient underwent allogeneic stem cell transplantation after cytoreduction with alemtuzumab. Among the 17 patients with staging after first-line treatment, 8 (47%) achieved a complete response, 3 (18%) achieved a partial response, 1 (6%) had stable disease and 5 (29%) had progression of disease. Median overall survival among the 26 patients was 420 days with a median follow-up of 426 days (range 21–3532). Recent studies have identified both TET2 and DNMT3A mutations in a small number of AITL samples as well as hematopoietic stem cells from the same patients. From the 26 cases, we identified 28 TET2 mutations (13 frameshift, 8 nonsense, 5 missense, 2 splice site) in 17 (65.4%) cases, including 5 cases with 2 TET2 mutations and 2 cases with 3 TET2 mutations. Among the latter, the fractions of mutant reads suggested that one mutation was present within all tumor cells and the other 2 mutations were present in subclones. Six of 17 cases with TET2 mutations also harbored mutations in DNMT3A, compared with 0 of 9 cases that lacked TET2 mutations (p=0.06). Median age for patients with wild-type TET2 was 56 years (range 30–83), 68 years for patients with one TET2 mutation (range 55–89), and 77 years for patients with 2 or 3 TET2 mutations (range 59–84; p Known gain-of-function mutations were identified at JAK3 V722I (n=2) and IDH2 R172K (n=1). Frame-shift mutations were recovered in ABCA7 (n=1), ARID1B (n=1), CREBBP (n=2), GNAS (n=1), IKZF2 (n=2), PRDM1 (n=2), SMARCA2 (n=1), and TNFRSF14 (n=1). Nonsense mutations were recovered in ALPK2 (n=1), CCND3 (n=2), and CHD8 (n=2). Only 2 cases harbored TP53 mutations (M133R, S261_splice). Additional missense variants were identified across a large number of genes, including several previously described in the COSMIC database. In conclusion, the genetics of AITL vary broadly across cases, with a small number of genes previously associated with NHL harboring recurrent mutations. Mutations of DNMT3A commonly occur in combination with one or more mutations in TET2. Mutation of either locus is associated with advanced age, supporting the hypothesis that the mutations are acquired over time within hematopoietic stem cells. Disclosures: Horwitz: Seattle Genetics: Consultancy, Research Funding; Allos: Consultancy, Research Funding; Celgene: Consultancy, Research Funding; Bristol-Myers Squibb: Consultancy; Genzyme: Consultancy; Kyowa Hakko Kirin Pharma: Consultancy; Johnson & Johnson: Consultancy; Infinity Pharmaceuticals, Inc.: Research Funding. Weinstock:Novartis: Consultancy, Research Funding.

Abstract 244 Clathrin assembly lymphoid myeloid leukemia (CALM) protein is implicated in clathrin dependent endocytosis (CDE) and the CALM gene is the target of the t(10;11)(p13;q14-21) CALM/AF10 translocation, which is observed in multiple types of acute leukemia. Although the translocation generally dictates poor prognosis, the molecular mechanisms by which the fusion protein exerts its oncogenic activity remains elusive. To determine the role of CALM and CDE in normal hematopoiesis and leukemogenesis, we generated and characterized both conventional (Calm+/−) and conditional (CalmF/FMx1Cre+) Calm knockout (KO) mutants. Furthermore, we determined the impact of Calm loss on leukemia cell growth in vitro and in vivo employing a series of leukemia cell lines and leukemia mouse models. Hematopoietic-specific Calm knockout mice (CalmF/FMx1Cre+) exhibited a hypocromatic anemia with increased serum iron levels. We observed significant reduction in mature erythroblasts/erythrocytes (TER119+CD71-) with concomitant increase in immature erythroblasts (TER119+CD71+) in the spleen of CalmF/FMx1Cre+ mice. The frequencies of erythroblasts in S phase were lower and the proportions of apoptotic (cleaved PARP positive) erythroblasts were increased in CalmF/FMx1Cre+ mice. Surface transferrin receptor 1 (Tfr1, CD71) levels were significantly up-regulated in Calm-deficient hematopoietic progenitors, and uptake of Alexa647-conjugated transferrin was abrogated in Calm-deficient erythroblasts, revealed by immunofluorescence analysis. Freez-etch electron microscopy analysis showed a defective clathrin coated vesicle (CCV) formation in Calm-deficient erythroblasts, indicating that Calm is indispensable for iron-bound transferrin internalization by regulating CCV formation, thereby critical for erythroid differentiation and hemoglobinization. CALM was highly expressed in leukemia/lymphoma cell lines and primary acute myeloid leukemia samples, although its expression was limited to erythroblasts in normal hematopoietic lineage cells. Treatment of leukemia cell lines with Desferoxamine (DFO), an iron chelator, led to a significant increase in Calm mRNA levels, suggesting that Calm expression is regulated by intracellular iron levels. Since highly proliferative leukemia cells demand iron as a cofactor for ribonucleotide reductase (RNR), we hypothesized that Calm is required for leukemia cell proliferation by regulating iron-bound transferrin internalization. To determine the effect of Calm inactivation in leukemia cells, we transduced a series of leukemia cell lines with a lentivirus-based ShRNA vector (pLKO-GFP), which allowed shRNA-expressing cells to be traced by green fluorescent protein (GFP). Calm shRNA transduced cells, but not cells transduced with scrambled shRNA, showed a proliferative disadvantage compared to non-transduced cells. To determine the effect of Calm deletion in leukemia cells in vivo, the CALM/AF10 oncogene was retrovirally transduced into either wild type (WT) or CalmF/FMx1Cre+ bone marrow (BM) cells and the cells were subsequently transferred to lethally-irradiated recipient mice. The Calm gene was deleted in donor cells via pIpC injections one month after transplant (before leukemia development) and survival curves generated. The recipients transplanted with the BM cells from CalmF/FMx1Cre+ mice showed a significantly delayed onset of leukemia and longer survivals compared to control (p=0.001), indicating that Calm is necessary for the development of CALM/AF10-induced leukemia. We next assessed whether Calm is required for the “maintenance” of leukemia in vivo. Leukemia cells were harvested from the primary recipients transplanted with the CALM/AF10-transduced CalmF/FMx1Cre+ BM cells (in which the endogenous Calm genes were intact) and transferred to the secondary recipients. The leukemic secondary recipient mice were then injected with pIpC and survival curves generated. Calm inactivation significantly delayed leukemia progression by blocking leukemia cell proliferation. Taken together, our data indicate that Calm is essential for erythroid development and leukemia cell proliferation by regulating TFR1 internalization. Since Calm inactivation significantly blocked the leukemia cell proliferation in vitro and in vivo, our findings may provide new therapeutic strategies for acute myeloid leukemia. Disclosures: Naoe: Kyowa-Hakko Kirin.: Research Funding; Dainipponn-Sumitomo Pharma.: Research Funding; Chugai Pharma.: Research Funding; Novartis Pharma.: Honoraria, Speakers Bureau; Zenyaku-Kogyo: Research Funding; Otsuka Pharma.: Research Funding.

Abstract 2630 Background: Recent evidence has shown that an increased number of tumor-associated macrophages is associated with decreased survival in patients with classic Hodgkin lymphoma. Methods: Eighty-one patients treated for relapsed and refractory classical Hodgkin lymphoma at Memorial Sloan-Kettering Cancer Center from 1995 to 2003 were identified. Patients received either standard ICE based chemotherapy or a risk stratified salvage therapy approach based on the number of pre-salvage therapy risk factors present (B symptoms, extranodal disease, and/or remission duration less than 1 year). Patients were also evaluated by functional imaging with gallium or fluorodeoxy glucose positron emission tomography (FDG-PET), as well as computed tomography (CT) prior to both salvage therapy and autologous stem cell transplant (ASCT). Patients with at least minimal response to salvage therapy proceeded to ASCT. Paraffin embedded tissue blocks were obtained from each patient. A tissue microarray (TMA) of tumor samples was constructed with triplicate 0.6mm tissue cores. A 4um section of the TMA was stained by immunohistochemistry with the anti-CD68 antibody (Ventana, CONFIRM anti-CD68 [KP-1]) on the Ventana Discovery XT instrument following manufacturerÕs protocol. Staining was scored based on the percentage of CD68 positive cells compared to the total number of cells in selected representative areas. The final percent of CD68 positivity for each case was based on the average of the cores available for examination. Scoring was performed independently by two individuals (CC and MA). Results: Scores for CD68 positive tumor associated macrophages ranged from 0–74%. Patients were grouped into two categories based on scores of 0–29%, and >30%. Forty three percent of patients (35/81) had scores of 29% or less. Fifty six percent (46/81) had scores of 30% or greater. In the intent to treat population, patients with relapsed and refractory Hodgkin lymphoma and CD68 scores > 30% had shortened overall survival (OS) compared with scores < 30%; 4.5 years versus 12.2 years, respectively (p=.048). Patients proceeding to salvage ASCT with CD68 scores > 30% also had inferior OS compared with patients who had scores < 30%; 4.69 years versus 12.7 years (p=.015). Increased levels of CD68 tumor-associated macrophages also impacted progression free survival (PFS) in patients undergoing salvage therapy and ASCT. CD68 levels > 30% were associated with an inferior PFS. Patients with CD68 scores < 30% had a median PFS that was not reached, compared with 1.1 years for patients with scores > 30% (p=.03). Conclusions: Pre-salvage therapy tumor biopsy specimens with elevated levels of CD68 positive tumor-associated macrophage were associated with poor outcomes. Scores > 30% conferred an inferior OS and PFS for patients with relapsed and refractory Hodgkin lymphoma undergoing salvage therapy and ASCT. We believe this score can be used to risk stratify salvage therapy in transplant eligible patients with relapsed and refractory Hodgkin lymphoma. In future studies we will correlate the scores of the initial tumor specimen with that of the relapsed tumor specimen. Disclosures: No relevant conflicts of interest to declare.

6626 Background: AML with t(8;16) represents a rare cytogenetic subtype of adult AML with distinctive morphologic and clinical features. It is associated with very poor prognosis and the molecular pathogenesis of AML is not understood. In several published reports, including the patients reported here, it was initially misdiagnosed and treated as acute promyelocytic leukemia (APL). Furthermore t(8;16) AML is reported to be more frequently found among therapy-related AMLs, which may make it more relevant to the patient population treated at tertiary cancer centers. Methods: Using a retrospective, multimodal diagnostic approach, we present a cohort of patients with t(8;16) AML, describing the clinical history, cytomorphology, immunophenotype and cytogenetics associated with this unique subclass of AML. Results: 14 patients Median age 51, 50% of patients have extramedullary manifestations, most commonly leukemia cutis, which seemed to precede systemic involvement. Cytomorphology from 7 patients showed blasts...

Abstract 119 The PI3K/AKT/mTOR pathway is frequently activated in lymphoma, but rapamycin-analog (rapalog) mTOR inhibitors have shown only modest benefits in clinical trials on lymphoma patients. To better understand the resistance of lymphomas to rapalogs, we have undertaken a study of the Pim family kinases, which signal in parallel to PI3K/AKT/mTOR and whose expression has been detected in multiple subtypes of non-Hodgkin lymphoma (NHL). The two pathways converge on activation of cap-dependent translation, suggesting new treatment possibilities by targeting this common downstream output. To assess the clinical relevance of Pim activity, we have quantified Pim1 and Pim2 expression in multiple NHL subtypes using tissue microarray (TMA) technology. We find common expression of Pim1, Pim2, or both proteins in diffuse large B-cell lymphoma (DLBCL, 65.5%), follicular lymphoma (58%), small lymphocytic lymphoma (76.5%) and mantel cell lymphoma (89.7%). Importantly, Kaplan-Meier survival analysis of clinical data linked to our DLBCL TMAs show a strong trend toward a worse overall survival when Pim expression is present in diagnostic tumor samples compared to Pim-negative tumors (p=0.0965). Studies in vivo demonstrate Pim's ability to accelerate oncogenesis in a manner similar to AKT in model systems specific to Burkitt's lymphoma (Eμ -Myc) and follicular lymphoma (VavP-Bcl2). Treatment studies in secondary recipient animals show that Pim promotes resistance to anthracycline chemotherapy like AKT. However, Pim tumors completely resist rapamycin in stark contrast to AKT tumors. To elaborate on these findings, we have employed a genetically defined system of rapamycin sensitivity lacking mTOR's upstream repressor TSC2. These studies show that Pim's ability to mediate rapamycin resistance is dependent on its ability to maintain inhibitory phosphorylation of the translation repressor 4EBP1. Specifically, a phosphorylation-site deficient mutant of 4EBP1 completely abrogates Pim's ability to maintain the viability of the TSC2 −/− cells. We have expanded on this finding using the drug silvestrol, which inhibits cap-dependent translation by targeting eIF4A. Silvestrol shows high potency against Pim-expressing TSC2 −/− cells (IC50 < 1 nM) and also against a panel of Pim-expressing lymphoma cell lines (IC50 1–10 nM). Indeed, targeting cap-dependent translation appeared more effective than the Pim kinase inhibitor SGI-1776 (IC50 1–10 μ M against lymphoma cell lines), which has significantly higher potency against Pim1 than Pim2. In conclusion, we have more clearly defined Pim kinase activity as a major mediator of oncogenesis in multiple NHL subtypes and as a likely negative prognostic marker in DLBCL. Our mechanistic and treatment studies provide a strong rational basis for targeting cap-dependent translation as a treatment strategy to bypass Pim activity and improve lymphoma patients' responses to both cytotoxic and rapalog therapies. Disclosures: No relevant conflicts of interest to declare.

Abstract 3111 Diffuse Large B cell lymphoma (DLBCL) is the most common form of non-Hodgkin's lymphoma. The Leukemia Lymphoma Molecular Profiling Project (LLMPP) collaboration has done extensive work regarding this disease including gene expression profiling to show that the cell of origin impacts the prognosis of the patient. Those cases which arise from an activated B cell (ABC) have a worse prognosis than those cases which arise from a germinal center B cell (GCB). We have used a subset of cases from our institution which have had gene expression profiling performed on frozen material and assigned ABC (n=18) or GCB (n=24) status (Rosenwald et al NEJM 2002, 346:1937). We have used matched formalin fixed paraffin embedded (FFPE) tissue block from these cases to obtain a microRNA profile utilizing the qNPA technology (High ThroughPut Genomics) as previously published (Roberts et al Lab Invest 2007, 87:979). This novel FFPE based RNAase protection assay measured 688 human microRNAs. Each microRNA was represented twice on the array and the values averaged for a signal value. The data were normalized to the total signal of the microarray. We were able to define a signature of increased microRNA for each DLBCL subtype as shown in Table 1. ABC subtype GCB subtype hsa-miR-155 hsa-miR-28-5p hsa-miR-196a hsa-miR-138 hsa-miR-501-3p hsa-miR-151-3p hsa-miR-656 hsa-miR-151-5p hsa-miR-1247 hsa-miR-182 hsa-miR-210 hsa-miR-613 The hsa-miR-155 is of importance in lymphoma as it regulates the generation of immunoglobulin class-switch in plasma cells (Turner et al Immunity 2007, 27:847) and was previously identified as a characteristic of ABC cell lines (Lossos et al Blood 2009,113:3754). The hsa-miR-210 has been shown to modulate the MYC antagonist MNT which allows for MYC expression (Grandori et all Cell Cycle 2009, 8:2756). This microRNA profile provides an opportunity to further explore the role microRNA plays in lymphoma biology and in particular, in DLBCL subtype determination. The success of the technique in FFPE tissue holds promise for further microRNA studies on archival material. Disclosures: No relevant conflicts of interest to declare.

Abstract 4256 Clathrin assembly lymphoid myeloid leukemia protein (CALM, also known as PICALM) is ubiquitously expressed in mammalian cells and implicated in clathrin dependent endocytosis (CDE). The CALM gene is the target of the t(10;11)(p13;q14-21) translocation, which is rare, but recurrently observed mutation in multiple types of acute leukemia. While the resultant CALM/AF10 fusion gene could act as an oncogene in vitro and in vivo in animal models, molecular mechanisms by which the fusion protein exerts its oncogenic activity remains elusive. Since CDE is implicated in the regulation of growth factor/cytokine signals, we hypothesized that the CALM/AF10 fusion oncoprotein could affect normal Calm function, leading to leukemogenesis. To determine the role of CALM and CDE in normal hematopoiesis, we generated and characterized both conventional (Calm+/−) and conditional (CalmF/F Mx1Cre+) Calm knockout mutants. While we didn't observe a gross defect in the heterozygous mutant (Calm+/−), homozygous deletion of the Calm gene (Calm-/-) resulted in late embryonic lethality. Total numbers of fetal liver (FL) cells were significantly reduced in Calm-/-embryos compared to that of control due to inefficient erythropoiesis. Proportions of mature erythroblasts (CD71-Ter119+) in FL were significantly reduced in the absence of the Calm gene. Furthermore, Calm deficient Megakaryocyte-Erythroid Progenitors (MEPs) gave rise to less CFU-E colonies when seeded in methyl cellulose plates, suggesting that Calm is required for terminal erythroid differentiation in a cell autonomous manner. To determine the role of Calm in adult hematopoiesis, we analyzed peripheral blood (PB), bone marrow (BM) and spleen of CalmF/F Mx1Cre+ mice after pIpC injection. CalmF/F Mx1Cre+ mice demonstrated hypochromic anemia, T-lymphocytopenia and thrombocytosis one month after pIpC injection. Levels of plasma transferrin and ferritin were intact in CalmF/F Mx1Cre+ mice, while plasma iron levels were increased, indicating that iron uptake is impaired in Calm deficient erythroblasts. We observed significant reduction of mature erythroblasts and erythrocytes in both BM and spleen with concomitant increase of immature erythroblasts (CD71+Ter119+) in CalmF/F Mx1Cre+ mice. The increased population mainly consists of CD71+Ter119+CD44+FSCdim polychromatophilic erythroblasts, and Benzidine staining of PB and splenic erythroblasts revealed reduced hemoglobinization in Calm deficient erythroblasts. To examine the global changes in transcriptome of CD71+Ter119+CD44+FSCdim polychromatophilic erythroblasts with or without the Calm gene, we compared mRNA expression profile by gene chip microarray analysis. Over 400 genes, including genes associated with iron metabolism and CDE pathway, were up- or down-regulated more than 1.5-fold in Calm deficient polychromatophilic erythroblasts as compared to control. Genes Set Enrichment Analysis (GSEA) revealed that multiple metabolic pathways were downregulated in Calm deficient polychromatophilic erythroblasts. Calm deficient CD71+Ter119+CD44+FSCdim polychromatophilic erythroblasts demonstrated a defect in cellular proliferation revealed by cell cycle analysis. Transferrin receptor 1 (TFR1, CD71) is highly expressed in rapidly dividing cells and erythroblasts, and uptake of iron-bound transferrin through TFR1 is the main pathway of iron intake to erythroid precursors. Since CDE is implicated in TFR1 endocytosis, we next examined surface expression levels of CD71 in Calm deficient erythroid progenitors and erythroblasts. While CD71 is normally expressed at low level in early stage of megakaryo/erythroid progenitors and highly expressed in CFU-E through polychromatophilic erythroblasts, its expression was dramatically up-regulated throughout the erythroid development in CalmF/F Mx1Cre+ mice. Up-regulation of surface CD71 expression was also evident in K562 erythroid leukemia cell lines upon ShRNA-mediated CALM knockdown. Taken together, our data indicate that CALM plays an essential role in terminal erythroid differentiation via regulating TFR1 endocytosis. Since iron is required for both erythroblast proliferation and hemoglobinization, Calm deficiency significantly impacts erythroid development at multiple levels. Disclosures: Naoe: Chugai Pharm. Co.: Research Funding; Zenyaku-Kogyo Co.: Research Funding; Kyowa-Kirin Co.: Research Funding; Dainippon-Sumitomo Pharm. Co.: Research Funding; Novartis Pharm. Co.: Research Funding; Janssen Pharm. Co.: Research Funding.

Abstract 1961 Poster Board I-984 LRF (Leukemia/Lymphoma Related Factor) is a transcriptional repressor originally identified as an interaction partner of the oncoprotein BCL6 (B cell Lymphoma 6). We previously found that LRF acts as a proto-oncogene by repressing tumor suppressor ARF (Alternative Reading Frame, also known as p19 in mice and p14 in humans) and is highly expressed in 60-80% of human Non-Hodgkin Lymphoma (NHL) cases (Maeda et al., Nature 2005). LRF was also found to be indispensable for hematopoietic stem cells (HSCs) to commit to the B cell lineage by opposing Notch function (Maeda et al., Science 2007). Considering that: 1) LRF is normally expressed in Germinal Center B cells (GCB) and overexpressed in NHL tissues and 2) LRF opposes Notch function to maintain normal B cell fate at HSC/progenitor levels, we explored the role of LRF in B cell development and its functional interaction with the Notch pathway in vivo. Upon T cell dependent (TD) immunization, GC formation was severely impaired in secondary lymphoid organs of B cell specific LRF conditional knockout mice (LRFflox/flox mb1-Cre+). While a GC reaction was robustly induced in control mice upon immunization, only few GCB cells were noted in secondary lymphoid organs of LRFflox/flox mb1-Cre+ mice. To assess functional significance of LRF loss in antigen response in vivo, titers of class-switched immunoglobulin (Ig) were measured in the serum; baseline serum titers of IgG1, IgG2b and IgG3 were perturbed, and the primary and secondary antibody response against the TD antigen was impaired in LRFflox/flox mb1-Cre+ mice. Absolute numbers of memory B cells and long-lived BM plasma cells were reduced in LRFflox/flox mb1-Cre+ mice 20 wk after immunization. To determine the cause of defective GC formation, apoptosis and proliferation of GCB cells were examined by FACS. While proportions of apoptotic (AnnexinV positive) GCB cells were similar, regardless of genotypes, LRF deficient GCB cells failed to proliferate upon antigen stimuli. Short-term kinetic analysis demonstrated 5-ethynyl-2'-deoxyuridine (EdU) incorporation was markedly decreased in LRF deficient GCB cells and that the proportion of GCB cells in S phase was reduced in LRFflox/flox mb1-Cre+ mice. In agreement with these findings, quantitative RT-PCR analysis in FACS-sorted GCB cells demonstrated up-regulation of p19Arf and p21, but not p53, mRNA levels in LRF deficient GCB cells. Up-regulation of p19Arf protein levels was also observed in Western Blots. Furthermore, microarray analysis and subsequent Gene Set Enrichment Analysis in FACS-sorted GCB cells showed signatures of defective proliferation, further implicating a critical role for LRF in GCB cell proliferation. Signals mediated by Notch2 are necessary for transitional B cells to commit to the marginal zone B cells (MZB). Inactivation of a component of the Notch pathway in mice resulted in no MZB development and increased follicular B cells (FOB). On the contrary, deletion of the MINT/SHARP gene, a suppressor of Notch signaling, lead to increase of MZB cells and concomitant reduction of FOB cells, indicating that Notch induces MZB cell fate at the transitional B cell stage. While B cell development in the BM was grossly normal, a reduction of FOB cells and a concomitant increase of MZB cells were observed in LRFflox/flox mb1-Cre+ mice. Since the phenotype was reminiscent of that seen in MINT/SHARP knockout mice and opposite to that observed in Notch2 knockout mice, we hypothesized that LRF antagonizes Notch2 mediated signal during the FOB vs. MZB fate determination process. To test this, LRF/Notch2 double knockout mice (LRFflox/flox Notch2flox/flox mb1-Cre+) were established and their mature B cell compartments analyzed. As expected, loss of the Notch2 gene led to an increase of FOB cells and decrease of MZB in LRFflox/flox mb1-Cre+ mice, suggesting that LRF regulates FOB vs. MZB fate in a Notch2 dependent manner. However, Notch2 deficiency did not restore GC formation in LRFflox/flox mb1-Cre+ mice. In summary, our genetic studies strongly indicate that the proto-oncogene LRF is required for normal mature B cell development and function via distinct mechanisms. We propose that LRF is necessary for mature B cell fate by blocking Notch2-mediated signals and plays a critical role in GCB cell proliferation via suppressing p19Arf mediated cell cycle arrests. Our findings provide a further rational for targeting LRF for the treatment of B cell malignancies as well as autoimmune diseases. Disclosures: No relevant conflicts of interest to declare.

e19522 Background: Gene expression studies have demonstrated that the proliferation signature is the most important factor in determination of outcome in MCL. However, visual assessment of immunohistochemical staining is not standardized. QIA of MIB-1/Ki67 staining may have a greater reliability for determination of the PI. We tested the agreement between manual counting (MC) and QIA and its prognostic value in patients with MCL. Methods: 105 patients with MCL underwent therapy for de novo MCL, of these, 66 had slides stained with MIB-1 for the analysis. The PI was determined in three ways: 1) estimation by a hematopathologist using a microscope (manual count, MC); 2) visual estimation by an independent hematopathologist after whole slide digitization; and 3) QIA software (color deconvolution algorithm). The survival analysis was restricted to 47 patients that were treated with sequential therapy followed by upfront consolidation with HDT/ASCR. Results: A median of 612,394 cells per slide were analyzed by QIA. There was a good correlation between the QIA and both the MC under the microscope and digital image (0.9061, P No significant financial relationships to disclose.

8095 Background: The IGF-1R (IGF receptor type 1) pathway is frequently deregulated in human tumors and has become a target of interest for anti-cancer therapy. We investigated predictive biomarkers of response to an anti-IGF-1R antibody (Ab) in vitro in NSCLC. Methods: We examined the growth inhibitory effects of R1507, a fully-humanized IgG1 anti-IGF-1R monoclonal Ab (Roche), against a panel of 22 NSCLC cell lines using CellTiter Blue assays. Phospho-receptor tyrosine kinase (pRTK) arrays and ELISAs were used to determine the status of IGF-1R and other RTKs. SNP arrays were used to determine IGF-1R copy number. Immunohistochemical (IHC) staining of total IGF-1R was performed with G11 (an anti-total IGF-1R Ab; Ventana) on a tissue microarray (TMA) containing 77 independent NSCLC tumor samples. Staining intensity was scored on a scale of 0 to 3+ by a pathologist (JF). Results: 5 of 22 NSCLC cell lines were moderately sensitive (25–50% growth inhibition) to R1507 alone. ELISA and pRTK array analysis demonstrated that pIGF-1R levels in the presence or absence of serum did not correlate with drug sensitivity. However, 4 of 5 sensitive lines displayed high levels of total IGF-1R vs 1/17 resistant lines (p=0.003 Fisher's Exact). SNP array analysis showed that sensitive lines also harbor relatively higher copy numbers of IGF-1R. There was no correlation with EGFR/KRAS mutational status. 48% of TMA NSCLC tumors had scores of 2+ or greater, while 5% were scored as 3+. Addition of erlotinib or paclitaxel to R1507 led to further growth inhibition in sensitive but not resistant lines. In one EGFR mutant lung adenocarcinoma cell line, R1507 and erlotinib co-treatment induced apoptosis, whereas treatment with either drug alone induced only cell cycle arrest. Apoptosis was mediated, in part, by the survival-related AKT pathway, as pAKT was significantly downregulated by R1507 but not erlotinib. Conclusions: In NSCLC cell lines, high levels of total IGF-1R are associated with moderate sensitivity to R1507. These results suggest a possible enrichment strategy for clinical trials with anti-IGF-1R therapy. [Table: see text]

Introduction: Optimal management of mantle cell lymphoma (MCL) remains undefined. However, recent data support the conclusion that upfront consolidation with high dose therapy with autologous stem cell rescue (HDT/ASCR) can improve the PFS (Dreyling, et al., 2005; Geisler, et al., 2008) and possibly OS (Geisler, et al. 2008). Based on gene expression profiling data, the proliferation signature has emerged as a critical determinant of prognosis in MCL (Rosenwald, et al., 2003). Immunohistochemical assessment of proliferation as estimated by MIB-1 (Ki-67) staining has been shown to predict outcome in patients with MCL treated with chemotherapy or immunochemotherapy. With conventional chemotherapy, the reported MIB-1 staining cutoff associated with poor prognosis reportedly varies from 10–30%. Since 1994 we have treated patients (pts) with MCL with sequential chemotherapy followed by HDT/ASCR. The current analysis is a retrospective review of our treatment program focused on the impact of upfront consolidation with HDT/ASCR on the prognostic significance of proliferation as measured by MIB-1 immunohistochemistry. Methods: Seventy-nine patients underwent upfront consolidation with HDT/ASCR for MCL. Fifty-two patients had evaluation of proliferation by MIB-1 immunohistochemistry. MIB-1 expression was estimated visually and assigned a percentage. Patients were binned into quintiles of MIB-1 expression: 0–20% (n=32, 61.5%); 21–40% (n=5, 9.6%); 41–60% (n=5, 9.6%); 61–80% (n=5, 9.6%); and 81–100% (n=5, 9.6%). Outcomes were analyzed by the method of Kaplan-Meier and comparisons were by log-rank. Results: The EFS and OS at 5 years was 65.1% and 61.4% respectively for the subset of patients with available MIB-1 staining (n=52); the outcomes for the entire group (n=79) did not differ from the subset. Outcomes (PFS, OS) were evaluated by MIB-1 quintile with successive cutoff values of 20%, 40%, 60% and 80%. MIB-1 cutoff values of 40% or less were not predictive of outcome. However, when the cutoff was 60% or 80%, both the PFS and OS were significantly worse for the high proliferation group. Nineteen percent of patients had a proliferation fraction of greater than 60%. For patients with MIB-1 staining of >60% the PFS was 4.2 years and was not reached for the group with staining ≤60% (p=0.004). Similarly, overall survival was significantly different, (p=0.031). Conclusions: These findings suggest that upfront consolidation with HDT/ASCR can partially overcome the adverse impact of proliferation on outcome and the adverse outcomes are seen in a subgroup of approximately 20% of patients with MIB-1 staining of greater than 60%.

PML was discovered as a gene frequently involved in the t(15:17) translocation of acute promyelocytic leukemia (APL). Since then, consequences of PML-loss of function in APL and other solid tumors have been extensively studied. Numerous pathways influenced by this tumor suppressor, including the PI3K pathway, have been identified but their importance for the pathogenesis APL is still unclear. We have now discovered that PTEN, the main tumor suppressor opposing PI3K signaling, is aberrantly delocalized to the cytoplasm in APL, but not in AML. Moreover, we show that drugs such as arsenic trioxide and retinoic acid, currently used for the treatment of APL and known to target the oncogenic PML-RARalpha fusion protein, rescue the nuclear localization of PTEN. This discovery has critical biological implications since nuclear localization of PTEN is central to its tumor suppressive activity. Utilizing an APL cell line, genetically engineered cells and transgenic mice, we have delineated the mechanism by which PML sustains nuclear PTEN. PML opposes the activity of a novel PTEN-deubiquitinating enzyme, HAUSP, towards PTEN through a framework involving the adaptor protein DAXX (death domain-associated protein). We demonstrate that PML-RARalpha translocation results in overactive HAUSP and as a consequence, nuclear-excluded PTEN. In support of this notion, we show that HAUSP is overexpressed in human cancer, (e.g. in prostate cancer) which is in turn associated with a nuclear exclusion phenotype of PTEN. Overall our results describe a novel role for PML in the maintenance of PTEN function that is perturbed by oncogenic cues in human leukemia and solid tumors. Significantly, our findings open new avenues for the design of targeted therapies focused on potentiating the activity of the PML-PTEN tumor suppressive network.

The proto-oncogene LRF, encoded by the Zbtb7a gene, is a transcriptional repressor that belongs to the POK (POZ/BTB and KrŸppel) protein family. Along with its oncogenic property, recent evidence has shown that POK proteins play distinct roles in hematopoiesis and immune system development. Conditional inactivation of the LRF gene in mouse hematopoietic stem cells (HSCs) results in the development of CD4/8 double positive (DP) T cells in bone marrow (BM) at the expense of B cell development (Maeda et al. Science 2007). While LRF acts as a master regulator of B versus T lymphoid lineage fate decision by suppressing Notch-mediated signals, it is unclear as to which Notch genes LRF targets and whether LRF is required for the maintenance of HSCs per se. To address these questions, we analyzed HSC/progenitor population of conditional LRF knockout mice (LRFF/FMx1-Cre) as well as LRF/Notch1 double conditional knockout mice (LRFF/FNotch1F/FMx1-Cre). In the absence of Notch1, LRF deficient HSCs/lymphoid progenitors (LRFF/FNotch1F/FMx1-Cre) could successfully give rise to early B cells (Pro B, Pre B and immature B). There were no abnormal DP-T cells seen in the BM, suggesting that LRF primarily targets Notch1 at the HSC/progenitor stages to maintain normal lymphoid development. However the loss of the LRF gene did not rescue the phenotype of Notch1F/FMx1-Cre mice (Radtke et al. Immunity 1999). Immature B cell development in the thymus was still observed in LRFF/FNotch1F/FMx1-Cre mice, suggesting that LRF acts genetically upstream of Notch1 during the early lymphocyte development. Notably, LRFF/FNotch1F/FMx1-Cre mice still exhibit a block of terminal erythroid differentiation and macrocytic anemia as seen in LRFF/FMx1-Cre mice. Thus, LRF is required for erythropoiesis via Notch-independent mechanisms. To further identify distinct HSC/progenitor compartments, we performed multicolor-FACS analysis utilizing antibodies for SLAM family members (CD41, CD48 and CD150), c-Kit, Sca-1, Flt3, IL7R-α, Vcam-1 and lineage markers (Lin). Remarkably, no Flt3 positive HSC/progenitors were observed in LRFF/FMx1-Cre mice. While IL7R-α+ T cell precursors (IL7Rα+Lin-Sca1+c-Kit+Flt3-), which were previously reported as common lymphoid progenitors (Maeda et al. Science 2007), existed abundantly. Absolute numbers of the long-term HSCs (LT-HSCs), defined as CD150+CD48-Flt3-Vcam-1+IL7Rα-LSK (Lin-Sca1+c-Kit+), were significantly reduced in LRFF/FMx1-Cre mice one month after pIpC injection. At the same time, CD150+CD48high+Flt3-Vcam-1-IL7Rα-LSK cells, which are likely T-committed lymphoid precursors, are increased in LRFF/FMx1-Cre mice. To investigate the presence of a population of quiescent HSC/progenitors, we treated LRFF/FMx1-Cre mice with 5-fluorouracil (5-FU), a S phase-specific cytotoxic chemotherapeutic agent, and examined recovery of HSCs in BM. LT-HSCs in LRFF/FMx1-Cre mice did not repopulate as many as their counterpart one month after 5-FU treatment. Our data indicates that LRF deficient HSCs are unable to maintain its quiescent status and are on the state of cell differentiation toward T cells due to the high Notch activity. In fact, loss of the Notch1 gene partially rescued reduced LT-HSCs numbers seen in LRFF/FMx1-Cre mice.

The cancer-testis antigens (CTA) are highly immunogenic antigens expressed in various tumors but not in normal tissues (except during gametogenesis), making them an attractive target for cancer immunotherapy. Expression of CTAs such as MAGE-A3, MAGE-C1 (CT7), MAGE-C2 (CT10), NY-ESO1 and the SSX antigens has been previously reported in multiple myeloma (MM). To date, however, these reports have included a heterogeneous group of newly diagnosed and relapsed/refractory patients, all in different stages of treatment. Therefore, the extent and prognostic significance of CTA expression, and of de novo immune responses against CTA in newly-diagnosed MM patients are not known. We now report on both CTA expression and antibody responses in MM patients at diagnosis and on their prognostic significance. From 8/00-11/04, we treated 67 newly-diagnosed, symptomatic patients with a thalidomide, doxorubicin, and dexamethasone-based induction regimen. (Brit J Haematol2006;132:155). Median age was 58; 54% were ISS stage I, 28% ISS II, and 18% ISS III. Nine of 63 tested (14%) had deletion 13q by FISH, while 24% had soft tissue involvement by MM. Responses to induction therapy included 10 (15%) CR, 16 (24%) VGPR, 26 (39%) PR, 6 (8%) stable or progressive disease, and 9 (13%) inevaluable. Post-induction 54 underwent autoSCT and 9 also underwent alloSCT.. Median overall survival (OS) has not been reached with 61% alive at median follow up of 65 months. Cryopreserved pre-treatment bone marrow plasma cells were used to assess CTA expression by RT-PCR. Pre- and post-treatment sera were used to assess antibody (Ab) responses against CTA proteins by ELISA. Fifty-two patients had sufficient RNA for PCR, and 46 had baseline serum for ELISA. OS of these groups did not differ significantly from the entire cohort. At least 1 CTA was expressed in 77% of cases, including MAGE-A3 (52%), SSX1 (40%), CT7 (29%), CT10 (25%), NY-ESO1 (21%), and SSX5 (17%). Three or more CTA were expressed in 29% of cases. Individually MAGE-A3 or NY-ESO1 expression at diagnosis conferred a poorer prognosis (MAGE-A3: median OS 66 mos. vs. not reached, p=0.02 by log-rank; NY-ESO1: median OS 65 mos. vs. not reached, p=0.09). These poorer outcomes were independent of ISS stage, presence of del 13q, or response to induction therapy. No other CTA was associated with an OS difference, nor was the total number of CTA expressed prognostically significant. Baseline Ab responses, all at titers > 1:1600, were noted to NY-ESO1 in 6/46 (13%) patients, 5 of whom also had Ab to the NY-ESO1 homologue LAGE-1. Ab responses were also noted to CT7 (n=2), CT10 (n=1) and SSX4 (n=1). No Ab responses were noted to MAGE-A3. The effect of induction therapy on antibody titers was inconsistent, with increases, decreases, and no changes seen. Interestingly, 2 of the 6 NY-ESO1 Ab+ patients had no NY-ESO1 expression in bone marrow plasma cells. Both, however, had extensive soft tissue (ST) plasmacytomas, suggesting another source of NY-ESO1 antigen. Presence of NY-ESO1 Ab correlated significantly with baseline ST involvement, with 67% of Ab+ patients having ST disease compared with 20% of Ab− patients (p=0.05). NY-ESO1 Ab+ patients also had significantly poorer OS (med 21 mos. vs. not reached, p=0.009), independent of other prognostic factors. In sum, CTA expression is frequent in newly diagnosed MM patients, and expression of MAGE-A3 or NY-ESO1 is associated with worse long-term survival. Spontaneous antibody responses against NY-ESO1 are seen in untreated patients, and are associated with ST involvement and poorer survival. Further exploration of biologic differences between CTA+ and CTA-MM, as well as immunotherapeutic strategies which target these antigens, are warranted.

LRF (Leukemia/Lymphoma Related Factor, also known as Pokemon, FBI-1, OCZF and ZBTB7a) was originally identified as an interaction partner of the oncoprotein BCL6. LRF can act as a proto-oncogene by repressing the tumor suppressor ARF and cooperates with BCL6 in MEF (mouse embryonic fibroblasts) immortalization. It is highly expressed in human Non-Hodgkin Lymphoma (NHL) cases, in the pathogenesis of which BCL6 is known to be involved (Maeda et al. Nature 2005). Inducible inactivation of the LRF gene in mouse Hematopoietic Stem Cells (HSCs) results in complete block of early B cell development at the HSC/progenitor stages and concomitant development of double positive (DP) T cells in the bone marrow (BM) (Maeda et al. Science 2007). While these findings clearly illustrate key roles of LRF in normal and malignant B cell development, it is not fully identified as to which B cell stages LRF is required during normal B cell development. To elucidate the role of LRF in B cells in vivo, we established and characterized B cell-specific LRF conditional knockout (KO) mice. We took advantage of mb-1 Cre knock-in mice, in which Cre expression is restricted to the B cells after the ProB cell stage. B cell compartments in the BM (PreProB, ProB, PreB and immatureB) are grossly normal in LRFF/ Fmb1-Cre mice. The LRF gene was efficiently eliminated in BM CD19+ B cells revealed by quantitative real-time PCR assay. Furthermore, LRF protein was not detected in purified CD19+ B cells, but seen in CD19-non-B cells, confirming the specific inactivation of the LRF gene in B cells. Thus, despite its critical role at the HSC/progenitor stages, LRF was found to be dispensable for the survival of normal BM B cells. These findings are consistent with the fact that GSI treatment (Maeda et al. Science 2007) or Notch1 loss (Lee and Maeda, unpublished) rescues the defects in early B cell development seen in LRFF/FMx1-Cre+ mice. Notch signaling is necessary for the transitional B cells to commit to the marginal zone B cells (MZB). Inactivation of the component of the Notch pathways in mice results in no MZB development. On the contrary, deletion of the MINT/SHARP gene, a suppressor of Notch signaling, leads to increase of MZB cells and concomitant reduction of follicular B (FOB) cells, indicating that Notch induces MZB cell fate at the transitional B cell stage. Given that LRF is a potent Notch suppressor at the HSC/progenitor stages, we hypothesized that LRF opposes Notch pathway in mature B cells as well. To test this hypothesis, we characterized mature B cell development in LRFF/Fmb1-Cre mice. While transitional B cells were largely unaffected in LRFF/Fmb1-Cre mice, we observed a slight but statistically significant reduction of follicular (FO) B cells (B220+CD19+AA4.1-CD1d-CD23+) and concomitant increase of MZB cells (B220+CD19+AA4.1-CD1d+CD23-) as seen in MINT/SHARP knockout mice. Thus, LRF may also oppose Notch pathways at the branching point for the FOB vs. MZB fate decision. Finally, to determine the role of LRF in Germinal Center (GC) formation in vivo, we characterized secondary lymphoid organs of LRFF/Fmb1-Cre mice after antigen stimulation. Both spleen and Peyer’s Patches were analyzed two weeks after immunization with Chicken Gamma Globulin (NP-CGG). While a GC reaction was robustly induced in control mice upon immunization, GC formation was significantly impaired in LRFF/Fmb1-Cre mice as revealed by immuno-histochemical analysis (IHC) and FACS. Only few GC cells (B220+CD19+FAS+CD38-PNA+) were observed in spleens, and the absolute numbers of GC cells were drastically reduced in LRFF/Fmb1-Cre mice. Residual LRF-deficient GC B cells were mostly negative for CXCR4, which is predominantly expressed in proliferating centroblasts within GCs, suggesting that LRF-deficient GC B cells may have defects in cellular proliferation in response to antigen stimuli. Our data indicates that LRF plays key roles in mature B cell development in the secondary lymphoid organs, but dispensable for the maintenance of early BM B cells.

Replying to: H. E. Gee, et al. , 10.1038/nature07362 (2008) Gee et al. contest that microRNA-10b is not a prognostic marker for metastasis risk in breast cancer1. However, their observations do not bear on the pro-metastatic roles of microRNA-10b (miR-10b) as we described them2, nor do they undermine our conclusions1.

The natural history of de novo cardiac amyloidosis is poorly described, a limitation that makes clinical decision-making difficult given the growing number of therapies for light-chain (AL)-amyloidosis (AL). We identified all patients presenting to our center from 3/99 to 8/07 at, or within 4 months of, diagnosis with symptomatic cardiac amyloidosis, and analyzed the baseline and post-treatment factors influencing survival for those with AL. During that period, 34.5% (157/455) of amyloid patients were diagnosed with de novo cardiac amyloidosis: 112 men and 45 women with a median age of 62 (31–83). Heart biopsies were obtained and were positive in 39% of men (44/112) and 31% of women (14/45). AL was diagnosed in 86% of patients (n=135) and hereditary (n=7) and senile cardiac (n=15) in the rest. Eight patients underwent heart transplant for hereditary (2 men) and AL (5 men, 1 woman). AL was diagnosed in 81% of men and 98% of women (p=0.005, chi square). We analyzed survival from diagnosis based on intention-to-treat. Ninety percent of patients with AL (n=122) received chemotherapy. Hematologic responses were scored as complete (CR), partial (PR, > 50% reduction) or non. Patients received IV melphalan with stem cell transplant (SCT, n=45), oral melphalan and dexamethasone (MDex, n=42) or other regimens (n = 35). Selection for SCT was based on age and extent of organ disease. Thirty-eight patients (28%) subsequently received second- and third-line therapies. The median survival for all patients was 10 months (range, 1 to 94). Baseline predictors of survival included age, gender and troponin I. Patients ≤ 60 years old had a median survival of 22 and those > 60 of 8 months (p=0.007). Women had a median survival of 24 and men of 8 months (p=0.02). Patients with troponin I ≤ 0.10 lived a median of 21 and those with levels > 0.10 of 9 months (p=0.01). Median ages at diagnosis of the 91 men and 44 women with AL were 59 (31–83) and 63 (38–82) respectively (p=0.19), and there were no differences in baseline albumin, alkaline phosphatase, CRP, brain natriuretic peptide, troponin I or clonal free light chains. There was a significant difference in serum creatinine (medians of 1.35 and 1.00 in men and women, p < 0.01). Overall, 60% of patients responded to chemotherapy: 55% of men and 74% of women (p=0.04, chi-square). Responders lived a median of 40 and non-responders 7 months (p

Background: The effective management of lymphoma depends upon an accurate and precise pathologic diagnosis. However, the classification of lymphoma continues to evolve. Reports addressing the role of second opinion expert pathology review have found varying impact, and little is known regarding the predictors of a change in diagnosis. Furthermore, the impact of the World Health Organization (WHO) classification of lymphomas over the 5 years following their formal publication has not been formally assessed. Methods: All outside pathology is reviewed at Memorial Sloan-Kettering Cancer Center (MSKCC) before a clinical opinion is finalized. We performed a chart review of all externally referred lymphoma cases from 1/1/01 to 6/30/01 and from 1/1/06 to 6/30/06 with second opinions from MSKCC hematopathology. Statistical analysis was performed using Chi-square or Fisher’s exact test for univariate analysis and logistic regression for multivariate analysis. Results: 719 patients (365 in 2001, 354 in 2006) met inclusion criteria. Diagnostic revisions were classified as major or minor; major changes were those that would lead to management changes as per National Comprehensive Cancer Network guidelines. 122 patients (18% in 2001, 16% in 2006) had a major diagnostic revision and an additional 22 (4% in 2001, 2% in 2006) had confirmation of major revisions rendered previously at second opinion from another National Cancer Institute Comprehensive Cancer Center (CCC). This did not change significantly by era, with 79 major revisions (22%) in 2001 and 65 (18%) in 2006 (P=NS). An additional 55 patients [24 (7%) in 2001, 31 (9%) in 2006] received minor revisions. Common categories of major revision included changing from nondiagnostic/ambiguous to definitive [6 in 2001, 8 in 2006], definitive to nondiagnostic [9 in 2001, 9 in 2006], malignant to benign [1 in 2001, 6 in 2006], indolent B-cell lymphoma (BCL) to aggressive BCL [15 in 2001, 8 in 2006], and aggressive BCL to indolent BCL [4 in 2001, 1 in 2006]. Major diagnostic revision was significantly associated with additional immunohistochemistry (IHC) testing in 2001 (OR=2.3; 95%CI 1.3, 4). In 2006, additional IHC (OR=1.8; 95%CI 1, 3.4), repeat biopsy (OR=3.1; 95%CI 1.2, 8.0), and skin biopsy (versus lymph node biopsy; OR 3.3; 95%CI 1.6, 7.0) were significantly associated with major revision. Two of the 7 patients reclassified as benign received revisions based on additional IHC, whereas 7 of the 14 patients reclassified as malignant were revised due to either additional IHC (4) or repeat biopsy (3). No effect was seen by biopsy type, nor were patient gender, age, race or ethnicity associated with odds of major revision. Of cases seen first at another CCC, 12% in 2001 and 16% in 2006 received major revisions, compared to 19% (2001) and 16% (2006) of other cases; these differences were not statistically significant. Conclusion: The rate of clinically meaningful diagnostic revisions at second opinion expert pathology review was high for patients seen at MSKCC, and remained so despite five years of increased familiarity with the WHO classification schema. These data confirm the fact that an appropriate evaluation, including detailed IHC and an adequate biopsy specimen, plays a central role in the accurate diagnosis of lymphoma. The high rates of diagnostic revision reported here lend support to the routine application of expert second opinion hematopathology review.

Background: Proteasome inhibitors (PI) affect a variety of intracellular processes, including the cell cycle, NF-kB and other signal transduction pathways, and the balance of bcl-2-related proteins, yet it remains unknown which of these targets are most crucial in determining response to PI treatment in non-Hodgkin Lymphoma (NHL). Responses to bortezomib (btz) have been seen in different NHL subtypes in several phase 2 studies. The consistency of response suggests that there may be biological characteristics that determine individual tumor responses to PIs. We used tissue microarray (TMA) technology to determine if pre-treatment tumor characteristics can predict outcome of btz treatment in NHL. Methods: TMAs were constructed using paraffin-embedded blocks of pre-treatment tissue from patients enrolled in a phase 2, CTEP-NCI sponsored trial of btz in NHL. Microsections were stained for 18 proteins with reported relevance to proteasome function and/or lymphoma prognosis. All stains were evaluated and scored by a central pathologist. Staining was correlated with quality and duration of response (DOR) or time to treatment failure (TTF). The best response was treated as a binary outcome (PR/CR vs SD/POD). Correlation between each marker and response was assessed by univariate analysis. The exact Wilcoxon rank-sum test was used to compare DOR and TTF with respect to the binary marker scores. Results: This analysis includes 35 of 75 patients treated with btz between 11/6/01 and 2/23/07 for whom pre-treatment tissue blocks could be obtained. Response data were available for 32 of these patients. Diagnoses include: FL (N=10), SLL (N=4), MCL (N=14) and MZL (N=4). Expression of the cell-cycle inhibitor p27 was more common in responding vs non-responding pts (p=0.024), and was also significantly associated with longer DOR (p=0.020) and longer TTF (p=0.021; figure). Presence of Bcl-6 was significantly associated with shorter DOR (p=0.049) and shorter TTF (p=0.048; figure), but not response. Conclusions: Using TMA analysis, we have shown that NHL patients with pre-treatment p27 staining were more likely to respond to btz, and had a longer DOR and TTF than those without baseline staining. Conversely, patients with baseline Bcl-6 expression had a shorter DOR and TTF than those without. Although the number of patients in this analysis is small, the data will help us generate hypotheses about disease and treatment biology that can be tested in larger studies. We are currently collecting samples from a broader pool of patients treated with btz at MSKCC to verify these results. TTF with bortezomib by (A) p27 and (B) Bcl-6 status. Figure. Figure.

Systemic light-chain amyloidosis (AL) is a protein deposition disorder and a monoclonal plasma cell disease. Treatment of AL has focused on the reduction or elimination of the clonal plasma cells that make amyloid-forming light chains. High-dose melphalan and stem cell transplant (MEL SCT) is an effective therapy for selected AL patients. At 3 months post-SCT, response of the clonal plasma cell disease can be reliably scored; the achievement of a complete response (CR= immunofixation negative) is usually associated with extended survival while no response (NR= < 50% reduction) is not. Many factors likely contribute to this distribution of responses. In order to identify factors specific to clonal plasma cells, gene expression profiles (GEP; Affymetrix U133 PLUS 2.0) were obtained using FACS-sorted CD138+/DAPI- plasma cells (>95% pure) from untreated AL patients before SCT. Supervised analyses were then performed based on responses at 3 months post-SCT, comparing the CR (n=4) and NR (n=5) groups. The basic analysis was a t-test, filtering genes for differential expression at p

LRF (Leukaemia/Lymphoma Related Factor, formerly described as Pokemon, FBI-1 and OCZF, encoded by the Zbtb7a gene) is a transcriptional repressor that belongs to the POK (POZ/BTB and Krüppel) protein family. We recently reported that LRF is a proto-oncogene, which is highly expressed in Non-Hodgkin’s Lymphoma tissues (Nature. 433, 278–85). To elucidate its function in fetal and adult lymphopoiesis, we analyzed LRF knockout mice (Zbtb7a−/−) and the conditional knockout mutants (Zbtb7a flox/−Mx-1cre+), respectively. Zbtb7a −/− mice are embryonic lethal due to severe anemia around 16.5 d.p.c. Absolute number of mature B cells was markedly decreased in 15.5 d.p.c Zbtb7a−/− fetal livers (FL), while total number of the earliest immature B cells (Lin-AA4.1+CD19−B220+) was comparable to wild type (WT) littermates. Flow-Sorted Zbtb7a−/− FL Hematopoietic Stem Cells (FL-HSCs) did not give rise to ProB cells in vitro in an OP9 cell culture system. Furthermore, competitive repopulation assay suggested that the defect in B cell development in Zbtb7a−/− FL was cell-autonomous. Conditional inactivation of LRF in adult mice resulted in a significant decrease of B220+ B cells in the peripheral blood (PB). Absolute numbers of both ProB and PreB cells in the BM were drastically reduced in Zbtb7a flox/−Mx-1cre+ mice after pIpC injections, while PreProB cells were rather accumulated. Zbtb7a flox/−Mx-1cre+ PreProB cells did not give rise to ProB cells in vitro in OP9 cell culture. Unexpectedly, Zbtb7a flox/−Mx-1cre+ PreProB cells ectopically expressed T cell genes (e.g. pTCRα, Notch1, Notch3, Hes1, Gata3, TCF1), while they lacked B-cell specific gene expression (e.g. E2A, Ebf1, Pax5, Rag, VpreB1). In agreement with this finding, Zbtb7a flox/−Mx-1cre+ PreProB cells efficiently differentiated into DP-T cells upon 6 days of culture on OP9-DL1 cells, which overexpress Notch1 ligand Delta-like1. We did not observe a gross defect in the T cell compartment in PB and Thymus of Zbtb7a flox/−Mx-1cre+ mice. However, we observed an accumulation of CD4/8 double positive (DP) T cells in their BM. DP-T cells consisted of nearly 30% of the BM mononuclear cells (MNCs) in Zbtb7a flox/−Mx-1cre+ mice. Since the phenotype of LRF conditional knockout mice is reminiscent of that of ICN1 (Intracellular domain of Notch1) overexpression in the mouse BM, we hypothesized that LRF could oppose Notch1 signaling pathway at the early lymphoid progenitor stage. We found that pTCRα, a Notch1 target gene, is highly up-regulated in Zbtb7a flox/−Mx-1cre+ CLPs and that LRF transcriptionally represses mouse pTCRα promoter activity. Our finding strongly indicates that LRF loss at the early lymphoid progenitor stage causes aberrant de-repression of T-cell specific genes, which results in the block of B cell development and generation of T cells in the BM. We therefore propose that LRF is essential in instructing the early lymphoid progenitors into B cell lineage by repressing T cell-instructive signal produced by Notch.

The PML tumor suppressor controls key pathways for growth suppression, induction of apoptosis and cellular senescence. PML loss occurs frequently in hematopoietic and solid tumors through unknown post-translational mechanisms. PML loss often correlates with tumor progression. Casein kinase 2 (CK2) is a stress activated serine/threonine protein kinase that is oncogenic and frequently over-expressed in human tumor of multiple histological origins including non Hodgkin’s lymphomas, acute leukemias and multiple myeloma. In addition, CK2 over-expression due to gene amplification has been reported to be a powerful adverse prognostic factor in non-small cell lung cancer. We recently reported that PML undergoes ubiquitin/proteasome mediated degradation in immortalized and tumor derived cell lines (Cell. 2006, 126:269). PML degradation depends on direct CK2 phosphorylation of Ser517 in the PML C-terminal degron. PML mutants that are resistant to CK2 phosphorylation display increased tumor suppressive functions in assays measuring apoptosis, replicative senescence and in xenograft models. Now, we report the identification and characterization of novel cellular and oncogenic stress signaling pathways that control the CK2/PML degradation pathway. This analysis allowed us to determine the signaling cascades upstream of CK2. In addition, we found an inverse correlation between CK2 kinase activity and PML protein levels in cancer-derived cell lines and primary specimens. Significantly, CK2 pharmacological inhibition enhances PML tumor suppressive property in vivo and in vitro. These data identify a key post-translational mechanism that controls PML protein levels and provide therapeutic means toward PML restoration through CK2 inhibition. The implications of these new findings during the physiologic response to cellular stress and in the pathogenesis of hemopoietic malignancies will be discussed at the meeting.

Systemic light-chain (AL) amyloidosis is both a protein deposition disorder and a monoclonal plasma cell dyscrasia. Effective treatment depends on reduction of the clonal plasma cells that produce the toxic light chains. Although high-dose melphalan with autologous stem cell transplant (MEL SCT) is effective for AL patients, the mobilization and collection of peripheral blood stem cells with G-CSF can be problematic despite their lack of prior chemotherapy and minimal marrow infiltration with plasma cells. In addition, unusual toxicities of G-CSF mobilization such as hypotension have been noted in AL patients. As new therapies show efficacy in AL, the role of stem cell mobilization and MEL SCT will likely be re-evaluated, increasing the importance of understanding the variability in stem cell mobilization. Recent work has shown that the mechanism of G-CSF mobilization likely involves the sympathetic nervous system (Cell2006;124:407). Interestingly, amyloid is well known to cause autonomic neuropathy. In addition, involvement of the heart in amyloidosis has also been shown to cause cardiac adrenergic denervation. If G-CSF mobilization depends upon the sympathetic nervous system, then the type of organ involvement could predict the yield of stem cell collection in AL patients mobilized with G-CSF. With these concerns in mind, we retrospectively assessed the results of stem cell mobilization and collection in all patients with AL mobilized with G-CSF (6ug/kg bid) for 4 days, with collections beginning on day 5 with a collection target of 10x106 CD34+ cells per kg in up to 4 leukaphereses. At initial evaluation, we assessed patients using standard criteria to identify the dominant organ of involvement (Am J Hematol2005;79:319) and to determine candidacy for risk-adapted MEL SCT (Blood2002;99:4276). Over a 7-year period 105 patients with AL were mobilized and collected as described, including 40 with kidney (K), 21 with liver/GI or soft-tissue (L/GI/ST), 31 with cardiac (H) and 13 with peripheral nervous system (PNS) involvement. We recorded the number of leukaphereses per patient and the dose of CD34+ cells/kg collected per leukapheresis. There were no significant differences in the number of leukaphereses per group. The K, L/GI/ST and H groups had a median of 2, and the PNS group of 3, collections. There were significant differences in CD34+ cells/kg collected based on dominant organ involvement: patients with H or PNS involvement had significantly fewer CD34+ cells/kg collected (see figure below; K vs H p=0.03, K vs PNS p=0.02, L/GI/ST vs H p=0.07, L/GI/ST vs PNS p=0.04; two tailed Mann-Whitney). Moreover, 15/44 H+PNS patients failed to collect at least 4x106 CD34+cells/kg compared to 7/61 K+L/GI/ST patients (p Figure Figure

14520 Background: Inactivation of the Pten tumor suppressor gene, leading to constitutive activation of the PI3K/AKT/mTOR pathway, is correlated with resistance to EGFR-targeted therapies. This trial tested the concept that inhibition of mTOR by RAD 001 (NOVARTIS) will restore sensitivity to EGFR inhibition by gefitinib (ASTRA ZENECA), in patients with progressive PC and GBM. Methods: Phase I was designed to determine a safe and tolerable dose and the pharmacokinetics (PK) of RAD (30/ 50/ 70 mg po weekly) with a fixed dose of gefitinib (250 mg po qd) in patients with PC and GBM. Phase II evaluated the proportion of PC patients with no change or a decline in PSA at 12 weeks, without clinical or radiographic progression. FDG PET imaging and immunohistochemistry were included as correlative studies. Results: 12 patients (2 GBM, 10 PC) were treated in Phase I, and 22 of 27 PC patients have been treated in Phase II. (The GBM Phase II results are reported separately). 20/32 (63%) of PC patients had received prior chemotherapy. No dose limiting toxicities were observed in Phase I, and 70 mg of RAD 001 weekly with gefitinib 250 mg qd was studied as the Phase II dose. PK parameters estimated during a 3 week single agent lead-in phase and during subsequent combined therapy suggested no clinically relevant PK interactions between RAD and gefitinib. The most common drug-related grade 3/4 toxicities were lymphopenia (28%) and elevated ALT (7%). Serial FDG PET scans showed > 25% decline in SUV uptake in 9/23 evaluable patients during the first week of treatment; of these, 3 patients showed no progression at 12 weeks. Overall, 5/29 PC patients showed no progression at 12 weeks, 3 (60%) of whom had received prior chemotherapy. Conclusions: Combination therapy with RAD 70 mg weekly and gefitinib 250 mg daily appears to be safe and is associated with modest activity in metastatic prostate cancer. Support: Novartis Pharmaceuticals, Astra Zeneca Pharmaceuticals. [Table: see text]

Autologous stem cell transplant (SCT) with high-dose melphalan is a standard therapy for selected patients with AL. Treatment-related mortality (TRM), however, remains 10% to 15%. Risk-adapted melphalan (MEL) aims to reduce TRM (Blood2002;99). Over the past 6 years we have used risk-adapted MEL at 200, 140 and 100 mg/m2. Dosing was based on age (MEL200 if ≤ 60, MEL140 if 61 to 71, MEL100 if > 71) for patients with no cardiac involvement, involvement of 1 or 2 major organ systems (of kidneys, liver/GI tract and peripheral nervous system) and creatinine clearance (CrCl) ≥ 51ml/min. Similarly, with cardiac involvement and/or CrCl < 51, MEL140 was used for those < 61 and MEL100 for those 61 to 71. Patients < 41 were treated with MEL200 as a rule. Patients with advanced involvement of the heart or ≥3 organ systems, or with cardiac involvement and age ≥ 71, were not SCT candidates. We followed TRM, defined as death during G-CSF mobilization or within 100 days of SCT, and also asked if survival at 3 months post-SCT was affected by achievement of complete hematologic response (CR) or by clonal Ig VL germline gene use. Of the 303 amyloid patients seen in that period, 94 were mobilized with G-CSF for SCT, 50M/44W with a median age of 57 (range, 32–73) at a median of 3 months from diagnosis (1–43). None had prior SCT and 71% (n=67) were untreated for plasma cell disease, including 45 who entered a clinical study employing adjuvant oral therapy after SCT. Fifty percent had dominant renal amyloid, 28% (n=27) cardiac, and 71% (n=67) single organ involvement. Clonal lambda disease was present in 83% (n=78) and serum free light-chain abnormalities in 95% (57/60). Patients were assigned to MEL200 (n=35), MEL140 (n=42) or MEL100 (n=17). There were 8 early deaths, 4 related to treatment (RTT), 1 in mobilization and 3 in SCT, and 4 related to progression of disease before SCT. All were associated with cardiac amyloid. Seven of the 8 were in the MEL100 group (RTT=3) and 1 was MEL140 (RTT=1). Overall TRM was 4.3% (4/94) with 3 in the MEL100 group. With a median follow-up of 22 months (1–77), survival is 81% and median survival has not been reached. From mobilization, non-cardiac patients had significantly better survival [88% (59/67) vs 63% (17/27), p

Introduction: Approximately twenty percent of patients with Hodgkin’s lymphoma (HL) relapse or have primary refractory disease. About 50% of these patients achieve long-term remissions after high-dose chemoradiotherapy and autologous stem cell transplantation (HDT/ASCT). At MSKCC, ICE (ifosfamide, carboplatin, etoposide) was incorporated as second-line chemotherapy prior to HDT/ASCT in a comprehensive treatment program. In addition to chemosensitive disease, a clinical prognostic model that emerged from this study identified 3 risk factors - B symptoms at relapse, extranodal disease, and complete remission duration of less than 1 year (Blood. 2001 Feb 1;97(3):616–23). This model was used to intensify treatment according to the number of risk factors, with stratification overcoming the significance of poor prognostic features (Blood. 2003 Nov 16;102(11), abstract #403). Methods: To further identify important prognostic factors, we evaluated pre-ICE biopsy specimens of patients enrolled on one of 3 IRB-approved clinical trials of HDT/ASCT. Prior studies showed that overexpression of bcl-2 and p53 have negative impact on outcome with primary therapy. We sought to determine if our comprehensive second-line program could overcome these poor prognostic features. We performed immunohistochemistry staining for bcl-2, bim (a bcl-2 family marker), and p53; samples were considered positive if any Reed-Sternberg (RS) cells stained for bcl-2 or bim, and if more than 50% stained for p53, at any staining intensity. Results: Seventy one patients had sufficient tissue available. Thirty five patients (49%) had disease progression and 28 (39%) died. Median PFS was 4.8 years, median OS was not reached, and median follow-up was 5.7 years. Bcl-2 was overexpressed in 19(27%), bim in 22 (32%), and p53 in 20 (29%) patients. Expression of bcl-2, bim, or p53 had no significant association with PFS or OS. Five-year PFS rates for positive vs. negative cases were 52.6% vs 50% for bcl-2, 54.5% vs 50% for bim, and 50% vs 51% for p53 (all p=NS). The 3 factor clinical model (B symptoms at relapse, extranodal disease and complete remission duration of less than 1 year) remained highly significant (0/1 vs 2/3 factors) for PFS and OS (p=0.002 and p=0.0003, respectively). Conclusion: Despite the evidence that p53 and bcl-2 overexpression may predict a worse prognosis with initial treatment, it appears that the approach of incorporating ICE and HDT/ASCT may overcome the significance of these biological markers at relapse. Further studies will focus on other pathways that are thought to play a role in relapsed/refractory HL outcomes. Bim is a novel pro-apoptotic marker from the bcl-2 family that is expressed on RS cells and suggests a role in the pathogenesis of HL. Future studies will focus on its role in both initial and relapsed/refractory setting.

Background: POKEMON (for POK, Erythroid, Myeloid ONtogenic factor) has recently been shown to be a novel proto-oncogene, playing a key role in cellular transformation and repression of ARF. Pokemon overexpression leads to overt oncogenic transformation both in vitro and in vivo in transgenic mice. Because these transgenic mice developed T-cell lymphoma we sought to expand our original screening analyses. We intially showed POKEMON expression in diffuse large B-cell lymphoma (DLBCL) and follicular lymphoma (FL) and co-expression of BCL6 and POKMEON showed higher proliferation and predicted for better overall survival in DLBCL (Nature 433, 278–285). Nodular lymphocyte predominant Hodgkin Lymphoma (NLPHL) is another tumor that expresses BCL6 and shows localization to L&H tumor cells. We therefore sought to compare POKEMON protein expression in NLPHL and Classical Hodgkin Lymphoma (cHL) cases. Design: Stains for POKEMON were performed with the anti-POKEMON hamster monoclonal antibody (clone 13E9). Sections of reactive human tonsil were stained as controls showing localization to squamous epithelium and reactive germinal centers as well as paracortical regions. For NLPHL, the cohort comprised of 19 males, 2 females; for cHL 16 males, 14 females. 20 NLPHL and 30 cHL tissue biopsies were stained using whole sections. Intensity and percent positivity of malignant L&H cells and Reed-Sternberg (RS) cells as well as surrounding reactive lymphocytes was scored. Expression in T-cell lymphomas was expanded to include a total of 84 cases analyzed by tissue microarray (TMA) that included: 8 ALCL (Anaplastic Large Cell Lymphoma), 11 AILT (Angioimmunoblastic T-cell Lymphoma), 17 T-ALL (T Lymphoblastic Lymphoma), 43 PTCL (Peripheral T-cell Lymphoma), 3 T/NK (Nasal type T/NK cell Lymphoma), and 2 PTCL, NOS (Not Otherwise Subclassified). Tumor cells in T-cell lymphomas and HL were scored and defined as 0 (negative), 1 (scattered 50%, strong positive) with a nuclear localization pattern. Results: For NLPHL, 18/20 (90%) cases showed diffuse strong positivity in >50% of malignant L&H tumor cells whereas in cHL cases, 6/30 (20%) cases showed diffuse strong positivity in >50% of malignant RS tumor cells for POKEMON protein expression. In T-cell lymphomas, POKEMON expression was strongest in ALCL (6/8, 75%) and AILD (7/11, 74%) compared to the other subtypes. In contrast, BCL6 protein expression was positive in 12/20 (60%) cases which showed weak, scattered to diffuse strong positivity in L&H tumor cells and 1/30 (3%) with scattered weak reactivity in cHL malignant RS cells Conclusion: POKEMON is co-expressed with BCL6 in malignant L&H tumor cells in NLPHL but not in cHL. Pokemon’s critical role in cellular transformation makes it an attractive target for therapeutic intervention in NLPHL.

Background: The diagnosis of amyloidosis is relevant in patients with monoclonal gammopathies (MG) and unexplained organ dysfunction, or in those with peripheral neuropathy or a family history of amyloidosis. Tissue diagnosis is essential to clinical evaluation but biopsy of involved viscera can be morbid and delay therapy. The abdominal fat pad aspirate (FPA) is a minimally invasive way to obtain a diagnostic specimen. The FPA is available for light microscopy (eg, Congo red staining, immunohistochemistry (IHC)) and electron microscopy (EM). At our center, we are referred 200 patients a year with MG as well as patients with neuropathy and family histories of amyloidosis. The FPA is performed by cytopathologists as part of the diagnostic evaluation on clinician request. All Congo red (CR) stained slides with controls are reviewed by a cytopathologist using polarized light and, on request, cases are also sent for transthyretin IHC or EM. Design: This study is a retrospective review of our experience with FPA in regards to sensitivity, specificity, and predictive values. A true positive FPA is when the FPA is CR+ and the diagnosis is confirmed by other biopsy or by non-invasive testing, while a false positive is CR+ but the diagnosis not confirmed. A true negative is when the FPA is CR- and the diagnosis not confirmed, while a false negative is CR- but the diagnosis confirmed. We reviewed 136 consecutive FPAs plus the clinical and pathologic reports for this study. Results: There were 133 patients (90M, 43W) with a median age of 61 years (range, 34–81); most were seen between 2000 and 2005. Of 109 CR+ FPA, 3 patients are still in evaluation and 106 have been evaluated. The diagnosis was confirmed in 85 including 63 with CR+ other biopsies and 22 with non-invasive testing and clinical confirmation (true positives), giving a positive predictive value of 80% (85/106). Of these, 75 had systemic AL, 7 had mutant transthyretin genes, 1 had secondary (AA) amyloid, 1 had a CR+ plasmacytoma associated with MG, and 1 (with CR+ kidney biopsy) could not be typed. In 21 patients the diagnosis was not confirmed (false positives): 12 not confirmed on non-invasive testing, 7 had CR- other biopsies, and 2 had negative EM. Of these 21, 19 had MG and 2 peripheral neuropathy without MG. There were 27 CR- FPA, all evaluable. Four had CR+ other biopsies (false negatives). Of the 23 true negatives, 2 had amyloid confined to soft-tissue plasmacytomas. The negative predictive value of FPA then is 85% (23/27). FPA were also analyzed by EM (n=8). Results were concordant with the CR stain (n=4 CR+, n=2 CR−), discordant with it (n=1 CR+ but EM negative) or were unsatisfactory (n=1). Conclusions: The sensitivity of FPA in this series is 95.5% (85/89) and the specificity 52.3% (23/44). The specificity is low because of the prevalence of amyloidosis in those suspected of it. Among those on whom FPA was performed, 80% had CR+ FPA, a rate much higher than at other centers. The utility of EM and organ biopsies as gold standards in equivocal cases is supported by our data; however, such approaches are not usually necessary given the sensitivity of the FPA. In sum, the FPA is a convenient and useful tool in the evaluation of patients suspected of amyloidosis.

BACKGROUND: The serum free light chains assay (FLC) has recently become commercially available for use in the management of multiple myeloma. This assay is more sensitive for the detection of monoclonal light-chains than previously available tests and has therefore found many unique applications. It can detect two-thirds of non-secretory multiple myelomas that were previously missed by immunofixation; it permits a more sensitive detection of monoclonal FLCs in AL amyloidosis; and it may allow earlier identification of disease recurrence in multiple myeloma and amyloidosis patients after treatment. We have sought to examine the usefulness of this assay as an early predictor of final response in patients with symptomatic multiple myeloma undergoing initial therapy. METHODS: We have performed an analysis of FLC assay in forty-five patients who were enrolled in an IRB approved phase II clinical trial whose main objectives were to determine the efficacy and toxicity of the administration of doxorubicin and dexamethasone, followed by thalidomide and dexamethasone in patients with untreated multiple myeloma (ASH 2004). The response to treatment was assessed based on the EORTC consensus criteria. Using Fisher’s exact test, we have evaluated the association between the status of the FLC ratio after cycle one and cycle two (i.e. normalized or persistently abnormal) and the status of response at the completion of 5 cycles of treatment, in patients with an abnormal FLC ratio at baseline (normal range 0.26 – 1.65). RESULTS: Of the 42 patients evaluable for response, 37 had an abnormal free light chain ratio at baseline (i.e. < 0.26 or > 1.65). Normalization of the FLC ratio after cycle one or cycle two occurred in seven out of 15 patients who achieved near complete remission (nCR) or complete remission (CR), and only in one of 22 patients who achieved partial remission (PR), stable disease (SD), or progression of disease (POD). Among the eight patients who had normalization of the FLC ratio after one or two cycles of treatment, seven achieved nCR or CR and one achieved PR. Normalization of the FLC ratio after one or two cycles of treatment was significantly associated with the achievement of CR or nCR (p=0.003). CONCLUSION: Normalization of the free light chain ratio after the first or the second cycle of treatment is highly predictive of achievement of nCR or CR at the completion of treatment. If this observation is confirmed in larger studies and for other regimens, the assessment of the free light chain ratio after two cycles may become an important milestone in the decision making for patients with multiple myeloma undergoing initial therapy. Since the aim of therapy is to achieve nCR or CR, the addition of alternative drugs might be advisable at an early stage of treatment in patients whose free light chain ratio remains abnormal.

In order to treat patients with symptomatic amyloidosis, the amyloidosis must be typed with confidence. Immunohistochemical techniques for light-chain isotype identification of amyloid are not reliable, and techniques to type fibrils extracted from clinical specimens are neither widely available nor validated. Retrospectively, investigators in the United Kingdom have shown that hereditary amyloidosis can be misdiagnosed as AL in 10% of cases because family history is often not relevant and because a monoclonal gammopathy (MG) and an hereditary variant can be present in the same patient (NEJM2002;346). The most common types of amyloidosis in the USA are AL and hereditary, the latter due to mutations in proteins such as transthyretin (TTR), fibrinogen A α-chain (FnAα), apolipoprotein AI or AII, and lysozyme. The genes for these proteins are not routinely examined in all cases of likely AL amyloidosis, and clinical genetics laboratories do not offer screening for all of them. The most common hereditary variant in the USA is the V122I transthyretin mutation found in 4% of African-Americans. In addition, a common presentation of both AL and hereditary amyloidosis is peripheral neuropathy. Furthermore, in the case of FnAα nephropathy, patients usually have isolated renal amyloid without marrow involvement. To minimize the risk of misdiagnosis of AL at our center, we prospectively tested four categories of patients with a tissue diagnosis of amyloidosis for hereditary variants whether or not MG was present: African-Americans, patients with dominant peripheral neuropathy (PN), patients with isolated renal amyloidosis (RA) without marrow involvement, and patients referred for hereditary testing or lack of MG. Testing was by PCR amplification and sequencing of genomic DNA. From 6/1/02 to 8/1/05, we evaluated 178 patients referred for amyloidosis, and 30% (n=54) were screened according to this algorithm: 20 African-Americans (16 with MG), 16 with PN (11 with MG), 7 with RA without marrow amyloid (all with MG), 7 referred for hereditary testing and 4 without MG. Of those with amyloidosis and monoclonal gammopathies, 6% (2/34) had both a monoclonal gammopathy and heterozygosity for a mutant TTR: a 45 year-old African-American man (V122I) with cardiac amyloid and a 59 year-old man (F64L) with polyneuropathy. Of the 9 African-American and PN patients without MG, 4 had mutant TTR. Of 7 sent for hereditary testing, 6 had mutant TTR, one of whom also was found to have undiagnosed stage I lambda light-chain myeloma. Of 4 without MG, 2 had senile cardiac amyloid, 1 AA and 1 had amyloid that could not be typed. For those with mutant TTR and MG, TTR tissue-staining resolved the type of amyloid. These results justify further study of screening for hereditary variants in patients with apparent AL, including those with AL associated with multiple myeloma. Myelotoxic therapies such as stem cell transplant are not appropriate treatments for patients with hereditary amyloidosis. These results also highlight the need for the development, validation and dissemination of reliable techniques for identifying fibrils extracted from tissue, and raise the question as to whether or not hereditary amyloid proteins increase the risk of developing MG.

Risk-adapted dosing of melphalan (MEL) has been proposed as a means of reducing treatment-related mortality (TRM) in patients with AL undergoing SCT but recent retrospective studies show poorer response rates in patients receiving < 200mg/m2 of MEL (Ann Intern Med2004;140; BMT2004;34). We have prospectively studied a combined approach using risk-adapted MEL followed by adjuvant D ± T with the goals of achieving low TRM and optimal hematologic and organ response rates. To be eligible, patients with systemic AL with ≤ 2 organ systems involved and no advanced cardiac disease had to be untreated and diagnosed within 12 months of enrollment. Patients received MEL200, 140, or 100 mg/m2 with autologous SCT based on age, renal function, and cardiac involvement (Blood2002;99). Those not achieving a hematologic complete response (CR) at 3 months post-SCT were treated with 9 months of D (20mg/m2, 1–3 pulses monthly) + T (50–200mg nightly), or only D if they had prior DVT or neuropathy. Forty-five patients (51% men) enrolled, a median of 57 years old (range 34–73) and 2 months from diagnosis. Dominant organ involvement was renal in 58% (n=26), cardiac in 24% (n=11), and liver/GI or peripheral nervous system in 18% (n=8). A third (n=15) had 2 organ systems involved. Twenty-one percent (8/38) had elevated serum troponin I levels (>0.06ng/ml) and 53% (19/36) had elevated serum brain natriuretic peptide levels (>100pg/ml) at baseline. Dose-assignments were MEL200 (n=15), MEL140 (n=24) and MEL100 (n=6). TRM was 4.4% (2/45), with 6 patients out of hospital and well < 100 days post-SCT. At 3 months post-SCT, 63% of patients had hematologic responses (8 CR, 15 PR). Persistent clonal plasma cell disease was found in 29 patients; 1 refused and 2 were too ill for D+T, 17 received D+T and 9 D. Ten completed 9 months of adjuvant therapy, with 13 discontinuing and 3 still on therapy. Reasons for discontinuation were progressive disease (n=4), DVT/PE (n=2), RSV pneumonia (n=2), femoral avascular necrosis (n=1), anxiety (n=1) and intolerance (n=3). Thirty-five percent (9/26) on adjuvant therapy had improvements in hematologic response at 12 mos post-SCT, including 4 PR to CR (2 D+T, 2 D) and 5 stable disease (SD) to PR (3 D+T, 2 D). The hematologic response rate at 12 months was 79% (22/28; 9 CR, 13 PR) with no significant difference based on dose of MEL. Forty-six percent (13/28) had organ responses at 12 months, while 32% had stable and 21% worsened organ function. Only patients who achieved a hematologic response at 12 months had organ responses (59% (13/22) of PR/CR patients vs. 0% (0/6) with stable or progressive disease, p=0.01). With a median follow-up of 18 months (range 1–33), overall survival is 76% and median survival is not yet reached. In summary, risk-adapted MEL has a low TRM and adjuvant D±T can improve hematologic response in a third of patients with persistent clonal plasma cell disease post-SCT, albeit with moderate toxicity. Further study of such combined approaches in AL SCT patients is warranted.

We recently reported a role of the POK transcription factor Pokemon in oncogenesis (Nature. 433, 278–85). To elucidate its function in fetal and adult hematopoiesis, we studied the hematopoiesis in Pokemon knockout mice (Pokemon−/−) and the conditional knockout mutants (Pokemonflox/flox) respectively. We generated the Pokemonflox/floxMx−1cre+ mice, in which Exon 2 of the Pokemon genomic locus can be excised in vivo upon pIpC (polyinosinic-polycytidylic ribonucleic acid) injections. Pokemon−/− mice died around 16.5 d.p.c due to severe anemia. Flow cytometric analyses (FACS) revealed a significant decrease of mature erythroblasts and profound increase of AnnexinV positive apoptotic TER119+CD71+ cells. Pokemon is therefore indispensable for definitive erythropoiesis by blocking apoptosis in late differentiation stage. Since the loss of the tumor suppressor Arf reverted the senescence phenotype of Pokemon−/− MEFs, we next examined whether the loss of Arf rescues the anemia phenotype of Pokemon−/− mice. Although Arf mRNA level is elevated in Pokemon−/− TER119+CD71+ erythroblasts, Pokemon−/−Arf−/− embryos were also lethal due to anemia, which indicates that the apoptosis is caused by Arf independent mechanism. Conditional inactivation of Pokemon upon pIpC injections in adult mice resulted in macrocytic anemia. FACS analyses of BM and spleen demonstrated a significant decrease of TER119SP cells. Spleen weights of Pokemonflox/floxMx−1cre+ mice were three times more than those of control mice due to the compensatory expansion of TER119+CD71+ cells. Conditional inactivation of Pokemon in adult mice also resulted in a significant decrease of B220+ B lymphocytes in peripheral blood (PB). Furthermore, total B220+ splenocytes were severely decreased and immunohistochemical analyses of spleens displayed the lack of mature germinal center formations. The proB cells were barely detectable, while PreProB cells were accumulated in Pokemonflox/floxMx−1cre+ mice BM. We therefore concluded that Pokemon plays a key role in early B cell development, especially at the transition stage from PreProB to ProB cells. Regarding T lymphocytes, the numbers of CD4/8 SP cells in Pokemonflox/floxMx−1cre+ PB were similar to those of control mice. No major immuno-phenotypical differences were observed in thymocytes. Two months after pIpC injections, Pokemonflox/floxMx−1cre+ mice displayed a moderate decrease of Gr1+Mac1+ myeloid lineage cells in PB and BM. However, in vitro colony forming assays, taking advantage of hematopoietic stem cells from Pokemonflox/floxMx−1cre+ mice, displayed a clear block of differentiation at the promyelocyte stage in CFU-GM. Furthermore, FACS analyses of progenitor populations showed the decrease of GMPs (granulocyte/macrophage progenitors) in Pokemonflox/floxMx−1cre+ mice. Our result indicates that the proto-oncogene Pokemon is a master regulator of both fetal and adult hematopoiesis. To further elucidate the mechanism, GeneChip microarray analyses were performed utilizing the RNA prepared from the flow-sorted TER119−CD71+ 12.5 d.p.c FL erythroblasts. Candidate target genes were found and the results were also confirmed by real-time PCR. The details of possible target genes will be discussed further at the meeting.

POKEMON (for POK, Erythroid, Myeloid ONtogenic factor, also known as LRF) was originally isolated as a homologue of PLZF (Promyelocytic Leukemia Zinc Finger) which can physically interact with the BCL6 (B Cell Lymphoma 6) proto-oncogene. Analysis of Pokemon−/− mice demonstrates that this transcriptional repressor plays a key developmental role whose function is required for cellular differentiation in multiple tissues such as skin, adipose tissue and hematopoietic lineage cells. Taking advantage of Pokemon−/− MEFs (Mouse Embryonic Fibroblasts), we found that Pokemon is indispensable for cellular transformation mediated by oncogenes (such as BCL6, E1A, H-RasV12, c-Myc, or T-Ag) and Pokemon overexpression overcomes oncogene induced premature senescence and apoptosis leading to oncogenic transformation by repressing p19ARF tumor suppressor. Furthermore, Pokemon−/− MEFs showed a premature senescence phenotype in early passage which was rescued by inactivation of p19ARF. To explore the oncogenic role of Pokemon in vivo, we generated Lck/Eμ Pokemon transgenic mice in which Pokemon is over expressed in T and B lineage cells under the control of Lck/Eμ promoter/enhancer. Strikingly, about 50% of the mice developed T lymphoblastic lymphoma (CD4+, CD8+, CD44 + and TdT+) after a latent period of 2 to 4 months. As Pokemon expression proved essential for cellular transformation and is oncogenic when overexpressed in vivo, we analyzed the levels of POKEMON expression in human cancers, performing Tissue Microarray Analysis (TMA). We first analyzed human B-cell lymphomas as POKEMON is normally co-expressed with BCL6 in the B-cell within the germinal center, and may therefore cooperate with BCL6 in lymphomagenesis. We analyzed and scored a total of 123 cases of diffuse large B-cell lymphoma (DLBCL) and 290 cases of follicular lymphoma (FL) for POKEMON protein expression levels. POKEMON was highly expressed in 25–35% and moderately expressed in 40–50% of both DLBCL and FL clinical samples. Strikingly, as previously shown for BCL6, POKEMON positivity predicted a better prognosis of overall survival in DLBCL. Interestingly, the vast majority (approximately 80%) of BCL6 positive DLBCLs and FLs were found to be POKEMON positive. By contrast, not all the POKEMON positive lymphomas expressed BCL6. Importantly, coexpression of POKEMON and BCL6 in DLBCL predicted an even better clinical outcome, identifying a clinically distinct DLBCL subtype.

High-dose melphalan (M) with autologous stem cell transplant (ASCT) is an effective therapy for AL, but treatment-related mortality (TRM) remains high, and hematologic complete responses (CR) occur in a minority of patients. In this study, we asked whether a risk-adapted approach to IV M dosing could decrease TRM, and whether adjuvant D +/− T could improve hematologic and amyloid-involved organ response rates. Other objectives of the trial are to study prospectively the prognostic significance of the serum free light chain (FLC), troponin I, and brain natriuretic peptide (BNP) assays. Newly diagnosed untreated AL patients are risk-stratified upon enrollment. Low-risk patients (1 or 2 major organs involved, no advanced cardiac disease) receive M 100, 140, or 200 mg/m2 with ASCT based on age, presence of cardiac involvement, and renal function. High-risk patients (≥3 organs involved or advanced cardiac disease) receive 2 cycles of M 40 mg/m2 without ASCT. Patients with persistent clonal plasma cell disease at 3 months receive 9 months of adjuvant D+T or D alone (if history of deep venous thrombosis or neuropathy). Since 9/02, 38 patients (median age=57.5 (range 34–73), 68% males) have enrolled. Median time from diagnosis to enrollment is 1.5 months (range 0.5 – 7). Organ involvement includes 13 (34%) cardiac, 24 (63%) renal (12 with renal only), 13 (34%) liver/GI tract, and 10 (26%) peripheral nervous system. Thirty-two (84%) had baseline elevated serum FLC with abnormal κ:λ ratio. Thirty-four patients have been treated--3 high-risk and 31 low-risk (5 at M 100, 16 at 140 and 10 at 200 mg/m2). Treatment-related mortality is 7.4% (2/27) with an additional 7 patients alive

From 1993–2001 we treated 186 patients (pts) with relapsed or primary refractory diffuse large B cell lymphoma (DLBCL) on IRB-approved clinical trials with ifosfamide, carboplatin, etoposide (ICE)-based second-line chemotherapy (SLT). Pts with chemosensitive disease received high dose chemoradiotherapy (HDT) and ASCT. In the context of these studies we reported the following: Pts achieving a complete response to ICE, pre-HDT/ASCT, have an improved overall survival (OS) (Journal of Clinical Oncology 1999, 12: 3776–85), pts with chemosensitive primary refractory disease have the same OS as relapsed pts (Blood 2000, 96: 2399–2404); and the second-line age-adjusted IPI (SAAIPI) predicts OS (Blood 2003, 102: 1989–96). In an attempt to further delineate prognostic factors in this setting tissue microarrays (TMA) were constructed as previously described by our group (Hum Pathol 2002: 968–74). All patients had a repeat biopsy prior to initiating SLT confirming active DLBCL; this sample was used for the TMA. Adequate tissue was available on 88 of these pts. TMA sections were stained for the following immunohistochemical markers and analyzed: MIB-1 (Ki-67), MUC1, MDR, p53, bcl-2, CD10, bcl-6, and MUM1. MIB-1 was scored in quartiles and for statistical purposes grouped as 1–2+ (50% tumor cells positive). For all other markers, a positive score was based on >20% of tumor cells staining positive. The "molecular phenotype"(MP), germinal center (GC) vs non-GC has an impact on progression-free survival in untreated DLBCL; we also evaluated whether the MP of the pre-ICE biopsy specimen could predict survival in pts with relapsed or primary refractory DLBCL undergoing SLT/HDT/ASCT. A GC phenotype is defined as either CD10 positive or bcl-6 positive and MUM 1 negative (Hans et al Blood2004; 103: 275–282). At a median follow-up of 6.5 years the the actuarial OS is 39.7% and 49% for patients receiving HDT/ASCT. In this subset of 88 pts, we confirmed that the SAAIPI predicted outcome: good risk pts having an OS of 52% and poor risk 26%, p Conclusion: A non-GC "molecular phenotype" does not predict for a poor outcome in pts with relapsed or primary refractory DLBCL treated with ICE-based SLT followed by HDT/ASCT. MARKER ANALYSIS FOR 88 PTS WITH RELAPSED/REFRACTORY DLBCL MARKER OVERALL SURVIVAL P VALUE MIB 1 0–2+ 38.6% MIB 1 3–4+ 41% 0.9 MUC 1 neg. 39.2% MUC 1 pos. 40.6% 0.5 MDR neg. 34.5% MDR pos. 40.6% 0.6 p53 neg. 39.5% p53 pos. 41% 0.6 Bcl-2 neg. 37.1% Bcl-2 pos. 42.2% 0.9 CD10 pos. 39.7% CD10 neg. 40% 0.8 bcl-6 pos. 33% bcl-6 neg. 44% 0.5 MUM 1 neg. 36% MUM 1 pos. 41.8% 0.8

Introduction: Follicular lymphoma (FL) is the second most common subtype of non-Hodgkin’s lymphoma (NHL) diagnosed annually in the United States. FL is regarded as an indolent NHL but has a clinically heterogeneous course. Various prognostic systems have been described for follicular lymphoma (e.g. International Prognostic Index (IPI), WHO histological grade, etc.) but none have been satisfactory for identifying patients with high-risk follicular lymphoma. Recently, the FLIPI has been proposed as a useful prognostic index for follicular lymphoma. In this report, we retrospectively analyze patients with follicular lymphoma with respect to the FLIPI and directly compare this prognostic index to the WHO/REAL histological grade. Methods: We retrospectively identified patients seen at MSKCC who had archived diagnostic or relapsed biopsy specimens available for pathological review. Patients were included if there was sufficient clinical information available and if review of the specimen confirmed follicular lymphoma according to the WHO/REAL classification system. Clinical information was collected for all patients and their archived pathology was reviewed by 2–3 pathologists independently. The WHO/REAL follicular lymphoma grade was assigned by consensus of at least 2 pathologists. Adverse FLIPI risk factors (RF) included age ≥60, stage III/IV, abnormal LDH, >4 nodal sites, hemoglobin < 12 mg/dl. Patients were stratified into low-risk (LR; 0,1 RF), intermediate-risk (IR; 2 RF), high-risk (HR; >2 RF). Survival analysis were performed by Kaplan-Meier and the log-rank method was used to test for signficance. Results: In all, 260 patients are included in the analysis. The demographics of the patients are as follows: median age at diagnosis was 56 with 38.8% ≥60 years old; 52.3% were male and 47.7% female; 12.3% had a KPS 4 nodal sites of involvement. Sixty-two percent of biopsies were at diagnosis and 38% at relapse. By FLIPI, 128 patients (49%) had LR disease, 76 (29%) had IR and 56 (22%) had HR disease. LR patients had a median survival and 10 year survival of 16.5 years and 76%, respectively; IR patients, 12.4 years and 52%; and HR patients, 5.4 years and 24% (p Conclusion: The WHO/REAL grade does not improve upon the abilility of the FLIPI to risk stratify patients with follicular lymphoma. Furthermore, the FLIPI is superior to the WHO/REAL histolgical grade in identfiying patients with high-risk follicular lymphoma.

We describe the case of a patient treated with 2-chloro-2'-deoxyadenosine, CdA or Cladribine for hairy cell leukemia who subsequently developed an Epstein Barr virus (EBV)-positive polymorphous large B-cell lymphoma (p-LBCL). The time interval between Cladribine therapy and development of p-BCL was 11 months and morphologically resembled an EBV-positive post transplant lymphoproliferative disorder (PTLD). Molecular genetic studies for EBV-clonality by Southern blot hybridization showed a clonal population of infected cells, implying that this was an EBV induced lesion. The chronology of events suggest that Cladribine, a purine analog which has been previously described to induce long-lasting immunodeficiency, can, in some cases, weaken the host defense mechanism to a level at which an innocuous EBV infection may transform the normal lymphoid cells into an aggressive neoplasm. Unlike most methotrexate-related lymphoproliferative disorders (LPDs), which undergo spontaneous remission after discontinuation of therapy, LPDs secondary to purine analogs often fails to resolve after discontinuation of therapy and requires additional therapy. Our patient was treated with rituximab following the diagnosis of p-LBCL, with the goal of improving the pancytopenia to permit chemotherapy. However, the patient failed to show any dramatic improvements in counts, developed systemic symptoms and progressive ascites. He expired 3 weeks after a second dose of rituximab. Cladribine is a potent immunosuppressive agent and should be included with the list of immunosuppressive agents that may be associated with EBV-related B-cell lymphoproliferative disorders.