Citations

Abstract Background Peripheral T-cell lymphomas (PTCLs) are aggressive non-Hodgkin lymphomas with poor overall survival rates. The genetics of PTCLs are poorly understood. We have described t(6;14)(p25.3;q11.2) translocations in PTCLs involving the T-cell receptor-alpha (TRA@) and interferon regulatory factor-4 (IRF4) genes and overexpressing IRF4 protein (also called MUM1). IRF4/MUM1 is a lymphoid transcription factor involved in differentiation and growth, the expression of which is tightly regulated in normal T cells. However, nearly half of PTCLs constitutively express IRF4/MUM1 and this expression is associated with adverse overall survival in several PTCL subtypes. Because IRF4/TRA@ translocations are rare among PTCLs (<1%), we investigated the presence of other genetic abnormalities of IRF4 in PTCLs and examined their association with IRF4/MUM1 protein expression. Methods Studies were performed on paraffin tissue sections of PTCL specimens diagnosed by standard WHO criteria. IRF4/MUM1 immunohistochemical staining (IHC) was performed using the MUM1p antibody and considered positive if nuclear staining was present in >30% of tumor cell nuclei. Fluorescence in situ hybridization (FISH) was performed using breakapart probes constructed from bacterial artificial chromosome DNA that hybridized to loci flanking the IRF4 locus on 6p25.3. Cases with 3 or more fusion signals were considered to have extra copies of the IRF4 gene. In a subset of cases, PCR and Sanger sequencing was performed on DNA extracted from frozen PTCL tumor tissue with primer sets that amplified exons 1 to 10 of the IRF4 gene. Data were analyzed using the chi-square test or t test. Results PTCLs from 277 patients (175 M:102 F; median age, 61 y) were examined by both IHC and FISH. The subtype distribution was: PTCL, not otherwise specified (NOS), 93; anaplastic large cell lymphoma (ALCL), 89; angioimmunoblastic T-cell lymphoma (AITL), 27; cutaneous T-cell lymphoma (CTCL), 43; and cytotoxic PTCL, 25. IRF4/MUM1 was positive in 116 cases (42%). Extra copies of IRF4 were identified in 18 PTCLs (7%), 83% of which were IRF4/MUM1 protein-positive. In contrast, 39% of cases without extra copies of IRF4 were IRF4/MUM1 protein-positive (p=0.0002). The mean percentage of tumor cells with IRF4/MUM1 staining was 69% in cases with extra copies of IRF4, compared to 30% in cases without extra copies (p=0.0001). Extra copies of IRF4 were seen in 13% of ALCLs, 4% each of PTCLs, NOS and cytotoxic PTCLs, 2% of CTCLs, and 0% of AITLs (p=0.02). Two IRF4/MUM1-positive cases had an IRF4/TRA@ translocation (1%). We selected 30 IRF4/MUM1-positive PTCLs for Sanger sequencing. In this subset, we identified a previously undescribed non-synonymous variant in exon 5 (S309R) in 1 case (3%) without an extra copy of IRF4. Conclusions Extra copies of IRF4 are seen in ∼7% of PTCLs, are significantly associated with IRF4/MUM1 expression, and are preferentially observed in ALCLs. Both IRF4/TRA@ translocations and IRF4 mutations are rare in PTCLs. These data suggest that genetic abnormalities of the IRF4 gene – specifically extra copies of IRF4 – may contribute to IRF4/MUM1 expression in some PTCLs. Since IRF4/MUM1 expression is seen in ∼42% of PTCLs, these data also imply the existence of other regulatory mechanisms. Characterizing these mechanisms may be important clinically, since IRF4/MUM1 has been proposed as a therapeutic target in human cancers but drugs that directly inhibit its function are currently not available. Disclosures: No relevant conflicts of interest to declare.

Abstract Background Platelet dense granule (DG) deficiencies, including Hermansky-Pudlak syndrome, are relatively common and frequently under-diagnosed hereditary platelet hypofunctional disorders. Among various laboratory tests, platelet whole mount (WM) transmission electron microscopy (TEM) is considered the gold standard method for diagnosing DG deficiencies. Nevertheless, platelet DG TEM testing is still largely a research tool and has not been thoroughly validated and standardized as a clinical test. Furthermore, reference ranges (RR) of DG count/platelet (DGC) have not been adequately established. Aim Our goals were to validate and standardize platelet WM TEM testing and establish reference ranges for platelet DGC. Methods Based on previously established methods (Blood, 33:598-606), we first optimized and standardized various pre-analytical, analytical and post-analytical procedures including sample matrix, sample stability, platelet-rich plasma (PRP) preparation, platelet mounting procedure and DG counting criteria. Then whole blood samples in ACD-B tubes were collected from a total of 128 healthy donors (64 males and 64 females). Platelet DGC, percentage of platelets without any DG (empty platelets), platelet count (PC), mean platelet volume (MPV) and immature platelet fraction (IPF) were measured. In addition, platelet aggregometry analyses were performed on citrated PRP samples from the same donors. Standard statistical analyses were employed in this study. Results DGC and platelet ultra-structure were stable when whole blood samples were stored at room temperature (RT) for up to 4 days. Reproducible DGC and adequate preservation of platelet ultrastructure depended on an optimized mounting method. The DG counting criteria were developed based on previous publications and guidance from Dr. James G. White. Using the same WM TEM images and DG counting criteria, agreements among different technologists (n=5) improved from 60% to 95%. Finally, DGC reference range study using 128 healthy donor samples (age range 18, 72 years) and scoring 100-200 platelets from each donor showed that the DGC distribution of each donor sample was left-skewed with a median of 15% of empty platelets (range: 3, 40). The median DGC was 2.6 with a range of 1.0—5.1. Reference ranges established at the lower 95th percentile for DGC empty platelets is ≤ 31% (95% CI 28, 34) and for the mid-95th percentiles of DGCs between 1.5 and 4.0. DGCs were weakly associated with MPV and IPF (ρ between 0.22 and 0.28); and inversely associated with the percentage of empty platelets (ρ between -0.26 and -0.25). DGC showed no significant association with PC, platelet maximum aggregation or primary slope of aggregation, nor age or gender. Conclusion We validated platelet WM TEM procedures and for the first time established an adult reference range for platelet DGC. The results of association analyses implied that immature platelets may have slightly more DGs. Disclosures: No relevant conflicts of interest to declare.

Abstract Background: Prothrombin time (PT) and activated partial thromboplastin time (aPTT) are screening tests of hemostasis commonly ordered for anticoagulant therapy monitoring, evaluation of bleeding, screening for acquired or inherited deficiencies of coagulation factors, and preoperative screening. Evaluation of a prolonged PT/aPTT is best accomplished in an algorithmic fashion consisting of mixing studies and factors assays or inhibitor testing as indicated. However, there is little data available on frequency of conditions that may result in abnormalities of PT/aPTT. The Special Coagulation Laboratory (SCL) at our institution offers an algorithmic approach termed Prolonged PT/aPTT Profile (PT/aPTT profile). Aim: To review our institutional experience with PT/aPTT profile testing (indications and outcomes), and to provide clinical correlation of abnormal PT/aPTT tests on patients seen at our institution in an effort to explain common causes of altered hemostatic testing in outpatient clinical practice. Methods: In this retrospective study, following IRB approval, we reviewed the medical records of patients who underwent outpatient PT/aPTT profile testing in the SCL at Mayo Clinic in Rochester, Minnesota between 2010 and 2017. Data abstracted included patient demographics and clinical presentation, indications for testing, use of anticoagulants, medical diagnoses which could potentially influence coagulation, and results of hemostatic testing (Table 1). Statistical analyses were completed using SAS 9.4 (SAS Institute Inc., Cary, North Carolina). Continuous variables are presented as median (range) while discrete variables are summarized as frequency (percentage). Comparisons of clinical characteristics between groups are made using the rank-sum test for continuous variables. Statistical significance is defined as a 2-tailed P value of less than 0.05. Results: A total of 107 patients met our study criteria, 58 (54 %) male, median age 55 years (range 1-89). Indications for testing included prior prolongation of PT (48%), PTT (29%), PT and aPTT (8%) on outside coagulation testing performed in a non-SCL setting, clinical bleeding episodes (10%) and preoperative screening (5%). Patients with established medical diagnoses known to influence coagulation testing including solid tumors, hematologic malignancies, liver disease, renal disease, autoimmune and rheumatologic conditions and pregnancy were documented along with other baseline patient characteristics (Table 1). SCL PT/aPTT profile testing revealed normal results in 29 of 107 (27%) patients on repeat testing while 78 patients warranted further evaluation of which 27 (35%) had an isolated prolonged PT, 28 (36%) had a prolonged APTT and 23 (29%) had prolongations of both PT and aPTT. After further workup and clinical correlation, causes of abnormal outpatient PT/aPTT profile testing were found to be related to an underlying congenital condition in 16 (21%) patients - most commonly factor VII deficiency (9%) and factor VIII deficiency (4%) - acquired conditions in 53 (68%) patients - most commonly anticoagulant therapy effects (31%) and liver disease (21%) - and of unclear etiology in 7 (9%) cases (Table 2). Specifically, the most common cause of an isolated prolonged PT in our study was found to be vitamin K deficiency (37%) whereas elevations in aPTT and both PT/aPTT were most frequently attributed to the effects of anticoagulant therapy (29% and 52% respectively). Other causes of prolongation on PT/aPTT profile testing following clinical correlation were also identified (Table 2). Discussion: While PT and aPTT are commonly ordered tests, there is little evidence regarding epidemiology of testing indications, distribution of test results, and the clinical correlation of test results. Our study demonstrates that abnormalities in hemostatic testing can be attributed to a wide range of causes including artefactual prolongation influenced by pretest variables or marked abnormalities in secondary hemostasis due to inherited and acquired conditions. Strengths of this study include the SCL algorithmic approach to testing to further elucidate coagulation cascade abnormalities to properly characterize conditions that pose severe hemostatic risk while limitations of this study reflect the experience of a single institution and possible referral bias. Disclosures No relevant conflicts of interest to declare.

Abstract Abstract 775 Introduction: WW has been an initial strategy for advanced stage FL because randomized trials showed no clear benefit for immediate initiation of therapy with single agent alkylators when compared to deferring therapy (Brice, JCO 1997; Ardeshna, Lancet 2003), although a recent trial raises questions about WW in the era of rituximab (R) based frontline therapies (Ardeshna, ASH Meeting 2010). Methods: The NLCS is a prospective, multicenter, observational study collecting data on 2,738 newly diagnosed pts with FL diagnosed from 2004–2007 at 265 US sites (80% non-academic). Initial management decisions were made by the treating physician and were categorized as WW (for pts who did not receive therapy in the 90 period following diagnoses and were coded by physicians as WW) or active therapy (AT) including: R alone, R-chemotherapy (RChemo), or other. Baseline clinical factors and treatment setting were compared for WW and AT pts using multivariable logistic regression. Kaplan-Meier (KM) estimation and Cox models were used to evaluate progression-free survival (PFS), time to 2nd and 3rd line management strategy (TT2, TT3), transformation-free survival (TFS), PFS following 1st AT, and overall survival (OS). Due to differences between WW and AT groups, adjusted Cox regression models are presented for the primary analysis of study endpoints controlling for sex, community vs. academic practice site, FL grade, and FLIPI. All time to event analyses were calculated from the 90 day period following diagnosis to allow enrollment in WW, and PFS was calculated from the starting point to the clinician's recording of progressive disease (PD) or death. Results: Among 1,737 pts presenting with stage III/IV FL in the analysis population, 237 underwent WW and 1,500 received AT immediately following diagnosis including: 241 pts treated with R, 1046 with RChemo, 26 with radiation, and 187 other therapies. Multivariable logistic regression demonstrated that the following pt characteristics were predictive of selection of WW rather than initial AT: age > 60 years, no B-symptoms, FL grade 1 or 2, ECOG PS = 0, ≤1 extranodal sites involved, and LDH ≤ULN (all p < 0.05). Median PFS and TT2 were 27 and 35 months for WW, and 64 months and not reached for AT. With a median follow-up of 55 months, 17% of WW pts and 19% of AT pts have died, with no significant difference in OS between WW and AT (p = 0.31). Compared with WW, RChemo improved PFS: adjusted Hazard Ratio (HR) = 0.36, TFS: HR=0.65, TT2: HR=0.48, and TT3: HR=0.53, all p<0.01. For most of the 145 WW pts that later initiated an AT, PD was recorded as the reason for treatment (83%). ATs following WW were: R (51, 35%), RCVP (21, 14%), RCHOP (16, 11%), R + other chemo (23, 16%), investigational (15, 10%), chemo alone (9, 6%), radiation (8, 6%), BMT (1, 1%), or other therapy (1, 1%). The median PFS for R was 42 months when given at diagnosis and 55 months when given following WW. The median PFS for RChemo was 71 months when given at diagnosis and 37 months when given following WW. After adjustment for FLIPI, FL grade, sex, practice setting, and type of treatment, WW adversely affected PFS following 1st AT, H=2.02, p <0.0001. Conclusions: FL pts in the US who undergo initial WW are treated at a median of nearly 3 years. Median PFS following RChemo, but not R is shorter when utilized after period of WW than at diagnosis – albeit in different clinical circumstances. Additional follow-up is needed to assess the tradeoff of this time without therapy with as yet no observable differences in OS. Disclosures: Sinha: CELGENE: Research Funding. Friedberg:Genentech: Consultancy, Membership on an entity's Board of Directors or advisory committees. Link:Genentech: Consultancy. Flowers:Spectrum: Consultancy, Research Funding; Millennium/Takeda: Consultancy, Research Funding; Celgene: Consultancy, Research Funding; Genentech/Roche (unpaid): Consultancy; Novartis: Research Funding; Seattle Genetics: Consultancy.

Abstract Abstract 2835 Background: Extrinsic growth and survival signals provided by nonneoplastic cells present within the tumor microenvironment promote the growth and survival of CLL cells. For example, bone marrow stromal cells, blood-derived “nurse-like” cells and follicular dendritic cells have been shown to promote the survival of CLL cells via the provision of both soluble and cell-contact dependent factors. More recently, monocytes were convincingly shown to support the survival of CLL cells during in vitro culture, and many of the factors (e.g. BLyS/BAFF, SDF-1, CD14) previously implicated in supporting the growth and survival of CLL cells are produced by monocytes. To explore the potential clinical implications of these in vitro findings, we evaluated the relationship between the absolute monocyte count (AMC) at diagnosis and disease progression in a well defined cohort of CLL patients. We hypothesized that the presence of a relative monocytosis at presentation in patients newly diagnosed with CLL may create a nurturing environment for CLL cells and be associated with more rapid disease progression. Methods: Previously untreated patients diagnosed with CLL/SLL within 12 months of presentation to the Mayo Clinic (from 1/1/2000-8/31/2010) were identified (n=1396). The AMC was obtained from routine complete blood counts (CBC) with automated or manual differentials that were performed within 3 months of diagnosis (n=947). The primary study objective was to determine whether the AMC is associated with time to first treatment (TTFT) in CLL. The secondary objective was to determine whether the AMC had independent prognostic value after adjusting for CD38/CD49d/Zap-70 expression, IGHV gene mutation status, and cytogenetic (11q-, 17p-) findings by FISH analysis. Results: The median age was 66 years and the median AMC was 550/μL (380–812/ μL, 1st-3rd quartile) at diagnosis (normal range: 300–900/μL). No significant difference in the median AMC was observed by age, gender; CD38, CD49d, or Zap-70 expression; IGHV gene mutation status or cytogenetic findings. Patients were followed for a median of 3.2 years (range: 0–10.3), with disease progression requiring the initiation of treatment observed in 268 (28%) patients. On multivariate analysis using a Cox proportional hazards model which incorporated age, gender and Rai stage, the AMC (as a continuous variable) was associated with TTFT with a hazard ratio (HR) of 1.099 (per unit change; 95% CI, 1.003–1.205; p=0.04). The optimal cutpoint for the association between AMC and TTFT was determined to be ≥1, 170/μL, a threshold that remained an independently significant predictor of TTFT after adjusting for age, gender and Rai stage with a HR of 5.1 (95% CI, 2.56–10.10, p<0.0001). The median TTFT was 1.0 year for patients with an AMC ≥1, 170/μL and 7.4 years for those with an AMC <1, 170/μL (p<0.0001). In order to evaluate whether the AMC had independent prognostic value after adjusting for other prognostic biomarkers, CD38, CD49d, Zap-70, IGHV gene mutation status and cytogenetic findings (11q- or 17p-), were individually incorporated into the multivariate analysis. In each case, the AMC remained an independent predictor of TTFT, with HRs ranging from 4.31 to 5.43 (p<0.0001 for all analyses). Conclusions: The AMC at the time of diagnosis as measured by a routine CBC is associated with accelerated disease progression in CLL/SLL independent of other prognostic biomarkers. These results suggest that the in vitro observation that nonneoplastic cells present in the tumor microenvironment promote CLL cell survival may have direct clinical implications for the prognostic evaluation of CLL patients and further substantiate the pathogenic role of monocytic cells in CLL. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 971 Unlike the case in acute myeloid leukemia, there is limited information on the prognostic impact of IDH mutations in myelodysplastic syndromes (MDS). In the current study of 277 patients with MDS, IDH mutations were detected in 34 (12%) cases: 26 IDH2 (all R140Q) and 8 IDH1 (six R132S and two R132C). Mutational frequency was 4% (2 of 56) in refractory anemia with ring sideroblasts (RARS), 12% (16 of 130) in refractory cytopenia with multilineage dysplasia, 14% (2 of 14) in MDS-unclassifiable, 14% (6 of 42) in refractory anemia with excess blasts (RAEB)-1, and 23% (8 of 35) in RAEB-2. Normal karyotype was noted in all but one IDH1-mutated cases and 13 IDH2-mutated cases. Multivariable analysis identified presence of mutant IDH1 (p=0.0004; HR 4.0, 95% CI 1.9–8.8), revised International Prognostic Scoring System (IPSS-R) risk category (p<0.0001), and red cell transfusion need (p=0.002) as independent predictors of inferior survival (Figure 1). In a similar multivariable analysis, mutant IDH1 was the only variable associated with shortened leukemia-free survival (p=0.001; HR 7.0, 95% CI 2.3–20.8). The presence of IDH2 R140Q did not affect overall (p=0.54) or leukemia-free (p=0.81) survival (Figure 2). The current study suggests a powerful adverse prognostic effect for mutant IDH1 in MDS. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 85 Background: Follicular lymphoma (FL) is an indolent non-Hodgkin lymphoma (NHL) with a highly variable natural history. Clinical variables describing either patient- or tumor-specific characteristics, many of which are included in the Follicular Lymphoma International Prognostic Index (FLIPI), are used to risk-stratify patients at the time of diagnosis. However, characteristics of non-malignant cells present within the tumor microenvironment provide important prognostic information. For example, both immunohistochemistry and gene expression profiling studies demonstrate that lymphoma-associated macrophages (LAMs) are prognostically important in FL. As LAMs are monocyte-derived, we sought to determine whether the absolute monocyte count (AMC) obtained from a CBC at diagnosis is an adverse prognostic factor in FL. Methods: Consecutive patients with FL who were evaluated and treated at the Mayo Clinic, Rochester between 1998 and 2007 were considered for study participation. Study approval was granted by the Mayo Clinic Institutional Review Board. The primary objective of the study was to determine the significance of the AMC, obtained at the time of diagnosis from a standard complete blood count with differential, in predicting overall survival. The secondary study objective was to determine whether the AMC is an independently significant prognostic factor when compared with the FLIPI. Results: The median follow-up was 4.2 years (range: 2 months to 11 years) for the 355 patients included in this study. At diagnosis, 43%, 34% and approximately 23% of patients were classified as low-, intermediate- or high-risk by the FLIPI. The median AMC at diagnosis was 510/μL, with 8% of patients presenting with an absolute monocytosis. At diagnosis, 45% of patients were initially observed, 42% received combination chemotherapy or immunochemotherapy alone, while the remaining 13% underwent surgery or radiation. The AMC was evaluated as both a continuous and dichotomized variable on univariate analysis. The optimal cut-point for the AMC was determined to be 570/μL by ROC analysis and was used when analyzing the AMC as a dichotomized variable. A relative elevation in the AMC was associated with inferior overall survival, whether analyzed as a continuous or dichotomized variable, with hazard ratios (HR) of 1.002 (HR per unit change in AMC; 95% confidence interval 1.001–1.002; p=0.0002) and 3.4 (95% confidence interval 2.0–6.0; p<0.0001), respectively. The median OS for patients with an AMC <570/μL at diagnosis has not been reached, whereas a median OS of 10.2 years (95% confidence interval 7.8–11.2 years; p<0.0001) was observed for those patients with an AMC ≥570/μL. For comparison, median OS has not been reached for high-risk patients identified by the FLIPI. Among those patients who were initially observed, the median time to progression (TTP) was 31 months (95% confidence interval 25–37 months) for those with an AMC <570/μL at diagnosis. In contrast, patients with an AMC ≥570/μL experienced a significantly shorter TTP following initial observation, with a median TTP of 18 months (95% confidence interval 8–28 months; p=0.03). When adjusting for prognostic factors included in the FLIPI on multivariate analysis, the association between the AMC as a dichotomized variable and survival remained significant with a HR of 2.6 (95% confidence interval 1.5–4.6; p=0.001). When the FLIPI alone (high vs. low/intermediate) was analyzed on multivariate analysis, the AMC remained an independent predictor of survival (p<0.0001). Furthermore, the AMC provided additional prognostic information when superimposed upon the FLIPI. For example, a subset of patients classified as high-risk by the FLIPI were observed to have an OS comparable to low/intermediate risk patients identified by the FLIPI upon further risk-stratification by the AMC, with a 6-year OS of 80%. In contrast, high-risk FLIPI patients with an AMC ≥570/uL were observed to have a median OS of 4.2 years (95% confidence interval 2.7–7.2) and 6-year OS of 47%. Conclusions: The AMC is associated with overall survival in FL and is independent from the FLIPI on multivariate analysis. While the AMC did not appear inferior to the FLIPI in identifying high-risk patients, it may be most helpful when used in conjunction with the FLIPI. These results further support the central role of nonneoplastic myeloid-lineage cells in FL biology. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 97 Background: There is no consensus or standard of care for front-line therapy for FL as demonstrated by our initial publication of the National LymphoCare Study (NLCS) showing that strategies used in newly diagnosed FL patients (pts) in the US included: watchful waiting (17.7%), rituximab (R) with chemotherapy (51.9%), R alone (13.9%), radiation therapy (5.6%), chemotherapy (3.2%), or a clinical trial (6.1%). Given the lack of observational data comparing the effectiveness of front-line R-chemotherapy regimens, we examined the outcomes for patients with stage III/IV FL receiving R with cyclophosphamide, vincristine, and prednisone (RCVP), R with cyclophosphamide, doxorubicin, vincristine, and prednisone (RCHOP), or R with a fludarabine-based regimen (RFlu) as first-line therapy. Methods: The NLCS is a prospective, multicenter, observational study collecting data on 2,738 previously untreated patients (pts) with FL diagnosed from 2004 to 2007 at 265 sites in the US. Initial management decisions were made by the treating physician without protocol-specified treatment assignments. Descriptive statistics for the baseline characteristics of pts who received RCVP, RCHOP, and RFlu were compared using Pearson Chi-square tests. A generalized logistic model was used to identify baseline factors of clinical interest (FLIPI, sex, practice setting [academic vs. community]) that were correlated with treatment. Kaplan–Meier estimation was used to evaluate progression-free survival (PFS) and overall survival (OS) for the treatment regimens; comparisons for the 2-year landmark estimates were obtained using a Wald statistic. PFS was defined as the number of days from diagnosis up to and including the date of progressive disease (PD) as assessed by the treating physician or death from any cause. Pts who had not yet experienced PD at the time of analysis were censored at the date of the most recent response assessment. To evaluate the effects of treatment on PFS and OS, Cox proportional hazards models were used controlling for FLIPI, sex, community or academic practice setting, and maintenance R or observation following treatment. Results: Of the 2,738 pts enrolled in NLCS, 926 pts were identified as having stage III/IV FL and having received the regimens of interest. Within this cohort 60% received RCHOP (n=558), 27% received RCVP (n=247) and the remaining 13% received RFlu (n=121). Baseline characteristics were correlated with induction treatment and whether a patient received subsequent R maintenance. Women (odds ratio [OR]=1.50, 95%CI 1.05–2.13) and pts aged >60 years (OR=2.36, 95%CI 1.64–3.38) more commonly received RCVP than RCHOP, and pts with LDH > upper limit of normal (OR=1.75, 95%CI 1.02–3.03) more commonly received RCHOP than RFlu. A greater % of pts receiving RCVP (61%) had poor-risk FLIPI (3–5) vs. RCHOP (49%) or RFlu (44%). In this cohort, 60% of RCVP, 43% of RCHOP, and 46% of RFlu pts received maintenance R. The overall response rates for RCVP, RCHOP, and RFlu were 88%, 94%, and 96%, respectively, and 2-year PFS was 72% (95% CI 66–78%) for RCVP, 78% (95%CI 74–81%) for RCHOP, and 76% (95%CI 67–-83%) for RFlu. There were no significant differences in PFS between treatments after adjustment (p= 0.07). However, with a median follow-up of 58 months there were significant differences in unadjusted OS (p=0.03), with 2-year OS probabilities of 88% (95%CI 84–92%) for RCVP, and 91% (84–95%) for RFlu, and 94% (92–96%) for RCHOP (Figure A). After adjustment for sex, FLIPI, practice setting, and maintenance use, the OS benefits of RCHOP persisted over RCVP HR=1.67 (95%CI 1.15–2.42) but not RFlu HR=0.94 (0.53–1.68). The benefits of RCHOP were even more pronounced in poor risk FLIPI pts (Figure B). Conclusions: First-line R-chemotherapy regimens in clinical practice have high overall response rates and high 2-year PFS and OS. With 4.75 years of follow-up in this large pt cohort, these regimens appear similar in response rates and early PFS and OS with suggestions that RCHOP may provide benefits over RCVP particularly for poor risk patients. Disclosures: Sinha: Celgene: Research Funding. Friedberg:Genentech: Consultancy; astellas: ; Lilly:; Abbott/Trubion:; Seattle Genetics: Honoraria; Cephalon: Consultancy. Hirata:Roche: Equity Ownership; Genentech Inc/Roche: Employment. Flowers:Seattle Genetics: Consultancy; Millennium/Takeda: Consultancy, Research Funding; Celgene: Consultancy, Research Funding; Genentech/Roche (unpaid): Consultancy; Novartis: Research Funding; Spectrum: Consultancy, Research Funding.

Abstract Systemic mastocytosis (SM) is a rare hematologic neoplasm characterized by abnormal growth and accumulation of tissue mast cells (MC) in various organ systems, including bone marrow (BM). Indolent and advanced forms of SM have been described. Whereas patients with ISM have a normal or near normal life-expectancy, patients with advanced SM, including those suffering from mast cell leukemia (MCL) have a poor prognosis. In these patients, neoplastic MC are usually resistant against conventional drugs and various targeted drugs. In rapidly progressive aggressive SM (ASM) and MCL, polychemotherapy followed by allogeneic hematopoietic stem cell transplantation (alloHCT) has been proposed. However, outcome of alloHCT in advanced SM is unknown, and it also remains uncertain whether clinically relevant graft-versus-SM (GVSM) effects may occur in these patients, as only sporadic case reports have been published. We performed a retrospective multi-center analysis to evaluate the outcome of alloHCT in patients with advanced SM. Fifty-four advanced SM patients receiving SCT in 32 transplantation centers in Europe and America were identified between 1990 and 2013. The median patient age was 45 years. Donors were: HLA identical siblings (31), unrelated donors (URD) (15), umbilical cord blood donors (UCB) (2), and haploidentical donors (1). In 5 patients, stem cell source was not defined (5). Thirty-four patients received myeloablative conditioning (MAC) and 18 received reduced intensity conditioning regimens (RIC). In 2 patients, conditioning regimen was not specified. Indications for alloHCT were SM with an associated clonal hematologic non-mast cell lineage disease (SM-AHNMD) (n=32), MCL (n=13, including one with MCL-AHNMD), 8 with ASM and 1 with myelomastocytic leukemia (MML). The most prevalent AHNMD was acute myeloid leukemia (AML, n=16). With follow-up of 35-6180 (median 365) days, SM responses (defined as ≥50% decrease in BM mast cells ± decrease in serum tryptase ± regression of other organ manifestations) were observed in 39 patients (72%), including complete responses (CR) documented in 12 patients (22%). Eleven patients had stable disease, whereas 4 patients (7%) progressed immediately after alloHCT (primary resistance). In addition, 10 patients progressed (5 of them within 100 days) after an initial response. Progression was most frequently seen in MCL patients (n=6, 50%). In the AHNMD group, only 8 patients relapsed/progressed (25%). The overall survival (OS) and SM progression-free survival (PFS) at 1 year were 63% and 50% for all patients, 77% and 68% for SM-AHNMD, 63% and 50% for ASM, and 25% and 17% for MCL, respectively. The strongest predictive variable associated with inferior survival was a diagnosis of MCL. Other factors associated with poor outcome were: Karnofsky performance status ≤70%, ≥2 SM regimens given before alloHCT (e.g., steroids, cladribine, chemotherapy, tyrosine kinase inhibitor), donor source (alternative donors-UCB and haploidentical compared to sibling or URD), SM progression within the first 100 days, normal cytogenetics (compared to t(8;21) (q22;q22), and RIC (compared to MAC). The following variables were not associated with poor outcome: patient and donor age, recipient-donor sex match status, graft source (BM vs. peripheral stem cells), BM mast cell percentage at time of alloHCT, and CR status of AML or SM response at time of alloHCT. This largest multi-center analysis of results in advanced SM provides evidence for clinical efficacy of alloHCT, presumably because of a GVSM effect of alloHCT (achieving CR, and response to donor-lymphocyte infusions and RIC alloHCT). However, responses varied among different SM categories: while patients with SM-AHNMD enjoyed excellent outcomes, the OS for MCL patients in general, was poor. Nevertheless it is remarkable that 3 of 13 patients with MCL – an otherwise fatal disease with a median survival of &lt;12 months – became long-term survivors after alloHCT. In summary, our results support development of a prospective alloHCT study, with the aim to optimize therapy and to improve overall outcome in advanced SM. Based on our pilot study, alloHCT should also be considered in practice for eligible and fit patients with SM-AHNMD (if treatment of AHNMD component needs alloHCT or SM component is aggressive), rapidly progressing ASM, MCL and MML when a suitable HLA-matched sibling donor or URD is available. Disclosures: Vercellotti: Sangart Inc.: Research Funding; Seattle Genetics: Research Funding. Akin:Novartis: Consultancy. Valent:Novartis: Consultancy, Honoraria, Research Funding.

Abstract Abstract 3929 Poster Board III-865 In addition to providing prognostic information, gene expression profiling studies have helped to elucidate the putative cell of origin (e.g. activated or germinal center B cell) in diffuse large B-cell lymphoma. With the exception of angioimmunoblastic T-cell lymphomas, which are thought to derive from follicular helper T cells, little is known about the cell of origin for the most common peripheral T-cell lymphoma (PTCL), PTCL-unspecified (PTCL-U). Following appropriate antigenic stimulation, naïve CD4+ T cells may terminally differentiate into effector cells, defined by the production of specific cytokines, including IL-4/IL-10 producing Th2 cells. We have previously demonstrated that IL-10, produced by malignant T cells, is systemically elevated in the serum of patients with PTCL and contributes to the suppression of host anti-tumor immunity. Murine studies have demonstrated that the transcription factor GATA-3 is both critically important for IL-10 transcription and is required for the differentiation of Th2 cells. Therefore, we hypothesized that a subset of PTCL-U may be Th2-derived and express GATA-3. To examine this possibility, PTCL-U cases with adequate paraffin-embedded tissue were identified from a single institution (n=46) and immunohistochemically stained for GATA-3 expression. More than 10% of tumor cells expressed GATA-3 in 17 (37%) of the 46 cases examined. Follow up data sufficient to determine overall survival was available for 43 of these patients, in whom the expression of GATA-3 was associated with a poorer overall survival. Patients with GATA-3 negative tumors (n=27) had a median overall survival (OS) of 3.8 years, whereas patients with GATA-3 positive tumors (n=16) had a median OS of 0.7 years (p=.002). Among low-intermediate risk patients, as defined by the International Prognostic Index (IPI), GATA-3 expression was associated with inferior median overall survival (<12 months, p<.001). IL-10 production may explain the adverse impact of GATA-3 expression on OS in PTCL-U, as GATA-3 plays a central role in IL-10 transcription in normal T cells. To determine whether it may have a similar role in malignant T cells, we identified a PTCL cells line (SU-DHL-1) that highly expressed GATA-3 by intracellular flow cytometry. We next transfected these cells with a negative control or GATA-3 specific siRNA and demonstrated that transfection with GATA-3 siRNA significantly decreased GATA-3 expression in these cells. Following knockdown of GATA-3 expression, we measured both cell proliferation by 3H-TdR incorporation and IL-10 production by ELISA. No significant difference in cell proliferation was observed (n=4, p=0.2), whereas an approximately 50% reduction in IL-10 production was observed (n=3, p=.008) following inhibition of GATA-3 expression. Therefore, GATA-3 expression in malignant T cells may promote the production of IL-10, thus impairing host anti-tumor immunity. Immunomodulatory drugs (IMiDs), like lenalidomide, may downregulate GATA-3 expression in normal T cells. Therefore, we sought to determine whether lenalidomide may inhibit IL-10 production in normal and malignant T cells. T cells obtained from normal donors were activated with anti-CD3 and treated with either DMSO or lenalidomide (10 μM). Intracellular staining for both IL-10 and IL-2 was performed 3 days later. In DMSO control treated cells, 1.0% of cells expressed IL-10 and 12% expressed IL-2. In contrast, in lenalidomide treated cells, only 0.3% of cells expressed IL-10 and 17% expressed IL-2 (representative of at least 4 independently performed experiments). Similar results were obtained using T cells from a patient with Sezary Syndrome. We next treated SU-DHL-1 cells with DMSO or lenalidomide (0.1, 1.0 and 10 μM) and measured both cell proliferation by 3H-TdR incorporation and IL-10 production by ELISA. Significant inhibition of SU-DHL-1 proliferation was only observed when cells were treated with 10 μM lenalidomide. In contrast, significant inhibition of IL-10 production was observed in the presence of 1 μM and 10 μM lenalidomide (p<.01). Collectively, the data presented demonstrates that GATA-3 promotes the production of IL-10 by malignant T cells and is an adverse prognostic factor in PTCL-U. Given its immunomodulatory properties, including inhibition of IL-10 production, lenalidomide represents a promising therapeutic agent in PTCL-U. Disclosures: No relevant conflicts of interest to declare.