Abstract: Infections represent one of the major concerns regarding the utilization of ruxolitinib (RUX) in patients with myelofibrosis. With the aim to investigate epidemiology, outcome and risk factors for infections in RUX‐exposed patients, we collected clinical and laboratory data of 446 myelofibrosis patients treated with RUX between June 2011 and November 2016 in 23 European Hematology Centers. After a median RUX exposure of 23.5 months (range, 1‐56), 123 patients (28%) experienced 161 infectious events (grades 3‐4 32%, fatal 9%), for an incidence rate of 17 cases per 100 pts/y. The rate of infections tended to decrease over time: 14% of patients developed the first infection within 6 months, 5% between 6 and 12 months, 3.7% between 12 and 18 months, 3.4% between 18 and 24 months, and 7.9% thereafter ( P   〈  .0001). Respiratory tract infections were more frequently observed (81 events, 50%), and bacteria were the most frequent etiological agents (68.9%). However, also viral (14.9%) and fungal infections (2.5%) were observed. In multivariate analysis, previous infectious event (HR 2.54; 95% CI, 1.51‐4.28; P  = .0005) and high international prognostic score system category (IPSS) (HR 1.53; 95% CI, 1.07‐2.20; P  = .021) significantly correlated with higher infectious risk. On the contrary, spleen reduction ≥50% from baseline after 3 months of treatment ( P  = .02) was associated with better infection‐free survival. Taken together, these findings reinforce the concept of disease severity as the most important risk factor for infections, and describe, for the first time, that a positive therapeutic effect in reducing splenomegaly may also reduce subsequent infectious complications.

Abstract: The A-Myb transcription factor is structurally related to the c-myb proto-oncogene and is involved in the control of proliferation and/or differentiation of mature B lymphocytes. We have shown previously by PCR analysis that A-myb is preferentially expressed in CD38+CD39−sIgM− mature B cells. We demonstrate here, using in situ hybridization, that A-mybexpression is restricted to the dark zone of human tonsils and lymph nodes. Furthermore, we show that A-Myb expression is cell cycle regulated both in tonsillar B cells and in Burkitt’s lymphoma cell lines, being detectable only in the S and G2/M phases of the cell cycle and not in G0/G1 phase. Strong proliferation of resting human B cells induced in vitro by a variety of physiologic signals, including anti-μ, CD40 ligand, IL-2, IL-4, IL-6, IL-13, IFN-γ, TNF-α, anti-CD19, and anti-CD20, failed to induce A-myb expression, suggesting that proliferation alone is not sufficient for A-myb expression in the absence of induction of a true centroblast phenotype. Finally, we show that differentiation of germinal center B cells in vitro toward either memory or plasma cells is accompanied by rapid down-regulation of A-myb expression. We conclude that A-myb is a marker of centroblasts generated in vivo.

Abstract: Background: Response to ruxolitinib (RUX), the only JAK1/2 inhibitor commercially available for the treatment of Myelofibrosis (MF) may vary among patients (pts) and is largely unpredictable at therapy start. Therefore, pts' selection is based only on clinical needs. Aims: To evaluate the impact of pre-treatment clinical/laboratory factors, as well as RUX dose, on response to RUX in a cohort of "real-life" MF pts. Methods: A multicenter observational study on WHO-defined MF was conducted in 18 Italian Hematology Centers. Data were extracted from a database that included retrospective data on pts treated before January 2015. Subsequently, data were prospectively collected and updated at a 6-month interval.Response to RUX was evaluated according to IWG-MRT criteria. Results: Between June 2011 and Apr 2016, 408 pts with PMF (54.4%), or postET (27.7%) / post-PV (17.9%) were treated with RUX. At RUX start, baseline characteristics were (median): age, 68.5 y (range, 26.5-89); ≥65 y, 63.5%; male, 56.4%; hemoglobin (Hb), 10.7 g/dL (7-16.7); transfusion-dependence 27.9%; PLT, 256×109/L (50-1887); PLT 〈 100×109/L, 9.6%; spleen enlargement, 96.6% (spleen length ≥10cm: 64.2%); total symptoms score (TSS), 20 (12-70).International Prognostic Score System (IPSS) was intermediate (intm)-1 (15.7%), intm-2 (47.3%), high (37%). Molecular data were available for 332 pts (81.4%) and was positive in 81% (JAK2V617F), 6.3% (CALR), 1% (MPLW515K/L); 2.7% (triple negative). 30 pts (9%) were JAK2V617Fnegative but did not receive further molecular evaluation. Karyotype was abnormal in 55 (26%) out of 210 evaluable pts (unfavorable: 8.1%). Median follow-up from MF diagnosis was 3.8 yr (0.3-29.6) and median RUX exposure was 20 mos (3-56.2). Overall, 152 out of 365 (42%) pts with spleen ≥5cm achieved a spleen response at any time during RUX therapy. At 3 and 6 mos, the response was achieved by 26.6% and 34.4% of evaluable pts, respectively. In univariate analysis, pre-treatment factors negatively correlating with spleen response were: transfusion dependence, platelet count ≤200x109/l, spleen palpable ≥10 cm below costal margin, grade 3 marrow fibrosis, intm-2/high IPSS risk and interval between MF diagnosis and RUX start ≥2y. Three variables remained significant in multivariate regression logistic analysis: large splenomegaly (HR: 2.05, 95%CI: 1.1-3.7; p=0.02), time-interval ≥2y (HR: 1.78, 95%CI:1.0-3.1; p=0.04) and transfusion dependency (HR: 1.95, 95%CI:1.0-3.7; p=0.04). Spleen response significantly correlated with the average RUX dose in the first 12 wks, with pts treated with doses ≥10 mg BID having better response rates (47.3% vs 26.6% if dose 〈 10 mg BID, HR:2.36, 95%CI:1.3-4.3, p=0.005). 360 pts had a TSS 〉 10 at RUX start and 319 (88.6%) achieved a symptom response. In multivariate analysis, factors associated with worse responses were: transfusion dependency (HR: 3.15, 95%CI 1.5-6.4, p=0.001) and a baseline TSS 〉 20 (HR: 6.7, 95%CI 3.2-13.8, p 〈 0.001). RUX titrated dose 〈 10 mg BID during the first 12-wks of therapy negatively correlated with symptoms response (HR: 2.6, 95%CI 1.05-6.7, p=0.037). Drug-related anemia (acquisition of transfusion dependency or Hb 〈 10g/dl in pts with a previous Hb≥10) was observed in 127/291 (43.6%) evaluable pts. The probability to develop anemia was significantly higher in females (HR: 1.63, 95%CI 1.03-2.57, p=0.036). Notably, anemia was not influenced by RUX 12-wks titrated dose (41.8% in pts with RUX titrated dose 〈 10 mg BID vs41.5% with higher doses). 80 (19.6%) pts discontinued RUX because of: lack/loss of response (28.8%); drug-related toxicity (27.5%, specifically: thrombocytopenia, 16.2%; infection, 6.3%; anemia, 5%); disease progression with/without acute evolution (8.8%); death (13.8%); allogeneic transplant (8.8%); 2ndneoplasia (3.8%); other unrelated causes (8.5%). Summary/Conclusion: In a real-life setting, IWG-MRT-defined spleen and symptoms response rates were observed in 42% and 88.6% of evaluable pts, respectively. Disease severity (in terms of transfusion dependency and large splenomegaly) and a delay in RUX start ≥2yr from diagnosis identified pts with lower spleen response rates. Titrated doses 〈 10mg BID significantly correlated with poorer spleen and symptoms responses. Overall, these data point out the importance of an early treatment and of an effective (≥ 10 mg BID) titrated dose in order to achieve better therapeutic results. Disclosures Palumbo: Novartis: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Celgene: Speakers Bureau; Shire: Honoraria; Roche: Honoraria. Bonifacio:Novartis: Research Funding; Bristol Myers Squibb: Consultancy; Amgen: Consultancy; Ariad Pharmaceuticals: Consultancy; Pfizer: Consultancy. Tiribelli:Bristol-Myers Squibb: Consultancy, Speakers Bureau; Ariad Pharmaceuticals: Consultancy, Speakers Bureau; Novartis: Consultancy, Speakers Bureau. Latagliata:Novartis: Consultancy, Honoraria; Bristol Myers Squibb: Honoraria; Celgene: Honoraria; Janssen: Consultancy, Honoraria; Shire: Honoraria. Vitolo:Roche: Membership on an entity's Board of Directors or advisory committees, Other: Honoraria for lectures; Janssen: Membership on an entity's Board of Directors or advisory committees, Other: Honoraria for lectures; Gilead: Other: Honoraria for lectures; Takeda: Other: Honoraria for lectures. Fanin:Novartis: Speakers Bureau. Merli:Teva Pharmaceuticals Industries: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; Roche: Membership on an entity's Board of Directors or advisory committees, Research Funding. Cavo:Janssen-Cilag: Consultancy, Honoraria; Amgen: Consultancy, Honoraria; Celgene: Consultancy, Honoraria; Millennium: Consultancy, Honoraria; Bristol-Myers Squibb: Consultancy, Honoraria. Breccia:Ariad: Honoraria; Bristol Myers Squibb: Honoraria; Novartis: Consultancy, Honoraria; Celgene: Honoraria; Pfizer: Honoraria.

Abstract: Background. The prognostic role of cell of origin profile (COO) assessed by immunohistochemistry (IHC) is controversial in Rituximab era. FIL conducted a phase III randomized trial aimed at investigating the benefit of intensification with high dose therapy plus autotransplant compared to R-dose-dense therapy as first line in young DLBCL at poor risk (aa-IPI 2-3). Clinical results were reported (Vitolo, ASH 2012). The aim of BIO-DLCL04 was to correlate the biological markers with PFS. Patients and Methods. From 2005 to 2010, 412 untreated DLBCL at aa-IPI 2-3 were enrolled. Central histology revision was mandatory and 13 patients were excluded due to different histologies. Biological markers were analyzed on DLBCL NAS; COO analysis was performed by IHC and cases were classified in germinal center (GC) and non-GC according to Hans' algorithm; COO determined by gene expression profile using the NanoString® nCounter® Analysis System based on 20-gene assay (Lymph2Cx) using formalin fixed paraffin embedded tissue is ongoing; BCL2, BCL6 and MYC anomalies were tested by IHC; final analysis by fluorescent in situ hybridization (FISH) is ongoing. Cases were deemed positive if at least 30% of lymphoma cells were stained with each antibody (with the exception of at least 40% for MYC). Results. At the time of this analysis, 223 DLBCL NAS were analyzed: 131 non-GC and 92 GC; BCL2, BCL6 and MYC anomalies were tested in 196, 74 and 107 cases respectively. Clinical characteristics for non-GC vs GC were: median age 51 years for both, male 49% vs 45%, aa-IPI 3 15% vs 25%, bone marrow involvement (BM) 16% vs 24%. R-HDC was performed in 45% of non-GC patients and in 49% of GC. Complete response was recorded in 105 (80%) non-GC patients and in 62 (67%) GC. At a median follow-up of 49 months, the 3-year PFS for non-GC vs GC was 75% (95% CI: 67-82) vs 57% (95% CI: 46-67) with crude hazard ratio, HR 0.55 (0.35-0.87), p.01 and adjusted (for age, gender, aa-IPI, BM) aHR 0.56 (0.35-0.88), p.013. No significant differences by treatment were reported. Overexpression of MYC by IHC had a relevant prognostic impact, with aHR 1.84 (0.99-3.44), p.054. By IHC, 3-years PFS for double negative vs single BCL2 or MYC overexpression vs double positive, was 85% vs 68% vs 51% respectively, with an aHR for double expressors compared to double negative of 3.91 (1.13-13.53), p.031. At the time of the present report, FISH analysis was conducted in 88 cases: 43 were triple negative, 37 single hit and 8 double/triple hit. By FISH, 3-years PFS for triple negative vs single hit vs double/triple hit was 74% vs 84% vs 25% respectively, with an aHR for double/triple hit compared to triple negative of 5.73 (2.05 to 16.02), p.001. Conclusions. In conclusion, with the limit of the analysis performed by IHC based on Hans' algorithm, BIO-DLCL04 showed an unexpected better outcome for non-GC compared to GC, irrespective of treatment arm. The ongoing analysis conducted by Nanostring will be more informative. The overexpression of MYC was an unfavourable risk factor, mainly if associated with BCL2 overexpression, irrespective of type of treatment. Moreover, double/triple hit patients represent a subgroup with extremely poor prognosis. High dose therapy plus autotransplant was not able to reverse the inferior outcome of neither double expressors nor double hit patients and new strategies are deemed for these poor prognosis patients. Disclosures No relevant conflicts of interest to declare.

Abstract: Background: Peripheral T-cell lymphomas (PTCL) have an aggressive clinical course with a poor 5-year overall survival with conventional therapy. Autologous stem cell transplantation (autoSCT) and allogeneic stem cell transplantation (alloSCT) can improve long-term disease control in first and second remission, respectively. Prognostic factors are not able to discriminate patients with chemorefractory disease. Preliminary evidence from gene expression profiling (GEP) and immunoistochemical study have suggested that the majority of PTCL-not otherwise specified (PTCL-NOS) can be subdivided in two different groups based on GATA3 and Tbet (transcription regulators of T-helper 2 and T-helper 1 lymphocytes) expression with a poorer prognosis in the former group. Because GEP studies are not feasible in routine clinical practice, we investigated whether expression changes of those proteins could be evaluated by immunohistochemistry (IHC) and used to predict prognosis. Material and Methods: We collected paraffin tissues from 63 consecutive patients (pts) with a diagnosis of PTCL [subtypes: PTCL-NOS (n=32), ALK-negative anaplastic large cell lymphomas (n=15, ALCL ALK-negative), angioimmunoblastic (n=16, AITL)] treated at different Italian Institutions. Sections were analyzed for Ki-67, GATA3, Tbet and CCR4 expression by IHC. Results were expressed as percentages of positive tumor cells. Histology was centrally reviewed. Median age was 58 years (range, 18-76 years); 20 of 63 (32%) pts were characterized by intermediate-high/high IPI; 47 of 63 (74%) pts were candidated to transplantation whereas 16 (25%) were not due to age ≥65 years (n=14) or limited stage/IPI 0 (n=2). Fifty-nine pts (93%) received anthracycline-based induction chemotherapy. Twenty of 47 (43%) pts underwent autoSCT in first (n=11) or subsequent remissions; 23 of 47 (49%) underwent alloSCT in first remission (n=3) or at relapse (n=20). The median follow-up of alive pts was 30 months (range, 8-250). Results: Expression of GATA3 was highly variable, but was significantly higher on average in PTCL-NOS [34% positive tumor cells (range, 1-99%)] than in other subtypes [11% (range,1-97%) in ALK-negative ALCL (p=0.027); 12% (range, 1-82) in AITL (p=0.0025)] ; The mean value of Ki67 expression was higher in ALK-negative ALCL [87% (range, 70-95%)] as compared to other subtypes [46% (range,15-90%) in PTCL-NOS (p 〈 0.0001)]; 42% (range, 15-70%) in AITL (p 〈 0.0001)]; Tbet and CCR4 expression was not significantly different among the subtypes. In PTCL-NOS (n=32), ROC curve analysis identified a cut-off of 27% for GATA3 with 16 pts showing a GATA3 expression higher than the cut-off value. While no significant difference in baseline clinical characteristics was observed among the two groups, pts with GATA3 ≥27% showed a significantly lower CR rate following induction [19% versus 62% (p=0.02)]. Cases with high GATA3 expression were significantly associated with a reduced 5-year PFS and OS as compared to those with low expression [PFS: 6% (95%CI:1%-24%) versus 49% (95%CI:22%-71%), (p=0,004); OS: 39% (95%CI:13%-65%) versus 72% (95%CI:35%-90%) (p=0,04) respectively] . In multivariate analysis including age, IPI score ≥2 and sex, high GATA3 expression retained a strong prognostic value in terms of PFS (p 〈 0.01) and OS (p=0.04). In others hystotypes, high GATA3 expression did not predict the outcome. Conclusions: Our analysis identifies GATA3 ICH expression as a strong prognostic and predictive biomarker among PTCL-NOS. Pts with high GATA3 expression values were characterized by chemorefractory disease and poorer outcome even with transplantation strategies. Validation of this biomarker in a larger series of patients is ongoing. Disclosures No relevant conflicts of interest to declare.

Abstract: Abstract 3412 The phase II explorative study of intermittent Imatinib (IM) treatment (InterIM) in elderly patients with Ph + chronic myeloid Leukemia (CML) who achieved a stable complete cytogenetic response (CCgR) after at least 2-years standard IM therapy (any dose between 300 and 800 mg/day) was started in April 2008 and closed for the enrollment in August 2009, since more than 78 patients required by statistics were included into the study. The main objective of the study was to investigate if after 12 months (trial time) the CCgR achieved with standard (daily administration) IM therapy could be maintained with InterIM. For this purpose, the CgR status was assessed by Interphase Fluorescence In Situ Hybridization (I-FISH) on peripheral blood (≥ 200 cells counted) every 3 months. When I-FISH (% Ph + nuclei) increased more then 1%, chromosome banding analysis (CBA) on bone marrow was performed to confirm the loss of CCgR and to check for additional cytogenetic abnormalities (ACA). At the present time, out of the 95 patients who were enrolled, 82 patients were evaluable and out of them 77 (94%), 73 (89%), 71 (87%) and 70 (85%) completed 3, 6, 9 and 12 months of the treatment program, respectively. Therefore, the great majority of patients completed the study core and at the end of 2010 all the patients are expected to complete the trial time (12 mo). During the first 12 months of InterIM, 1% to 11% of the evaluable patients at 3, 6, 9 and 12 months showed an I-FISH 〉 1% Ph+ nuclei (Figure 1). Figure 1 Distribution of patients according to I-FISH Figure 1. Distribution of patients according to I-FISH Totally, eleven (13%) out of 82 patients treated with InterIM showed an I-FISH 〉 1% and they were checked by CBA on bone marrow (Figure 2). Out of them only 3 cases, that means 4% of the 82 evaluable patients, lost the CCgR and resumed standard IM therapy (daily administration), but none completed 3 months of therapy. All the patients lost the MMR and increased several folds the BCR-ABL transcript levels. Two pts had a low risk Sokal and one a high risk; age was 66, 69, 77 years; time from diagnosis was 29, 91 and 100 months; duration of IM therapy was 29, 83 and 84 months; the IM dose was 400mg in all cases. Figure 2 Cytogenetic and molecular response in 11 cases who showed I-FISH 〉 1% + nuclei and who were checked by CBA on bone marrow. Black boxes shows the 3 cases who lost the CCgR Figure 2. Cytogenetic and molecular response in 11 cases who showed I-FISH 〉 1% + nuclei and who were checked by CBA on bone marrow. Black boxes shows the 3 cases who lost the CCgR As concern as molecular response, 99% of the patients had a major molecular response (MMR= 〈 0.001-0.1 BCR-ABL/ABLISX 100) at the baseline. The proportion of the patients who maintained the MMR after 3, 6, 9 and 12 months of InterIM was 95%, 92%, 91%, 84%, respectively. Interestingly, we found a weak but significant correlation between the % of BCR-ABL + nuclei and the BCR-ABL transcript levels in the patients who completed the trial time (12 mo) (r=0.27; p=0.001). In conclusion, the results of the InterIM study core (12 months), clearly show that Intermittent Imatinib (IM) treatment (InterIM) is sufficient to maintain the complete cytogenetic response (CCgR) previously achieved with standard IM therapy in elderly (≥ 65 years) Ph+ CML patients. The risk to loose the CCgR has been very low (4%), while the benefit either in terms of reduction of IM dose and of costs of therapy or in terms of compliance (data not shown) was very high. Acknowledgments: This work was supported in part by CML-Leukemia Net and Progetto Regione Lombardia. Disclosures: No relevant conflicts of interest to declare.

Abstract: Primary testicular lymphoma (PTL) has poor prognosis with failures in contralateral testis, CNS, and extranodal sites. To prevent these events, we designed an international phase II trial (International Extranodal Lymphoma Study Group 10 [IELSG-10]) that addressed feasibility and activity of conventional chemoimmunotherapy associated with CNS prophylaxis and contralateral testis irradiation. The trial was conducted by the IELSG and the Italian Lymphoma Foundation. Patients and Methods Fifty-three patients (age 22 to 79 years) with untreated stage I or II PTL were treated with six to eight courses of rituximab added to cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) every 21 days (R-CHOP21); four doses of intrathecal methotrexate (IT-MTX) and radiotherapy (RT) to the contralateral testis (30 Gy) for all patients and to regional lymph nodes (30 to 36 Gy) for stage II disease. Results All patients received R-CHOP21, 50 received CNS prophylaxis, and 47 received testicular RT. With a median follow-up of 65 months, 5-year progression-free survival and overall survival rates were 74% (95% CI, 59% to 84%) and 85% (95% CI, 71% to 92%), respectively. Ten patients relapsed or progressed: two in lymph nodes, five in extranodal organs, and three in the CNS. The 5-year cumulative incidence of CNS relapse was 6% (95% CI, 0% to 12%). No contralateral testis relapses occurred. Ten patients died: lymphoma (n = 6), secondary leukemia (n = 2), heart failure (n = 1), and gastric cancer (n = 1). Grade 3 to 4 toxicities were neutropenia, 28%; infections, 4%; and neurologic, 13%. No deaths occurred as a result of toxicity. Conclusion This international prospective trial shows that combined treatment with R-CHOP21, IT-MTX, and testicular RT was associated with a good outcome in patients with PTL. RT avoided contralateral testis relapses, but CNS prophylaxis deserves further investigation.