Abstract: Abstract 4271 The achievement of Complete Cytogenetic Response (CCyR) (Ph+ cells 0%) with Imatinib treatment still remains the most important objective in Chronic Myelogenous Leukemia (CML) patients. According to The European Leukemia NET guidelines, CCyR should be achieved within the 12th month of treatment while at 3 months of treatment the goal should be complete haematological response. To address the prognostic role of the early achievement of CCyR, we revised 108 chronic phase CML patients [M/F 57/51, median age 54.9 years, interquartile range (IR) 40.8 – 68.1] treated with front-line Imatinib at our Institution from June 2002 to June 2008 who had evaluable karyotype after 3 months of treatment. At onset, median WBC and PLT counts were 78.4 × 109/l (IR 34.0 – 135.9) and 399 × 109/l (IR 282 – 585), respectively. Sokal risk score was low in 52 patients (48.1%), intermediate in 49 (45.4%) and high in 7 (6.5%); a short pre-treatment phase ( 〈 3 months) with Hydroxyurea was administered to 94/108 patients (87%). After 3 months of Imatinib treatment, 84 patients (77.7%) achieved CCyR while 24 patients (22.3%) still presented Ph+ metaphases (median value 40%, IR 20 - 80) at cytogenetic analysis. At univariate analysis, factors with a negative prognostic impact on achievement of CCyR at 3 months were palpable spleen enlargement (p=0.002), WBC count at onset 〉 100.0 × 109/l (p=0.01) and pre-treatment with Hydroxyurea (p=0.032); on the contrary, sex, age, Sokal risk score and PLT value did not appear to affect early CCyR achievement. Among the 84 patients in CCyR after 3 months, there were 10 failures during follow-up (6 cytogenetic relapses, 2 molecular relapses and 2 evolution to blastic phase); among the 24 patients who did not achieve early CCyR, there were 12 failures during follow-up (9 primary resistances and 3 cytogenetic relapses). The difference in the incidence of failures during follow-up in the 2 groups was highly significant (p 〈 0.001). In conclusion, the achievement of CCyR at 3 months seems unrelated to traditional prognostic factors (Sokal); furthermore, in our experience it appears to have an important prognostic role, as patients not achieving it showed a significantly higher rate of failures during the follow-up. Disclosures: No relevant conflicts of interest to declare.

Abstract: The DASISION and ENESTnd controlled clinical trials have changed the front-line treatment of Chronic Myelogenous Leukemia (CML) leading to the advent of dasatinib and nilotinib in this setting: however, both these company-sponsored trials had many exclusion criteria, with a possible selection bias compared to the real-life CML population. To address the impact of these exclusion criteria on the 1st line treatment in the current clinical practice, we revised 207 unselected newly diagnosed chronic phase CML patients [M/F 108/99, median age 58.8 years, interquartile range (IR) 42.3 – 70.2] treated with front-line imatinib from June 2002 to June 2013 at our Institution and evaluated how many of them would have been excluded from enrolment in the 2 trials. Among these 207 patients, 28 patients should have been excluded by both trials due to polycomorbidities (12 cases), severe cardiopathy (5 cases), age 〉 80 with frailty (3 cases), drug abuse (2 cases), severe liver impairment, Rendu-Osler disease, active prostatic cancer, chronic obstructive broncopulmonar disease (COPD) + peripheral arterial obstructive disease (PAOD), COPD + arrhythmia, refusal to any marrow examination (1 case each). In addition, 8 patients should have been considered not eligible only for the DASISION due to isolated COPD and 19 patients should have been considered not eligible only for the ENESTnd due to isolated diabetes (10 cases), arrhythmia (4 cases), acute myocardial infarction 〉 6 months before CML diagnosis (2 cases), chronic pancreatic disease (2 cases), PAOD (1 case). On the whole, 36/207 patients (17.4%) would have been considered not eligible for the DASISION trial and 47/207 (22.7%) for the ENESTnd trial. As expected, these patients potentially not eligible for DASISION and ENESTnd were significantly older and with the imatinib treatment had a worse follow-up in terms of Complete Cytogenetic Response (CCyR), Major Molecular Response (MMolR) and Overall Survival (OS) compared to patients potentially eligible, as shown in the Table. In conclusion, our data highlight that an important fraction of newly diagnosed patients in a real-life setting would have been excluded by the 2 controlled trials whose results are the current mainstay of the 1st line treatment in CML: thus, an automatic transposition of those results into the current clinical practice should be regarded with caution. Disclosures: No relevant conflicts of interest to declare.

Abstract: Imatinib mesylate was given to 26 Philadelphia positive (Ph+) chronic myeloid leukemia (CML) patients who were in late chronic phase (CP) and in stable complete cytogenetic response (CCR) after interferon-alfa (IFN-α), but showed persistent positive residual disease at PCR analysis under this treatment. At diagnosis median age was 40 years (range 21–64) and according to Sokal’s score, 18 patients were low risk and 8 were intermediate risk. Median IFN treatment was 88 mo.s (range 15–202) and median CCR duration was 73 mo.s (range 10–148). Imatinib was administered at the standard dose of 400 mg/die, after stopping IFN for 1 week. Residual disease was measured on bone marrow (BM) cells at baseline, before starting Imatinib, at 3, 6, 12, 18 mo.s and at the last follow-up (median 32 mo.s, range 21–49), by assaying BCR-ABL transcripts using quantitative PCR (RQ-PCR). The copy number (CN) of BCR/ABL and ABL transcript were derived by the interpolation of CT values to the appropriate standard curve, and the result, for each sample, was expressed as ratio of BCR/ABL mRNA copies to ABL mRNA x 100 (normalized copy number - NCN). Imatinib treatment resulted in a progressive and consistent decline of residual disease in all but one patient, from a median of 0.89 at baseline to 0.01 at the end of follow-up. Major molecular response (BCR/ABL levels 〈 0.1) was reached in 20 patients (77%) and BCR/ABL transcripts were undetectable in 13 (50%). Achievement of molecular response was significantly correlated with post-IFN baseline transcript level (mean 1.194 for patients achieving complete molecular response vs 18,97 for those who did not; p 〈 0.001), but not with other clinical/biological patient characteristics. In all patients, imatinib was well tolerated with no side effects requiring drug dose reduction or dose discontinuation. Albeit obtained from an unusual subset of selected patients with favourable prognosis, and likely particularly sensitive to imatinib, present results confirm the efficacy of combining Imatinib and IFN-α and further support investigating treatment approaches employing these two drugs.

Abstract: Background Early occurrence of severe (≤ 10 g/dl) anemia has been reported in about 5% of newly diagnosed patients with Chronic Myeloid Leukemia (CML) treated with imatinib and is generally transient: in patients responsive to imatinib, however, severe late anemia seems to occur in a higher rate and seems to be more common in elderly patients. Methods To highlight this issue, we revised retrospectively 81 CML patients aged 〉 60 years treated at our Institution with frontline imatinib for at least 24 months who achieved a durable complete cytogenetic response (CCyR). The main clinical features of the patients at diagnosis were as follows: male/female 42/39 (51.9%/48.1%), median age 70.5 years [interquartile range (IQR) 64.5 - 76.5], median Hb 12.8 g/dl (IQR 11.2 - 13.7), median WBC 42.5 x 109/l (IQR 25.4 - 78.9), median PLTS 448 x 109/l (IQR 282 - 717). According to Sokal risk classification, 12 patients (14.8%) were low risk, 58 (71.6%) intermediate risk, 11 (13.6%) high risk. Severe late chronic anemia was defined as the presence of persistent ( 〉 6 months) and otherwise unexplained (creatinine level 〈 2 mg/dl, normal iron balance, bilirubin level 〈 2 mg/dl, folate and Vitamin B12 in the normal range) Hb levels ≤ 10 g/dl which occurred 〉 6 months from imatinib start. Results On the whole, a condition of late chronic anemia occurred in 22 out 81 patients (27.2%) at different intervals from imatinib start: the incidence was 13.5% (11/81 patients) at the 12th month, 14.8% (12/81 patients) at the 24th month, 12.3% (9/73 patients) at the 36th month, 7.0% (4/57 patients) at the 48th month and 5.9% (3/51 patients) at the 60th month. Seven out 22 patients (31.8%) needed packed red cell transfusions during the follow-up. Clinical features at diagnosis of patients who had severe late chronic anemia compared to the remaining 59 patients without anemia are shown in the Table: Among the 22 patients with severe late chronic anemia, 6 were treated with subcutaneous recombinant alpha-erythropoietin (EPO) at the standard dosage of 40,000 UI weekly, after a median time from imatinib start of 44.7 months (absolute range 26.8 - 73.8): in all 6 patients, baseline endogenous EPO levels were 〈 100 mUI/ml. All 6 patients achieved erythroid response, which was complete (achievement of stable Hb levels 〉 12 g/dl) in 4 patients and partial (stable increment 〉 1.5 g/dl of Hb levels with Hb levels 〈 12 g/dl) in 2 patients: one patient had a relapse of anemia after 24.1 months from EPO start and stopped EPO treatment, the remaining 5 patients are still in response and in treatment with EPO. No EPO-related toxicity was observed. Cumulative 4-year Event-Free Survival (EFS) for patients with severe late chronic anemia was 69.7% (CI95% 49.3 - 90.1) compared to 86.2% (CI95% 77.4 - 95.0) for patients without anemia (p=0.075): cumulative 4-year Overall Survival (OS) for patients with severe late chronic anemia was 89.6% (CI95% 75.9 - 100) compared to 94.8% (CI95% 89.1 - 100) for patients without anemia (p=0.204). Conclusions The occurrence of a late severe chronic anemia during long-lasting treatment with imatinib has been observed in 〉 25% of our responsive elderly patients, with a trend for a worse EFS: its occurrence seems more common in very elderly patients with lower Hb levels at diagnosis. Results with EPO are encouraging, but larger studies are warranted to define the role of such an approach in treating this common late complication of prolonged imatinib therapy. Table Table. Disclosures Breccia: Celgene: Honoraria; Novartis: Consultancy, Honoraria; Bristol Myers Squibb: Honoraria; Ariad: Honoraria; Pfizer: Honoraria. Latagliata:Novartis: Consultancy, Honoraria; Bristol Myers Squibb: Honoraria; Celgene: Honoraria; Janssen: Consultancy, Honoraria; Shire: Honoraria.

Abstract: Abstract 3769 Anemia has been reported in about 5–8% of newly diagnosed patients with Chronic Myelogenous Leukemia (CML) treated with imatinib in the first 6 months of therapy, and has been generally regarded as transient like neutropenia and thrombocytopenia: in patients responsive to imatinib, however, it is still unclear if the prolonged treatment could induce the appearance of a late chronic anemia. To highlight this issue, we revised 104 CML patients (M/F 54/50, median age at diagnosis 58.0 years, interquartile range 43.3 – 70.2) treated at our Institution with 1st line imatinib therapy for at least 36 months and in stable complete cytogenetic response (CCyR). At the 36th month of imatinib treatment, a late chronic anemia (defined as Hb levels 〈 11 g/dl for 〉 6 months) was present in 16/104 patients (15.4%): the anemia was severe (Hb 〈 10 g/dl) in 10 patients (9.6%) and moderate/mild (Hb ≥ 10 〈 11 g/dl) in 6 patients (5.8%). Clinical features at diagnosis of patients who had late chronic anemia at the 36th month compared to the remaining 88 patients without late anemia are shown in the Table: Hb 〈 11 g/dl at the 36th month Hb ≥ 11 g/dl at the 36th month p M/F 6/10 48/40 0.209 Median age (yrs) (IR) 68.0 (58.9 – 75.4) 56.6 (43.1 – 69.1) 0.016 Sokal Score: Low 3 44 0.002 Int 8 39 High 5 5 Median Hb (g/dl) (IR) 11.6 (10.4 – 13.3) 12.7 (11.2 – 13.8) 0.098 Median WBC (× 109/l) (IR) 79.0 (33.2 – 136.6) 60.5 (31.8 – 112.4) 0.423 Median PLTS (× 109/l) (IR) 624 (299 – 835) 412 (296 – 557) 0.122 All patients with late chronic anemia had a low reticulocyte count and 7/16 a condition of iron deficiency with reduced serum ferritin but no clinical and instrumental sign of chronic blood loss: in this latter group, oral iron supplementation was always ineffective. Four out 16 patients (25%) needed 1 or more red cell transfusions during the follow-up. At landmark analysis from the 36th month of imatinib treatment, there was no significant difference in the cumulative 4-year overall survival for patients with and for those without late chronic anemia (80%, 95% CI 44.8 – 100%, vs 94%, 95%CI 85.8 – 100%, respectively; p=0.068). In conclusion, the occurrence of a late chronic anemia during long-lasting treatment with imatinib has been observed in 〉 15% of our responsive patients and was severe in about 10% of cases: its occurrence seems more common in elderly patients with higher Sokal score at diagnosis. Late chronic anemia does not seem to affect OS, but the real impact should be evaluated on a large cohort of patients. Disclosures: Tafuri: Sigma Tau Pharmaceuticals: Research Funding.

Abstract: According to the World Health Organization (WHO) 2008/2016 criteria for classification of myeloid neoplasms, a platelet (PLT) count ≥ 450X109/l, thus reduced from the previous WHO 2001 level ≥ 600 x 109/l, was considered the new PLT threshold for the diagnosis of Essential Thrombocythemia (ET). Aim of the study was to validate in a setting of current clinical practice this important diagnostic change and compare clinical and hematological features at diagnosis and during follow-up of patients with PLT ≥600 x 109/l versus patients with PLT 〈 600 x 109/l. We retrospectively analyzed data from 264 patients with ET according to WHO 2008/2016 criteria, enrolled in our center from 1/2008 to 12/2017. Patients were divided into Group A (G-A) (PLT ≥600 x 109/l at diagnosis) (199 patients - 75.4%) and Group B (G-B) (PLT ≥ 450 x 109/l 〈 600 x 109/l at diagnosis) (65 patients - 24.6%) and compared for clinical features at the onset, clinical course and follow-up. Main features and commonly recognized pro-thrombotic risk factors at diagnosis of the entire cohort as well as of G-A and G-B are reported in the Table 1. Among clinical features, only the median value of leukocytes was significantly higher in G- A [9.1 x 109/l, interquartile range (IQR) 7.8-10.3 vs 7.4 x 109/l, IQR 6.0-9.6; p = 0.001]. Among pro-thrombotic risk factors, only the median cholesterol value was significantly lower in the G-A [187 mg/dl (IQR 164-220) vs 204 mg/dl (RIQ 177-238); p = 0.048] . Cytostatic treatment was administered in 175 patients (71.1%) of entire cohort at different intervals from diagnosis, with a significantly higher rate in patients of G-A (76.9% versus 49.2%, p 〈 0.001). After a median follow-up of 37.5 months (IQR 19.8 - 60.7), 13 thrombotic events (4.9%) were recorded in the entire cohort (7 episodes in the G-A and 6 episodes in the G-B), with a 5-year Cumulative Incidence of Thrombosis (CIT) significantly higher in the G-B [79.6% (95%CI 59.6 - 99.6) versus 95.4% (95%CI 91.8 - 99.0); p=0.047] (Figure 1). No patient evolved in myelofibrotic phase, 2 patients evolved in blastic phase (BP) after 42 and 58 months, respectively [1 patient (0.5%) in the G-A and 1 patient (1.3%) in the G-B; p=0.40). At the last follow-up, 4 patients (1.5%) died (1 from BP, 1 from cerebral hemorrhage, 2 from unavailable cause), 15 (5.7%) were lost to follow-up and 245 (92.8%) are still alive and currently followed at our Institute. The 5-year Overall Survival (OS) of the entire cohort was 96.2% (IC95% 92.2 - 100), without differences between the two groups [96.3% (95% CI 92.0 - 100) in the G-A versus 96.7% (IC95% 91.7 - 100) in the G-B; p=0.898] . Our data indicate a substantial homogeneity among ET patients regardless of the PLT number at diagnosis, thus confirming the usefulness of 2008/2016 WHO diagnostic criteria. Furthermore, the counterintuitive lower CIT observed in G-A, due to a larger use of cytostatic treatments and/or to an acquired Von Willebrand phenomenon when PLT levels 〉 1.000 x 109/l, highlights how thrombotic risk is unrelated to PLT value and leads to consider the administration of adequate cytostatic therapy even in patients with relatively lower PLT count at diagnosis. Disclosures Breccia: Novartis: Honoraria; Pfizer: Honoraria; Incyte: Honoraria; BMS: Honoraria. Foà:INCYTE: Other: ADVISORY BOARD; JANSSEN: Other: ADVISORY BOARD, Speakers Bureau; GILEAD: Speakers Bureau; CELTRION: Other: ADVISORY BOARD; ABBVIE: Other: ADVISORY BOARD, Speakers Bureau; CELGENE: Other: ADVISORY BOARD, Speakers Bureau; AMGEN: Other: ADVISORY BOARD; ROCHE: Other: ADVISORY BOARD, Speakers Bureau; NOVARTIS: Speakers Bureau.

Abstract: Background : Unlike to Essential Thrombocythemia, Polycythemia Vera (PV) is rare in women of childbearing age, with only few previous pregnancies reported in literature; a positive outcome of pregnancy has been described in about 2/3 of cases and maternal complications in about 1/4 of delivery. Aims and Methods : We retrospectively analyzed outcome and complications in a series of 11 pregnancies in 7 females with PV. These patients were diagnosed and managed between 2000 and 2017 in 3 Italian Centers. PV diagnosis was perfomed or revised according to WHO 2016 criteria. Different therapeutic approaches have been assessed prior and during each pregnancy. Results : Median age at diagnosis and at conception were 24.3 and 30.1 years, respectively. According to molecular status, all patients presented JAK2V617F mutations. Four patients (57.1%) had two pregnancies. All patients received phlebotomies as required, with a median hematocrit (Ht) level at the time of conception of 45.2% (range: 41.4-46.8). Median values of WBC and PLT at conception were 9.9 x 109/l (range: 6.7-18.3) and 510 x 109/l (250-917). Only one patient presented a history of thrombosis (2 TIA) occurred before first pregnancy. Three (42.8%) females presented cardiovascular risk factors at the time of conception, while a thrombophilic predisposition was documented in 3 patients (42.8%). Among the 11 pregnancies, 8 (72.7%) ended with a full term delivery, while 2 (18.2%) were complicated by a fetal loss in the first trimester and by an Intrauterine growth retardation with preterm delivery. It is worth of note that the fetal loss was reported in the patient with prior arterial thrombosis and concomitant signs of myeloproliferation (mild leukocytosis, Ht level 〉 45% and PLT level 〉 900 x 109/l), while the intrauterine growth retardation was reported in a patient with Ht level 〉 45%, despite the typical hemodiluition of pregnancy. The remaining pregnancy was characterized by maternal complications, consisting of an extra-tubal pregnancy with consequent laparoscopic surgery approach and fetal death. The global live birth rate was 81.8 %. Anti-thrombotic treatment was administered in all but one pregnancy, consisting of low-dose acetil salicylic acid (ASA) already started since PV diagnosis and continued during pregnancy in 6 pregnancies (54.5%), or a combination of ASA and Low Molecular Weight Heparin (LMWH) during the second and last trimester of pregnancy and post-partum in the remaining 4 pregnancies (36.3%). Two patients were also treated with interferon-alfa during their pregnancy. One patient was receiving hydroxyurea at the time of conception, which was immediately stopped. Conclusions : Our data indicate that PV patients may develop pregnancy complications in about 1/4 of cases (27.2%), but the risk of fetal loss (18.2%) is lower than recently reported in similar series of PV females. The vast majority of our patients received ASA +/- LMWH during pregnancy and post-partum, suggesting an important role of anti-thrombotic treatment in reducing incidence of fetal complications. The history of prior thrombosis and the signs of myeloproliferation at conception and at delivery seem to be associated to development of fetal complications. Considering the current guidelines for the management of PV, in particular the more aggressive control of Ht (with Ht target 〈 45%), it is possible that the large application of this approach also in PV pregnancies could further improve the rate of fetal complications. Larger collaborative multicenter studies, in order to better clarify the optimal management of pregnancy in PV, are warranted. Disclosures Breccia: Pfizer: Honoraria; BMS: Honoraria; Novartis: Honoraria; Incyte: Honoraria. Gambacorti-Passerini:Pfizer: Consultancy, Honoraria, Research Funding; BMS: Consultancy.

Abstract: Background Deferasirox (DFX) is widely employed as iron chelation therapy (ICT) in the current clinical practice in patients with myelodysplastic syndromes (MDS) and chronic transfusion need. The efficacy of DFX in reducing median ferritin levels in different cohorts of these patients has been reported in many trials, but the lack of worldwide accepted criteria of individual response to ICT makes it difficult to appreciate its clinical relevance for any single patient. Aim To highlight the clinical impact of ICT with DFX in a large real-life cohort of MDS patients, based on different individual ferritin variation during treatment. Methods A retrospective cohort of 301 consecutive MDS patients [M/F 187/114 (62.1%/37.9%)] of any age followed in 20 hematological Centers in Italy was analyzed: the main features at diagnosis are reported in the Table 1. Individual response to ICT was categorized as complete response (CR) (ferritin levels 〈 500 ng/ml), partial response (PR) (ferritin levels 〈 1,000 ng/ml), ferritin improvement (FI) (ferritin reduction 〉 50% of baseline value but with levels 〉 1,000 ng/ml), ferritin stability (FS) (ferritin levels without changes from baseline during ICT) or no ferritin response (NR) (ferritin levels increasing during ICT). Results ICT was started after a median period from diagnosis and from transfusion start of 21.0 months [interquartile range (IQR) 8.9 - 44.3] and 11.3 months (IQR 7.1 - 21.7), respectively, with a median burden of red cell transfusions at baseline of 22 units (IQR 14 - 35). The main features of patients at baseline of ICT are reported in the Table 1. Starting DFX dose was 〈 10 mg/Kg in 38 patients (12.7%), 10 - 14 mg/Kg in 110 patients (36.6%), 15 - 19 mg/Kg in 57 patients (18.9%) and ≥ 20 mg/Kg in 96 patients (31.9%). As to individual response, 4 patients (1.3%) were too early for evaluation ( 〈 6 months of DFX treatment): in addition, 16 patients (5.4%) discontinued ICT behind 6 months from start, due to early toxicity (10 patients, 7 for gastro-intestinal toxicity and 3 for skin toxicity) or other reasons (unrelated death, AML evolution, transplant procedure). Among the remaining 281 patients, 37 (12.3%) achieved a CR, 65 (21.6%) a PR, 23 (7.6%) a FI, 112 (37.2%) a FS and 44 (14.6%) a NR. Five-year overall survival (OS) of the whole cohort from ICT start was 43.9% (95%CI 37.1 - 50.7). Five-year OS according to ICT response was 74.8% (95%CI 57.9 - 91.7) in patients with CR, 51.7% (95%CI 37.6 - 65.8) in patients with PR, 50.6% (95%CI 28.2 - 73.0) in patients with FI, 38.6% (95%CI 27.0 - 50.2) in patients with FS and 21.1% (95%CI 5.2 - 37.0) in patients with NR (p=0.002) (Figure 1). Five-year cumulative incidence of AML evolution (CIE) of the whole cohort from ICT start was 27.1% (95%CI 20.3 - 33.9). Five-year CIE according to ICT response was 7.6% (95%CI 0 - 18.0) in patients with CR, 27.0% (95%CI 13.0 - 40.5) in patients with PR, 38.3% (95%CI 15.5 - 61.7) in patients with FI, 20.8% (95%CI 10.4 - 31.2) in patients with FS and 57.7% (95%CI 31.9 - 83.5) in patients with NR (p=0.003) (Figure 2). Notably, no statistical difference was observed for both OS and CIE among patients achieving PR, FI or FS. Conclusions Present data highlight the clinical relevance of individual response in MDS patients receiving ICT with DFX. In particular, achievement of CR seemed related to a better OS and a lower CIE, while patients with NR had a significant worst OS and CIE: furthermore, the achievement of stable ferritin levels was associated with similar OS and CIE than PR and FI and thus should be considered as a response. Disclosures Latagliata: Celgene: Honoraria; Janssen: Honoraria; Novartis: Honoraria; Pfizer: Honoraria. Oliva:Novartis: Consultancy, Speakers Bureau; Celgene Corporation: Consultancy, Honoraria, Speakers Bureau; Apellis: Consultancy. Pilo:Novartis: Other: Advisory board. Molteni:Celgene: Membership on an entity's Board of Directors or advisory committees. Balleari:Celgene: Membership on an entity's Board of Directors or advisory committees. Breccia:Novartis: Honoraria; BMS: Honoraria; Pfizer: Honoraria; Incyte: Honoraria; Celgene: Honoraria. Foà:Incyte: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Celltrion: Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy, Speakers Bureau; Roche: Consultancy, Speakers Bureau; Incyte: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Roche: Consultancy, Speakers Bureau; Celltrion: Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Pfizer: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Amgen Inc.: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Shire: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Pfizer: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Abbvie: Consultancy, Speakers Bureau; Shire: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Amgen Inc.: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Finelli:Novartis: Consultancy, Speakers Bureau; Janssen: Consultancy, Speakers Bureau; Celgene Corporation: Consultancy, Research Funding, Speakers Bureau.

Abstract: Abstract 1777 Poster Board I-803 Introduction The proteasome, a multicatalytic enzyme complex, plays an important role in the activation of nuclear factor kappa B (NF-kB), through degradation of its suppressor IkB. NF-kB can intervene in oncogenesis with its capacity to regulate the expression of genes that modulate apoptosis, cell survival as well as proliferation, inflammation, angiogenesis and tumour metastasis. Bortezomib (Velcade, formerly PS-341) is a proteasome inhibitor with documented antitumor activity in multiple myeloma and other lymphoid malignancies and has also been investigated in myeloid malignancies, such as acute leukemias (AL) and myelodysplastic syndromes (MDS), indicating some therapeutic activities. We performed a Phase II study (EudraCT number 2004-002935-23) to investigate the efficacy and safety of bortezomib in patients with MDS. Patients and methods: Ten patients were enrolled in the study: there were 6 females and 4 males, median age was 64 years (range 44-70), median duration of MDS phase was 12 months. All patients but one received prior erythropoietin therapy without efficacy. 8/10 patients required transfusional therapy at the time of enrolment (median 4 units per month). Results According to WHO classification, 5 patients were refractory cytopenia with multilineage dysplasia (RCMD), 1 patient was pure refractory anemia (RA), 1 patient was RCMD with ringed sideroblasts (RCMD-RS), 3 patients were refractory anemia with excess of blasts type 1 (RAEB-1). According to IPSS stratification, 5 patients were classified as low risk, 3 patients as intermediate-1 and 2 patients as intermediate-2 risk. Six patients presented cytogenetic aberrations at baseline (isolated del(5q), 2 patients; del(5q) plus t(3;12), 1 patient; del(5q) and monosomy 7, 1 patient, isolated trisomy 8, 1 patient; trisomy 8 and trisomy 11, 1 patient). WT1 gene expression was tested in all cases: 9/10 patients displayed elevated WT1 expression level in bone marrow (median 219 copies, range 20-2682) and 7/10 also in peripheral blood (median 31 copies, range 11-2247). Bortezomib was administered at the dose of 1.3 mg/mq with a 1, 4, 8, 11 day schedule, every 28 days, for a maximum of 8 cycles. Four patients discontinued therapy after 1, 3, 3 and 4 cycles respectively, due to vascular disease not related to the drug in the first two cases and to disease progression in the other two (both patients with intermediate-2 risk), whereas 6 patients performed all planned eight cycles. As to safety, grade 3/4 neutropenia was recorded in 4 patients and grade 3/4 thrombocytopenia in 6 patients. Non-hematological side effects were recorded in 7 patients: diarrhea in 1 patient, fever in 3 patients, skin rash in 2 patients and pneumonia in 1 patient during neutropenic phase. As to response, 3 patients (50% of evaluable cases) achieved a minor erythroid response (according to IWG criteria) and 3 patients had a stable disease. We did not reveal modifications as to bone marrow blasts and cytogenetic changes with respect to pre-treatment findings. As to outcome, 7/10 patients are still alive at a median follow-up of 24 months. WT1 decrease was recorded in the 3 patients with minor erythroid response, both in bone marrow (from median 109 copies to median 14 copies at the end of treatment) and in peripheral blood (from median 70 copies to median 9 copies at the end of treatment), and in 1 patient with stable disease. In the remaining patients with stable disease, median bone marrow WT1 copies increased from 659 copies at baseline to 890 copies at the end of treatment, and to median 736 copies at baseline to 788 copies at the end of treatment in peripheral blood. Conclusions Present results indicate that bortezomib as single agent in MDS, seems to have some efficacy in terms of hematological improvement and to also affect the WT1 gene expression. Further studies on larger series of MDS patients are needed to confirm our observation. The combined use of this drug with other agents might potentiate its therapeutic activity. Disclosures No relevant conflicts of interest to declare.