Finazzi, Guido

Abstract Abstract 277 Background: Current information on the natural history of polycythemia vera (PV) is mostly derived from early studies based on the PV Study Group diagnostic criteria. The World Health Organization (WHO) diagnostic approach to PV is currently the gold standard. The main objective of the current study was to describe survival, disease complication rates and risk factors in WHO-defined PV. Methods: Under the auspices of an International Working Group, seven international centers of excellence for myeloproliferative neoplasms submitted a total of 1,430 patients, locally diagnosed with “WHO-defined PV”. The two principle investigators (AT and TB) reviewed all the cases and selected 1,263 patients who were age 18 years or older, diagnosed after 1970, and whose submitted data included diagnostically essential information. Results I: Presenting Features: Median age was 60 years (range, 18–95; 51% females). The number of patients diagnosed between 1970–1990, 1991–2000, and 2001–2011 were 136 (11%), 337 (27%) and 790 (62%), respectively. Arterial/venous thrombosis history before or at diagnosis was documented in 142/73 and 80/42 patients, respectively. Other features at diagnosis included pruritus (37%), microvascular disturbances (29%), palpable splenomegaly (37%), abnormal karyotype (13%), and leukoerythroblastosis (6%). History of hypertension (45%), hyperlipidemia (17%), diabetes (9%), and tobacco use (15%) was also obtained. Results II: Clinical Course: To date, 228 (18%) deaths, 42 (3%) leukemic progressions, and 101 (8%) fibrotic transformations have been recorded. Post-diagnosis arterial or venous thrombosis occurred in 139 (11%) and 85 (7%) patients, respectively. Cytoreductive therapy had not been used in 342 (27%) patients whereas 154 (12%) patients were exposed to two or more cytoreductive drugs. Results III: Risk Factors: Multivariable analysis identified advanced age, leukocytosis, venous thrombosis history and abnormal karyotype as significant (p<0.01) risk factors for overall survival (OS). Receiver operating characteristic (ROC) plots delineated ages 69 (AUC 0.83) and 59 (AUC 0.61) years and leukocyte count 13 × 109/L (AUC=0.61) as the best discriminant levels in this regard; HR (95% CI) were 11.4 (7.7–16.9) for age ≥70 years, 3.8 (2.6–5.7) for age 60–69 years, and 2.0 (1.5–2.6) for leukocyte count ≥15 × 109/L. Results IV: Risk Stratification: HR-based risk scores were assigned to age ≥70 years (3 points), age 60–69 years (2 points) and leukocyte count ≥15 × 109/L (one point), in order to create four risk groups: low (zero points; n=492; median survival 322 months), intermediate-1 (one or 2 points; n=353; median survival 208 months; HR 3.8, 95% CI 2.5–5.8), intermediate-2 (3 points; n=286; median survival 132 months; HR 10.7, 95% CI 6.9–16.5), and high (4 points; n=56; median survival 90 months; HR 22.7, 95% CI 13.9–37.1) (Figure 1). Similar results were obtained when analysis was restricted to the 1,011 JAK2 mutation-positive patients (Figure 2). The particular risk model was equally robust in predicting leukemia-free survival (p=0.001). Conclusions: Age and leukocyte count matter the most in predicting survival in PV. The current study suggests additional prognostic value from karyotype and venous thrombosis history. Disclosures: Vannucchi: Novartis: Honoraria. Gisslinger:Novartis: Speakers Bureau; Celgene Austria: Speakers Bureau; AOP-Orphan: Speakers Bureau.

Abstract Abstract 5148 INTRODUCTION. Patients presenting with a clinical picture of essential thrombocythemia (ET) can actually have an early/prefibrotic myelofibrosis (PMF), according to current WHO criteria, in about 18% of cases. Laboratory tests which are significantly different in early/prefibrotic PMF as compared with histologically confirmed ET (WHO-ET) include decreased gender-matched hemoglobin (Hb), increased white blood cell (WBC), platelet (PLT) counts and lactate dehydrogenase (LDH) values. AIM. To evaluate sensitivity (SE) and specificity (SP) of blood cell counts and LDH, at presentation, for the diagnosis of early/prefibrotic PMF vs. WHO-ET. METHODS. Five hundred thirty-six cases (50%) who had complete laboratory data measured at diagnosis constituted the exploratory set of our study and were derived from an international ET database. The discriminatory ability of Hb, WBC, PLT and LDH in correctly classifying patients in the early/prefibrotic PMF or WHO-ET groups was initially tested by plotting their Receiving Operating Characteristic (ROC) curves and comparing the relative Areas Under the Curve (AUC) with the value of 0.50 (which stands for the completely useless of the test). Three parameters with statistically significant discriminatory power were chosen (Hb, WBC and LDH) and thresholds searched in order to guarantee at least 90% of SE or SP. Finally, a diagnostic algorithm was designed. The validation set of this analysis was constituted by 321 patients with WHO-ET (n=62) or early/prefibrotic PMF (n=259) diagnosed by the same pathologist who confirmed the training set cohort and collected in the Institute for Pathology, University of Cologne, Germany. SE and SP for the same parameters and thresholds as well as the final diagnostic algorithm were applied to this set of patients to demonstrate the results' reproducibility. RESULTS. Sensitivity and specificity to recognize early/prefibrotic PFM have been evaluated by ROC curves. The best performance was found for LDH (AUC = 0.7059). WBC and Hb had super imposable curves, with AUC of 0.6279 and 0.6257, respectively. The worst performance was registered for PLT count: its AUC was only 0.5628, not significantly different from the reference value of 0.50 (p=0.154). Thresholds of Hb, WBC and LDH were searched to achieve at least 90% of SE or SP. HB < 12 g/dL for women or <13 g/dL for men, and WBC >= 13 x109/L had higher SP (92% and 91%, respectively). High SP is highly related to the presence of early/prefibrotic PMF (true positives). On the contrary, LDH < 200 mU/mL and WBC < 7 x109/L had good sensitivity (91% and 94%, respectively). High SE is highly related to the absence of early/prefibrotic PMF (true negatives). By applying these SE and SP values in a step-by-step algorithm, nearly half of patients (48%) could be classified as WHO-ET or early/prefibrotic PMF, assuming at each step a margin of error of less than 10%. For the remaining 50% of patients, laboratory results didn't allow to suspect or exclude the presence of early/prefibrotic PMF. In the validation set of 321 patients classified by WHO 2008 as true ET or early/prefibrotic PMF (Cologne cohort) SP of anemia was 84%, WBC < 7 x109/L or >= 13 x109/L had 91% and 81% of SE and SP, respectively. LDH values < 200 mU/mL had 85% of SE. By applying the same flow-chart, 46% of patients were classified as WHO-ET or early/prefibrotic PMF. CONCLUSIONS. The present study provides clinicians with laboratory parameters that should increase suspicion of early/prefibrotic PMF in a patient with a working clinical diagnosis of ET. In fact, while patients presenting clinically with ET can now be discriminated as true ET or early/prefibrotic PMF by adopting the WHO 2008 criteria that require bone marrow histology, an algorithm including baseline anemia, WBC count and LDH, allows this differentiation in about 50% of patients with a good approximation. However, for a definitive proof, bone marrow histology is still an integral part for final diagnosis. Disclosures: Vannucchi: Novartis: Honoraria.

Abstract Abstract 3846 INTRODUCTION There is strong evidence indicating that the clear-cut separation of prefibrotic primary myelofibrosis (PMF) from essential thrombocythemia (ET) by consequent application of the World Health Organization (WHO) 2008 criteria is reflected in different, well defined clinical pictures and divergent prognoses (Barbui et al, JCO 2011,Thiele et al, Blood 2011). Very recently the specific risk profile for arterial and venous thrombosis in WHO- diagnosed ET was published (Carobbio et al, Blood 2011). In PMF so far all published data on vascular events were exclusively based on overt disease manifestations until now there are no data available regarding prefibrotic stages. Consequently, we aimed to evaluate the corresponding risk profile of patients with WHO-diagnosed prefibrotic PMF. METHODS A total number of 264 patients with WHO-defined prefibrotic PMF derived from either the Medical University of Vienna or an International Database (Barbui et al, JCO 2011) were studied. Nonfatal thrombotic events considered in this study were reported as rates per 100 patient-years and included transient ischemic attacks, thrombotic cerebrovascular accidents, coronary artery disease, myocardial infarction, peripheral arterial disease, deep vein thrombosis of peripheral vasculature, pulmonary embolism, and abdominal large vein thrombosis. Cardiovascular risks factors considered, comprised of arterial hypertension, diabetes mellitus and tobacco use. To evaluate risk factors for total thrombosis and for arterial and venous events in particular multivariate Cox- regression analysis including the co-variables sex, age, previous thrombotic event, laboratory parameters measured at diagnosis and need for cytoreductive and/or antiplatelet therapy during follow-up was calculated. P values less than 0.05 were considered as statistically significant. RESULTS After a median follow- up of 5.8 years (range0.0 – 27.2), the total rate of non fatal thrombotic events was 2.1% patient-years (95% CI, 1.5–2.8); the incidence of arterial events was higher (1.7% patient- years) than of venous events (0.6% patient-years).Considering thrombosis in general a higher white blood cell (WBC) count enhances the risk significantly (p=0.005; HR 1.15). This is also true in arterial events in particular (p=0.047; HR 1.12). A lower platelet count is associated with a higher risk for thrombotic events; for thrombosis in general this association is of borderline significance (p=0.056; HR 0.99), for arterial thrombosis in particular of significance (p= 0.042; HR 0.99). A lower hemoglobin level is associated with a higher risk for venous thrombosis (p=0.007; HR 0.59). CONCLUSION Leukocytosis appears as a risk factor for thrombosis in general and also for arterial thrombosis in particular in WHO-diagnosed prefibrotic PMF. Moreover, higher platelet counts seem to decrease significantly the risk for thrombotic events in general and arterial thrombosis in particular. Anemia is associated with a higher risk for venous thrombosis. These observations are partly in line with recently published findings in WHO-diagnosed ET (Carobbio et al,2011) and might indicate the existence of a specific risk profile for thrombotic events in prefibrotic PMF. This is the first study reporting data on the risk profile for thrombosis in WHO-diagnosed prefibrotic PMF. Certainly these findings ask for validation in a larger patient population. Disclosures: Gisslinger: Novartis: Speakers Bureau; Celgene Austria: Research Funding, Speakers Bureau; AOP-Orphan Pharmaceuticals AG: Speakers Bureau. Vannucchi:Novartis: Honoraria.

Abstract Background and purpose of the study: Several reports have documented that histone deacetylase inhibitors induce neoplastic cells to undergo growth arrest, differentiation and/or apoptotic cell death. Among these agents, Givinostat inhibits proliferation of cells bearing the JAK2V617F mutation and has shown clinical significant activity with good tolerability in patients with chronic myeloproliferative neoplasms, including polycythemia vera (PV). In two phase II studies [Rambaldi et al. Br J Haematol 2010;150:446, Finazzi et al. Br J Haematol 2013;161:688] the maximum administered dose (150 mg/day) was generally well tolerated. Assuming a linear relationship between dose and efficacy, greater clinical efficacy could be expected with increased doses of Givinostat. Methods: This is a two-part, multicenter, open label, non-randomized, phase Ib/II study to assess the safety and tolerability, maximum tolerated dose (MTD) and preliminary efficacy of Givinostat in patients with JAK2V617F positive PV. Part A (i.e. phase Ib) is the dose escalation portion of the study, while Part B assesses the preliminary clinical efficacy of Givinostat at the MTD defined in Part A. Eligible patients for this study include JAK2V617F positive PV patients, who have active and/or uncontrolled disease, defined as: (i) Hematocrit (HCT) ≥ 45% or HCT <45% normalized by phlebotomy, and (ii) PLT > 400 x109/L, and (iii) WBC > 10 x109/L. Study therapy is administered in 28 day cycles (cycle is defined as 4 weeks of treatment). Disease response has been evaluated according to the clinico-haematological European LeukemiaNet (ELN) criteria [Barosi et al. Blood 2009;113:4829] at Cycles 3 and 6. Disease-related symptoms have been reported directly by the patients who completed the MPN-SAF QoL form. The study lasts up to a maximum of 24 weeks of treatment. However, after completion of the trial, all patients achieving clinical benefit are allowed to continue treatment with Givinostat in a long-term study (ClinicalTrials.gov ID: NCT01761968). In Part A, the Cycle 1 has been used to define the MTD of Givinostat. Only dose limiting toxicities (DLTs) experienced during Cycle 1 were considered for dose escalation decisions. Part B is the cohort expansion part of the study to assess the preliminary clinical efficacy of Givinostat at the MTD in evaluable patients Results: In Part A, 12 patients with active and/or uncontrolled disease, were treated with Givinostat, 6 patients at DL1 (100 mg b.i.d.), 3 patient at DL6 (150 mg/die) and 3 patients at DL0 (50 mg b.i.d.). Only one DLT .(grade 3 dyspepsia) was observed in the 6 patients treated at DL1 during the Cycle 1. Givinostat treatment was generally well tolerated: one patient developed grade 4 and two patients grade 3 thrombocytopenias at DL1 after the Cycle 1. Complete/partial response was observed in 100% of evaluable patients at Cycle 3 (Figure). At that time, an important reduction of WBC and PLT count (about 50% in both cases), an HCT control without phlebotomy, a net decrease of MPN-SAF total score and JAK2V617F allele burden (about 50% and 30%, respectively) have been noted and generally maintained at Cycle 6. One out of 3 (33%) patients with splenomegaly at baseline normalized the spleen at Cycle 6. PK/PD evaluations are currently under analysis. Part B started on December 2015 and is currently recruiting patients. At the time of ASH Meeting, a presentation of the preliminary results of the interim analysis (IA) is planned. The objective of this IA is to evaluate the response rate of the first 12 treated and evaluable patients. Conclusions: Taking into account the continuous schedule of the drug as prescribed in this current study, the daily dosage of 200 mg has been identified as the MTD of Givinostat in PV patients. Givinostat treatment is generally well tolerated and shows preliminary clinical efficacy in PV patients with an active/uncontrolled disease. Response rate of patients in Part A C3D28 = Cycle 3 Day 28, or three months of treatment. C6D28 = Cycle 6 Day 28, or six months of treatment. Two patients dropped-out the study Part A, due to Informed Consent Form withdrawal, while 1 patient left the study for AE (i.e. grade 3 dyspepsia, DLT). At C3D28, 1 Complete Response (CR) and 8 Partial Response (PR) have been observed. At C6D28, 1 Complete Response (CR), 7 Partial Response (PR) and 1 No Response (NR) have been observed Figure Figure. Disclosures McMullin: Novartis: Honoraria, Speakers Bureau. Vannucchi:Novartis: Honoraria, Speakers Bureau. Mesa:Ariad: Consultancy; CTI: Research Funding; Galena: Consultancy; Celgene: Research Funding; Promedior: Research Funding; Novartis: Consultancy; Incyte Corporation: Research Funding; Gilead: Research Funding. Tibes:Italfarmaco S.p.A.: Consultancy, Honoraria. Bettica:Italfarmaco S.p.A.: Employment. DI Tollo:Italfarmaco S.p.A.: Employment.

Abstract Background In the recent International Prognostic Score for Thrombosis in essential thrombocythemia (IPSET-thrombosis), age and history of thrombosis were confirmed as independent risk factors for future thrombosis and the study also identified independent prothrombotic role for cardiovascular (CV) risk factors and JAK2 V617F mutation (Barbui et al. Blood 2012). Methods In the current study, we re-analyzed the original IPSET-thrombosis data in 1019 patients with WHO-defined ET in whom JAK2 mutational status was available, in order to quantify the individual contributions of JAK2 mutations and CV risk factors in conventionally-assigned low and high risk ET, as well as in age- versus thrombosis-defined high risk status. Results After a median follow-up of 6.8 and 5.0 years in conventionally-assigned low- and high-risk patients, respectively, the overall annual rate of total thrombosis (108 events) in conventionally-assigned low- and high-risk patients was 1.11%-pt/y (CI 0.81-1.52) and 2.46%-pt/y (CI 1.94-3.11) respectively (p=0.001), and the difference was mainly due to a higher frequency of arterial thrombosis in high-risk patients (p<0.001).The influence of JAK2 mutational status and CV-risk factors on the rate of thrombosis in conventionally assigned low- and high-risk groups is presented in the table. Table 1. Additional risk factors N (%) Event Rate % pts/yr (95% CI) P-value P-value P-value trend Low risk 506 (50) 39 1.11 (0.81-1.52) None 200 (40) 7 0.44 (0.21-0.92) ref Cardiovascular risk factor 36 (7) 3 1.05 (0.34-3.25) 0.220 0.227 JAK2V617F 213 (43) 21 1.59 (1.04-2.44) 0.001 0.217 Both 52 (10) 8 2.57 (1.29-5.15) <0.001 ref <0.001 High risk 513 (50) 69 2.46 (1.94-3.11) None 111 (22) 10 1.44 (0.78-2.68) ref Cardiovascular risk factor 44 (9) 4 1.64 (0.62-4.37) 0.909 0.067 JAK2V617F 222 (43) 30 2.36 (1.65-3.38) 0.168 0.082 Both 136 (27) 25 4.17 (2.82-6.17) 0.011 ref 0.005 The number of major arterial and venous thrombosis was reported as rates per 100 patient-years and the difference among groups was assessed by Mantel Cox log-rank test i) Conventionally-assigned low-risk group. Amongst 506 patients, 200 (40%) displayed neither JAK2 mutation nor CV risk factors and their annual rate of thrombosis was 0.44%, as opposed to 1.05% in the presence of CV risk factors (P=NS), 1.59% in the presence of JAK2 mutation (p=0.001) and 2.57% in the presence of both CV risk factors and JAK2 mutation (P<0.001). There was no significant difference when low-risk patients with both JAK2 mutation and CV risk factors were compared with either those with CV risk factors only (p=0.227) or those with JAK2 mutation only (p=0.217). ii) Conventionally assigned high-risk group: The absence or presence of one or both of the aforementioned additional risk factors for thrombosis were documented in 111 (22%), 44 (9%), 222 (43%) and 136 (27%) patients, respectively, with corresponding annual rates of thrombosis at 1.44%, 1.64%, 2.36% and 4.17% (Table). High-risk patients with both risk factors had a significantly higher risk of thrombosis compared to their counterparts with the absence of JAK2 mutations and CV risk factors (p=0.011). Additional analysis revealed limited enhancement of thrombosis risk by either JAK2 mutations or CV risk factors or both in patients whose high-risk status was defined by the presence of thrombosis history, regardless of age (P=NS). In contrast, the presence of JAK2 mutations, with or without CV risk factors, might have affected thrombosis risk in patients where high-risk status was defined by age alone (p=0.05). Conclusions The current study suggests the possibility of considering four risk categories in ET: "very low risk" group (age ≤60 years and without thrombosis history, JAK2 mutations or CV risk factors); "low risk" (age ≤60 years and without thrombosis history but with JAK2 mutations or CV risk factors); "intermediate risk" (age>60 years but without thrombosis history or JAK2 mutations); and "high risk" (thrombosis history at any age or JAK2 -mutated patients who are older than 60 years of age). Treatment recommendations for each one of the above-mentioned new risk categories should be examined in the context of prospective controlled studies. Disclosures Barbui: Novartis: Speakers Bureau. Vannucchi:Novartis Pharmaceuticals Corporation: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Shire: Speakers Bureau; Baxalta: Membership on an entity's Board of Directors or advisory committees. Buxhofer-Ausch:AOP Orphan: Research Funding. De Stefano:Novartis: Research Funding, Speakers Bureau; Janssen Cilag: Research Funding; Shire: Speakers Bureau; GlaxoSmithKline: Speakers Bureau; Bruno Farmaceutici: Research Funding; Roche: Research Funding; Amgen: Speakers Bureau; Celgene: Speakers Bureau. Gisslinger:Janssen Cilag: Honoraria, Speakers Bureau; AOP ORPHAN: Consultancy, Honoraria, Research Funding, Speakers Bureau; Geron: Consultancy; Sanofi Aventis: Consultancy; Celgene: Consultancy, Honoraria, Research Funding, Speakers Bureau; Novartis: Honoraria, Research Funding, Speakers Bureau.

Abstract Introduction Patients with masked polycythemia vera (mPV) present with hemoglobin (HB) values borderline inferior to those required by the World Health Organization (WHO) diagnostic criteria (HB >16.5g% and >18.5 g% in female and male respectively) but are suspected to have PV for presenting clinico-hematological features. The diagnostic HB values required by the WHO guidelines have been considered surrogate of increased red cell mass (RCM). However, in both males and females HB values are not always equal to an absolute erythrocytosis and conversely a HB concentration below the required cut off level could be associated with an increased red cell mass. We believe that bone marrow (BM) morphology may play an important role in differentiating PV from other MPN entities and reactive changes. In this study we investigated JAK2V617/Exon12-positive patients showing BM features characteristic for PV and classified as mPV or overt PV according to their hemoglobin levels and evaluated their respective outcomes including thrombosis, progression to myelofibrosis (MF), acute leukemia (AL) and overall survival with the purpose to support the diagnosis of PV also in patients presenting with lower HB levels than required by WHO criteria. Materials and Methods A clinicopathological database of patients who were diagnosed and treated for PV was created by clinicians and hematopathologists from seven international centers of excellence for MPNs. Eligibility criteria for entry included the presence of JAK2V617F (or Exon12) mutation, EPO measurement, no evidence of subnormal ferritin levels and the availability of representative, treatment-naive BM biopsies (hematoxylin-eosin staining and silver impregnation after Gomori or Gordon-Sweet). All BM biopsies were centrally re-reviewed by one of the authors (J.T.) who was completely blinded to outcome data. In both mPV and overt PV patients, phlebotomy and cytoreductive therapy were used with the same frequency (p=0.101). Results and Discussion Among 397 mutated patients, with PV BM morphology as defined by WHO, 257 (65%) met the full WHO-2008 criteria. The remaining 140 patients (35%) were classified as mPV for levels of HB at diagnosis, ranging from 16.5 to 18.5 g/dL in men and from 15.0 to 16.5 g/dL in women, and frequent presence of subnormal EPO levels (65%). At baseline, mPV patients were more males, had more previous arterial thrombosis and presented more frequently with platelet counts exceeding the required level of WHO-defined ET (450 x 109/L), often mimicking clinically this entity at onset. After a median follow-up of 3.8 (0-29.8) and 4.5 (0-21.1) years in mPV and overt PV, respectively, time to the first thrombotic event was superimposable in the two groups while time to progression to MF and AL was significantly different .These combined events were recorded in 10% of mPV and 5.8% of overt PV, corresponding to an annual rate of 1.60 % pts/year in mPV and 0.97 % pts/year in overt PV, respectively (p=0.010). In addition, an inferior overall survival was documented in mPV patients in comparison with overt PV (p=0.011). The annual rate of death in mPV was almost twice that of overt PV, mainly due to an excess of hematological transformation (overt MF and AL). Multivariable analysis identified age older than 65 years (hazard ratio (HR) 6.63, P<.0001), WBC >10 x109/L (HR 2.99, P=.005) and diagnosis of mPV (HR 2.38, P=.036) as independent risk factors for survival in these patients. These results show that in JAK2 mutated patients with borderline HB levels, BM features may be a valid support to confirm the suspicion of PV diagnosis. Of interest, recent data demonstrated that all patients with increased RCM also had BM morphology typical of PV. It is tempting to speculate that the clinico-hematologic phenotype of mPV may represent an early stage of a JAK2 mutated MPN, presenting as a variant of PV, with a more rapid progression to MF and AL. The recognition of these early PV patients has an important clinical impact since, in spite of not meeting the full WHO criteria, mPV share the same vascular risk as overt PV. In conclusion, these data suggest that mPV is not necessarily only an early form of classical PV but a distinct entity with disease outcomes akin to primary myelofibrosis. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 2849 Background We have previously reported on the natural history of polycythemia vera (PV), focusing primarily on overall and leukemia-free survival (ASH Annual Meeting Abstracts. 2011;118(21):277-). In the current study, we present, on behalf of the International Working Group for Myeloproliferative neoplasms Resarch and Treatment (IWG-MRT), our analysis regarding risk factors for thrombosis. Methods Under the auspices of IWG-MRT, seven international centers of excellence for myeloproliferative neoplasms participated in the current study. The two principle investigators (AT and TB) reviewed all the cases and selected 1,545 patients who met the 2008 WHO criteria for PV, were age 18 years or older, diagnosed after 1970, and whose submitted data included diagnostically essential information. Results I: Presenting Features Median age was 61 years (range, 18–95; 51% females). Arterial and venous thrombosis history before or at diagnosis was documented in 246 (16%) patients and 114 (7.4%) patients, respectively. Major hemorrhage hemorrhage before or at diagnosis was documented in 17 (4.5%) patients. Other features at diagnosis included pruritus (36%), microvascular disturbances (28.5%), palpable splenomegaly (36%), abnormal karyotype (12%), leukoerythroblastosis (6%), increased LDH (50%), thrombocytosis (53%), extreme thrombocytosis (platelets >1 million mm3; 4%) leukocytosis (49%), JAK2 V617F (95%), other JAK2 mutations (3%), subnormal serum erythropoietin (Epo) level (81%), and endogenous erythroid colonies (EEC; 73%). History of hypertension (46%), hyperlipidemia (18.3%), diabetes (8.4%), and tobacco use (16%) was also obtained. Results II: Clinical Course To date, 347 (23%) deaths, 50 (3%) leukemic progressions, and 138 (9%) fibrotic transformations have been recorded. Overall, cytoreductive treatment was not used in 416 (27%) patients and the remaining were exposed to different agents based on physician discretion. Post-diagnosis arterial or venous thrombosis occurred in 184 (12%) and 137 (9%) patients, respectively. Results III: Risk Factors for thrombosis Arterial and venous thrombosis-free survival, from time of diagnosis, were separately analyzed using the occurrence of thrombosis as the endpoint (uncensored variable) and last follow-up or death before thrombosis as the censored variable. In univariate analysis, the following were significantly associated with post-diagnosis arterial thrombosis: advanced age, leukocyte count, presence of a leukoerythroblastic smear (LES), history of hypertension and history of arterial thrombosis before or at diagnosis; multivariable analysis using all these five parameters identified arterial thrombosis history (RR 2.5, 95% CI 1.6–4.0; p<0.0001), LES (RR 2.3, 95% CI 1.3–4.2; p=0.005) and history of hypertension (RR 1.6, 95% CI 1.1–2.4; p=0.02) as independent predictors of post-diagnosis arterial thrombosis. Only two parameters predicted post-diagnosis venous thrombosis, in univariate analysis, and both remained significant during multivariable analysis: abnormal karyotype (RR 3.1, 95% CI 1.7–5.4; p=0.0001) and history of venous thrombosis (RR 2.4, 95% CI 1.2–4.9). Of note, the type of JAK2 mutation or presence of either subnormal Epo or EEC did not influence either arterial or venous thrombosis. Results IV: Risk Stratification for arterial and venous thrombosis The figures below illustrated arterial or venous thrombosis-free survival of patients stratified by the absence of all risk factors or presence of one or ≥2 risk factors. For arterial thrombosis, the presence of ≥2 risk factors clearly delineated a high risk group (RR 3.1, 95% CI 1.9–5.0) whereas the presence of one (RR 2.4, 95% CI 1.4–4.2) or two risk factors (RR 10.1, 95% CI 3.6–28.2) for venous thrombosis delineated an intermediate and high risk group, respectively. Conclusions: History of arterial thrombosis and venous thrombosis are key risk factors, respectively, for recurrent arterial and venous thrombosis in PV. In addition, abnormal karyotype is a strong independent risk factor for venous thrombosis and the presence of leukoerythroblastosis and hypertension, for arterial thrombosis. This information allows for a simple and practical risk stratification and raises interesting pathogenetic implications that require further clarification. Disclosures: Vannucchi: Novartis: Membership on an entity's Board of Directors or advisory committees. Gisslinger:Novartis: Consultancy, Research Funding, Speakers Bureau; Celgene: Consultancy, Research Funding, Speakers Bureau. Passamonti:Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Sanofi-Aventis: Honoraria, Membership on an entity's Board of Directors or advisory committees.

Abstract Background Current risk stratification in Polycythemia Vera (PV), is designed to estimate the likelihood of thrombotic complications. In ECLAP large prospective study of 1638 PV cases diagnosed by Polycythemia Vera Study Group (PVSG) criteria, risk factors were age over 65 years and prior thrombotic events. After the JAK2V617F discovery, diagnosis has undergone substantial revision by the World Health Organization (WHO). Consequently, current risk factors determination should be described in patients diagnosed by the most recent recommendations and requires a large sample size with reliable and sufficiently long clinical course documentation. Accordingly, the current study includes 1545 patients with strictly WHO-defined PV diagnosed and followed-up in seven centers from Italy, Austria and the United States, belonging to the International Working Group for Myeloproliferative Neoplasm (MPN) Research and Treatment (IWG-MRT). Objective and design 1) Describe the rate of thrombosis in this population; 2) Report the sites and frequency of recurrences after a first arterial or venous event; 3) Identify separately risk factors for arterial versus venous thrombosis and develop practically useful prognostic models. Results 1. Presenting features The characteristics of the cohort recruited reflected well the heterogeneity of epidemiologic and clinical features that are found in the routine clinical practice of PV. Arterial and venous thrombosis history before or at diagnosis was documented in 246 (16%) and 114 (7.4%) patients, respectively. The frequencies of arterial and venous thrombosis were lower than those recorded in the ECLAP4 (27% and 11% respectively) but comparable to those reported in the recent Cyto-PV2 (arterial 17%, venous 12%) studies. The discrepancy might be partly related to earlier diagnosis in the post-ECLAP JAK2mutation era. 2. Thrombotic events after diagnosis Median follow-up was 6.9 years and treatment included aspirin (84%) and cytoreductive agents (73%), in addition to phlebotomy. Post-diagnosis total major thrombosis rate was 2.62% pt/yr, lower than reported in the ECLAP trial (4.4% pt/yr) but comparable to the Cyto-PV study (2.7% pt/yr) where management of cardiovascular risk factors was more intensive than in ECLAP. Arterial or venous thrombosis occurred in 184 (12%, rate: 1.59 % pts/yr) and 137 (9%, rate: 1.05 % pts/yr) patients, respectively. In addition, prior arterial events were associated in 75% of cases with subsequent arterial events and in the remaining 25% with venous thrombosis including splanchnic vein events. Moreover, patients with prior venous events had recurrences in venous district in 61% of cases and in the arterial district in 39%. 3. Risk factors for arterial versus venous thrombosis and prognostic models In multivariable analysis, arterial thrombosis was predicted by previous arterial event (HR 1.7, 95% CI 1.2-2.4), and hypertension (HR 1.6, 95% CI 1.2-2.2) while predictors of venous events were previous venous thrombosis (HR 2.6, 95% CI 1.5-4.4), and age ≥ 65 years (HR 1.7, 95% CI 1.2-2.5). We used the above-mentioned risk factors in order to develop prognostic models allowing to identify patients at different probability to develop arterial or venous thrombosis. For arterial thrombosis, the absence of previous arterial thrombosis and hypertension defined lower-risk (rate 1.27 % pts/yr) while the presence of one or both identified higher-risk patients (rate 2.04 % pts/yr). Previous venous event and/or age ≥ 65 years identified patients at higher risk of venous (rate 1.62 % pts/yr) thrombosis and, if absent, patients at lower risk (rate 0.78 % pts/yr). Conclusion The overall rate of thrombosis was half than in ECLAP and should be considered in future clinical trials. Of note, patients with a prior venous thrombosis had a more frequent subsequent recurrence in the venous district but a proportion of them (40%) had also arterial events; conversely, recurrences after previous arterial event were also registered in the leg and non-leg veins in 30 % of cases, indicating that in PV, venous and arterial thrombosis may share some common pathogenetic mechanisms. These findings may have practical implications and suggests antiplatelet therapy and anticoagulant therapy as the favorite choice to prevent arterial and venous thrombosis, respectively, but also suggests studies of antithrombotic drug combination. Disclosures No relevant conflicts of interest to declare.