Tefferi, Ayalew

Abstract Background: A mutation in the Janus tyrosine kinase 2 gene (JAK2 V617F) has been recognized in Philadelphia chromosome-negative myeloproliferative disorders, including PV (∼95% of patients), ET (∼50%) and PMF (∼50%). INCB018424 is a potent and orally bioavailable selective JAK2 inhibitor with >80-fold selectivity against a broad panel of kinases, including JAK3. INCB018424 potently inhibits JAK2 V617F mediated signaling and malignant cell survival in vitro and in vivo in mice. Preclinical safety studies with INCB018424 demonstrated an excellent safety profile with no off-target toxicities. Methods: A phase I/II trial of INCB018424 given orally BID is being conducted in patients with PMF and Post-PV/ET MF. Both JAK2 V617F and JAK2 wild type patients are eligible. PK and PD data are being collected. Responses are being evaluated using the International Working Group (IWG) consensus criteria for treatment response in myelofibrosis with myeloid metaplasia, for the IWG for Myelofibrosis Research and Treatment (IWG-MRT) (Tefferi et al., Blood, 108, 1497–1503, 2006). Results: The initial dose of 25 mg PO BID resulted in a rapid and marked reduction in splenomegaly. All 3 patients in the first cohort of patients achieved a reduction in spleen size that is consistent with a clinical improvement response provided it is sustained for 8 weeks; spleen sizes of 25, 22, and 7 cm below the left costal margin have decreased to 8, 10, and 0 cm in the first month of therapy. At two months follow-up, the patient with a baseline spleen size of 22 cm is now down to 2 cm. All 3 patients also noted significant symptomatic improvement. The second cohort of 3 patients started therapy at an increased dose (50 mg PO BID) and at one week follow-up, the initial two patients (one JAK2 wild type, one JAK2 V617F) had decreases in spleen size from 22 cm to 17 cm and from 22 cm to 16 cm, respectively. No dose-limiting toxicities or other significant adverse events occurred in any patients to date. PK/PD analyses demonstrated that administration of INCB018424 resulted in plasma drug concentrations sufficient to markedly inhibit JAK2, as shown by suppression of phosphorylated STAT3 (a substrate of JAK2) in whole blood cells in all 3 patients in the first cohort. The 3 patients in the first cohort were all JAK2 V617F mutation positive: percentages of JAK2 V617F positive whole blood cells before therapy were 79%, 49% and 91%, and after one month of therapy, they were 59%, 48% and 78%, respectively. Conclusions: Initial results show marked reduction in splenomegaly and symptomatic improvement, without significant toxicity. The percentage of blood cells with JAK2 V617F mutation appears to be decreasing in patients on therapy. Updated results on current and future patients will be presented at the meeting.

AbstractRecent observations have underscored the biologic relevance of intratumoral angiogenesis and its potential impact on prognosis. Increased bone marrow angiogenesis has been demonstrated in a variety of hematologic disorders, including multiple myeloma. The extent and prognostic significance of bone marrow angiogenesis in 114 patients with myelofibrosis with myeloid metaplasia (MMM) was investigated. A control group of 44 patients without bone marrow disease, 15 patients with polycythemia vera, and 17 patients with essential thrombocythemia was also studied. Bone marrow microvessel density was assessed by a semiquantitative method, visual microvessel grading, and 2 separate quantitative methods, visual count and computerized image analysis. Angiogenesis estimation by all 3 methods was highly comparable. On visual microvessel grading, a grade 3 or 4 increase in bone marrow angiogenesis was demonstrated in 70% of patients with MMM, 33% of patients with polycythemia vera, 12% of patients with essential thrombocythemia, and 0% of normal controls. In a multivariate analysis, increased angiogenesis in MMM correlated significantly with increased spleen size and was found to be a significant and independent risk factor for overall survival. Increases in marrow angiogenesis correlated with hypercellularity and megakaryocyte clumping. In contrast, these 2 features were inversely proportional to reticulin fibrosis, whereas increases in marrow angiogenesis were independent of reticulin fibrosis. These preliminary findings suggest that neo-angiogenesis is an integral component of the bone marrow stromal reaction in MMM and may provide useful prognostic information and a rationale for the therapeutic investigation of anti-angiogenic agents.

Abstract We review the clinical manifestations and long-term outlook of patients with chronic natural killer (NK) cell lymphocytosis. After reviewing more than 1,500 peripheral blood lymphoid flow cytometry reports and molecular genetics data from patients with suspected large granular lymphocyte (LGL) proliferation, we identified 10 patients (median age at diagnosis, 60 years; range, 35 to 76 years; male:female ratio, 3:2) with persistent (greater than 6 months) increase in phenotypically determined NK cells (CD3-CD16+). Southern blot analysis performed on 9 patients showed no clonal T-cell receptor gene rearrangements. Disease duration was measured from time of initial recognition of LGL or NK cell excess (greater than 40% of the lymphocyte fraction). Clinical data from these 10 patients were compared with those from 68 patients with T-cell LGL (T-LGL) leukemia. Currently, all patients are alive (median disease duration, 5 years; range, 0.8 to 8 years). Associated disease manifestations included pure red blood cell aplasia, recurrent neutropenia, recurrent neutropenic sepsis, and vasculitic syndromes, all of which were responsive to immunosuppressive therapy. No patient had palpable lymphadenopathy or splenomegaly. Compared with the patients with T-LGL leukemia, patients with chronic NK cell leukemia has similar lymphocyte counts, associated conditions, treatment responses, and survival but had less neutropenia and anemia.

Abstract Background The timing of SCT for treatment of AML after first relapse is controversial, i.e.,whether to proceed to SCT at the time of relapse or to proceed to chemotherapy re-induction first in an attempt to reach a second remission. The purpose of this study is to determine the optimal time of allogeneic SCT in AML patients in first relapse by comparing outcomes of allogeneic SCT between patients untreated and patients post re-induction chemotherapy. Method This retrospective study was approved by the Mayo Foundation IRB. Between February 1983 and March 2003, 60 patients with AML in untreated first relapse (REL1) or post re-induction chemotherapy (CT) underwent allogeneic SCT in Mayo Clinic, Rochester. Study outcome included overall survival (OS), disease free survival (DFS), and transplant related mortality (TRM). The outcome was compared between the patients who underwent allogeneic SCT at REL1 and CT. Survival was analyzed using Kaplan-Meier (KM) and groups were compared statistically with the log-rank test. Result Of the 60 patients receiving allogeneic SCT , 25 patients and 35 patients underwent SCT at REL1 and CT, respectively. In the group of CT, 23 patients (66%) achieved CR2 while 12 patients (34%) had refractory disease. Forty-nine of 60 patients received cyclophosphamide and total body irradiation (Cy/TBI) as a conditioning regimen and the remainder received busulfan and cyclophosphamide( Bu/Cy). Patients in REL1 were older at SCT (median age 38 vs 31 years p= 0.024 ) and had lower percentage blast cells in bone marrow at first relapse (median 15% vs 50% p=0.00095) than in CT group. Difference in other baseline characteristics were not significant. KM analysis showed that the two groups did not show statistically differences in OS (p=0.43), DFS (p=0.68), and TRM (p=0.25). However, there were trends toward longer median OS (26.2 vs 5.4 months) and DFS (11.9 vs 3.3 months) with REL1 versus CT. OS at 1 year was 56 % for REL1 and 34.3 % for CT and at 5 year 32.1 % and 28.6 %, respectively. DFS at 1 year was 48 % for REL1 and 31.4 % for CT and at 5 years 22.9 % and 24.5 % respectively. TRM at 1 year was 25% for REL1 and 40 % for CT. A univariate Cox Proportional Hazard analysis for survival showed that unfavorable cytogenetics at diagnosis and relapse, antecedent MDS, and shorter duration of CR1 were significantly associated with a higher risk of death. Higher percentage of blast cells in bone marrow at first relapse was not significantly associated with a higher risk of death. In a multivariate model, unfavorable cytogenetics at first relapse and shorter duration of CR1 were both significantly associated with an increased risk of death. In this study, even after adjusting for other factors in a multivariate model, the difference between REL1 and CT groups was not statistically significant. Conclusion This study suggests that outcome of allogeneic SCT was comparable between REL1 and CT. Patients with AML in first relapse may proceed to allogeneic SCT because there appear to be no statistical differences in REL1 versus re-induction chemotherapy before SCT. REL1 patients tended to have a better outcome than CT patients.

Abstract A favorable prognosis and a low rate of leukemic transformation has been attributed to the 5q- syndrome, a myelodysplastic syndrome (MDS) characterized by macrocytic anemia, hypolobulated micromegakaryocytic hyperplasia, and an interstitial deletion of chromosome 5. We examined the characteristics and outcome of 43 consecutive patients in our institution strictly defined by morphologic criteria and a solitary 5q- cytogenetic defect. The median age at diagnosis was 68 years, with a clear female predominance (7:3). Eighty percent of the patients were red blood cell transfusion-dependent at diagnosis and all untransfused patients had macrocytic indexes. In contrast, significant neutropenia or thrombocytopenia was rare. The French-American-British (FAB) class distributions were RA (72%), RARS (7%), RAEB (16%), and RAEB-IT (5%). At a median follow-up of 31 months, 56% of the patients survive, with a projected median survival of 63 months. The incidence of acute leukemia was 16% and was uniformly fatal. Clinical hemosiderosis occurred in 28% of the patients, resulting in two deaths. Neither survival nor the risk of leukemic transformation was predictable from initial clinical parameters, including FAB classification, Bournemouth score, and degree of aneuploidy. The lack of significant neutropenia and thrombocytopenia seemed to account for a very low incidence of infection and bleeding resulting in a prognosis equal or superior to historical patients with MDS. Therapeutic endeavors, including the use of corticosteroids, androgens, cis-retinoic acid, pyridoxine, and danazol, were largely unsuccessful.

Abstract We used the X-linked restriction fragment length polymorphism (RFLP)- methylation strategy to study the clonal basis of the myelodysplastic syndrome (MDS) in seven patients. RFLP-methylation analysis was performed on cell populations from bone marrow (BM) aspirates and peripheral blood using probes specific for the hypoxanthine phosphoribosyltransferase (HPRT) or phosphoglycerate kinase (PGK) gene regions. Density gradient centrifugation methods were used to separate granulocytes and monocytes, and T lymphocytes were positively selected by CD2 (a pan-T marker) immunoconjugated magnetic beads. Cell populations from BM aspirates in 6 of the 7 patients with MDS showed a monoclonal pattern of X-inactivation. The neutrophilic and T- lymphocytic cell fractions were analyzed in 4 of the 6 patients, and the monocytic cell fraction in one of these, and all fractions analyzed showed a similar monoclonal pattern. In 2 of the latter 4 patients, both of whom had normal karyotypes, DNA from a skin biopsy showed a polyclonal pattern. Our data suggest that MDS is a clonal disorder, even in the absence of detectable cytogenetic abnormalities, and that the abnormal clone is capable of myeloid, monocytic, and lymphoid differentiation.

Two prospectively studied patients with polycythemia vera (PV) whose platelet counts showed marked periodic fluctuation during treatment with hydroxyurea (HU) are reported. Cycle lengths in both were approximately 28 to 30 days. In one patient, the cyclic process was no longer evident when treatment with HU was withheld, and it reappeared on treatment rechallenge. Circulating thrombopoietin (TPO) levels fluctuated out of phase with the platelet count despite markedly reduced TPO-receptor (c-Mpl) expression in bone marrow megakaryocytes. These observations suggest that the cyclic phenomenon may be related to both a transient state of HU-induced depletion of megakaryocytes and a concentration-dependent mitigation by TPO of the HU effect on megakaryocytes and their precursors. It is conceivable that the affected patients harbor a megakaryocyte progenitor pool whose apoptotic activity is differently modulated by either HU or high concentrations of TPO.

Abstract Abstract 3891 Poster Board III-827 Age over 65 years, hemoglobin level lower than 10 g/dL, WBC count greater than 25 × 109/L, peripheral blood blasts equal to or greater than 1%, and presence of constitutional symptoms provide a reliable prediction of survival in patients with primary myelofibrosis (PMF) at diagnosis [Cervantes et al. Blood. 2009 Mar 26;113(13):2895-901]. In fact, based on the presence of 0 (low risk), 1 (intermediate risk-1), 2 (intermediate risk-2) or greater than or equal to 3 (high risk) of the above factors, 4 risk groups with no overlapping in their survival curves were defined. To investigate whether the acquisition of these prognostic factors during follow-up may predict survival anytime, we studied 525 PMF patients regularly followed at each participating Institution from 1980 to 2008. The study was carried out within the IWG-MRT, and an ad hoc database was developed for data collection. Patients with post-polycythemia vera and post-essential thrombocythemia myelofibrosis, and patients with pre-fibrotic myelofibrosis were excluded from this study. Overall, 68 patients (26% of subjects who were below the age of 65 at diagnosis) passed the 65-year age threshold during the study period, 158 (47% of patients without this risk factor at diagnosis) developed anemia, 72 (15% of patients without this risk factor at diagnosis) presented marked leukocytosis, 102 (27% of patients without this risk factor at diagnosis) had more than 1% peripheral blast cells, and 54 (14% of patients without these symptoms at diagnosis) had constitutional symptoms. All variables resulted significant when analyzed as time-dependent covariates in a multivariable Cox proportional hazard regression model, and all of them were therefore included in the subsequent models. Each variable was assigned an integer weight close to the corresponding hazard ratio (HR), and the sum of values obtained was used to allocate the patients into 4 risk categories (low, intermediate-1, intermediate-2, high). We developed 2 separate models, one for all patients (DIPSS) and the other one for patients below the age of 65 (age-adjusted DIPSS), both applicable during the course of the disease. Cumulative survival can be estimated with the Kaplan Meier method. The increased risk of death when changing risk category was estimated as HR. According to DIPSS, the HR was: 4.13 (95% CI: 1.73-9.82; P < .001) if the risk category shifted from low to intermediate-1, 4.61 (95% CI: 3.18-9.82; P < .001) from intermediate-1 to intermediate-2, and 2.54 (95% CI: 1.94-3.31; P < .001) from intermediate-2 to high. According to age-adjusted DIPSS, the HR was 3.97 (95% CI: 1.5-10.5, P = .005) if the risk category shifted from low to intermediate-1, 2.84 (95% CI: 1.46-5.54; P = .002) from intermediate-1 to intermediate-2, and 1.81 (95% CI: 1.08-3.04; P = .025) from intermediate-2 to high. In conclusion, this study shows that age over 65 years, hemoglobin level lower than 10 g/dL, WBC count greater than 25 × 109/L, peripheral blood blasts equal to or greater than 1%, and the presence of constitutional symptoms predict survival independently and in a time-dependent manner in patients with PMF. Both DIPSS and age-adjusted DIPSS allow a reliable prognostic assessment of PMF patients at any time during the clinical course of the disease, and are therefore useful for clinical decision making. Disclosures: No relevant conflicts of interest to declare.

Recent studies have shown decreased megakaryocyte expression of the thrombopoietin receptor (c-mpl) in patients with polycythemia vera (PV) but not in those with reactive erythrocytosis. We examined the diagnostic utility of this observation in 22 patients with PV, 7 patients with secondary erythrocytosis (SE), and 10 normal controls. Commercial antibodies against c-mpl were used with standard immunoperoxidase methods. Megakaryocyte c-mpl staining intensity was uniformly moderate-to-strong in the healthy controls and in all the patients with SE. In contrast, staining intensity in 9 patients with PV (41%) was uniformly weak. Furthermore, in 12 of the remaining 13 patients with PV, the c-mpl staining pattern in each case was heterogeneous and was associated with weak staining intensity in more than 20% of the megakaryocyte population. These preliminary data suggest that c-mpl immunostains may complement bone marrow histopathology in distinguishing PV from nonclonal causes of erythrocytosis.

Abstract Abstract 4971 Background Marked splenomegaly severely compromises quality of life in primary myelofibrosis (PMF). Preliminary results from the use of small molecule JAK2 inhibitors suggest prominent activity in reducing spleen size. In order to put such salutary effect of investigational drug therapy in context, it is important to determine the value of conventional therapy in this regard. Methods Study population was selected from the Mayo Clinic PMF database and consisted of patients treated with hydroxyurea (HU) as first-line therapy for symptomatic, marked splenomegaly (palpable at >10 cm below the left costal margin) and in whom follow-up information was available to assess response. Palpable spleen size measurements were recorded before and after 3, 6 and 12 months of HU therapy. Spleen response was defined as minimal (< 25% reduction), minor (25-50%), or major (>50%). Standard statistical procedures were used to correlate spleen response with other clinical and laboratory parameters, including cytogenetic and JAK2V617F mutational status. Results 46 consecutive patients (median age, 62 years; range 38-75; 30 males) met the above stipulated eligibility criteria for study inclusion. Pre-treatment IPSS risk categories were low, intermediate-1 (int-1), int-2 and high in 4, 13, 10 and 19 patients, respectively; 7 patients were transfusion-dependent. Quantitative JAK2V617F and pre-treatment cytogenetic information were available in 25 and 36 patients, respectively; 15 (60%) patients were positive for JAK2V617F and mutant allele burden was > 50% in 6 patients; 17 (47%) patients carried either favorable (n=5) or unfavorable (n=12) cytogenetic abnormalities. Overall response rate (major and minor) was 35% (16 of 46 patients) and major response rate 17%. Overall response rates were 10%, 67% and 33% in patients with undetectable JAK2V617F or mutant allele burden of < or ≥ 50%, respectively (p=0.04). Overall response rates in patients with normal, favorable and unfavorable karyotype were 32%, 40% and 33%, respectively (p=0.9). Response was not affected by age, sex, HU dose or pretreatment status regarding spleen size, IPSS score or transfusion need. Conclusions The current study provides background information on the value of conventional chemotherapy for controlling PMF-associated marked splenomegaly. The study also suggests a positive effect of JAK2V617F on treatment response to HU, a detail that needs to be considered when assessing the corresponding effect from a JAK2 inhibitor therapy. Disclosures Off Label Use: Hydroxyurea use in myelofibrosis.