ALBERT

Description: Discovery of somatic mutation of JAK-2 (G1849T that produces JAK-2V617F) in the hematopoietic cells of patients with Philadelphia chromosome negative myeloproliferative disorders (Ph−MPDs) was a watershed event that not only provided new insights into the pathobiology of polycythemia vera, essential thrombocytosis and primary myelofibrosis but also identified a potential target for therapy. Herein we report the results of preclinical studies designed to characterize the activity of a novel inhibitor of JAK-2. The compound, SGI-1252, developed by SuperGen (Dublin, CA) incorporates with high affinity into the ATP-binding site of JAK-2. SGI-1252 was tested against a panel of 75 kinases and was found to have significant activity against only FLT-3, TYK-2 and the SRC family members, ABL, LCK, YES, in addition to JAK-2 and JAK-1. SGI-1252 has an IC50 for JAK-2 of 5.4 nM with an IC50 for JAK-2V617F of 19.7 nM. The inhibitor also effectively blocks the activity of JAK-1 (IC50 14.8 nM) but has little JAK-3 inhibitory activity (IC50 1,700 nM). SGI-1252 is a potent inhibitor of STAT-5 phosphorylation (EC50 76.2 nM) and was also found to block the JAK-2 dependent expression of the anti-apoptotic protein, BCL-XL (EC50 778 nM). Drug treatment of a murine cell line (FDCP) transfected with either human wild-type JAK-2 or JAK-2G1849Tgenerated IC50 values of 83 nM and 108 nM, respectively, and SGI-1252 treatment of human cell lines, HEL, UKE-1 and SET-2, that express mutant JAK2 in different copy numbers, gave IC50 values of 472 nM, 83 nM and 63 nM, repectively. When tested in ex-vivo expanded native human erythroid progenitor cells from 17 patients with Ph−MPDs (10 PV and 7 MF), SGI-1252 showed an IC50 of ~100 nM, regardless of the JAK-2V617F allele burden. Using a flow cytometric assay, SGI-1252 was shown to induce apoptotic cell death in a concentration dependent manner. Luminex technology allows for concurrent quantitative analysis of multiple proteins from the same tissue source, and this technology was used to investigate simultaneously the effects of SGI-1252 on total and phospho ERK1/2, total and phospho STAT3, phospho STAT5, caspase 3, cleaved PARP and GAPDH (control) in untreated and drug treated cells at IC50 and IC80 concentrations. Significant in vivo efficacy of SGI-1252 was also observed using HEL and MV-4-11 xenograft models when compared to treatment with vehicle or daunorubicin. Using a murine model, we found that SGI-1252 has high oral bioavailability and is well tolerated with a five-day repeat maximum dose of at least 900 mg/kg. Together, these studies demonstrate that SGI-1252 is a potent inhibitor of JAK-2 dependent proliferation in both JAK-2V617F positive cell lines and in ex vivo expanded erythroid progenitors derived from patients with JAK-2V617F positive Ph−MPDs. Moreover, our studies show that the effects of SGI-1252 are mediated by blocking both JAK-2 dependent anti-apoptoic pathways and JAK-2 dependent proliferative pathways. Using the orally available form of the compound, pharmacokinetic, pharmacodynamic and toxicity studies in mice suggest that serum concentration of the drug well above the predicted therapeutic range can be achieved without significant hematological toxicity. Based on these preclinical experiments, SGI-1252 appears to be an excellent candidate for phase I/II studies in patients with Ph−MPDs.

Description: PNH is an acquired disorder of clonal hematopoiesis in which the affected hematopoietic stem cells (HSC’s) have a mutation of the X-linked gene PIG-A. Mutant PIG-A explains the clinical manifestations of PNH that result from deficiency of glycosyl phosphatidylinositol anchored proteins but does not explain why the mutant clones are selected or expand. The peripheral blood of patients with PNH is a mosaic of normal and abnormal cells, and molecular studies have demonstrated that the abnormal cells may be derived from multiple, discrete PIG-A mutant clones. Further, the contribution of the mutant clones to hematopoiesis varies greatly, but usually one clone is dominant. Together, these observations suggest that factors independent of mutant PIG-A influence clonal expansion. To investigate the basis of clonal expansion and clonal dominance, the pattern of gene expression among T cell clones developed from an informative patient with PNH was compared. The samples analyzed represented the following 3 categories of interest: (1) PIG-A mutant, dominant; (2) PIG-A mutant, non-dominant; (3) PIG-A normal. All samples were developed at the same time. Total RNA from the clones was amplified using the RiboAmp kit (Ambion). The antisense RNA was labeled, and the transcription profile of the dominant PIG-A mutant clone was compared with that of both the non-dominant PIG-A mutant clone and the PIG-A normal clone using microarray analysis. RNA from the original samples was used to verify results by PCR. Components of the AP-1 family of transcription factors (particularly FOSB) were differentially expressed by the PIG-A mutant dominant clone compared to both the PIG-A mutant non-dominant and the PIG-A normal clones. The major downstream effect of AP-1 upregulation appears to be on prostaglandin synthesis as the dominant clone also differentially expressed COX-2. Our experiments also showed that the pool of GPI-AP deficient T cells from a second patient with PNH but not the pooled GPI-AP positive T cells from the same patient expressed COX-2. Further, COX-2 expression was not observed in cultured T cells derived from any of 3 volunteer donors. COX-2 is mainly the inducible form of cyclooxygenase, with expression regulated by inflammatory cytokines. Studies by others using knockout mice suggest that COX-2 plays a key role in accelerated hematopoiesis following marrow injury. COX-2 also appears to function as an anti-apoptosis factor in lymphoproliferative disease, and inhibition of COX-2 has been reported to activate apoptosis through both defined and novel, BCL-2 independent, mitochondrial pathways. We have discovered COX-2 overexpression in the PIG-A mutant, dominant T cells through comparative gene expression analysis. COX-2 appears to be vital to hematopoietic recovery after marrow insult, conceivably as a consequence of anti-apoptotic activity. We hypothesize that an undefined marrow insult leads to clonal selection of PIG-A mutant HSC’s because of deficiency of one or more GPI-AP. Clonal expansion and clonal dominance, however, may be dependent on the pattern of expression of genes other than PIG-A (e. g., COX-2) that enhance proliferation, survival or both. Studies designed to evaluate further the role of COX-2 in clonal expansion are in progress.

Description: cMPL is a gene encoding for the thrombopoietin receptor that is essential for thrombopoiesis and contributes to pluripotent hematopoietic stem cell expansion. A gain of function cMPL mutation, MPLW515L, was identified in myeloid cells from patients with primary myelofibrosis (PMF). Subsequent studies identified a second gain of function mutation, MPLW515K, in PMF and essential thrombocytosis (ET). The prevalence of MPLW515L and MPLW515K mutations was 5% in PMF and 1% in ET. No mutant cMPL was detected in Polycythemia Vera (PV). The methods utilized in these assays were sensitive to mutant frequencies of 〉3–5%. We developed a rapid, sensitive, quantitative real time PCR assay based on a unique primer design wherein allelic discrimination was enhanced by the synergistic effect of a mismatch in the −1 position, and a locked nucleic acid nucleoside at the −2 position of the allele-specific primers. An assay of similar design can detect G1849T mutation of JAK2 in

Description: Clonality assays, based on X-chromosome inactivation, discriminate active from inactive alleles. They are useful in the diagnosis and assessment of therapeutic response in clonal hematologic disorders, especially in absence of an identifiable somatic mutations. Skewing of X-chromosome allelic-usage, based on preferential methylation of one of the HUMARA alleles, was reported as evidence of clonal hematopoiesis in ~30% elderly women, precluding the use of this assay in elderly patient (〉65 years of age). Using a quantitative, transcriptionally-based clonality assay, we did not detect clonal hematopoiesis in 〉200 healthy non-elderly women. In view of the susceptibility of aging hematopoietic stem cells to epigenetic dysregulation, we reinvestigated the issue of clonality in forty elderly women (age 65–92, mean 81.3 years), using a novel, quantitative qPCR transcriptional clonality assay. In this assay, mRNA transcribed from five X-chromosome polymorphic genes expressed in peripheral blood neutrophils is quantified by real time, allele specific RT-PCR. We did not detect clonal hematopoiesis in any of the elderly women. However, using HUMARA assay, 30% of these elderly women were detected to have monoallelic methylation of the HUMARA gene locus, consistent with previously reported literature. We concluded that our novel transcriptional clonality assay is suitable for evaluation of clonal hematopoiesis in all women including elderly women (Swierczek S, Agarwal N et al. Blood 2008, July 18 epub). Using this novel assay, we detected clonal hematopoiesis in 31 out of 32 well characterized patients with myeloproliferative disorders: polycythemia vera (all fourteen patients were clonal by our assay and all were JAK2V617F positive), essential thrombocytosis (nine out of ten patients were clonal by our assay, one out of ten patients was cMPLW515L positive, seven out of ten were JAK2 V617F positive; however one subject with low JAK2 V617F allelic burden was polyclonal by our assay), and primary myelofibrosis (all eight patients were clonal by our assay and two of them were positive for JAK2V617F). In addition, we detected clonal hematopoiesis in 4 patients with unexplained anemia (two eventually evolved in to myelodysplastic syndrome), and in one patient with persistent leukocytosis (eventually found to be cMPLW515L positive). Using our assay we did not detect clonal hematopoiesis in 10 patients with reactive or secondary erythrocytosis, thrombocytosis or leukocytosis. We conclude that our novel transcriptional clonality assay is suitable for detection of clonal hematopoiesis in patients with clonal hematologic disorders, especially in patients lacking known somatic mutation. Studies to detect an emerging clone in milieu of polyclonal hematopoiesis (such as seen in PNH or early stages of clonal hematological disorders) by comparison of X-chromosome allelic usage ratio in myeloid cells and in T lymphocytes are underway.

Description: GNPAT (chromosome 1q42.2) encodes the peroxisomal enzyme glyceronephosphate O-acyltransferase. In a previous study, DNA of men with hemochromatosis and HFE p.C282Y homozygosity and either markedly increased iron stores or normal or mildly increased iron stores were evaluated with exome sequencing. Positivity for the GNPAT polymorphism p.D519G (rs11558492) was significantly greater in men with markedly increased iron stores (McLaren CE et al., Hepatology 2015;62:429-39). This result suggests that the p.D519G is a candidate modifier of iron phenotypes in p.C282Y homozygotes. To learn more, we examined associations of p.D519G, age, iron-related variables, and daily alcohol consumption with iron stores in p.C282Y homozygotes classified by extremes of iron overload phenotypes. We defined markedly increased iron stores as serum ferritin 〉1000 µg/L and either hepatic iron 〉236 µmol/g dry weight or mobilizable iron 〉10 g by induction phlebotomy (men and women). Normal or mildly elevated iron stores were defined as serum ferritin

Description: Abstract 157 In PNH, red blood cells (RBCs) lack key complement control proteins, CD55 and CD59 and are therefore sensitive to complement activation and intravascular hemolysis. These regulatory proteins function at two different steps in the complement cascade; CD55 (decay accelerating factor, DAF) controls the formation and stability of the APC C3 and C5 convertases, while CD59 (membrane inhibitor of reactive lysis, MIRL) blocks formation of the cytolytic membrane attack complex (MAC). The intravascular hemolysis of PNH can be inhibited in vivo by eculizumab, a humanized mAb that binds complement C5, thereby preventing formation of the MAC. However, PNH patients treated with eculizumab continue to manifest evidence of ongoing hemolysis as they remain anemic with an elevated reticulocyte count and low serum haptoglobin concentration, and approximately 50% of eculizumab-treated patients require transfusion. This observation is consistent with the hypothesis that, in patients treated with eculizumab, PNH RBCs undergo extravascular hemolysis as a consequence of C3 opsonization because eculizumab does not compensate for deficiency of DAF. Recent studies (Risitano et al., Blood, 2009) support this hypothesis as patients undergoing treatment with eculizumab were found to have a positive Coombs test for C3 but not IgG, and flow cytometry demonstrated C3 activation and degradation products bound to the PNH RBCs. This process appeared clinically relevant as transfusion requirement correlated with the percentage of C3 opsonized PNH RBCs. These observations suggest that blocking the APC C3/C5 convertase would be a better way to treat the hemolysis of PNH because this approach has the advantage of blocking both extravascular hemolysis by inhibiting C3 opsonization and preventing intravascular hemolysis by inhibiting MAC generation. We have developed a mAb 3E7 and its deimmunized chimeric humanized derivative H17 that specifically block the APC C3/C5 convertase by binding to a neoepitope expressed when complement C3 is activated. In vitro, 3E7/H17 prevents APC-mediated lysis of rabbit RBCs in human serum and blocks deposition of human C3 activation fragments on APC activator substrates such as zymosan (Mol Immunol, 2006; J Immunol, 2007). We now report that mAb H17/3E7 blocks lysis in acidified normal human serum (aNHS) (a process mediated by the APC) of RBCs from patients with PNH (n=5). Representative results for patients 1 and 2 are as follows: 60% and 40% of RBCs were lysed after a one hour incubation at 37°C; lysis was reduced to 10% and 6%, respectively, at 80 ug/ml of mAb H17, and to 1% lysis (both patients) at 170 ug/ml of mAb H17. We also showed that mAb H17/3E7 blocks deposition of C3 activation fragments on PNH RBCs. After lysis in aNHS, blood samples from PNH patients were probed with Al488 mAb 1H8, specific for C3b/iC3b/C3dg. Flow cytometry experiments revealed C3 fragment deposition on lysed cells corresponding to 30,500 molecules of equivalent soluble fluorochrome (MESF) compared to a background signal of 225 MESF on unlysed RBCs in the same sample. Addition of mAb H17 blocked C3 fragment deposition not only on the unlysed cells but also on the small number of recovered ghosts . Importantly, mAb H17/3E7 inhibits the APC specifically. PNH RBCs, opsonized with IgM in serum from a patient with chronic cold agglutinin disease, were lysed in NHS by the classical complement pathway, and this lysis was not inhibited by mAbH17/3E7. Together, these experiments demonstrate that both hemolysis and C3 opsonization of PNH RBCs can be inhibited by a novel mAb that specifically blocks the APC C3/C5 convertase while leaving intact the classical pathway of complement. These findings suggest an approach to therapy of PNH in which both intravascular and extravascular hemolysis can be inhibited while preserving the important immune functions of the classical pathway of complement. Disclosures: No relevant conflicts of interest to declare.

Description: The clinical hallmark of paroxysmal nocturnal hemoglobinuria (PNH) is chronic intravascular hemolysis that is a consequence of unregulated activation of the alternative pathway of complement (APC). Intravascular hemolysis can be inhibited in patients by treatment with eculizumab, a monoclonal antibody that binds complement C5 thereby preventing formation of the cytolytic membrane attack complex of complement. However, in essentially all patients treated with eculizumab, persistent anemia, reticulocytosis, and biochemical evidence of hemolysis are observed; and in a significant proportion, their PNH erythrocytes become opsonized with complement C3. These observations suggest that PNH patients treated with eculizumab are left with clinically significant immune-mediated hemolytic anemia because the antibody does not block APC activation. With a goal of improving PNH therapy, we characterized the activity of anti-C3b/iC3b monoclonal antibody 3E7 in an in vitro model of APC-mediated hemolysis. We show that 3E7 and its chimeric-deimmunized derivative H17 block both hemolysis and C3 deposition on PNH erythrocytes. The antibody is specific for the APC C3/C5 convertase because classical pathway–mediated hemolysis is unaffected by 3E7/H17. These findings suggest an approach to PNH treatment in which both intravascular and extravascular hemolysis can be inhibited while preserving important immune functions of the classical pathway of complement.

Description: PNH is a clonal hematopoietic stem cell disorder and transplantation is potentially curative. However, the pathobiology of PNH is different from malignant clonal myelopathies such as acute leukemia, and the optimal approach to transplant is unresolved. Somatic mutation of PIGA underlies deficiency of glycosyl phosphatidylinositol anchored proteins (GPI-APs) that accounts for the hemolytic anemia of PNH but does not explain the clonal expansion required for clinical manifestation of the disease to become apparent. Recently we identified two patients whose PIGA mutant cells had a concurrent rearrangement of chromosome 12 that resulted in ectopic expression of HMGA2, the architectural transcription factor deregulated in many benign tumors. These observations suggest that aberrant HMGA2 expression, in concert with mutant PIGA, accounted for clonal hematopoiesis in these patients and support the view of PNH as a benign tumor of the bone marrow. One of the patients (a 31 y.o. female) underwent syngeneic SCT because of persistent transfusion-dependent hemolysis and thromboembolic complications that occurred despite prophylactic anticoagulation. At presentation, flow cytometry showed that 95% of PMN were GPI-AP deficient. Marrow analysis revealed erythroid hyperplasia and a karyotype of 46,XX,ins(12)(p13q15q13) in 20/20 (100%) metaphases. The cytogenetic abnormality was shown to be an intrachromosomal insertion that deregulated expression of HMGA2. A 14 bp deletion in exon 2 of PIGA was identified in PMN DNA. Quantitative PCR showed mutant PIGA and der(12) in 91.8% and 93.9%, respectively of bone marrow cells, demonstrating that the two mutations coexisted in the same cells. Neither mutant PIGA nor der(12) was identified in hematopoietic elements of the syngeneic donor, confirming their somatic origin. Donor CD34+ cells were mobilized with filgrastim to yield 5 × 106 cells/kg recipient weight. Because of the complex molecular abnormalities and the anticipated absence of graft vs. tumor effect, a preparative regimen of targeted busulfan (1mg/kg × 16 doses) and cyclophosphamide (50 mg/kg × 4 doses) was given. Engraftment occurred on Day 9 with no subsequent hemolysis. At 6 and 15 months post-transplant, high sensitivity flow cytometry showed no GPI-AP deficient RBCs or PMNs. Moreover, PCR revealed no mutant PIGA or der(12) post-transplant, demonstrating, for the first time, a molecular complete remission of PNH following SCT. These results reveal important concepts in PNH pathobiology and transplant efficacy. 1) As the donor was a syngeneic twin, graft vs. tumor effect was not required for eradication of the PNH clone. 2) Prior to transplant, essentially all hematopoiesis was derived from the double mutant clone, but the marrow ablative regimen completely eradicated the clone without the need for pretransplant cytoreduction. Together, these observations add further support to the concept of PNH as a benign tumor of the marrow and suggest a rational approach to SCT.