Gilliland, D. Gary

Abstract Since the identification of the JAK family of intracellular tyrosine kinases and their downstream signaling partners, the STATs (Signal Transducers and Activators of Transcription), evidence has accumulated that activation of JAK signaling pathways can contribute to proliferation and survival of cancer cells. A direct link to oncogenesis was made when an activated form of JAK2 (JAK2 V617F) was identified in the majority of patients with myeloproliferative disorders (MPDs) including approximately half of patients with myelofibrosis (MF). Introduction of this JAK2 mutation into the bone marrow of mice induced erythrocytosis [Wernig et al. (2006) Blood107:4274] supporting JAK2 as a relevant target for treatment of myeloproliferative disorders. We have identified a novel small molecule inhibitor of JAK2, XL019, which may prove to have utility in treating MPDs. XL019 is a potent and reversible inhibitor of the JAK2 enzyme, with a Ki of 2 nM, and shows excellent selectivity (minimum of 50-fold selectivity against >120 protein kinases tested including other JAK family members). XL019 downregulates STAT signaling in cell lines expressing both wild type and activated forms of JAK2. IC50s for inhibition of STAT5 phosphorylation by XL019 ranged from 623 nM (HEL92.1.7) to 3398 nM (KG-1) in tumor cell lines. XL019 showed increased potency in primary human erythroid cells in vitro, where the IC50 for inhibition of EPO-stimulated phospho-STAT5 was 64 nM. XL019 inhibits proliferation in cell lines harboring activated or overexpressed JAK2, including certain lines derived from patients with Hodgkin’s Lymphoma (L-1236, 928 nM IC50), AML (MV4-11, 992 nM IC50), essential thrombocythemia (SET-2, 386 nM IC50), and erythroleukemia (HEL92.1.7, 6777 nM IC50). Preclinical single-dose pharmacodynamic studies demonstrate a potent effect of XL019 on JAK-STAT signaling in HEL92.1.7, CFPAC-1 and DU 145 xenograft tumors. Twice daily dosing (bid) led to substantial tumor growth inhibition in the DU145 and HEL models (maximum tumor growth inhibition of 86% and 60%, respectively), accompanied by increases in tumor cell apoptosis (4 – 4.4 fold increase) and decreases in tumor microvasculature (44% reduction in DU 145 xenograft tumors). Based on these encouraging data, XL019 is being evaluated in subjects with primary or post PV/ET MF in a Phase I dose escalation study, which has completed enrollment of the first cohort. The primary objectives of this study are to determine the safety and tolerability of XL019 when administered orally once daily for 21 days in 28 day cycles. Secondary objectives include determination of the pharmacokinetics and pharmacodynamics of XL019, and to evaluate response using the International Working Group for MF consensus response criteria. We have developed a flow-based mechanistic assay for JAK2 activity, which measures GM-CSF-stimulated phospho-STAT5 in peripheral granulocytes in whole blood samples. Additional pharmacodynamic assessments for this study include JAK2 V617F allele burden, erythropoietin-independent colony formation, cytokine levels, and changes in bone marrow histology. Our initial findings regarding the pharmacokinetics, pharmacodynamics, tolerability and efficacy of XL019 in MF patients will be presented.

Abstract Background: The recent discovery of JAK2V617F, as an activating mutation in the majority of patients with myeloproliferative neoplasms, has spurred development of JAK2-selective small molecule inhibitors for the treatment of these diseases. TG101348 is a potent (enzyme IC50=3 nM) and selective (35- and 334-fold selectivity over JAK1 and JAK3, respectively) orally bioavailable JAK2 inhibitor that is effective in treating a murine model of Polycythemia vera (PV) (Cancer Cell. 2008 13:311). TG101348 selectively inhibits growth of hematopoietic colonies harboring V617F, MPLW515K, or JAK2 exon 12 mutations (Leukemia. 2008 Mar 20, Epub). Methods: A Phase I dose-escalation study of TG101348 is ongoing for patients with high- or intermediate-risk (with symptomatic disease) primary myelofibrosis (PMF) and post-PV or post-essential thrombocythemia (ET) myelofibrosis (Cancer.2007109:2083). TG101348 was administered orally once daily in 28-day cycles in cohorts of at least 3 patients. Additional eligibility criteria included absolute neutrophil count of ≥1 × 109/L and platelet count of ≥50 × 109/L, but a JAK2 mutation was not required. Primary objectives were determination of safety and tolerability, as well as pharmacokinetic behavior of TG101348. The secondary objectives were evaluation of treatment response (Blood.2006108:1497), pharmacodynamic activity, and drug effect on V617F allele burden and plasma cytokine levels. The spleen response was documented as change in palpable spleen tip from the costal margin. For patients achieving less than a complete remission after 3 cycles of treatment at their initial dose, step-wise escalation was permitted at each cycle if a cohort had been shown to have tolerated that dose. Results: Fifteen patients (9 males; median age=66 years; range=53 to 79 years) have been enrolled in the study to date – 10 with PMF (8 V617F+) and 5 with post-PV MF (all V617F+). The median follow-up to date is 13 weeks (range 1–28). Treatment was discontinued in 1 patient, in cohort 1, on the 2nd day of dosing after it was discovered that the patient had baseline QTcF prolongation (Grade 1). In the remaining 14 patients, no dose-limiting toxicities or other significant non-hematological adverse events have been observed to date. The pharmacokinetic parameters of TG101348 in plasma were well characterized following single daily oral doses of 30 mg, 60 mg, 120 mg or 240 mg. Dose-proportional pharmacokinetics were observed following single and multiple doses of TG101348. Treatment response data are as follows: Cohort 1 (starting dose=30 mg; 3 patients; median follow-up=26 weeks): 2 of the 3 patients had improvement in constitutional symptoms including fatigue, night sweats and pruritus. No other benefits were noted. These patients have now been dose escalated to either 120 or 240 mg/day. Cohort 2 (starting dose=60 mg; 3 patients; median follow-up=18 weeks): 1 patient had a 40% decrease in spleen size (18 to 10 cm). All 3 patients have now been dose escalated to 120 or 240 mg/day. Cohort 3 (starting dose=120 mg; 3 patients; median follow-up=13 weeks): 2 of the 3 patients reported improvement in constitutional symptoms including fatigue, night sweats, and pruritus. 1 patient had a 25% decrease in spleen size (28 to 21 cm). 2 patients have been dose escalated to 240 mg/day. Cohort 4 (starting dose=240 mg; 3 patients; median follow-up=8 weeks): 1 patient developed Grade 3 anemia (baseline Grade 1) at the end of the first cycle. 1 patient reported decreased fatigue. All 3 patients have experienced a 20 to 53% reduction in spleen size; baseline measurements were 10, 13, and 17 cm. Cohort 5 (starting dose=360 mg; 2 patients; median follow-up=1 week). Both patients had decreased splenomegaly within 1 week of therapy, including 1 with disappearance of palpable splenomegaly from a baseline of 20 cm; this patient also reported significant improvement in constitutional symptoms. Conclusions: Preliminary findings indicate that TG101348 is well tolerated in patients with myelofibrosis. Although it is too early to make valid conclusions, a decrease in spleen size has been noted at higher dose levels. Similarly, improvement in constitutional symptoms has been reported by some patients. Updated results on current and future patients, including data on pharmacokinetics, pharmacodynamics, and changes in V617F allele burden and plasma cytokine levels will be presented at the meeting.

Abstract Abstract 1280 Poster Board I-302 Philadelphia-chromosome negative myeloproliferative neoplasms (MPNs) including polycythemia vera (PV), essential thrombocytosis (ET), and primary myelofibrosis (PMF) are defined as clonal hematopoietic stem cell disorders. These disorders show an inherent tendency for transformation into leukemia (MPN-blast phase) which is hypothesized to be accompanied by acquisition of additional genomic lesions. We, therefore, obtained a comprehensive profile of genomic alterations associated with leukemic transformation by using single-nucleotide polymorphism (SNP) array in 88 MPN patients, as well as 71 cases with MPN-blast phase, and correlated these findings with their clinical parameters. A relatively high number of genomic alterations was found in MPN after leukemic transformation with 4.6 ± 0.6 abnormalities per sample compared to only 1.4 ± 0.2 changes per patient in chronic phase (p<0.001). Compared to the cytogenetic data, SNP-chip analysis detected about 47% additional chromosomal changes in the MPN samples, and 31% more in the MPN-blast phase cases, whereas SNP-array allelokaryotyping practically captured all cytogenetic abnormalities in our study population. Several additionally altered regions were detected in patients with MPN-blast phase compared to chronic phase, including both deletion and copy-number neutral-loss of heterozygosity (CNN-LOH) on chromosome 12p (9%) and 21q (9%), involving ETV6 and RUNX1. Notably, deletion and CNN-LOH on 17p involving TP53 were diagnosed in 18% of MPN-blast phase samples, which was highly associated with preceding treatment with alkylating agents (p=0.016). Moreover, trisomy 8, as well as amplification of 8q24.21 involving the MYC gene, were detected in 13% of patients with MPN-blast phase who were almost exclusively negative for the JAK2V617F mutation. Genome-wide inspection of further critical regions with promising new candidate genes involved in the evolution to the MPN-leukemic phase included deletion and CNN-LOH on 7q22.1 (SH2B2) in 18%, duplication/amplification on 19p13.2 (PIN1, ICAM1, CDC37) in 13% and 21q22.2 (ERG) in 9% of MPN patients with blast crisis. In contrast, we detected a decreased frequency of JAK2V617F in MPN-blast phase samples (52%) compared to chronic phase (71%). Also, the percentage of patients with homozygous mutant JAK2 as a result of CNN-LOH was lower in the MPN-blast phase (43%) compared to the chronic phase (53%). Taken together, the data suggest that gain-of-function mutation of JAK2 is not a perquisite for leukemic transformation. Remarkably, CNN-LOH on either 7q or 9p was related to decreased survival after leukemic transformation (p=0.02 and p=0.012, respectively). Given the variety of allelic imbalances, our data suggest that MPN-blast phase appears to be a heterogeneous disease prone to have evolved multiple mechanisms to provide a proliferative advantage to the abnormal leukemic clone. Our analysis of MPN genomes in the chronic compared to the leukemic stage provided new prognostic insights, as well as novel causative genes which might be involved in the transformation to MPN-blast phase. Disclosures No relevant conflicts of interest to declare.

Abstract Recently, a gain-of-function MPL mutation, MPLW515L, was described in patients with JAK2V617F-negative myelofibrosis with myeloid metaplasia (MMM). To gain more information on mutational frequency, disease specificity, and clinical correlates, genomic DNA from 1182 patients with myeloproliferative and other myeloid disorders and 64 healthy controls was screened for MPL515 mutations, regardless of JAK2V617F mutational status: 290 with MMM, 242 with polycythemia vera, 318 with essential thrombocythemia (ET), 88 with myelodysplastic syndrome, 118 with chronic myelomonocytic leukemia, and 126 with acute myeloid leukemia (AML). MPL515 mutations, either MPLW515L (n = 17) or a previously undescribed MPLW515K (n = 5), were detected in 20 patients. The diagnosis of patients with mutant MPL alleles at the time of molecular testing was de novo MMM in 12 patients, ET in 4, post-ET MMM in 1, and MMM in blast crisis in 3. Six patients carried the MPLW515L and JAK2V617F alleles concurrently. We conclude that MPLW515L or MPLW515K mutations are present in patients with MMM or ET at a frequency of approximately 5% and 1%, respectively, but are not observed in patients with polycythemia vera (PV) or other myeloid disorders. Furthermore, MPL mutations may occur concurrently with the JAK2V617F mutation, suggesting that these alleles may have functional complementation in myeloproliferative disease.