ALBERT

Description: Abstract 1026 Poster Board I-48 Background: FLT3 mutations (ITD or D835 point mutation) are frequently observed in patients (pts) with AML and they confer an adverse prognosis, particularly among pts with diploid karyotype. This has made FLT3 an important target for drug development in AML. Several FLT3 inhibitors are currently being developed (eg, sorafenib, PKC-412, AC-220, CEP-701, IMC EB10, sunitinib). Results from early trials with many of these agents suggest they have clinical activity in the treatment of MDS and AML, although most responses are represented by a marked decrease in blast counts, with few complete remissions(CR). Whether these responses ultimately improve long-term outcome of pts, and whether they may be particularly beneficial for pts with FLT3 mutations compared to those with FLT3 wild-type (WT) is being investigated. Aims: To ascertain outcomes of patients given treatment with FLT3 inhibitors, alone or in combination with other therapies, and to compare outcomes in those patients with FLT3 mutations (ITD or D835) versus those with FLT3-WT. Methods: We reviewed the records of patients with MDS and AML who were enrolled on clinical trials with FLT3 inhibitors at our institution. We compared patient outcomes in those who received a FLT3 inhibitor in both FLT3 positive and FLT3 negative patients. Pts were classified as receiving FLT3 inhibitors 1) as part of their initial therapy, 2) as first salvage, or 3) as second salvage or beyond. Results: A total of 128 pts were included: 51 (40%) with FLT3-WT, 56 (44%) with FLT3-ITD, 11 (9%) with D835, and 10 (8%) had both FLT3-ITD and D835. The overall median age was 62 yrs (range, 17-88); by FLT3 status, median age was 70 yrs (35-88) for FLT3-WT pts and 58 yrs (17-81) for FLT3 mutated. Sixty-four pts (50%) were female. Twenty-three (18%) pts received FLT3 inhibitors as part of their induction therapy (18 FLT3-WT, 5 FLT3 mutated; median age 74 yrs); 22 (17%) as first salvage (4 FLT3-WT, 18 mutated; median age 67 yrs); and 83 (65%) as second or later salvage (29 FLT3-WT, 54 mutated; median age 59 yrs). Nine pts overall, all of whom were FLT3 mutated, achieved either CR (n=6) or CRp (n=3) with FLT3 inhibitors. Eight of the nine CR/CRp have been lost with a median CR duration of 8 months (mo) (3-12+). After a median follow-up of 3.5 mo, 115 (90%) pts have died, including 47 (92%) FLT3-WT, and 68 (88%) FLT3 mutated. The median survival is 3.8 mo for the total population. Survival by mutation status and timing of FLT3 inhibitor therapy is presented in table 1. Conclusions: Despite the inferior outcome expected for pts with FLT3 mutations, and the low rate of CR/CRp with FLT3 inhibitors, these results suggest that therapy with FLT3 inhibitors has the potential to improve the outcome of pts with FLT3 mutations. Additional studies incorporating these agents in AML therapy are warranted. Disclosures: Off Label Use: Sorafenib has not been FDA approved for use in MDS and AML. Kantarjian:Novartis: Research Funding. Cortes:Ambit: Research Funding; Novartis: Research Funding; ImClone: Research Funding.

Description: Clofarabine (CLO) is a second-generation nucleoside analog with activity in acute leukemias. Whereas it has received FDA approval in children with relapsed acute lymphoblastic leukemia, the focus of clinical research in adults has shifted to AML. To improve single agent activity, various CLO combinations are being studied. Here we present the results of two dose-finding phase I studies, exploring combinations of CLO with idarubicin (IDA) [CI] and CLO+IDA+Ara−C [CIA] in patients (pts) with relapsed or refractory AML and high-risk MDS. Eligibility for CI required previous Ara-C and primary refractory disease, or relapsed AML with first remission duration (CRD1) 〈 12 mos. Eligibility for CIA required either no previous Ara-C or Ara-c with CRD1 ≥ 12 mos. Maximum tolerated dose (MTD) was determined by “3+3” method. Forty-four pts (CI 23; CIA 21) have been accrued and are evaluable. The median age of pts on CI was 56 yrs (range 24–71), 9 were primary refractory and 14 in first relapse (median CRD1 2 mos. [0–9]). Fifteen pts had abnormal cytogenetics (including 6 with -5, -7, or 11q23). Sixteen pts had intermediate- or high-dose Ara-C-based prior therapy, 2 relapsed from allogeneic transplant (SCT). The median age of pts on CIA was 56 yrs (23–78), 8 were primary refractory and 13 in first relapse (median CRD1 9 mos [0–61]). Twelve pts had an abnormal karyotype (-5, -7, or 11q23 in 6 pts). Twelve pts received intermediate dose Ara-C-based prior therapy and 2 failed unrelated donor SCT. The starting dose level of the CI group was CLO 22.5 mg/m2 iv daily x 5d and IDA 12 mg/m2 iv daily x 3d. Three DLTs occurred (diarrhea, rash, ↑ bilirubin) requiring dose de-escalation to CLO 15 mg/m2/d x 5d and IDA 8 mg/m2/d x 3d. Gradual re-escalation led to CLO 30 mg/m2/d x 5d and IDA 10 mg/m2/d x 3d where 2/5 pts experienced DLTs (↑ SGPT, ↑ bilirubin, headache). MTD was defined by the next lower dose level: CLO 22.5 mg/m2/d x 5d and IDA 10 mg/m2/d x 3d. Three (13%) CR occurred. The CIA group started at CLO 22.5 mg/m2/d x 5d, IDA 8 mg/m2/d x 3d, and Ara-C 1g/m2/d x 5d. Two pts developed DLTs (diarrhea, ↑ bilirubin, renal failure) requiring de-escalation to CLO 15 mg/m2/d x 5d, IDA 6 mg/m2/d x 3d, and Ara-C 0.75 g/m2/d x 5d. Re-escalation to CLO 30mg/m2/d x 5d, IDA 6mg/m2/d x 3d, and Ara-C 0.75 g/m2/d x 5d revealed DLTs in 2/6 pts (diarrhea, mucositis, ↑ bilirubin). MTD was thus defined at next lower dose level of CLO 22.5 mg/m2/d x 5d, IDA 6 mg/m2/d x 3d, and Ara-C 0.75 g/m2/d x 5d. Ten (48%) responses (9 CR, 1 CRp) occurred. The higher response rate of CIA may be due to the difference in pt characteristics between the two trials. A phase II study is now in progress adaptively randomizing patients with AML relapse to CI versus CIA versus CLO (40 mg/m2/d x 5d) plus Ara-C (1 g/m2/d x 5d) (as established in an earlier study) to assess activity of the combinations.

Description: Background: Although arsenic trioxide (As2O3) is active in vitro against diverse hematological cancers, clinical data show activity only in acute progranulocytic leukemia (APL); there is little if any single-agent activity in multiple myeloma and other hematological cancers. This discordance may be because relatively little As2O3 can be given consequent to its substantial toxicity (especially QTc-prolongation) or because the mode of action (MOA) in APL (differentiation) is inoperative in other hematological cancers (or both). ZIO-101 (S-dimethylarsino-glutathione), a new organic arsenic, is active against diverse cancers in vitro and in animal models including AML and MM. Cell-killing by ZIO-101 is mediated by mitochondrial-disruption and apoptosis-induction rather than the differentiation MOA of As2O3. ZIO-101 can be given at doses ≥ 50-fold higher than As2O3 and achieves 5–10-fold higher intracellular concentrations at equimolar extracellular As concentrations. Gene-expression profiling data suggest different cellular responses to ZIO-101 and As2O3. These features make ZIO-101 attractive for evaluation in AML and MM. Methods: 2 phase-1 studies evaluating safety, activity and pharmacokinetic (PK) profile of ZIO-101 in 21 subjects with advanced AML (N=8) or MM (N=13). Median age is 58 y (range, 41–85 y). Median N of prior therapies is 5 (range, 2–12) including 4 subjects failing prior As2O3 therapy. Starting dose was 78 mg/me2/d IV for 5 consecutive d every 4 w. Results: Subjects received a median of 2 cycles (range, 1–6). Therapy was well-tolerated; adverse events ≥ grade-2 occurring in 〉 25% of subjects included neutropenia, hypokalemia and hyperglycemia. There was no clinically-important renal, liver or heart toxicity nor QTc-prolongation. Maximum tolerated dose (MTD) was 420 mg/me2/d. Pharmacokinetic (PK) studies at this dose showed a tmax=1 h (SD±0.9), Cmax=1.06 μg/mL (SD±0.07 μg/mL), t1/2=17.8 h (SD±1.4 h) and AUC0-∞=25.9 μg·h/mL (SD±0.8 μg·h/mL). 4 subjects with AML had stable disease (SD) after 1 cycle and received 2–4 more cycles before progressing. Blood myeloblasts decreased substantially in 1 subject and completely resolved in 2. Bone marrow myeloblasts decreased in 1. 1 subject with prior myelodysplastic syndrome (MDS) stopped requiring frequent RBC transfusions. 1 subject with rapidly-progressing As2O3- and bortezumib-resistant MM has stable disease (SD) 〉6 mo. Conclusions: These early data suggest activity of ZIO-101 in advanced AML and MM. The MTD is 420 mg/me2/d, ≥ 50-fold higher than As2O3. Plasma levels exceed the IC50 for AML and MM cell in vitro and animal models. Clinically-important QTc-prolongation was not seen. Some subjects failing As2O3 responded to ZIO-101 indicating efficacy of a higher dose, different MOA or both. Because of these favorable results phase-2 studies in hematological cancers are in progress.

Description: Treatment of AML is guided by cytogenetics, age, patients’ medical condition and increasingly, new molecular markers. However, no definitive algorithms and biomarkers predicting response to induction chemotherapy exist. We recently showed that distinct protein profiles of AML correlate with cytogenetics, FAB classes and treatment outcome for a class/group of patients as a whole (Tibes, ASH 2005; Kornblau, ASH 2006 and manuscript in submission). To further individualize therapy, we applied a set-based decision modeling methodology based on the Rough Set Theory (RST) to our initial proteomic dataset of 34 (phospho)-proteins for 96 samples from 73 patients (Tibes, ASH 2005). The first model was built based on partitioning discretized protein expression levels of proteins for the 96 samples into 2 outcome categories: complete remission (CR) vs. refractory to treatment; we generated reducts for all proteins and computed pseudo-cores. Several proteins (incl. phospho-NPM1, PTEN, phospho-AKT, p53, cyclin D1 etc.) were able to distinguish outcomes in terms of CR and refractoriness. A second model based on discretized clinical condition attributes (e.g., blast %, FAB, cytogenetics, age, sex, blood counts etc.) yielded the most important clinical reducts. In a third model, protein expression and clinical attributes were combined to generate decision rules to predict outcome to induction therapy for each patient. For proof-of-concept purposes, the data set was divided into a training set and a test set. Prediction accuracy for CR vs. refractory state was in the 75% range for individual patients based on their pre-treatment clinical and proteomic attributes. In conclusion, predictive models obtained with RST yield reducts corresponding to a small but significant number of proteins (6–8) and clinical parameters which can further be used to derive association rules, capable of accurately predicting a patient’s response to induction therapy. The availability of several dozens of distinct decisions rules within each group (CR vs. refractory) based on protein expression and clinical characteristics, allows an individualized approach for each patient. This preliminary study is now being expanded on an independent proteomic dataset of 52 proteins for 258 AML samples (Kornblau, ASH 2006) to be presented at the Annual Meeting. The clinical utility lies in the fact that measuring expression levels of our protein biomarker profiles can be adopted to routine clinical testing (e.g., immunohistochemistry on diagnosis marrows) and potentially help to direct patients towards standard chemotherapy, upfront treatment on a clinical trial or evaluation for early transplant depending on the chance of response and relapse. Lastly, this model can be applied to predict other clinical outcomes (relapse vs. continuous CR), as well as to derive predictive signatures of gene expression datasets.

Description: Sequence-specific single-stranded DNA binding protein 2 (SSBP2) is a candidate tumor suppressor gene, located at chromosome 5q13.3, that is thought to regulate hematopoietic growth and differentiation. Loss of SSPB2 expression is associated with deletions of chromosome 5 and replacement in cell lines induces partial differentiation and loss of clonogencity. Using a highly specific antibody, we studied the level of expression and prognostic relevance of SSBP2 protein expression (along with 51 other total and phospho proteins as described in ASH abstract #107, 2006) in leukemia-enriched samples from 423 patients (258 newly diagnosed, 47 primary refractory, 118 relapse 1, 2 or refractory) using high-throughput reverse phase protein array technology (RPPA). Levels of SSBP2 were lower, equal and higher than that of normal CD34+ cells in 2.7%, 58.7% and 11.6% of cases respectively. All 7 cases with low levels were female and 21/30 with high levels were male. The expression of SSBP2 was not equally distributed in the different FAB subtypes with the lowest median levels in M5 and the highest in M0, M1 and M2. Levels were significantly higher in favorable cytogenetics (t(8;21) and Inv (16) compared to intermediate (p=0.05) or unfavorable (p=0.008) cytogenetics. Median levels were lowest in cases with cytogenetic abnormalities on chromosome 5 or with an 11q23 abnormality. Levels were significantly higher in primary refractory and relapse samples compared to diagnostic samples. The levels of SSBP2 expression was strongly positively correlated with expression levels of anti-apoptotic proteins BCL2, Bcl-XL, caspase inhibitors SMAC, Survivin and XIAP, and inversely correlated with inactivated Bad-p112. Expression also was positively correlated with proliferation regulating proteins β-Catenin, Cyclin D, GSK3, Myc, P27, as well as VEGF receptor neuropilin and P53. Levels were strongly negatively correlated with activated signal transduction proteins (STP) including phosphorylated AKT (phosphorylated on threonine 308 or serine 473), pMEK, pERK2, pP38, pPKCα and as well as activated P70S6K. For outcome analysis martingale residuals were utilized to define an optimal cutpoint for dichotomization into groups with higher (n=130, 111 treated) vs. lower (n=128, 106 treated) SSBP2. Complete remission rates (58% vs. 63%, p=0.39), relapse rates (53% vs. 49%, P=0.59, and median remission duration (37 vs. 39 weeks, p=0.37) did not differ between patients with lower vs. higher SSBP2. However patients with higher SSPB2 levels had significantly longer median survival (40 vs. 31 weeks, p=0.015) compared to patients with lower SSBP2 levels. In summary, SSBP2 expression is heterogeneous in AML ranging from low in the poor and mixed prognosis subsets and high in the good prognosis t(8;21). Consistent with its demonstrated role in hematopoietic differentiation, the levels correlate negatively with expression of proliferation enhancing and STP. Nonetheless, the significance of positive correlation with antiapoptotic proteins is unclear at present. But the association with favorable longer overall survival suggests that therapeutic agents targeting the SSBP2 pathway function might have efficacy in AML.

Description: Deregulation of signal transduction pathways (STPs) may promote leukemogenesis by conferring cell proliferation and survival advantages in acute myelogenous leukemia (AML). Several agents targeting STPs are under development; however, redundancy and cross-talk between STPs could activate multiple downstream effectors and this could negate the effect of single-target inhibition. The frequency of concurrent activation of multiple STPs in AML and the prognostic relevance of STP activation in AML are unknown. STP protein expression (PKCα, ERK2, pERK2, AKT, and pAKT) was measured by Western blot in samples from 188 patients with newly diagnosed, untreated AML. In univariate and multivariate analysis high levels of PKCα, ERK, pERK, and pAKT, but not AKT, were adverse factors for survival as was the combination variable PKCα-ERK2&pERK2-pAKT. Survival progressively decreased as the number of activated pathways increased. Patients were more likely to have none or all 3 pathways activated than was predicted based on the frequency of individual pathway activation, strongly suggesting that cross-activation occurred. Simultaneous activation of multiple STPs is common in AML and has a progressively worse adverse effect on prognosis. It is thus likely that only combinations of agents that target the multiply activated STPs will be beneficial for patients with AML.

Description: We have used Reverse Phase Protein Arrays (RPPA) to perform proteomic profiling in Acute Myelogenous Leukemia (AML) focusing on cell cycle, apoptosis and signal transduction pathway proteins (ASH 2006, abstract #107). Protein expression signatures were derived from this dataset of 436 AML patients, analyzed for 30 total and 22 phopho- proteins. The predictive ability of these RPPA derived protein expression signatures has not been prospectively tested to determine if they are valid. This dataset presented an opportunity to validate this as there was a population of patients with known FLT3-ITD and D835 mutation status (n=297) and another population where the status was unknown (n=139), among which 55 had sufficient sample available for mutation analysis. Prior to performing the mutation analysis a predictive model was built using linear regression with part of the data utilized for training and the reminder for validation. The model was designed to predict for the presence of mutation, either ITD or D835, although there are differences int eh signature of each. The total population had 85 cases with FLT3-ITD and 15 with the D835 mutation. The optimal model that was developed, using 30%, 50% and 70% of the samples for training and the remainder for validation, had a median validation accuracy of 68%, 70% and 72% respectively. Prospective predictions of FLT3-ITD or D835 mutation status were then made for all samples lacking FLT3-ITD or D835 mutation data. Mutation analysis was then performed using PCR amplification followed by 2-D gel electrophoresis (FLT3-ITD) to evaluate for PCR product size, or sequencing (D835) on 55 samples. This revealed 9 cases with FLT3-ITD, 3 with a D835 mutation, 1 with both and 43 without mutation. Among these 55 cases the model correctly predicted that 8 of the 12 mutant cases would be mutant including 8 of 10 with a FLT3-ITD, but 0 of 2 with only the D835 mutation. Among the 43 wildtype cases 36 were accurately predicted to be wildtype, while 7 were incorrectly predicted to have the mutation mutant. This yields an overall accuracy (OA) of 80%, Sensitivity =66%, Specificity=90%, Positive Predictive Value (PPV) of 53%, False positive rate of (FPR) of 16%. Since most patients with FLT3-ITD have Diploid cytogenetics we also looked at the predictive accuracy of the protein expression signature in that population. Among 23 patients with Diploid cytogenetics the overall accuracy was OA) of 83%, Sensitivity =75%, Specificity=87%, Positive Predictive Value (PPV) of 75%, False positive rate of (FPR) of 13%. Since FLT3-ITD and D835 carry different prognostic impact, and had different protein expression signatures, greater accuracy might be achieved if separate models were developed for each mutation individually. The model demonstrated that RPPA derived protein expression signatures can accurately be used to predict mutation status providing the first prospective validation of protein expression signatures in AML.

Description: Functional activation of the PI3K/AKT signaling pathway provides survival signals to leukemic cells and pathway blockade may facilitate cell death. We have previously demonstrated that activation of AKT (phosphorylated on threonine 308 and serine 473 [pAKT308 & pAKT473]) pathway is a frequent event with adverse prognostic consequences in AML. To further explore the upstream regulation and downstream consequences of activation of this pathway in AML we assessed the level of expression and activation of PTEN/phosphoPTEN (pPTEN) (which controls AKT activation), P53 and DJ1 (which regulates PTEN), along with targets of activated AKT including proteins regulating proliferation (S6RP, Cyclin D1, GSK3, MTOR), and apoptosis (MCL, BAD, XIAP, BCL2). Total and phospho-protein levels from 258 leukemia-enriched samples of newly diagnosed, untreated AML patients were measured using high-throughput reverse phase protein array technology (RPPA). High P53 expression negatively correlated with PTEN levels (p=0.001). Almost all cases with very high P53 had unfavorable cytogenetics with complex karyotypes, and P53 mutations were found in 54% cases (see accompanying abstract). AKT activation, determined as high ratios of phospho to total (P/T) pAKT473 and pAKT308 were strongly correlated with loss of PTEN expression or increased PTEN phosphorylation,(p=0.0006 & 0.006) but not with increased levels of PTEN suppressor DJ1 (P=0.33 & 0.34). AKT activation (pAKT473 & pAKT308 P/T ratios) was associated with increased signaling through mTOR kinase (increased phosphorylation of S6RP on p236, p240, p244, p

Description: Protein expression and activation determines the pathophysiology of leukemic cells in Myelodysplasia (MDS) and Acute Myelogenous Leukemia (AML) and is dependent on endogenous changes (e.g mutation, methylation) and exogenous signals from stromal interactions, cytokines (CTKN) and chemokines. We have previously performed proteomics on primary AML sample (using reverse phase protein arrays) and wanted to assess how cytokines affect protein expression and phosphorylation. Prior studies of CTKN expression in AML and MDS have generally measured individual CTKNs, but not provided an overall assessment of CTKN expression. We measured the level of 26 CTKN (IL-1RA, 1B, 2, 4 5, 6, 7 , 8 , 9, 10,12, 13, 15, 17, Eotaxin, FGFB, G-CSF, GM-CSF, IFNγ, IP10, MCP1, MIP1α, MIP1β, PDGF, TNFα and VEGF) using multiplex cytometry (Bioplex™, Biorad) in serum samples from 176 AML (138 untreated (New), 38 relapsed (REL)) and 114 MDS patients (97 New, 10 post biological therapy, 7 REL) and 19 normal (NL) subjects. Individual CTKN expression was not correlated with clinical features (e.g. age, gender, cytogenetics, FAB, HB, WBC, platelet etc). The levels of IL -1β, 4, 5, 6, 7,10,12, 13, 17, IFNγ, FGFB and MIP1α were significantly lower and IL-8 and 15 higher in AML/MDS compared to NL. The expression profiles of AML and MDS were statistically indistinguishable whether analyzed individually or by unsupervised hierarchical clustering, except for IL-8 and 13 (higher in AML) and VEGF (higher in MDS). When CTKN were evaluated individually in new AML cases higher levels of IL4, 5 and 10 correlated significantly with remission attainment, and higher levels of IL8, Il1Ra, IP-10, Mip1β, VEGF and ILR, correlated significantly with shorter survival. No CTKN predicted remission attainment or survival in MDS. Unsupervised hierarchical bootstrap clustering using Pearson correlation and average linkage of CTKN expression relative to other CTKN expression, where high levels of one CTKN correlated with high expression of the other, revealed 6 highly reproducible expression patterns: IL-1β 4, 7, 10, 12, 13, G-CSF, IFNγ, MIP1α and PDGF IL 1ra, 6, 8 Eotaxin, IP-10, MIP1β and VEGF, IL2, 9, 15 and GMCSF, IL5 IL-7, FGF-Basic, TNFα and MCP1. Similar unsupervised clustering of the samples based on CTKN expression using average linkage also revealed 5 disease clusters and a NL sample cluster (containing all 19 NL samples). Average expression levels of each CTKN in these 5 clusters show diminished expression of most CTKN that had high expression in the NL samples, with each group showing increase in expression in a distinct subset of CTKN relative to NL. Remission attainment was strongly associated with cytokine signature (P=0.005). Additional CTKN are being studied (SCF, TGFβ, IL3). Comparison of CTKN expression patterns with proteomic profiling of expression and phosphorylation status is pending. In summary, this is the largest sample set studied for multiple CTKN expression in AML and MDS and the first assessment of many of these CTKN in these diseases. Most CTKNs showed different expression in AML and MDS compared to NL. Interestingly, CTKN expression in AML and MDS were similar. Many CTKN are predictive of outcome individually. CTKN signatures distinguish groups of patients and are predictive of outcome. Correlation with proteomic profiling may suggest CTKN to target in combination with other targeted therapies to maximally affect activated pathways.

Description: Abstract 2040 Poster Board II-17 The main challenge in the treatment of acute lymphocytic leukemia (ALL) is overcoming resistance to chemotherapy. Recent studies indicate that interactions between leukemia cells and bone marrow (BM) microenvironment promote leukemia cell survival and confer resistance to drugs commonly used to treat ALL. We investigated whether hypoxia was a contributing factor in the protective role of the BM microenvironment. We found that the Hypoxia-Inducible-Factor 1a (HIF-1a; a marker normally expressed by only a few hematopoietic cells) was expressed in 68% of BM biopsies from patients with B-lineage ALL (n=53). Expression of HIF-1α detected either histochemically (n=53, p=0.023) or by Reverse Phase Protein Arrays (RPPA, n=116, p=0.0013) inversely correlated with survival of patients with newly diagnosed B-lineage ALL. Silencing of HIF-1α with siRNA, or blockade of mTOR signaling with rapamycin derivatives, reduced expression of the glucose transporter Glut-1, diminished glucose flux, decreased glycolytic rate and ATP production and sensitized leukemic cells to the pro-apoptotic effects of chemotherapeutic agents under hypoxic conditions. In line with this findings, we observed a marked expansion of hypoxic BM areas in immunodeficient mice engrafted with the ALL cell line Nalm6 or with primary ALL cells, as detected by administration of the reductive 2-nitroimidazole compound pimonidazole (PIM), which forms stable adducts in hypoxic regions. Altogether, these findings provided a rationale for examining the effects of hypoxia-activated pro-drugs or HIF-1a inhibitors to eliminate ALL progenitor cells within hypoxic niches. To this end, we tested PR-104, a hypoxia-activated dinitrobenzamide mustard currently undergoing Phase II trials in solid tumors. Under hypoxic conditions, this agent is reduced to hydroxylamine and amine metabolites that in turn induce DNA cross-links and cell death (Patterson et al., Clin Can Res 2007). In vitro, 25μM PR-104 induced hypoxia-selective cell death in Nalm6 ALL cells with 80% Annexin V-positivity at 0.1% O2, 46% at 1%O2 compared to 13% at 21%O2. The anti-leukemic efficacy of PR-104 was next examined in the in vivo leukemia models. Administration of PR-104 (250 mg/kg IP TIW for two weeks) prolonged survival of NOD/Scid/IL2Rg-KO (NOG) mice injected with cells from primary refractory FLT3-mutated AML, and decreased leukemia burden as indicated by histopathological analyses of CD45 positive cells in the BM, spleen, lung and liver. Notably, analysis of PIM distribution indicated clearance of the hypoxic leukemic niches. In NOG mice injected with leukemic cells from an infant with MLL-rearranged B-lineage ALL, PR-104 at 200 mg/kg IP on days 1, 2 and 6 resulted in a dramatic decrease in the percentage of circulating leukemic CD45+ cells on day 15 (control, 92%±6%; treated, 9%±4%; n=7 mice/group). The therapeutic effect of the drug was also tested in a Nalm6-luciferase ALL model where PR-104 administration resulted in decreased tumor burden as determined by luciferase activity and prolonged survival of the PR-104 treated as compared to control mice (p=0.006). Similar to the models of human leukemia, analysis of BM sections of control mice showed extensive areas of hypoxia (PIM-positive) in close proximity to GFP-positive leukemia cells in contrast to the treated mice in which only discrete areas of PIM positivity were detectable. Altogether, these findings strongly suggest that targeting hypoxia is feasible and may increase the sensitivity of ALL cells to chemotherapy. If successful, this approach of targeting hypoxic microenvironment, alone or in combination with other chemotherapeutic or targeted agents, may significantly impact ALL therapy and ultimately improve patient survival. Figure 1. Co-localization of hypoxic PIM(+) areas with GFP positive HALMG tumor areas in bone marrow of control but not of PR-104 treated mice. Figure 1. Co-localization of hypoxic PIM(+) areas with GFP positive HALMG tumor areas in bone marrow of control but not of PR-104 treated mice. Disclosures: No relevant conflicts of interest to declare.