ALBERT

Description: Abstract 65 The canonical WNT-β-catenin pathway is essential to the cellular processes of self-renewal, growth and survival. Deregulated WNT-β-catenin in transformed hematopoietic progenitor cells inhibits the multi-protein degradation complex formed by axin, adenomatous polyposis coli (APC) and glycogen synthase kinase 3β (GSK3β). This results in the preservation, nuclear translocation and interaction of β-catenin with the T-cell factor (TCF)/lymphoid enhancer factor (LEF) transcription factor, which regulates the expression of genes such as cyclin D1, Myc, survivin and Axin. BC2059 (β-Cat Pharmaceuticals) is a potent small molecule, anthraquinone oxime-analog inhibitor of the WNT-β catenin pathway. Treatment with BC2059 mediates the degradation of β-catenin. In the present studies, we determined the activity of BC2059 in human cultured and primary CML and advanced MPN versus normal progenitor cells. Exposure to 50 to 100 nM of BC2059 induced cell cycle G1 phase accumulation and apoptosis (40 to 80%) of the cultured MPN cells HEL92.1.7 (HEL) and UKE1 cells expressing the mutant JAK2V617F, as well as of the CML K562 and LAMA-84 cells expressing BCR-ABL. BC2059 treatment also induced apoptosis of CD34+ primary MPN cells derived from the peripheral blood of patients with advanced MPN expressing mutant JAK2, as well as of primary CD34+ CML progenitor cells. In contrast, as compared to the untreated controls, BC2059 treatment did not induce apoptosis of normal CD34+ progenitor cells. Exposure to BC2059 resulted in marked down regulation of β-catenin protein levels and the activity of the LEF1/TCF4 transcription factor, which was accompanied with reduced levels of cyclin D1, MYC, survivin and up regulation of Axin 2 levels, as detected by immunoblot analyses of the cell lysates of BC2059-treated CML and MPN cells. We also determined the in vivo anti-MPN activity of BC2059. Following the tail vein infusion of HEL cells and establishment of MPN, NOD-SCID mice were treated with 15 or 20 mg/Kg of BC2059 administered b.i.w for three weeks via the tail vein. As compared to the control, BC2059-treated mice demonstrated significantly improved survival (p

Description: Abstract 4574 Background CTCLs are generally incurable with conventional therapies. In particular, advanced mycosis fungoides (MF), Sézary syndrome (SS) and gamma delta varieties of CTCL have poor survival rates and are often refractory to traditional chemotherapy. Allogeneic SCT may provide a GVL effect and improve outcomes for these patients. Methods A retrospective analysis was performed at the University of Pennsylvania to identify all patients with CTCL who underwent allogeneic transplantation. 12 patients were identified who were transplanted between 2004 to 2010. A chart review was performed to obtain data about demographics, diagnosis, staging, treatment, transplantation and outcomes. Results Median age at diagnosis was 49 yrs and M:F ratio was 5:7. Prior to transplantation, 4 had MF (stages IIB, IIIB, IVA1, IVB; 2 with nodal transformation), 4 had SS (one stage IVA1, three IVA2; 1 with nodal transformation), and 3 had gamma delta T-cell lymphoma (all T3b). Median time from diagnosis to transplantation was 3.3 yrs (range 0.5@02b97 yrs). Patients had received a median of 8 non-chemotherapy, and 2 chemotherapy-based treatment modalities before being transplanted. Only 3 patients were in complete remission (CR) at the time of conditioning and 9 had evidence of active disease. Reduced intensity conditioning (RIC) was used in 10 cases (Flu/Bu, Flu/Cy or Flu/Mel), and conventional myeloablative conditioning (Cy/TBI) was used in 2. GVHD prophylaxis consisted of calcineurin inhibitor and methotrexate in all patients. The median follow up for all pts is 6.6 months (range 1.4 to 37.1 months) and 11.2 months for surviving patients. All patients engrafted with an ANC 〉500 a median 13 days after SCT. Median donor chimerism at day 100 after SCT in 10 evaluable pts was 97%. 7 of 12 patients developed acute GVHD, 4 of whom had grade 3 GVHD. Two patients died within the first 100 days, from sepsis with active disease. At day 100, 7 of 10 evaluable patients were in CR, with an additional patient achieving CR shortly after; therefore transplant induced and maintained CR in 6 pts with active disease. 3 patients relapsed after achieving CR a median of 11.4 months (range 5.3–13.0 months) after SCT. 2 patients never achieved CR, and progressed within a month of transplantation. The median PFS for all pts was 31 wks, and 1 yr and 2 yr PFS were 48% and 32% without an obvious plateau. 2 year OS was 53% (Figure 1). Median OS is not reached. 6 patients have died from progression (5) and GVHD (1), 5 remain in CR and 1 is alive with active disease. Conclusion RIC SCT can provide long-term disease control in patients with advanced CTCL otherwise refractory to immunotherapy and chemotherapy. Given the limited TRM, consideration for earlier transplant should be given. Larger retrospective and ideally prospective studies will further define the role of allogeneic SCT in this disease. Disclosures: Rook: Therakos: Speakers Bureau; HY Biopharma: Consultancy. Kim:TenX: Research Funding; Biocryst: Research Funding; Genmab: Research Funding; Glouchester: Research Funding; Celgene: Research Funding; Eisai: Consultancy.

Description: Abstract 2611 Approximately 30% of acute myeloid leukemia (AML) patients have activating mutations in FLT3, commonly internal tandem duplication (ITD) mutations, which are associated with poor survival. Although FLT3 tyrosine kinase inhibitors (TKI) such as AC220 can induce remissions, resistance-causing mutations in FLT3-ITD are known to impair the in vitro activity of first and second generation FLT3 TKIs. DCC2036 is a unique switch-pocket, non-ATP competitive (allosteric) inhibitor with low nanomolar inhibitory concentration 50 (IC50) activity against a number of tyrosine kinases including FLT3 (1.7 nM), TRKA (7.0 nM), TIE-2 (2.7 nM) and BCR-ABL (2.0 nM). DCC2036 has shown promising activity in a phase I/II clinical trial in chronic myeloid leukemia (CML), where plasma concentrations of 350 nM of DCC2036 have been safely achieved. DCC-2036 has induced clinical and molecular remissions in patients with TKI-resistant CML expressing the ‘gate-keeper’ T315I BCR-ABL mutation, as well as demonstrated activity against mutations that cause BCR-ABL conformational escape resistant (Cancer Cell. 2011;19:556). Here, we evaluated the in vitro activity of DCC2036 against FLT3-ITD in cell line model systems. In the FLT3-ITD expressing human leukemia MV4-11 and MOLM-13 cells, treatment with DCC2036 (20 to 500 nM for 24 hours) dose-dependently induced cell cycle G1-phase accumulation with decline in the S and G2/M phases. Exposure to 50 to 500 nM DCC2036 for 48 to 72 hours also dose-dependently induced apoptosis of 30 to 80 % of MV4-11 and MOLM-13, as well as induced 30 to 50% apoptosis of patient-derived primary AML cells with FLT3-ITD (n =4). This was associated with dose-dependent decline in the levels of p-FLT3, p-STAT5, p-AKT, p-ERK1/2 and Bcl-xL levels but increase in the levels of BIM and p27. In contrast, treatment with DCC2036 induced significantly lower level of apoptosis (1000nM). DCC2036 retained some activity against the clinically relevant FLT3-ITD gatekeeper mutation F691L and F691I (IC50 49 nM and 34 nM), and was similarly active against the activation loop mutations Y842C and Y842H (IC50 26–28 nM). The activation loop mutations D835V and D835Y, which are commonly detected in patients with loss of response to AC220 and are hypothesized to destabilize the kinase inactive “DFG-out” conformation, were substantially less sensitive to DCC2036 (IC50 233 nM and 196 nM, respectively). Based on our previous findings (Blood. 2005;105:1768) that FLT3-ITD is a heat shock protein (hsp) 90 client oncoprotein, we also determined the effect of co-treatment with the non-geldanamycin hsp90 inhibitor AUY922 (5 to 10 nM) (Novartis Pharmaceuticals) against the cultured and primary FLT3-ITD expressing AML cells. Co-treatment with AUY922 significantly improved the activity of DCC2036 against primary AML cells (p 〈 .05). These findings demonstrate that DCC2036 exhibits potent activity against cultured and primary AML cells with FLT-3-ITD, as well as against cellular models of FLT3-ITD with AC220-resistant gatekeeper and select activation loop mutations. The molecular basis of resistance to DCC2036 conferred by activation loop mutations at D835 is under investigation. Co-treatment with DCC-2036 and the hsp90 inhibitor AUY922 exerted higher lethal activity against cultured and primary AML cells with FLT3-ITD. Disclosures: Wise: Deciphera Pharmaceuticals LLC: Employment. Reyes:Millennium, Sanofi Aventis: Consultancy. Berger:Deciphera Pharmaceuticals: Employment. Rutkoski:Deciphera Pharmaceuticals: Employment.

Description: Abstract 1939 Background: RIC SCT relies heavily on graft-versus-tumor alloreactivity, and relapse remains a major barrier to a favorable outcome. Early prediction of relapse would allow early intervention such as donor lymphocyte infusions (DLI), but factors that predict relapse and methods for detection of minimal residual disease are often disease specific and not standardized. The level of donor-recipient chimerism has been associated with both graft rejection and relapse, but the optimal timing, desired level, and predictive value of chimerism testing in RIC SCT are unclear. Furthermore, the clinical utility of whole blood (WB) vs T-cell chimerism is not well defined, particularly after RIC SCT. Methods: We aimed to assess the predictive value of early WB and T-cell chimerism on the incidence of relapse in SCT pts receiving a uniform RIC regimen for hematologic malignancies. Between August 2006 and May 2011, 120 consecutive patients (pts) underwent allogeneic SCT following conditioning with fludarabine 120mg/m2 and busulfan i.v. 6.4 mg/kg. Pts were not treated with pre-emptive DLI, but could receive DLI for relapse. Hematopoietic chimerism was determined by DNA genotyping of short tandem repeats. Chimerism was determined on WB and then on enriched T-cells, obtained by selection using CD3-labeled magnetic beads. We conducted a cumulative incidence analysis of relapse, using day 30 as a landmark to determine the predictive properties of chimerism studies obtained at that time. Results: The 1-yr cumulative incidence of relapse in this cohort was 48.3 ± 4.7%, and the median time to relapse was 102 days (range 16–566 days), highlighting the importance of early monitoring. 68 pts were in remission and had evaluable chimerism data on day 30; importantly, the 1-year incidence of relapse was not different between pts who did and did not have day 30 chimerism measured (50.2 ± 6.5% vs. 42.9 ± 7.2%, P=0.45). Median follow up was 238 days (range 75–1420). Median age was 61 (range 21–76) and 56% were male. Underlying diseases were AML (24), MDS (13), NHL (10), myelofibrosis (6), CTCL (4), Hodgkin (3), myeloma (3), CLL (2), aplastic anemia (2), CML (1). Pts received a peripheral blood stem cell graft (67) or bone marrow (1), harvested from a matched related (30) or unrelated (38) donor. A single antigen mismatch was present in 6 cases. GVHD prophylaxis was tacrolimus (55) or cyclosporine (13) based. Median WB donor chimerism at day 30 was 96% (range 31–100%). A cumulative incidence analysis of relapse from day 30 revealed that day 30 WB chimerism had a significant association with relapse (HR 0.97, 95% CI [0.95–0.99], P=0.0011), reflecting a 3% decrease in risk of relapse for each 1% increase in chimerism. Risk of relapse according to different day 30 chimerism levels is displayed in figure 1. Using a day 30 chimerism cutoff of 95%, we found a significant association with incidence of relapse (HR 0.29, 95% CI [0.15–0.57), P=0.0003). Pts who were alive without relapse at the end of follow-up had a significantly higher chance of 〉= 95% day 30 WB chimerism compared to pts who relapsed (88.9% vs. 45.5%, P=0.002). In 38 pts with myeloid diseases, there was a significant association between day 30 WB chimerism and risk of relapse (HR 0.97, 95% CI [0.96–0.99], P=0.00065), while in 22 pts with lymphoid diseases, an association did not reach statistical significance (HR 0.93, 95% CI [0.84–1.00], P=0.09). Median T-cell chimerism was 70% in the 49 evaluable pts with these measurements (range 26–99%). T-cell chimerism on day 30 did not predict relapse (HR 1.0, 95% CI [0.98–1.02], P=0.95). We analyzed potential associations between day 30 WB chimerism and various transplant and patient characteristics. A higher chimerism level was significantly associated with lower day 0 lymphocyte count (P=0.004) and lower preconditioning lymphocyte count (P=0.01), but was not associated with recipient or donor age, cell doses, busulfan levels, disease type and donor type. The day 0 lymphocyte count was also a strong predictor of relapse (HR 6.87, 95% CI [6.56–7.04], P=0.00035). Conclusions: Whole blood and not T-cell chimerism at day 30 is predictive of relapse after RIC SCT. WB chimerism is strongly associated with lower lymphocyte counts before and after the conditioning regimen. These data highlight the importance of adequate lymphodepletion and can be useful in designing future trials testing pre-emptive interventions to prevent relapse after RIC SCT. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 4519 Although the combination of fludarabine and busulfan is commonly used for non-myeloablative, reduced intensity (RIC), and myeloablative conditioning prior to sibling and unrelated allotransplantation, there is a wide spectrum of reported doses which range in fludarabine dose between 120 – 180mg/m2 and busulfan doses between 3.2 and 16mg/kg. Many reports include third agents or serotherapy for T-cell depletion, making retrospective comparisons difficult. Across the spectrum of published fludarabine/busulfan regimens, our investigation of fludarabine at 120mg/m2 falls into the lower range of reported flu doses, while 6.4mg/kg IV Busulfan (bu) has been studied at only a few centers. Very little busulfan pharmacokinetic data has been reported at this dose. We report our experience with fludarabine (30mg/m2/d × 4 days =120mg/m2) and IV busulfan (0.8mg/kg × 8 doses =6.4mg/kg) in 76 consecutive patients (pts) undergoing RIC with Peripheral Blood Stem Cell (PBSC) allografting for hematologic malignancy transplanted at our center from 2006 to March, 2010. GvHD prophylaxi included methotrexate in all pts with either CSA (37) or tacrolimus (38); 13 recipients of tacrolimus also received maraviroc through d30 on a phase I/II trial, and 1 pt received CSA/MMF. Therapy was not modified for unrelated donor source or mismatch. Median age was 60. Donor graft source was sibling PBSC in 35, sibling bone marrow (BM) in 1, unrelated PBSC in 39 and unrelated BM in 1. 47% (36/76) of patients had either AML (19) or MDS (17). The remaining diagnoses included CLL (4), MCL (3), CTCL (6), T-NHL (2), follicular NHL (7), HD (8), MM (3), MF (3), other (4). With a median follow-up of 8.2m (range 0–39m) for all patients, and median f/u for surviving pts of 10.7m (range 0–39m), probability of overall survival at 2 years is 35%. Early non-relapse mortality at 100 days was 9% (7/76). Disease progression accounted for 55% (24/44) of deaths to date. The incidence of grade II-IV acute graft vs host disease (GvHD) was 53% and there was no statistically significant difference in acute GvHD between patients receiving csa/mtx and tac/mtx. Of note, no difference in survival or incidence of GvHD was noted between recipients of 36 sibling vs 40 unrelated stem cells grafts. Donor chimerism of greater than 95% at day 100 was achieved in 39/52 (75%) patients and greater than 90% in 45/52 (86%) patients. Very limited pharmacokinetic data has been reported after RIC conditioning using busulfan; we therefore calculated steady state concentrations in 25 patients. Busulfan pharmacokinetic analysis following the initial dose achieved a concentration steady state between 600–900ng/mL in 56% (14/25) of patients, which is consistent with our full dose busulfan PK experience, but over 2 days instead of 4. In 25 pts, median busulfan level was 856 ng/mL., average busulfan level was 852 ng/mL with a range 610–1544ng/mL. Correlation of bu levels to outcomes will be presented. We conclude that fludarabine 120mg/m2 and IV busulfan 6.4mg/kg provides durable engraftment with acceptable treatment related toxicity in both sibling and unrelated recipients. There was no difference in GVHD between recipients of MTX with either CSA or Tac. Our data suggests that additional agents such as TBI, or ATG, may not be required during conditioning in the unrelated setting in light of the similar outcomes between sibs and unrelated recipients using a two-drug regimen. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 1011 Inhibition of lymphocyte trafficking early after allogeneic stem cell transplantation (SCT) may prevent GvHD without interfering with GvL activity. Animal models and genomic data in humans indicate that the interaction between CCR5 and its ligands CCL3, CCL4 and RANTES is pivotal in the pathogenesis of GvHD. Maraviroc (MVC; Selzentry®, Pfizer) is the first oral CCR5 antagonist in clinical use. The antiviral properties of MVC in HIV infection are known, but its effects on chemotaxis and immune function in patients without HIV infection have not been explored. We hypothesized that CCR5 inhibition early after allogeneic SCT would reduce lymphocyte chemotaxis and result in low rates of acute GvHD without impairing engraftment or antitumor activity. In vitro, MVC effectively and specifically inhibited CCR5 internalization and reduced RANTES-induced chemotaxis in concentrations achievable in humans, recapitulating a defect observed in homozygotes for the del32-CCR5 polymorphism. MVC had no effect on hematopoietic colony formation, T-cell mediated cytotoxicity and T-cell proliferation. Between May 2009 and March 2011, we enrolled 38 pts in a phase I/II study of reduced intensity conditioned (RIC) allogeneic SCT. Patients had high-risk features by age (median=62, range 21–74), donor source (matched related 34%, matched unrelated 50%, single-antigen mismatch 16%) and comorbidities (comorbidity index: low 55%, intermediate 34%, high 11%). Underlying diseases were AML (15), MDS (6), NHL (8), myelofibrosis (4), aplastic anemia, myeloma, CLL, Hodgkin, CML (1 each). Pts received fludarabine 120mg/m2 and IV busulfan 6.4 mg/kg followed by peripheral blood stem cells. In addition to standard GvHD prophylaxis with tacrolimus and methotrexate, MVC was given from day −2 to +30. Pharmacokinetic analysis on the first 13 pts identified 300 mg bid as the appropriate dose (Reshef, ASH 2010). MVC was well tolerated, and adverse events were similar to the expected toxicity observed in patients undergoing RIC SCT. The median time to ANC〉500/μL was 15 d (range 10–27) and to platelets〉20k/μL was 19 d (range 9–84). The median whole blood and T-cell donor chimerism at day 100 was 96.5% (range 0–100%) and 85% (range 0–100%) respectively. Median follow-up was 200 days (range 12–760). Among 35 evaluable patients, the cumulative incidences of any acute GvHD and grade III–IV acute GvHD at day 100 were 14.7 ± 6.2% and 2.9 ± 2.9%, respectively. Importantly, in the first 100 days, there were no cases of acute GvHD involving the liver or gut. At day 180, the rate of acute GvHD was 20.7 ± 7.1%, largely confined to the skin with low rates of GvHD in the liver (3 ± 3%) and gut (7.4 ± 5.3%). In evaluable pts who received a graft from their HLA-matched sibling (11), there was no GvHD before day 100 and only two cases of acute GvHD before day 180. We compared these results to a cohort of 38 well-matched consecutive patients treated at our institution with RIC SCT using an identical regimen but without MVC between 2009 and 2011. We observed a similar incidence of acute GvHD (all grades) at day 100 (14.7 ± 6.2 vs. 16 ± 6.1%; P=0.88), but a 64% decrease in the MVC group at day 180 (20.7 ± 7.1 vs. 45.4 ± 9%; P=0.03). The incidence rates of severe GvHD (grade III–IV) were 2.9 ± 2.9% in the MVC group vs. 5.5 ± 3.8% in the comparator group at day 100 (P=0.59) and 6.5 ± 4.5% vs. 18.1 ± 6.8% at day 180 (P=0.15). Treatment-related mortality in pts receiving MVC was low. At 1 year, non-relapse mortality rate was 7.6 ± 5.5% (control group: 15.7 ± 6.6%; P=0.35). Infectious complications were seen at a rate that is expected with RIC SCT. Recovery of lymphocyte counts and lymphocyte subsets was not impaired by MVC. We evaluated whether a protective effect against GvHD was associated with an increase in relapse. In the MVC group, the incidence of relapse was 34.2 ± 8.8% at day 180; this was not significantly different from the comparator group (43.9 ± 8.8%, P=0.44), implying preservation of the graft-versus-tumor effect with MVC. Rates of overall survival and relapse-free survival were similar in both groups. Pharmacodynamic testing revealed that sera from patients taking MVC prevented CCR5 internalization by RANTES and blocked T-cell chemotaxis in vitro, providing evidence for in vivo biological activity and supporting the hypothesized mechanism of action. In summary, inhibition of lymphocyte trafficking is a novel, specific and potentially effective strategy to reduce the incidence of acute GvHD. Disclosures: Off Label Use: Use of maraviroc in GvHD prophylaxis will be discussed. Frey:Pfizer: Speakers Bureau. Vonderheide:Pfizer: Research Funding. Porter:Pfizer: Research Funding.

Description: Donor leukocyte infusion (DLI) can induce potent graft-versus-leukemia (GvL) activity in patients with relapsed acute leukemia after allogeneic hematopoietic stem cell transplant (HCT), but outcomes are generally poor, with high disease-related mortality. There is little data comparing logistics and outcomes of unrelated (uDLI) versus sibling (msDLI) DLI for relapse after allogeneic HCT. We performed a single-center retrospective cohort study to assess differences in time to obtain DLI from unrelated versus sibling donors, and to compare outcomes from both populations between 2000-2011. Fifty-three patients had relapsed AML, ALL, or high-risk MDS after allogeneic HCT from 2000-2011 at the University of Pennsylvania. Median time from relapse to infusion administration was 56 days for uDLI and 40 days for msDLI patients (p=0.034). However the time from DLI request to infusion was not significantly different between groups (27 days for uDLI and 40 days for msDLI, p=0.20) implying that the longer time from relapse to DLI was due to a longer time interval to request DLI from an URD. 29/35 msDLI patients and 10/18 uDLI patients received chemotherapy prior to DLI. Administration of chemotherapy did not significantly affect time from relapse to DLI (msDLI p=0.26, uDLI p=0.71). 15/35 (43%) of msDLI patients and 8/18 (44%) developed acute GVHD (p=0.91). There was no significant difference in development of IBMTR Grade C/D GVHD among uDLI (29%) and msDLI (24%) patients, p=0.38. The overall survival for the entire group was 96 days with no significant difference between recipients of uDLI and msDLI (p=0.49). Median time to death post DLI was 95 days for uDLI and 97 days for msDLI patients (see Figure I). Death was primarily from disease (74%). At 1 year after DLI, 7/35 (20%) recipients of msDLI were alive and 3/18 (17%) recipients of uDLI were alive (p=0.54). For those patients with response to DLI (defined as improved chimerism, morphologic response, or hematologic response), median time to progression was 68 days in uDLI recipients and 167 days in msDLI recipients (p=0.38). For patients with relapsed acute leukemia and MDS after allogeneic HCT, the time from relapse to DLI was longer for recipients of uDLI compared to msDLI largely due to a longer time from relapse to DLI request. Once requested, there was no significant difference in time from request to infusion, implying this delay to not have an adverse impact on outcome of uDLI compared to MSDLI. Furthermore there was no difference in incidence of GVHD, relapse-free survival, or overall survival after uDLI versus msDLI. Outcomes unfortunately remain poor regardless of donor source. Our data support ongoing investigational efforts to improve outcomes for DLI recipients and overall supports feasibility of using both uDLI and msDLI in future studies. Figure 1 Figure 1. p=0.49 Disclosures No relevant conflicts of interest to declare.