ALBERT

Description: Current treatment of patients with hemophilia A often requires the frequent infusion of Factor VIII (FVIII) due to its short circulating half-life. A longer-acting FVIII molecule could profoundly impact patients’ lives by extending bleeding protection with a reduced frequency of infusions. Several strategies to prolong plasma concentrations of FVIII have been attempted. In particular, targeting domains on FVIII that bind to LRP, the putative clearance receptor, has been a popular strategy. We have investigated the use of site-directed pegylation of B-domain deleted (BDD) FVIII to evaluate the utility of PEG as a method to decrease FVIII clearance through steric hindrance of LRP binding, or other unknown clearance mechanisms, while minimizing decreases in vWF binding and in vivo activity. The evaluation of novel constructs required the development of in vivo pharmacokinetic models and a FVIII-dependent bleed model. We describe the development of an acute bleed model following uniform tail transection in the hemophilia A mouse that is FVIII dependent and allows the evaluation of the acute pharmacologic effects of FVIII or variants in vivo. Pharmacokinetic analysis of recombinant FVIII (rFVIII) and its variants was performed in rabbits over 32-hours and rFVIII or variants were measured using a modified Coatest® to differentiate endogenous rabbit FVIII from the administered human FVIII. For efficacy evaluations, hemophilia A mice were anesthetized with isoflurane and their pre-warmed tail was cut by a scalpel and placed into a new tube of warmed saline (37–40°C). Blood was collected over 40 minutes and blood loss was measured gravimetrically. Three modes of treatment were evaluated: prevention of bleeding (drug was administered 5 minutes before injury), treatment of an acute bleeding event (drug was administered 5 minutes after injury), and a delayed injury model (tail cut occurred at 20 or 24 hours after the drug administration). Over the course of 40 minutes control (C57BL6) mice demonstrated negligible bleeding (approximately 41 ± 8 μL) compared to 919 ± 26 μL in hemophilia A mice. A dose response curve was constructed for doses ranging from 0.1 to 5.0 IU of human rFVIII per mouse. Hemophilia A mice treated with 200 IU/kg of human rFVIII (5 IU/mouse) lost a similar volume of blood as control mice. The protective effect was rFVIII dose dependent over a range of 4–200 IU/kg (0.1–5 IU/mouse). In contrast, more rFVIII was required to stop an acute bleeding event when administered after the injury. In the delayed injury model, mice injured 24 hours after drug administration had a significantly larger mean blood volume loss compared to mice injured 20 hours post drug administration. Pegylated rFVIII constructs with longer half-lives also had increased activity over time compared to non-pegylated rFVIII in this mouse model. These results describe a superior hemophilia A tail bleed model that demonstrates FVIII-dependent bleeding reduction in response to acute hemorrhage over a 40 minute time course. This is the first demonstration of a hemophilia A mouse model in which all untreated animals uniformly bleed and all control animals demonstrate negligible bleeding. This model was used to evaluate the in vivo hemostatic efficacy of new rFVIII molecules that were designed to have superior pharmacologic and/or pharmacokinetic properties compared to rFVIII.

Description: Background: Idelalisib, a selective oral inhibitor of PI3Kd, is approved for the treatment of chronic lymphocytic leukemia (CLL) in combination with rituximab and as monotherapy for patients with follicular lymphoma who have received at least 2 prior therapies. Despite remarkable clinical efficacy, complete responses are rare, highlighting the need to identify more effective therapies, including combinations of novel agents. GS-4059 (ONO-4059) is an investigational next generation Bruton's tyrosine kinase (BTK) inhibitor with improved selectivity compared with ibrutinib. We report here the results of the combination of a PI3Kd inhibitor and GS-4059 in a diffuse large B-cell lymphoma (DLBCL) xenograft model, demonstrating supportive data for our ongoing combination trial in B-cell malignancies (NCT02457598). Additionally, we investigated preclinical orthogonal combination approaches for DLBCL. Methods: Growth inhibition was assessed using CellTiter-Glo Assay after 96 h incubation with idelalisib and GS-4059. CB17-SCID mice were irradiated, implanted subcutaneously with TMD8, and treated BID PO with the PI3Kd inhibitor GS-649443, GS-4059, or coformulated combination when tumors reached 200 mm3. Lysates from tumors or cell cultures were analyzed by Simple Western (Protein Simple). Synergy for antiproliferative effects was assessed using Chalice software (Horizon Discovery, Inc., Lehar et al., Nature Biotech, 2009). Results: Idelalisib and GS-4059 potently inhibited the ABC subtype DLBCL cell line TMD8, which is a B-cell receptor (BCR)-dependent line that exhibits chronic activated B-cell signaling due to mutations in CD79A/CD79B and MYD88 (Kim Y. et al., Hum Pathol, 2014). When a clinically relevant single concentration of idelalisib or GS-4059 was added in combination to a dose responsive effect of the other, a shift in EC50 on cell viability was seen. GS-4059 (50 nM) shifted the EC50 of idelalisib from 141 nM to 5 nM, a 28-fold shift. Idelalisib (1 µM) shifted the EC50 of GS-4059 from 27 nM to 2 nM, a 14-fold shift. Evaluation of downstream signaling pathways implicated in malignant B-cell survival and proliferation showed enhanced inhibition of pAkt S437, pBTK Y223, pErk1/2 T202/Y204, and MYC with a combination of idelalisib and GS-4059, more than either single agent alone. When TMD8 xenografts were treated with a PI3Kd tool compound, GS-649443, GS-4059 or a combination of the 2 inhibitors, a statistically significant decrease in tumor volume was seen as well as tumor regression, when compared with single agent effects (Figure 1A). Evaluation of TMD8 tumor lysates showed strong suppression of pAkt S437, pBTK Y223, pS6RP S235/236, and MYC in tumors treated with both GS-649443 and GS-4059 (Figure 1B). pS6RP S235/236 and MYC, in formalin-fixed paraffin-embedded (FFPE) TMD8 tumors, were profoundly inhibited in tumors treated with combination therapy compared to the monotherapies (Figure 1C). Since the combination of a PI3Kd inhibitor and GS-4059 led to TMD8 tumor regression, an effect correlated to strong down-modulation of MYC, the combination of idelalisib with a bromodomain and extra-terminal (BET) family inhibitor was explored as a potential new orthogonal combination approach for DLBCL. A panel of DLBCL cell lines was evaluated for inhibition of cell viability by idelalisib in combination with GS-5829, a BET inhibitor currently being evaluated in a phase 1 clinical trial. At clinically relevant concentrations, the combination of idelalisib and GS-5829 showed synergistic effects on cell viability in 2 of 6 ABC subtype, 4 of 5 GCB subtype, and 2 of 2 double-hit DLBCL cell. As compared with combination with other agents that inhibit the BCR pathway (GS-4059) or the Bcl-2 pathway (ABT-199), the broadest activity across cell lines was seen with the combination of idelalisib and GS-5829. Conclusion: Idelalisib and GS-4059 demonstrated synergistic inhibition of the TMD8 xenograft with concomitant inhibition of MYC. Screening of other targeted agent combinations in a panel of DLBCL lines revealed broad preclinical activity for the BET inhibitor GS-5829 in combination with idelalisib. This represents a potential orthogonal approach for a new therapeutic strategy for the treatment of B-cell malignancies. Figure 1A Figure 1A. Figure 1B Figure 1B. Figure 1C Figure 1C. Disclosures Meadows: Gilead Sciences: Employment. Yahiaoui:Gilead Sciences: Employment. Sorensen:Gilead Sciences: Employment. Cui:Gilead Sciences: Employment. Brockett:Gilead Sciences: Employment. Keegan:Gilead Sciences: Employment. Tannheimer:Gilead Sciences: Employment.

Description: Background: BAY 79-4980 refers to recombinant FVIII (rFVIII; Kogenate®-FS, KGFS) reconstituted with a PEGylated Liposome diluent. rFVIII binds non-covalently but with high affinity on the external surface of the PEGylated Liposome and retains procoagulant activity and normal vWF binding in vitro. In initial in vivo experiments in hemophilia A mice (FVIII −/−) subject to multiple tail clippings, hemostatic activity with BAY 79-4980 was augmented compared to FVIII alone. In the current studies, a more extensive pharmacologic analysis of BAY 79-4980 was performed in Hemophilia A mice using an acute FVIII dependent bleeding model (Landskroner et al). Objectives: To more thoroughly evaluate the pharmacologic properties of BAY 79-4980 using an acute FVIII dependent bleed model in the hemophilic A mouse. We examined BAY-4980 under three conditions: (1) Prophylaxis treatment, where drug is administered via the jugular vein and 5 minutes later the tail is transected; (2) On demand treatment, where drug is administered 3–5 minutes after the tail is severed and bleeding is extant; (3) PK/PD, where drug is administered via tail vein and 4 or 24 hrs is allowed to elapse before the tail is severed. Methods: Hemophilia A mice were anesthetized with isoflurane, the tail is then warmed and the right jugular vein is exposed. At t = 0 the tail is cut by a scalpel and the transected tail is placed back into a new tube of warmed saline (37–40°C). Blood is collected over 40 minutes into the plastic tubes and blood loss is measured gravimetrically. In all studies, BAY 79-4980 is compared with a similar dose of KGFS measured in chromogenic units/mouse. The data is analyzed in two ways; (1) the quantity (μl) of blood lost over 40 minutes and the frequency (% of mice) in the respective treatment group bleeding 〉150 μl. The 150 μl value is used as a cut-off value to determine a real bleeding event. This value was derived from the mean + 2.5 SD of bleeding in control (C57BL) mice administered 5% albumin. Results: With Prophylaxis treatment at 2.5 U/mouse, blood loss was 205 ± 59 μl with KGFS and 62 ± 16 μl with BAY 79-4980 (P = 0.12). Percentage of animals protected was not significantly different (73% versus 93%, respectively). With on demand treatment at 1 and 2.5 U/mouse, the results were 660 ± 69 μl with KGFS and 447 ± 69 μl with BAY 79-4980 (P = 0.02) at 1 U/mouse and 466 ± 107 μl with KGFS and 211 ± 52 μl with BAY 79-4980 at 2.5 U/mouse. The percent protected was 10% versus 36% at 1 U/mouse and 47% versus 60% at 2.5 U/mouse, respectively. When drug was administered at 2.5 U and 4 hours was allowed to elapse the results were 432 ± 102 μl with KGFS and 187 ± 70 μl with BAY 79-4980 (P = 0.04), and the percentage of mice protected was 47% and 73%, respectively. When drug was administered at a dose of 8 U/mouse and the tail was cut at 24 hours, the results were 822 ± 97 μl with KGFS and 395 ± 112 μl with BAY 79-4980 (P = 0.02), and the percentage of mice protected was 13% versus 53%, respectively (P = 0.05). Conclusions: These results demonstrate that BAY 79-4980 has good hemostatic activity when administered acutely compared to KGFS. It appeared that under certain conditions, BAY 79-4980 exhibited enhanced hemostatic benefit, even when administered acutely. When BAY 79-4980 was administered prophylactically and 4 or 24 hours was allowed to elapse, enhanced hemostatic activity was demonstrated compared to KGFS alone.

Description: Recombinant FVIII formulated in PEG-ylated liposomes (rFVIII-PEG-Lip) was reported to increase the bleed-free days from 7 to 13 days (at 35 IU/kg rFVIII) in severe hemophilia A patients. To understand the underlying mechanism, we sought to recapitulate its efficacy in hemophilia A mice. Animals treated with rFVIII-PEG-Lip achieved approximately 30% higher survival relative to rFVIII after tail vein transection inflicted 24 hours after dosing. The efficacy of rFVIII-PEG-Lip represents an approximately 2.5-fold higher “apparent” FVIII activity, which is not accounted for by its modestly increased (13%) half-life. The enhanced efficacy requires complex formation between rFVIII and PEG-Lip before the administration. Furthermore, PEG-Lip associates with the majority of platelets and monocytes in vivo, and results in increased P-selectin surface expression on platelets in response to collagen. Rotational thromboelastometry (ROTEM) analysis of whole blood from rFVIII-PEG-Lip–treated animals at 5 minutes up to 72 hours after dosing recapitulated the 2- to 3-fold higher apparent FVIII activity. The enhanced procoagulant activity is fully retained in plasma unless microparticles are removed by ultracentrifugation. Taken together, the efficacy of rFVIII-PEG-Lip is mediated mainly by its sensitization of platelets and the generation of procoagulant microparticles that may express sustained high-affinity receptors for FVIII.