Table S3. Molecular Docking Positive

Development and validation of the CCF method. A, Uptake of 0.01 μmol/L [3H]-EβG in OATP1B1-overexpressing HEK293 cells was evaluated at room temperature over 120 minutes in 96-well plates (n = 4 technical replicates, representative of n = 2 biological replicates; error bars represent SD). B, After saturation of cells at 1 hour, 100 μmol/L unlabeled EβG was spiked into wells, causing efflux of [3H]-EβG (n = 6 technical replicates across n = 2 biological replicates; error bars represent SD). C, After 1 hour of saturation with 0.01 μmol/L [3H]-EβG, 100 μmol/L positive (blue bars) and negative (gray bars) control substrates of OATP1B1 were spiked into wells and incubated for 30 minutes. Final intracellular radioactivity was measured and presented as relative to untreated control (n = 9 technical replicates across n = 3 biological replicates; error bars represent SD. ***, P < 0.001; ****, P < 0.0001; five compared with the control). D, Stimulated efflux of preloaded [3H]-EβG stimulated after the addition of 10 μmol/L TKI. Final intracellular radioactivity was measured and presented as relative to the efflux induced by an equimolar concentration of EβG, a known substrate and efflux inducer (n = 6 technical replicates across n = 2 biological replicates; error bars represent SEM). VC, vector control.

Abstract ID 131233Poster Board 060Objectives: Chemotherapy-related cognitive impairment (CRCI) is an adverse effect associated with both cytotoxic and targeted chemotherapy, including select tyrosine kinase inhibitors (TKIs). Based on recent clinical reports demonstrating that treatment with imatinib is associated with CRCI, we have been developing a preclinical model of imatinib-induced CRCI. In the development of this model, we noted a significant decrease in p-ERK staining in the hippocampi of mice chronically treated with imatinib. Based on these findings, we sought to identify an appropriate ex vivo or in vitro model to expand upon our in vivo findings. Based on findings with these models, we sought to determine physiologically relevant, steady-state concentrations of imatinib in murine plasma and brain by characterizing its pharmacokinetic (PK) profile.Methods: To determine if imatinib would decrease p-ERK in vitro, we utilized the SH-SY5Y cell line, a neuroblastoma cell line that we and others have successfully differentiated into a neuron-like state. We treated differentiated and undifferentiated SH-SY5Y cells with 1, 10, and 100 μM imatinib for 24 h, then we performed western blotting to determine the impact of imatinib on p-ERK (n=2). We also performed cell viability assays with imatinib-treated differentiated and undifferentiated SH-SY5Y cells and neurons isolated from mice. To characterize imatinib’s steady-state brain pharmacokinetics, we collected plasma and brain from female wild-type (WT) C57BL/6 mice at 0.25, 0.5, 1, 3, and 6 h after 7 days of receiving imatinib (100 mg/kg; p.o.) (n = 4-5/group). Concentrations of imatinib in plasma and brain were determined by a validated method based on liquid chromatography-tandem mass spectrometry, and pharmacokinetic parameter estimates were calculated with Phoenix WinNonlin. Follow-up studies were performed by inducing p-ERK activity in neurons isolated from murine hippocampi with 12-O-tetradecanoylphorbol-13-acetate (TPA) in the presence of vehicle, imatinib, or U0126.Results: Imatinib was cytotoxic to neurons in vitro and ex vivo when applied at concentrations >10 μM for >24 hours. Based on these data and an unexpected concentration-dependent increase in p-ERK in differentiated SH-SY5Y cells at 100 μM, we investigated imatinib’s brain accumulation and found the maximum concentration in the brain was around 1 μM. Using 1 μM imatinib, we found that imatinib decreased TPA-induced p-ERK activity.Conclusions: In the present study, we determined that imatinib reached concentrations as high as 1 μM in the brains of WT mice. These data suggest that concentrations beyond 1 μM for 24 h may not be physiologically relevant. Ex vivo experiments utilizing isolated murine hippocampal neurons and physiologically relevant concentrations and incubation times demonstrated that imatinib decreased p-ERK activity, consistent with our in vivo data.

Pharmacokinetics, tissue accumulation, and serum chemistry following pazopanib administration. A, Liver accumulation of pazopanib in WT (blue) or OATP1A/1B-KO mice (red) normalized to the plasma concentration of pazopanib at 24 hours (n = 5; error bars represent SD; ***, P < 0.001). B, Mean plasma concentration time profile of pazopanib in WT (blue) or OATP1A/1B-KO mice (red) after oral administration of pazopanib at a dose of 300 mg/kg (n = 5 technical replicates; error bars represent SD). C, Serum levels of AST and ALT 24 hours after a single 300 mg/kg oral dose of pazopanib in WT (blue) or OATP1A/1B-KO mice (red; n = 4; error bars represent SD; *, P < 0.05).

Previous studies have suggested that the incidence of vincristine-induced peripheral neuropathy (VIPN) is potentially linked with cytochrome P450 (CYP)3A5, a polymorphic enzyme that metabolizes vincristine in vitro, and with concurrent use of azole antifungals such as ketoconazole. The assumed mechanism for these interactions is through modulation of CYP3A-mediated metabolism, leading to decreased vincristine clearance and increased susceptibility to VIPN. Given the controversy surrounding the contribution of these mechanisms, we directly tested these hypotheses in genetically engineered mouse models with a deficiency of the entire murine Cyp3a locus [Cyp3a(-/-) mice] and in humanized transgenic animals with hepatic expression of functional and nonfunctional human CYP3A5 variants. Compared with wild-type mice, the systemic exposure to vincristine was increased by only 1.15-fold (95% confidence interval, 0.84-1.58) in Cyp3a(-/-) mice, suggesting that the clearance of vincristine in mice is largely independent of hepatic Cyp3a function. In line with these observations, we found that Cyp3a deficiency or pretreatment with the CYP3A inhibitors ketoconazole or nilotinib did not influence the severity and time course of VIPN and that exposure to vincristine was not substantially altered in humanized CYP3A5*3 mice or humanized CYP3A5*1 mice compared with Cyp3a(-/-) mice. Our study suggests that the contribution of CYP3A5-mediated metabolism to vincristine elimination and the associated drug-drug interaction potential is limited and that plasma levels of vincristine are unlikely to be strongly predictive of VIPN. SIGNIFICANCE STATEMENT: The current study suggests that CYP3A5 genotype status does not substantially influence vincristine disposition and neurotoxicity in translationally relevant murine models. These findings raise concerns about the causality of previously reported relationships between variant CYP3A5 genotypes or concomitant azole use with the incidence of vincristine neurotoxicity.

Abstract ID 128643Poster Board 259Background: Paclitaxel is an antineoplastic agent that is highly effective in treating a variety of cancers, but its use is limited by the development of adverse events, notably paclitaxel-induced peripheral neuropathy (PIPN). Preclinical models have been used to characterize the mechanisms underlying paclitaxel-induced toxicities. However, the potential impact of route- and strain-dependent pharmacokinetics (PK) and organ accumulation has largely been ignored, despite the importance of paclitaxel accumulation in the dorsal root ganglion (DRG) to development of PIPN. Based on dose-dependence of paclitaxel toxicities and observed differences in susceptibility to PIPN between strains, we hypothesized that paclitaxel PK and organ accumulation is strain- and route-dependent and that this results in variable toxicity phenotypes.Methods: We performed PK studies to evaluate paclitaxel plasma PK and organ accumulation. We administered 10 mg/kg paclitaxel to seven strains of mice (C57BL/6NTac, 129S6/SvEvTac, FVB, NSG, BALBc, CD2F1, and DBA) either IP or IV. Strains were selected based on their use in preclinical models, existing bioavailability data, or both. Concentrations of paclitaxel in plasma, DRG, and brain were determined by a validated LC-MS/MS method and analyzed using Phoenix WinNonlin. Assessment of transcriptional levels of transporters and enzymes relevant to the distribution and metabolism of paclitaxel in tissues of interest is ongoing. In vitro and ex vivo methods were used to determine whether paclitaxel is directly toxic to neurons.Results: Paclitaxel PK parameters differed significantly between strain and route of administration. AUC variation between strains was 1.7-fold for IV dosing and 2.0-fold for IP. Both brain and DRG levels showed significant strain-dependent differences [EE1] after administration, with 3.4-fold variation observed in brain and 2.0-fold variation observed in DRG. Ex vivo experiments demonstrated a concentration-dependent decrease in neuronal viability with paclitaxel treatment.Discussion: We show that paclitaxel plasma PK, DRG, and brain accumulation are route- and strain-dependent. We also show that paclitaxel exhibits concentration-dependent cytotoxicity to primary neurons at clinically achievable concentrations. These observations provide rationale to further investigate (i) if strain- and route-dependent paclitaxel PK variation are related to variation in toxicity phenotypes, (ii) if paclitaxel toxicities are a direct effect of cellular accumulation, and (iii) if paclitaxel toxicities are dependent on accumulation in specific cell types.

Nucleoside reverse transcriptase inhibitors (NRTIs) are the backbone of highly active antiretroviral therapy (HAART)—the current standard of care for treating human immunodeficiency virus (HIV) infection. Despite their efficacy, NRTIs cause numerous treatment-limiting adverse effects, including a distinct peripheral neuropathy, called antiretroviral toxic neuropathy (ATN). ATN primarily affects the extremities with shock-like tingling pain, a pins-and-needles prickling sensation, and numbness. Despite its negative impact on patient quality of life, ATN remains poorly understood, which limits treatment options and potential interventions for people living with HIV (PLWH). Elucidating the underlying pathophysiology of NRTI-induced ATN will facilitate the development of effective treatment strategies and improved patient outcomes. In this article, we will comprehensively review ATN in the setting of NRTI treatment for HIV infection.

Abstract Although the primary elimination pathway for most tyrosine kinase inhibitors (TKI) involves CYP3A4-mediated metabolism, the mechanism by which these agents are brought into hepatocytes remains unclear. In this study, we optimized and validated a competitive counterflow (CCF) assay to examine TKIs as substrates of the hepatic uptake transporter OATP1B1. The CCF method was based on the stimulated efflux of radiolabeled estradiol-17β-glucuronide under steady-state conditions in HEK293 cells engineered to overexpress OATP1B1. Of the 62 approved TKIs examined, 13 agents were identified as putative substrates of OATP1B1, and pazopanib was selected as a representative hit for further validation studies. The transport of pazopanib by OATP1B1 was confirmed by decreased activity of its target VEGFR2 in OATP1B1-overexpressing cells, but not cells lacking OATP1B1, consistent with molecular docking analyses indicating an overlapping binding orientation on OATP1B1 with the known substrate estrone-3-sulfate. In addition, the liver-to-plasma ratio of pazopanib in vivo was decreased in mice with a deficiency of the orthologous transporters, and this was accompanied by diminished pazopanib-induced hepatotoxicity, as determined by changes in the levels of liver transaminases. Our study supports the utility of CCF assays to assess substrate affinity for OATP1B1 within a large set of agents in the class of TKIs and sheds light on the mechanism by which these agents are taken up into hepatocytes in advance of metabolism. Significance: Despite the established exposure–pharmacodynamic relationships for many TKIs, the mechanisms underlying the agents’ unpredictable pharmacokinetic profiles remain poorly understood. We report here that the disposition of many TKIs depends on hepatic transport by OATP1B1, a process that has toxicologic ramifications for agents that are associated with hepatotoxicity.

Decitabine is a DNA methyltransferase inhibitor used in the treatment of acute myeloid leukemia and myelodysplastic syndrome. The notion that ongoing trials are presently exploring the combined use of decitabine, with or without the cytidine deaminase inhibitor cedazuridine, and other antileukemic drugs necessitates a comprehensive understanding of pharmacokinetic properties and an evaluation of drug-drug interaction liabilities. We report here the development and validation of a sensitive UHPLC-MS/MS method for quantifying decitabine in mouse plasma, which should be useful for such studies. The method involved a one-step protein precipitation extraction, and chromatographic separation on an XBridge HILIC column using gradient elution. The method was found to be robust, accurate, precise, and sufficiently sensitive (lower limit of quantitation, 0.4 ng/mL) to determine decitabine concentrations in microvolumes of plasma from mice receiving the agent orally or intravenously in the presence or absence of cedazuridine.

The addition of darolutamide, an androgen receptor signalling inhibitor, to therapy with docetaxel has recently been approved as a strategy to treat metastatic prostate cancer. OATP1B3 is an SLC transporter that is highly expressed in prostate cancer and is responsible for the accumulation of substrates, including docetaxel, into tumours. Given that darolutamide inhibits OATP1B3 in vitro, we sought to characterise the impact of darolutamide on docetaxel pharmacokinetics. We investigated the influence of darolutamide on OATP1B3 transport using in vitro and in vivo models. We assessed the impact of darolutamide on the tumour accumulation of docetaxel in a patient-derived xenograft (PDX) model and on an OATP1B biomarker in patients. Darolutamide inhibited OATP1B3 in vitro at concentrations higher than the reported C-max. Consistent with these findings, in vivo studies revealed that darolutamide does not influence the pharmacokinetics of Oatp1b substrates, including docetaxel. Docetaxel accumulation in PDX tumours was not decreased in the presence of darolutamide. Metastatic prostate cancer patients had similar levels of OATP1B biomarkers, regardless of treatment with darolutamide. Consistent with a low potential to inhibit OATP1B3-mediated transport in vitro, darolutamide does not significantly impede the transport of Oatp1b substrates in vivo or in patients. Our findings support combined treatment with docetaxel and darolutamide, as no OATP1B3 transporter based drug-drug interaction was identified.

ID 20840 Poster Board 250 Introduction: We performed single-cell RNA sequencing on samples from Acute Myeloid Leukemia (AML) patients with FLT3 mutations (FLT3+) pre- and post-treatment with the FLT3 inhibitor gilteritinib and identified in unresponsive patients: (i) upregulation of bone marrow (BM)-derived inflammatory cytokines, previously shown to promote disease progression and relapse in AML; and (ii) elevated expression of bone marrow kinase on chromosome X (BMX), a non-receptor tyrosine kinase from the Tec family of kinases. In FLT3+ cell lines and primary AML samples, we demonstrated that BMX gene knockout (KO) and inhibition reduced cytokine secretion. Mechanistic studies revealed that BMX kinase promotes AML cell-autonomous gilteritinib resistance through bypass signaling. Given the novel role of BMX in the cytokine network in vitro, we sought to develop a mouse model to evaluate AML cell/BM niche cell interactions in vivo to examine the mechanisms of BMX in promoting gilteritinib resistance. Methods: Genetic engineering was utilized to create a whole-body Bmx KO mouse model. To generate a syngeneic transplant model of murine Flt3+ AML, splenocytes (250,000 cells) from a female double mutant Npm1cA/+/Flt3ITD knock-in mouse (CD45.2) that produces spontaneous AML, were injected by tail vein injection (TVI) into male Bmx WT and KO mice. Starting 7 days after TVI, cohorts of mice were treated with vehicle or gilteritinib 30 mg/kg orally once daily (n=4-5 mice per cohort/per Bmx genotype). At survival endpoint, spleens were harvested and analyzed for CD45.2 cells (from vehicle-treated mice), gated for AML phenotypic markers CD11b and CD117 by flow cytometry, and murine cytokines were measured in plasma and spleen lysates by Luminex multiplex assay (from n=3-4 mice per cohort). Results: In our syngeneic transplant model of murine Flt3+ AML, gilteritinib treatment resulted in a 24-day median survival advantage over vehicle treatment (52 vs. 27.5 days) in Bmx WT mice, indicating this murine model represents a clinically relevant model of human FLT3+ AML that responds to a FLT3 inhibitor. However, no difference in survival was observed between Bmx KO and WT mice treated with gilteritinib. This is consistent with our published data showing an AML cell-autonomous role for BMX in gilteritinib resistance and suggests co-treatment with another drug to inhibit BMX or downstream mediators will be required. At study endpoint, Flt3+ AML cells were confirmed in spleens by expression of CD11b and CD117, compared to the normal spleen. We next sought to determine if microenvironment changes were occurring in Bmx KO vs. WT mice during gilteritinib treatment by assessing cytokine in plasma and AML cells in the spleen. We observed: (i) plasma CCLs/CXCLs and growth factors increased during gilteritinib treatment in Bmx WT mice but decreased in Bmx KO mice; and (ii) CCLs/CXCLs and other cytokines decreased in the spleens from Bmx KO vs. WT mice treated with gilteritinib. Conclusion: We generated a syngeneic transplant model of murine FLT3+ AML that mimics BMX-driven cytokine changes, which will be used to examine mechanisms underlying BMX-mediated AML cell/microenvironment niche cell interactions to promote gilteritinib resistance. This model will be used to evaluate a rational drug combination of gilteritnib + BMX inhibitor with translational potential in FLT3+ AML.

A recent article characterized dosing recommendations for cabozantinib in people living with HIV (PLWH) and cancer, a group that is often excluded from clinical trials. This study suggests cabozantinib is effective in cancers disproportionately impacting PLWH and has translational implications for the design of studies evaluating drug-drug interactions.

ID 55504 Poster Board 174 Background: Tyrosine kinase inhibitors (TKIs) belong to a class of commonly used chemotherapeutics that impair tumor growth by interfering with the function of protein kinases and their downstream signaling cascades. Although the primary elimination pathway for most TKIs has been well established and involves hepatic CYP3A4 metabolism, the mechanism by which these agents are taken up in hepatocytes remains unclear. Experimental challenges measuring the direct uptake of TKIs contribute to this knowledge deficit, especially a relatively high level of non-specific, extracellular membrane binding of TKIs. We hypothesized that we could overcome these challenges by using a competitive counterflow (CCF) assay and this methodology would allow us to systemically re-examine TKI transport by the hepatic uptake transporter OATP1B1. Methods: The CCF assay was based on the stimulated efflux of radiolabeled estradiol-17β-glucuronide (EβG) in HEK293 cells engineered to overexpress OATP1B1 or a vector and was validated using EβG as a positive control and glucose as a negative control. A total of 49 TKIs were examined for transport by OATP1B1. After identification of pazopanib as a positive hit, validation was performed by administering pazopanib to wild-type mice and mice deficient in all hepatic OATPs (OATP1A/1B-knockout mice) and measuring markers of drug-induced liver injury, including serum transaminases and histological evaluation of livers. Results: Of the 49 TKIs tested, 11 compounds (22%) caused efflux of radiolabeled EβG to a degree that was met a retrospectively determined cut off, suggesting that they are transported substrates of OATP1B1. These included the previously identified substrates axitinib and sorafenib, as well as the hepatotoxic TKI, pazopanib. Ensuing uptake studies with radiolabeled pazopanib in HEK293 cells with and without OATP1B1 overexpression confirmed that non-specific binding of TKIs compromises the use of conventional uptake assays for OATP1B1 substrate identification. In wild-type mice receiving a single oral dose of pazopanib (300 mg/kg), concentrations of the enzymes aspartate transaminase (AST) and alanine transaminase (ALT) were >200 IU/L, whereas in mice with complete deficiency of OATP1A and OATP1B isoforms, concentrations remained within the lower normal range (ALT, 25 IU/L; AST, 50 IU/L). In addition, vacuolization of hepatocytes in both the nucleus and cytoplasm in centrilobular to midzonal regions was only observed in wild-type mice, implicating OATP1B-type transport as a key mediator of pazopanib-induced liver injury. Conclusions: We have successfully optimized a CCF assay for identification of OATP1B1 substrates in the class of TKIs using an overexpressed HEK293 cell-based model. Validation studies in vivo implicate OATP1B-type transport as a contributor to pazopanib-induced liver injury and provide a rationale for the development of intervention strategies with OATP1B1 inhibitors to prevent this debilitating side effect. Keywords: OATP1B1, Pazopanib, Hepatotoxicity

A rapid, sensitive, and simple UHPLC-MS/MS method for the determination of the PARP inhibitor talazoparib in mouse plasma was developed and validated using [13C,2H4]-talazoparib as an internal standard (IS). The assay procedure involved extraction of talazoparib and the IS from plasma using a single-step deproteination and separation of the analytes was achieved on an ACQUITY UPLC RP18 HSS T3 column with a mobile phase gradient at a flow rate of 0.4 mL/min in a run time of 5 min. The calibration curve was linear (r2 > 0.99) over the concentration range of 0.5–100 ng/mL, and 10-fold dilution of samples could be accurately quantitated. The matrix effect and mean extraction recovery for talazoparib were between 93.7-109% and 87.7–105%, respectively. Precision and percent bias of quality control samples were always less than ±15%, indicating reproducibility and accuracy of the method. Talazoparib demonstrated bench-top stability at room temperature for 6 h, auto-sampler and reinjection stability at 4 °C for at least 24 h, and no significant degradation was observed after three freeze-thaw cycles. The developed method was successfully applied to pharmacokinetic studies involving serial blood sampling after oral administration of talazoparib to wild-type mice and animals with a genetic deficiency of the efflux transporters ABCB1 (P-gp) and ABCG2 (BCRP). Together, our results demonstrate the successful development of a suitable analytical method for talazoparib in mouse plasma and suggest that mice are a useful model to evaluate transporter-mediated drug-drug interactions involving therapy with talazoparib.

Supplementary Figures and Tables

Background: TP-0903 is a multi-kinase inhibitor designed to target AXL, but also inhibits additional kinases relevant to the progression of AML, including those regulating the cell cycle, such as Chk1/2. Pre-clinical studies of TP-0903 and decitabine (dec) showed additive cytotoxicity in vitro and prolonged survival in mouse models of TP53-mutant (TP53m) AML, a poor prognostic sub-group of AML. Here we report the clinical and safety results from the Leukemia and Lymphoma Society’s BeatAML phase 1b/2 (Ph1b/2) trial of TP-0903 in combination with dec (ClinicalTrials.gov NCT03013998). Aims: To establish the recommended dose of TP-0903 combined with dec and evaluate its efficacy in newly diagnosed older AML patients with TP53 mutations and/or complex karyotype (CK). Methods: Newly diagnosed AML patients ≥60 years with TP53m and/or CK (defined as ≥3 abnormalities) were consented, screened and assigned to this Ph1b/2 study by the BeatAML Master Trial. A standard 3 + 3 design was used for Ph1b to determine the initial recommended dose of TP-0903 (Days 1-21, 37 mg/day) in combination with dec (Days 1-10, 20 mg/m2) in a 28-day cycle. Ph2 followed a Simon’s 2-stage design in which, to move onto the 2nd stage, the 1st stage required 4 or more CR responses by the end of induction therapy (up to 3 cycles) or CRh/CRi/MLFS induction responses that converted to CR by 6 cycles. Nine additional patients were treated at 37 mg during Ph2, in which further assessments of safety, pharmacokinetic (PK) and pharmacodynamic data were used to amend the protocol and update the final recommended Ph2 dose of TP-0903 to 25 mg/day. Results: Of 27 total patients who started therapy with TP-0903, 15 received 37 mg and 12 received 25 mg. Two patients treated with 37 mg achieved CR and 3 achieved CRh, for a complete response (CR/CRh/CRi) rate of 33.3% (95% CI, 11.8 – 61.6). Three patients treated with 25 mg achieved CR, 1 achieved CRh, and 2 achieved CRi, for a complete response rate of 50% (95% CI, 21.1 – 78.9). Additionally, 4 patients treated with 37 mg were minimal residual disease (MRD) negative, measured by flow cytometry after 3 or 6 cycles of therapy, while only 2 patients treated with 25 mg obtained MRD- status. The median response duration for patients who achieved a complete response was 13.8 months (95%CI, 7.3 – Not Estimable (NE); median Follow-Up (FU), 16.7, 16.7 months) for those treated with 37 mg and 4.6 months (95%CI, 1.8 – NE; median FU, 4.4 months) for those treated with 25 mg. The median overall survival was 7.6 months (95%CI, 2.0 – 20.6; median FU, 17.1 months) for all patients treated at 37 mg and 7.5 months (95%CI, 1.1 – NE; median FU, 7.6 months) for those treated at 25 mg. Grade 3+ treatment-related adverse events were similar in patients treated with 37 mg and 25 mg TP-0903. The most common events in patients treated with 37 mg were neutropenia (33.3%), thrombocytopenia (26.7%), leukopenia (20%), and anemia (13.3%). Likewise, the most common events in patients treated with 25mg were neutropenia (50%), decreased lymphocyte counts (41.7%), febrile neutropenia (33.3%), leukopenia (33.3%), and thrombocytopenia (25%). Summary/Conclusion: Although the 1st stage in the Ph2 portion of the trial was terminated early after treating only 11 patients with 25 mg TP-0903 + dec, the trial still achieved 3 of the 4 required CR responses needed to move onto the 2nd stage. Additionally, 3 other patients in this group achieved responses of CRh (1) and CRi (2). This combination regimen is encouraging, considering the high risk and poor outcome of patients with TP53m and/or CK AML.Keywords: Phase I/II, Acute myeloid leukemia, Tyrosine kinase inhibitor

In recent years, various endogenous compounds have been proposed as putative biomarkers for the hepatic uptake transporters OATP1B1 and OATP1B3 that have the potential to predict transporter-mediated drug–drug interactions (DDIs). However, these compounds have often been identified from top–down strategies and have not been fully utilized as a substitute for traditional DDI studies. In an attempt to eliminate observer bias in biomarker selection, we applied a bottom–up, untargeted metabolomics screening approach in mice and found that plasma levels of the conjugated bile acid chenodeoxycholate-24-glucuronide (CDCA-24G) are particularly sensitive to deletion of the orthologous murine transporter Oatp1b2 (31-fold increase vs. wild type) or the entire Oatp1a/1b(−/−)cluster (83-fold increased), whereas the humanized transgenic overexpression of hepatic OATP1B1 or OATP1B3 resulted in the partial restoration of transport function. Validation studies with the OATP1B1/OATP1B3 inhibitors rifampin and paclitaxel in vitro as well as in mice and human subjects confirmed that CDCA-24G is a sensitive and rapid response biomarker to dose-dependent transporter inhibition. Collectively, our study confirmed the ability of CDCA-24G to serve as a sensitive and selective endogenous biomarker of OATP1B-type transport function and suggests a template for the future development of biomarkers for other clinically important xenobiotic transporters.

Acute myeloid leukemia (AML) with mutations in the tumor suppressor gene TP53 confers a dismal prognosis with 3-year overall survival of <5%. While inhibition of kinases involved in cell cycle regulation induces synthetic lethality in a variety of TP53 mutant cancers, this strategy has not been evaluated in mutant TP53 AML. Previously, we demonstrated that TP-0903 is a novel multikinase inhibitor with low nM activity against AURKA/B, Chk1/2, and other cell cycle regulators. Here, we evaluated the preclinical activity of TP-0903 in TP53 mutant AML cell lines, including a single-cell clone of MV4-11 containing a TP53 mutation (R248W), Kasumi-1 (R248Q), and HL-60 (TP 53 null). TP-0903 inhibited cell viability (IC50, 12–32 nM) and induced apoptosis at 50 nM. By immunoblot, 50 nM TP-0903 upregulated pChk1/2 and pH2AX, suggesting induction of DNA damage. The combination of TP-0903 and decitabine was additive in vitro, and in vivo significantly prolonged median survival compared to single-agent treatments in mice xenografted with HL-60 (vehicle, 46 days; decitabine, 55 days; TP-0903, 63 days; combination, 75 days) or MV4-11 (R248W) (51 days; 62 days; 81 days; 89 days) (p < 0.001). Together, these results provide scientific premise for the clinical evaluation of TP-0903 in combination with decitabine in TP53 mutant AML.

Glycochenodeoxycholate-3-sulfate (GCDCA-S) and chenodeoxycholate-24-glucuronide (CDCA-24G) are bile acid metabolites that potentially serve as endogenous biomarkers for drug-drug interactions mediated by the hepatic uptake transporters OATP1B1 and OATP1B3. We developed and validated a novel UHPLC-MS/MS method for the quantitative determination of GCDCA-S and CDCA-24G in mouse and human plasma with a lower limit of quantitation of 0.5 ng/mL. Chromatographic separation was achieved on an Accucore aQ column (50 mm x 2.1 mm, dp = 2.6 mu m) maintained at 20 degrees C and a gradient mobile phase comprising 2 mM ammonium acetate in water and methanol. The extraction recoveries of GCDCA-S and CDCA-24G were >80 %, and linear (r2 > 0.99) calibration curves ranged 0.5-100 ng/mL (CDCA-24G and GCDCA-S in mouse plasma) or 0.5-1000 ng/mL (GCDCA-S in mouse plasma). Values for precision (CV < 11.6 %) and accuracy bias (10.9 %) of analyte-spiked quality control samples verified that water was an acceptable matrix to prepare calibrators. This method was successfully applied to establish baseline activity of OATP1B1/OATP1B3 in humans and mice and establish the in vivo effects of OATP1B1/OATP1B3 inhibitors rifampin and micafungin.

7027 Background: TP-0903 is a multi-kinase inhibitor designed to target AXL, a receptor tyrosine kinase, and also inhibits cell cycle regulators such as Chk1/2 and other AML associated kinases. TP-0903 has shown prior anti-tumor activity at a safe dose in solid tumors. In pre-clinical AML studies, TP-0903 shows potent cytotoxicity in TP53 mutant ( TP53m) AML cell lines, an adverse prognostic genomic sub-group of AML. TP-0903 also had synergistic activity with decitabine (dec) in TP53m AML and prolonged survival in xenograft and genetically engineered mouse models. We report here on the initial safety and clinical results from the Leukemia and Lymphoma Society’s ongoing Beat AML phase 1b/2 (Ph1b/2) trial of TP-0903 in combination with dec (ClinicalTrials.gov NCT03013998). Methods: Newly diagnosed AML pts ≥60 years with TP53m and/or complex karyotype (≥3 abnormalities) were selected for a Ph1b/2 dose escalation study of TP-0903 combined with dec. Seven Ph1b pts were given TP-0903 every 28-day cycle from days 1-21 (Dose level (DL) 1 = 37 mg/day) and dec IV days 1-10 (20 mg/m2). A standard 3+3 design was used to evaluate the safety and tolerability. Nine additional patients enrolled onto Ph2 at DL1, but further assessments of safety, pharmacokinetics (PK) and correlative data was used to update the final recommended Ph2 dose (RP2D) of TP-0903 to DL-1 (25 mg/day) with dec. Results: At data cutoff (10Jan2022), 16 total pts were accrued. Ph1b treated 7 pts at DL1, 6 were DLT evaluable, and no DLTs were observed. Ph2 enrolled and treated 9 pts at DL1 before concerns of delayed count recovery led to the reduction of the Ph2 dose of TP-0903 to DL-1 (25 mg/day). For all 16 pts treated at DL1, 1 pt achieved CR, 4 pts CRh, and 1 pt CRi, for a composite CR (CR/CRh/CRi) rate of 37.5% (95% CI, 15.2-64.6), with 4 pts achieving MRD negativity by central flow cytometry. For the remaining 10 pts, 1 pt achieved MLFS (6%), 6 pts had stable disease (37.5%), 1 pt had treatment failure (6%), and 2 pts were not evaluable (12.5%) due to withdrawal of consent and death from early disease progression. Two pts (1 CR and 1 CRh) proceeded to stem cell transplantation. The most common grade 3 and above treatment-related AEs include decreased neutrophil counts (37.5%), platelet counts (31.3%), and anemia (18.8%). Finally, PK and correlative data analysis looking at soluble Axl and Gas6 also supported reduction to DL-1. Conclusions: Initial results with DL1 suggest that TP-0903/dec shows preliminary clinical activity in the prognostically poor TP53m/complex karyotype AML sub-group, with 4 pts achieving MRD negative status out of 6 patients who achieved a CR/CRh/CRi (66%). After further patients were treated on DL1, the toxicity profile and correlative data supported the de-escalation to DL-1 as the RP2D. The Ph2 study is ongoing to determine the clinical activity of this new RP2D (DL-1). Clinical trial information: NCT03013998.