Maximizing efficacy of the hypomethylating drug 5-aza-2'-deoxycytidine in human leukemia

5-aza-2'-deoxycytidine (DAC) is a cytidine analogue that strongly inhibits DNA methylation, and was recently approved for the treatment of myelodysplastic syndromes (MDS). To maximize clinical results with DAC, we investigated its use as an anti-cancer drug. We also investigated mechanisms of resistance to DAC in vitro in cancer cell lines and in vivo in MDS patients after relapse. We found DAC sensitized cells to the effect of 1-β-D-Arabinofuranosylcytosine (Ara-C). The combination of DAC and Ara-C or Ara-C following DAC showed additive or synergistic effects on cell death in four human leukemia cell lines in vitro, but antagonism in terms of global methylation. RIL gene activation and H3 lys-9 acetylation of short interspersed elements (Alu). One possible explanation is that hypomethylated cells are sensitized to cell killing by Ara-C. Turning to resistance, we found that the IC50 of DAC differed 1000 fold among and was correlated with the dose of DAC that induced peak hypomethylation of long interspersed nuclear elements (LINE) (r=0.94, P<0.001), but not with LINE methylation at baseline (r=0.05, P=0.97). Sensitivity to DAC did not significantly correlate with sensitivity to another hypomethylating agent 5-azacytidine (AZA) (r=0.44, P=0.11). The cell lines most resistant to DAC had low dCK, hENT1, and hENT2 transporters and high cytosine deaminase (CDA). In an HL60 leukemia cell line, resistance to DAC could be rapidly induced by drug exposure, and was related to a switch from monoallelic to biallelic mutation of dCK or a loss of wild type DCK allele. Furthermore, we showed that DAC induced DNA breaks evidenced by histone H2AX phosphorylation and increased homologous recombination rates 7-10 folds. Finally, we found there were no dCK mutations in MDS patients after relapse. Cytogenetics showed that three of the patients acquired new abnormalities at relapse. These data suggest that in vitro spontaneous and acquired resistance to DAC can be explained by insufficient incorporation of drug into DNA. In vivo resistance to DAC is likely due to methylation-independent pathways such as chromosome changes. The lack of cross resistance between DAC and AZA is of potential clinical relevance, as is the combination of DAC and Ara-C. ^

Pharmacogenetics of the human glutathione S-transferase P1 gene in the metabolism and therapeutic efficacy of cis-diamminedichloroplatinum

The human GSTP1 gene has been shown, conclusively, to be polymorphic. The three main GSTP1 alleles, GSTP1*A, GSTP1*B, and GSTP1*C, encode proteins which differ in the 3-dimensional structure of their active sites and in their function in phase II metabolism of carcinogens, mutagens, and anticancer agents. Although, it is well established that GSTP1 is over expressed in many human tumors and that the levels of GSTP1 expression correlate directly with tumor resistance to chemotherapy and inversely with patient survival, the significance of the polymorphic GSTP1 gene locus on tumor response to chemotherapy remains unclear. The goal of this project was to define the role and significance of the polymorphic GSTP1 gene locus in GSTP1-based tumor drug resistance and as a determinant of patient response to chemotherapy. The hypothesis to be tested was that the polymorphic GSTP1 gene locus will confer to tumors a differential ability to metabolize cisplatin resulting in a GSTP1 genotype-based sensitivity to cisplatin. The study examined: (a) whether the different GSTP 1 alleles confer different levels of cellular protection against cisplatin-induced cytotoxicity, (b) whether the allelic GSTP1 proteins metabolize cisplatin with different efficiencies, and (c) whether the GSTP1 genotype is a determinant of tumor response to cisplatin therapy. The results demonstrate that the GSTP1 alleles differentially protect tumors against cisplatin-induced apoptosis and clonogenic cell kill in the rank order: GSTP1*C > GSTP1*B > GSTP1*A. The same rank order was observed for the kinetics of GSTP1-catalyzed cisplatin metabolism, both in cell-free and cellular systems, to the rate-limiting monoglutathionyl-platinum metabolite, which was characterized, for the first time, by mass spectral analysis. Finally, this study demonstrates that both GSTP1 genotype and the level of GSTP1 expression significantly contribute to tumor sensitivity to cisplatin treatment. Overall, the results of this project show that the polymorphic GSTP1 gene locus plays a significant role in tumor sensitivity to cisplatin treatment. Furthermore, these studies have contributed to the overall understanding of the significance of the polymorphic GSTP1 gene locus in tumor resistance to cancer chemotherapy and have provided the basis for further investigations into how this can be utilized to optimize and individualize cancer chemotherapy for cancer patients. ^

Preclinical modeling of multi-drug cancer therapies

Anticancer drugs typically are administered in the clinic in the form of mixtures, sometimes called combinations. Only in rare cases, however, are mixtures approved as drugs. Rather, research on mixtures tends to occur after single drugs have been approved. The goal of this research project was to develop modeling approaches that would encourage rational preclinical mixture design. To this end, a series of models were developed. First, several QSAR classification models were constructed to predict the cytotoxicity, oral clearance, and acute systemic toxicity of drugs. The QSAR models were applied to a set of over 115,000 natural compounds in order to identify promising ones for testing in mixtures. Second, an improved method was developed to assess synergistic, antagonistic, and additive effects between drugs in a mixture. This method, dubbed the MixLow method, is similar to the Median-Effect method, the de facto standard for assessing drug interactions. The primary difference between the two is that the MixLow method uses a nonlinear mixed-effects model to estimate parameters of concentration-effect curves, rather than an ordinary least squares procedure. Parameter estimators produced by the MixLow method were more precise than those produced by the Median-Effect Method, and coverage of Loewe index confidence intervals was superior. Third, a model was developed to predict drug interactions based on scores obtained from virtual docking experiments. This represents a novel approach for modeling drug mixtures and was more useful for the data modeled here than competing approaches. The model was applied to cytotoxicity data for 45 mixtures, each composed of up to 10 selected drugs. One drug, doxorubicin, was a standard chemotherapy agent and the others were well-known natural compounds including curcumin, EGCG, quercetin, and rhein. Predictions of synergism/antagonism were made for all possible fixed-ratio mixtures, cytotoxicities of the 10 best-scoring mixtures were tested, and drug interactions were assessed. Predicted and observed responses were highly correlated (r2 = 0.83). Results suggested that some mixtures allowed up to an 11-fold reduction of doxorubicin concentrations without sacrificing efficacy. Taken together, the models developed in this project present a general approach to rational design of mixtures during preclinical drug development. ^

Bayesian adaptive designs for drug combinations in early phase clinical trials

Treating patients with combined agents is a growing trend in cancer clinical trials. Evaluating the synergism of multiple drugs is often the primary motivation for such drug-combination studies. Focusing on the drug combination study in the early phase clinical trials, our research is composed of three parts: (1) We conduct a comprehensive comparison of four dose-finding designs in the two-dimensional toxicity probability space and propose using the Bayesian model averaging method to overcome the arbitrariness of the model specification and enhance the robustness of the design; (2) Motivated by a recent drug-combination trial at MD Anderson Cancer Center with a continuous-dose standard of care agent and a discrete-dose investigational agent, we propose a two-stage Bayesian adaptive dose-finding design based on an extended continual reassessment method; (3) By combining phase I and phase II clinical trials, we propose an extension of a single agent dose-finding design. We model the time-to-event toxicity and efficacy to direct dose finding in two-dimensional drug-combination studies. We conduct extensive simulation studies to examine the operating characteristics of the aforementioned designs and demonstrate the designs' good performances in various practical scenarios.^

Novel Use of Dual Anti-Inflammatory Therapy to Overcome Drug Resistance and Improve Functional Recovery Following Spinal Cord Injury

Over 1.2 million Americans are currently living with a traumatic spinal cord injury (SCI). Despite the need for effective therapies, there are currently no proven effective treatments that can improve recovery of function in SCI patients. Many therapeutic compounds have shown promise in preclinical models of SCI, but all of these have fallen short in clinical trials. P-glycoprotein (Pgp) is an active transporter expressed on capillary endothelial cell membranes at the blood-spinal cord barrier (BSCB). Pgp limits passive diffusion of blood-borne drugs into the CNS, by actively extruding drugs from the endothelial cell membrane. Pgp can become pathologically up-regulated, thus greatly impeding therapeutic drug delivery (‘multidrug resistance’). Importantly, many drugs that have been evaluated for the treatment of SCI are Pgp substrates. We hypothesized that Pgp-mediated drug resistance diminishes the delivery and efficacy of neuroprotective drugs following SCI. We observed a progressive, spatial spread of Pgp overexpression within the injured spinal cord. To assess Pgp function, we examined spinal cord uptake of systemically-delivered riluzole, a drug that is currently being evaluated in clinical trials as an SCI intervention. Blood-to-spinal cord riluzole penetration was reduced following SCI in wild-type but not Pgp-null rats, highlighting a critical role for Pgp in mediating spinal cord drug resistance after injury. Others have shown that pro-inflammatory signaling drives Pgp up-regulation in cancer and epilepsy. We have detected inflammation in both acutely- and chronically-injured spinal cord tissue. We therefore evaluated the ability of the dual COX-/5-LOX inhibitor licofelone to attenuate Pgp-mediated drug resistance following SCI. Licofelone treatment both reduced spinal cord Pgp levels and enhanced spinal cord riluzole bioavailability following SCI. Thus, we propose that licofelone may offer a new combinatorial treatment strategy to enhance spinal cord drug delivery following SCI. Additionally, we assessed the ability of licofelone, riluzole, or both to enhance recovery of locomotor function following SCI. We found that licofelone treatment conferred a significant improvement in hindlimb function that was sustained through the end of the study. In contrast, riluzole did not improve functional outcome. We therefore conclude that licofelone holds promise as a potential neuroprotective intervention for SCI.

Design, Synthesis and Development of Transporter Targeting Agents for Image-guided Therapy and Drug Delivery

The purpose of this study was to design, synthesize and develop novel transporter targeting agents for image-guided therapy and drug delivery. Two novel agents, N4-guanine (N4amG) and glycopeptide (GP) were synthesized for tumor cell proliferation assessment and cancer theranostic platform, respectively. N4amG and GP were synthesized and radiolabeled with 99mTc and 68Ga. The chemical and radiochemical purities as well as radiochemical stabilities of radiolabeled N4amG and GP were tested. In vitro stability assessment showed both 99mTc-N4amG and 99mTc-GP were stable up to 6 hours, whereas 68Ga-GP was stable up to 2 hours. Cell culture studies confirmed radiolabeled N4amG and GP could penetrate the cell membrane through nucleoside transporters and amino acid transporters, respectively. Up to 40% of intracellular 99mTc-N4amG and 99mTc-GP was found within cell nucleus following 2 hours of incubation. Flow cytometry analysis revealed 99mTc-N4amG was a cell cycle S phase-specific agent. There was a significant difference of the uptake of 99mTc-GP between pre- and post- paclitaxel-treated cells, which suggests that 99mTc-GP may be useful in chemotherapy treatment monitoring. Moreover, radiolabeled N4amG and GP were tested in vivo using tumor-bearing animal models. 99mTc-N4amG showed an increase in tumor-to-muscle count density ratios up to 5 at 4 hour imaging. Both 99mTc-labeled agents showed decreased tumor uptake after paclitaxel treatment. Immunohistochemistry analysis demonstrated that the uptake of 99mTc-N4amG was correlated with Ki-67 expression. Both 99mTc-N4amG and 99mTc-GP could differentiate between tumor and inflammation in animal studies. Furthermore, 68Ga-GP was compared to 18F-FDG in rabbit PET imaging studies. 68Ga-GP had lower tumor standardized uptake values (SUV), but similar uptake dynamics, and different biodistribution compared with 18F-FDG. Finally, to demonstrate that GP can be a potential drug carrier for cancer theranostics, several drugs, including doxorubicin, were selected to be conjugated to GP. Imaging studies demonstrated that tumor uptake of GP-drug conjugates was increased as a function of time. GP-doxorubicin (GP-DOX) showed a slow-release pattern in in vitro cytotoxicity assay and exhibited anti-cancer efficacy with reduced toxicity in in vivo tumor growth delay study. In conclusion, both N4amG and GP are transporter-based targeting agents. Radiolabeled N4amG can be used for tumor cell proliferation assessment. GP is a potential agent for image-guided therapy and drug delivery.

The two -state model of receptor activation: The agonist and the efficacy

The Two State model describes how drugs activate receptors by inducing or supporting a conformational change in the receptor from “off” to “on”. The beta 2 adrenergic receptor system is the model system which was used to formalize the concept of two states, and the mechanism of hormone agonist stimulation of this receptor is similar to ligand activation of other seven transmembrane receptors. Hormone binding to beta 2 adrenergic receptors stimulates the intracellular production of cyclic adenosine monophosphate (cAMP), which is mediated through the stimulatory guanyl nucleotide binding protein (Gs) interacting with the membrane bound enzyme adenylylcyclase (AC). ^ The effects of cAMP include protein phosphorylation, metabolic regulation and transcriptional regulation. The beta 2 adrenergic receptor system is the most well known of its family of G protein coupled receptors. Ligands have been scrutinized extensively in search of more effective therapeutic agents at this receptor as well as for insight into the biochemical mechanism of receptor activation. Hormone binding to receptor is thought to induce a conformational change in the receptor that increases its affinity for inactive Gs, catalyzes the release of GDP and subsequent binding of GTP and activation of Gs. ^ However, some beta 2 ligands are more efficient at this transformation than others, and the underlying mechanism for this drug specificity is not fully understood. The central problem in pharmacology is the characterization of drugs in their effect on physiological systems, and consequently, the search for a rational scale of drug effectiveness has been the effort of many investigators, which continues to the present time as models are proposed, tested and modified. ^ The major results of this thesis show that for many b2 -adrenergic ligands, the Two State model is quite adequate to explain their activity, but dobutamine (+/−3,4-dihydroxy-N-[3-(4-hydroxyphenyl)-1-methylpropyl]- b -phenethylamine) fails to conform to the predictions of the Two State model. It is a weak partial agonist, but it forms a large amount of high affinity complexes, and these complexes are formed at low concentrations much better than at higher concentrations. Finally, dobutamine causes the beta 2 adrenergic receptor to form high affinity complexes at a much faster rate than can be accounted for by its low efficiency activating AC. Because the Two State model fails to predict the activity of dobutamine in three different ways, it has been disproven in its strictest form. ^