Application of population pharmacokinetic modeling in early clinical development of the anticancer agent E7820

The aim of this study was to assess the population pharmacokinetics (PopPK) of the novel oral anti-cancer agent E7820. Both a non-linear mixed effects modeling analysis and a non-compartmental analysis (NCA) were performed and results were compared. Data were obtained from a phase I dose escalation study in patients with malignant solid tumors or lymphomas. E7820 was administered daily for 28 days, followed by a washout period of 7 days prior to the start of subsequent cycles. A one compartment model with linear elimination from the central compartment was shown to give adequate fit, while absorption was described using a turnover model. Final population parameter estimates of basic PK parameters obtained with the PopPK method were (RSE): clearance, 6.24 L/h (7.1%), volume of distribution, 66.0 L (8.5%), mean transit time to the absorption compartment, 0.638 h (6.5%). The intake of food prior to dose administration slowed absorption (2.8-fold, RSE 13%) and increased relative bioavailability of E7820 by 36% (RSE 14%), although the effect on C (max) and AUC was not significant. Comparison with the NCA approach showed approximately equal PK parameter estimates and food effect measures, although specific advantages of PopPK included efficiency in use of data and more appropriate assessment of variability.

Assessment of Drug Interactions Relevant to Pharmacodynamic Indirect Response Models

The assessment of drug interactions for a simple turnover system when the basic pharmacodynamic response is governed by indirect mechanisms was explored. This report describes a diverse array of possible in vivo pharmacodynamic effects from a combination of two drugs acting by similar or different indirect mechanisms. Various conditions of pharmacodynamic drug combinations were explored mathematically and by simulation: (a) interactions of two drugs acting simultaneously either on the production (k(in)) or on the dissipation (k(out)) processes controlling the in vivo response by competitive (four cases) or non-competitive interaction (six cases); and (b) combinations of two drugs acting on separate k(in) and k(out) processes simultaneously (four cases). A range of different combinations of drug doses was used. Plasma concentration time profiles were generated according to monoexponential disposition. Pharmacodynamic response profiles were simulated using the above conditions and characterized by descriptors such as Area Between Effect (and Baseline) Curve (ABEC) values. The interaction of agents by competitive mechanisms produced net responses that were additive in nature. Response profiles for non-competitive interactions on the same process were both antagonistic (for two drugs with effects that oppose each other) and synergistic (for two drugs that produce the same response). On the other hand, response signatures for agents acting non-competitively on the production and dissipation factors by opposing mechanisms (e.g. inhibiting k(in) plus stimulating k(out)) showed dramatic changes in net effects and produced apparent drug synergy. Net indirect response profiles for joint use of two or more drugs measured by ABEC values may look "additive", "antagonistic", or "synergistic" depending on doses, their intrinsic potencies, the nature of interaction (competitive or non-competitive) as well as their mechanisms of action. These models may help explain changes in pharmacologic responses to two agents in a more rational and mechanistic fashion than older empirical methods.

Alpha 1-acid glycoprotein (AAG) and serum protein binding of methadone in heroin addicts with abstinence syndrome

OBJECTIVE

To quantify serum protein levels and protein-binding of methadone in vitro in heroin-addicted patients showing objective signs of heroin abstinence.

SUBJECTS AND METHODS

Serum samples were obtained from patients (n = 27) hospitalized to participate in a methadone detoxification program and from healthy volunteers (n = 21). The severity of the abstinence syndrome was assessed before blood sampling using a standardized scale. Concentrations of both albumin and alpha1-acid glycoprotein (AAG) were measured in all serum samples. The protein-binding of alpha1-methadone was determined by the ultrafiltration technique and the unbound concentration was measured by liquid scintillation counting.

RESULTS

The mean of the AAG concentrations was significantly increased in patients showing signs of withdrawal while the albumin concentrations did not change. Also, the unbound methadone was significantly decreased in this group when compared to the control. A positive correlation (Pearson r = 0.48; p < 0.005) indicates that AAG levels rise during abstinence as the score of withdrawal symptoms increases. Additionally, pooled data from all individuals show the binding of methadone to be related to AAG (r = 0.46; p < 0.05) levels and not to albumin.

CONCLUSIONS

The observed changes in protein-binding in abstinence individuals suggest the need for increased dosages of methadone when such patients are treated. Levels of AAG or protein-binding appear to be components of the interindividual variance observed in the response to methadone treatment, hence these variables could be included in future kinetic and dynamic studies.

The significance of lipoproteins in serum binding variations of propofol

The protein binding of propofol was investigated in vitro in isolated lipoprotein fractions (very low-density lipoprotein [VLDL], low-density lipoprotein [LDL], and high-density lipoprotein [HDL]) and in serum samples from the following subjects: healthy normolipemic volunteers (n = 16), hyperlipidemic subjects diagnosed with familiar polygenic hypercholesterolemia (n = 26) showing high levels of cholesterol, and elderly subjects (n = 15). Protein binding was determined by using ultrafiltration, and the concentration of unbound propofol was measured by using liquid chromatography. Levels of total cholesterol, triglycerides, VLDL cholesterol, LDL cholesterol, HDL cholesterol, albumin, and alpha1-acid glycoprotein were also measured. Propofol was extensively bound to the three lipoprotein fractions (88%+/-2% to VLDL, 93%+/-1% to LDL, and 91%+/-4% to HDL). The percentage of unbound propofol was significantly decreased (P < 0.0001) in hyperlipidemic (0.88%+/-0.20%) individuals whose levels of cholesterol and triglycerides were increased versus healthy subjects (1.26%+/-0.22%), whereas no significant difference was found in the elderly group (1.12%+/-0.23%). A positive relationship was found between serum protein binding of propofol and lipid levels. Multiple regression analysis, including all subjects, showed that changes in the levels of total cholesterol and triglycerides explained approximately 62% of the variability in the serum protein binding of propofol. These results stress the importance of triglycerides and cholesterol in the serum protein binding of propofol. We therefore suggest that these variations in lipid levels, and consequently in protein binding, may influence anesthetic practice with propofol.

IMPLICATIONS

We investigated the effect of serum lipids in the protein binding of propofol. We found that propofol binds extensively to all lipoprotein fractions. Propofol binding showed a significant relationship with the serum levels of cholesterol and triglycerides.

Serum protein binding of propofol in critically ill patients

BACKGROUND

Disease-induced modifications in the level of serum proteins may change the degree of binding of drugs highly bound to serum proteins.

METHODS

The serum protein binding of propofol, an intravenous anaesthetic agent was studied in vitro in samples of serum from healthy volunteers (n=28) and from critically ill patients (n=17). The free fraction was obtained by the ultrafiltration technique and was measured by high-performance liquid chromatography. Concentrations of serum albumin, alpha1-acid-glycoprotein and free fatty acids were also measured in all samples.

RESULTS

The percentage of free propofol was significantly increased (P<0.001) in critically ill patients (1.31 (1.06-2.25)%) vs control subjects (1.07 (0.49-1.47)%). Albumin levels were significantly decreased (P<0.001) in patients (16.3 (8.8-24.6) g x l(-1) vs 45.8 (31.4-55.5) g x l[-1]), while levels of alpha1-acid-glycoprotein were increased (P<0.001) (1.9 (0.9-2.8) g x l(-1) vs 0.9 (0.5-1.4) g x l[-1]), as were levels of free fatty acids (0.68 (0.50-1.14) mmol x l(-1) vs 0.37 (0.11-1.05) mmol x l(-1); P<0.05 ). No correlation was found between levels of alpha1-acid-glycoprotein or free fatty acids and the bound/free ratio of propofol. However, a linear relationship was found between levels of albumin and the bound/free ratio (r2=0.25; P<0.001).

CONCLUSION

In conclusion, in these critically ill patients, an increase in the percentage of free propofol occurs. The significance of this observation remains uncertain, but may be validated in future studies. However, the observation supports the common idea that potent drugs should be given with great care in critically ill patients.

Interaction between propofol and sulfisoxazole in mice an in vivo and in vitro study

The effect of sulfisoxazole on the propofol response has been investigated in an animal model. Doses of propofol (0, 50, 75 and 100 mg/Kg intraperitoneal route) were administered to control (n = 36) and sulfisoxazole pretreated mice (n = 36). The impairment of righting reflex and struggle response were evaluated before and 5, 10, 15, 20, 30, 40, 50 and 60 minutes after propofol administration. Ten minutes after administration of the different doses of propofol, total plasma concentration was measured in both groups by high performance liquid chromatography Protein binding displacement was evaluated in vitro by the ultrafiltration technique. Pretreatment with sulfisoxazole produced an important enhancement in the effect of propofol in both tests. This change was reflected in a significant increase in the area under the time-effect curve (p < 0.001) and in a shift of the log dose-effect relationship to the left. Sulfisoxazole itself did not produce any effect on either test. ED50 for the righting reflex was significantly reduced from 114 mg/kg to 64 mg/kg in sulfisoxazole pretreated groups and it fell from 87 mg/kg to 43 mg/kg for the struggle response test. No changes in the total plasma concentration and protein binding were observed. On the basis of these results, it was concluded that a clinical interaction could be expected but this cannot be explained by an alteration in the protein binding.