It is my long term view that novel formulations for optimum drug delivery can surface only upon detailed physicochemical characterization of the drug and therefore, an extensive investigation of the drug in aqueous solutions was carried out in order to match its properties to a carrier system. To accurately derive the kinetic and thermodynamic parameters governing the hydrolysis of the lactone ring at physiological pH, a derivative spectrophotometric technique was used for the simultaneous estimation of lactone and carboxylate forms of camptothecin (CPT). The hydrolysis of the CPT-lactone and the lactonization of CPT-carboxylate at 310.15 K followed a 1^st order decay with apparent rate constants equal to 0.0279 ± 0.0016 min^-1 and 0.0282 ± 0.0024 min^-1, respectively. The activation energy associated with the hydrolysis of the CPT-lactone and the lactonization of the CPT-carboxylate as calculated from the Arrhenius equation was 89.18 ± 0.84 kJ mol^-1 and 86.49 ± 2.7 kJ mol ^-1, respectively. The enthalpy and entropy of the thermodynamically favored hydrolysis reaction were on average 10.49 kJ·mol^-1 and 48.00 J·K^-1· mol^-1, respectively. The positive enthalpy and entropy values of the CPT-lactone hydrolysis indicate that the reaction is endothermic and entropically driven. The stability of CPT-lactone in the presence of human serum albumin (HSA) was also analyzed. Notwithstanding the much faster hydrolysis of the CPT-lactone in the presence of HSA at various temperatures, the energy of activation was determined to be similar to the one estimated in the absence of HSA, suggesting that HSA does not catalyze the hydrolysis reaction, but it merely binds, sequesters and stabilizes the CPT-carboxylate species.

Equilibrium hydrolysis studies of 7-ethyl-10-hydroxycamptothecin (SN38) in aqueous solutions revealed that SN38 is stable in its biologically active lactone form at pH < 5. The pK_a of the quinoline ring of the drug determined to be 2.4 ± 0.2, indicated that the drug is present exclusively as protonated lactone in acidic solutions of pH ≤ 0. The logarithm of the intrinsic partition coefficient P of the SN38-lactone free base was found to be 3.36 ± 0.05, indicative of a very lipophilic molecule. The intrinsic solubility of SN38-lactone free base was found to increase with temperature from 0.0997 μg mL ^-1 at 298.15 K to 0.225 μg mL^-1 at 318.15 K, with an estimated standard enthalpy of solution equivalent to 32.45 kJ mol ^-1. The apparent equilibrium solubility of the drug was found to increase with increasing temperature and decreasing pH from (0.111 ± 0.001) μg mL^-1 at pH 3 and 298.15 K to (2.159 ± 0.05) μg mL ^-1 at pH 1 and 318.15 K. The standard enthalpy of solution of the ionized SN38-lactone species and the apparent enthalpy of ionization were determined from corresponding van't Hoff plots to be 25.14 kJ mol^-1 and –8.53 kJ mol^-1, respectively. Contrary to the low intrinsic solubility, the solubility of SN38 in extremely acidic media increased by more than 4 orders of magnitude reaching saturated drug concentration of 6.211 mg mL^-1 in 6.25 M HC1. Loading the biologically active protonated drug into mesoporous silica material MCM-41 from acidic media proved to be a very promising system achieving maximum loading of 250 mg of SN38 per gram of MCM-41. It was also found that the equilibrium association constant K varies with the extent of drug adsorption. At low and high drug load, corresponding to 1 SN38 molecule bound for every 70 and 13 –SiO ₂–, K was determined to be 1253.5 M^-1 and 127.39 M^-1, respectively. The kinetics of SN38 release from MCM-4I in simulated gastric fluid and simulated intestinal fluid were characterized in terms of power law and can be described in terms of fickian diffusion. Overall, the extremely high loading capacity, efficient drug release and stability of the MCM-41 in aqueous media indicate a very promising system for the in vivo delivery of camptothecin derivatives.