Commentary on: Differences in nucleotide hydrolysis contribute to the differences between erythrocyte 6-thioguanine nucleotide concentrations determined by two widely used methods.

To the Editor: In their article, Shipkova et al. (1 ) present evidence to indicate that the concentration of erythrocyte thioguanine nucleotides (TGNs) measured by the method of Lennard (2 ) differ considerably from those measured by the method of Dervieux and Boulieu (3 ). The assay method studied by Shipkova et al. was superseded over a decade ago for studies of thiopurine metabolism (4 ). However, Shipkova et al. have made a valid observation. Both assays (2, 3) are based on the same general principle, hydrolysis of the thionucleotide back to the parent thiopurine before measurement of the liberated thiopurine by HPLC. Shipkova et al. (1 ) attribute the differences in the TGN concentrations measured, in part, to the hydrolysis protocols of the two methods. However, despite the differences recorded in the TGN concentrations measured, there was an excellent and highly significant correlation between the two methods. For the optimization of thiopurine therapy, Shipkova et al. (1 ) make a plea for method-independent therapeutic ranges and the standardization of analytical procedures. However, the former is difficult when the main obstacle is the lack of commercially available, quality-assured thiopurine standards. To investigate thionucleotide hydrolysis, we initially used a commercially available thioinosinic acid (6-mercaptopurine riboside 5 -monophosphate) preparation and studied 6-mercaptopurine yield (5 ). We eventually obtained a small sample of 6-thioguanosine 5 -monophosphate, from what was then Burroughs Wellcome, and confirmed the time and conditions of the acid hydrolysis to 6-thioguanine (6 ). These conditions were maintained in subsequent developments of our assay (2, 4) and reconfirmed in more recent times (7 ). That the hydrolysis conditions may differ when the in vivo product of drug metabolism is a mixture of mono-, di-, and triphosphates comes as no surprise. However, this hydrolysis procedure is not the only step that needs to be considered in the analysis of thionucleotide metabolites. The thiopurine thiol moiety is easily oxidized; thus, the thiopurine metabolite of interest is degraded (8 ). The spectrum of thiopurine metabolites that can be measured by our assay methodology (thioguanine, 8-hydroxymercaptopurine, mercaptopurine, 8hydroxythioguanine, thioxanthine, and thiouric acid plus methylthioguanine and the methylmercaptopurine metabolites) are safe from oxidation and stable for several days when back-extracted or reconstituted in 0.1 mol/L HCl. Our reversed-phase isocratic HPLC system (4 ) shows no deterioration in performance from repeat injections ( 1000) of 50 L of this acid. One of the most important features of our assay methodology, which is also reported in the method of Dervieux and Boulieu (3 ), is the quantification, in a single injection, of the TGNs and the methylmercaptopurine nucleotide metabolites of 6-mercaptopurine (4 ). These metabolites are used as compliance indicators in our studies of oral 6-mercaptopurine therapy (9 ). To conclude, quantification of TGN can differ among laboratories, and method-specific therapeutic ranges are required for the interpretation of the results generated. The lack of commercially available thiopurine metabolite standards hinders progress in standardization.