1. 2′,3′‐Dideoxyinosine (ddI) has potent activity against human immunodeficiency virus (HIV). It is converted within target cells to its active form dideoxyadenosine triphosphate(ddA‐TP). 2. In addition to the intracellular formation of ddA‐TP, ddI can be broken down to hypoxanthine, by purine nucleoside phosphorylase (PNP) and to uric acid, by xanthine oxidase. Since PNP is present in red blood cells we have examined the metabolism of [14C]‐ddI by human blood. 3. When incubated with whole blood at 37 degrees C, ddI was extensively metabolised, principally to hypoxanthine (50.4 +/‐ 12.5% formed at 6 h; mean +/‐ s.d.; n = 16). Small amounts of uric acid were formed (3.8 +/‐ 2.4%). 4. ddI breakdown was temperature dependent, being virtually negligible at 4 degrees C. Metabolism to hypoxanthine occurred within red blood cells. 5. The short half‐life of ddI in patients is probably the result of both hepatic and erythrocytic metabolism.

1 Zidovudine (ZDV) has proved unsuccessful in controlling disease progression over extended periods of time in patients with AIDS. Combination of ZDV with another reverse transcriptase inhibitor, dideoxyinosine (ddl) may improve the duration of effectiveness of antiretroviral therapy. The aim of this study was to investigate the possibility of a pharmacokinetic drug interaction between ZDV and ddl. 2 The pharmacokinetics of ZDV and ddl were determined in eight patients with AIDS who were randomised to receive ZDV 250 mg orally, ddl 250 mg orally or a combination of ZDV 250 mg plus ddl 250 mg orally on 3 study days separated by I week. 3 The administration of ZDV did not significantly alter ddl pharmacokinetics. The mean AUC was 6.8 ± 2.0 s.d. and 7.6 ± 2.5 s.d. gmol 1-1 h and oral clearance was 2766 ± 686 and 2660 ± 1297 ml min-' in the presence and absence of ZDV, respectively. 4 In the presence of ddl the elimination half-life of ZDV was increased significantly by 18% from 1.1 ± 0.3 to 1.3 ± 0.3 h (P < 0.05) and the mean AUC increased significantly by 35% from 4.8 ± 1.5 to 6.5 ± 1.5 gmol 1F1 h (P < 0.05). The clearance was decreased by 29% from 3518 ± 1123 to 2505 ± 575 ml min-1, but this difference was not significant. The renal clearance of ZDV was not altered by ddl. 5 Administration of ddl also resulted in a significant 22% increase in the AUC of GZDV, from 28.5 ± 15.7 to 34.9 ± 12.8 gmol F-' h (P < 0.05).6 Combination therapy with the nucleoside analogues ZDV and ddl may be the way forward in the treatment of advanced HIV disease, but the pharmacokinetic drug interaction described here should be taken into consideration.

Didanosine is a nucleoside analogue with potent inhibitory effects against the human immunodeficiency virus (HIV) [1]. It is converted within target cells to its active form ddA-triphosphate, which is thought to act as a chain terminator and inhibitor of reverse transcriptase [2]. In addition to the intracellular formation of ddA-TP, didanosine may be broken down to hypoxanthine and uric acid by the enzymes purine nucleoside phosphorylase (PNP) and xanthine oxidase [3]. As We have now investigated this possibility in a pharmacokinetic study of three men who were infected with HIV and who took ddl (250 mg twice daily) following zidovudine intolerance. Approval for the study was granted by the local Ethics Committee and all patients provided written informed consent. Each patient attended for study on two occasions separated by at least 2 weeks. After an overnight fast and following the insertion of an intravenous cannula, blood samples were taken at 15 min intervals for 2 h, half-hourly for another 2 h, and at 5 and 6 h after didanosine 250 mg.Each blood sample was divided into two aliquots, one was centrifuged immediately (3000 rev min-' for 10 min) and the other was left at room temperature and centrifuged when the pharmacokinetic study was completed. The separated plasma samples were exposed to a temperature of 580 C for 30 min to inactivate the human immunodeficiency virus [6] and then analysed for ddl using a commercial radioimminoassay (Sigma, London). Plasma samples were initially diluted 1:100 with blank plasma. The assay has a limit of detection of less than 1 nm. The interassay coefficient of variation was less than 10% at a concentration of 10 nm. Figure 1 shows the plasma drug concentrations for samples centrifuged immediately and those centrifuged after some delay, i.e. at the end of the study period. Although the latter were left for variable times (early samples left longer than later samples) this procedure reflects what will often happen in clinical studies.Plasma concentrations of ddl were consistently lower in samples whose centrifugation was delayed, with the greatest difference in early samples and little difference in later samples. 22% from 6.69 ± 1.12 to 5.21 ± 0.86 p,mol 1-1 h (mean + s.d; P < 0.001, Student's paired t-test; 95% CI of the difference 1.00 to 1.95). The maximum plasma drug concentration (Cmax) was reduced from 3.86 ± 0.66 to 3.10 ± 0.77 ,umol I' (P <0.01; 95% CI of the difference 0.22 to 1.30).The implications for pharmacokinetic studies involving ddI are clear. If blood samples are taken and left at room temperature (or higher) for any length of time, concentrations of ddl will be underestimated because of its continued breakdown in blood. Therefore, blood samples should be centrifuged immediately and the plasma separated to minimise breakdown by PNP.

The pharmacokinetics of zidovudine (ZDV) and dideoxyinosine (ddl) were investigated following administration alone and in combination to children with symptomatic HIV disease. The children were studied on three separate occasions and received ZDV 200 mg m-2, ddl 100 mg m-2 or a combination of ZDV 200 mg m-2 plus ddI 100 mg m-2. The administration of ddI did not significantly alter ZDV pharmacokinetics. The area under the curve (AUC) was 14.2 ± 4.9 and 15.8 ± 7.2 gimol 1-1 h and elimination half-life (t1/2,z) was 1.4 ± 0.4 and 1.2 ± 0.2 h in the absence and presence of ddl respectively. The peak concentration (Cmax), time to peak (tmax) and apparent oral clearance (CL/F) were also unchanged. The administration of ZDV had no significant effect on ddl Cmax, tmax, ti/2,z, or CL/F, however the AUC was reduced by 19% (5.9 ± 2.9 to 4.8 ± 2.7 ,umol 1-1 h; P < 0.05). This study suggests that ZDV and ddI may be co-administered to children with symptomatic HIV disease without concern of a clinically relevant pharmacokinetic drug interaction.

No abstract