Dapsone for Pneumocystis jirovecii pneumonia prophylaxis - applying theory to clinical practice with a focus on drug interactions.

Pneumocystis jirovecii pneumonia (PJP) is a potentially life-threatening infection that occurs in immunocompromised individuals. The incidence can be as high as 80% in some groups but can be reduced to less than 1% with appropriate prophylaxis. HIV-infected patients with a low CD4 count are at the highest risk of PJP. Others at substantial risk include haematopoietic stem cell and solid organ transplant recipients, those with cancer (particularly haematologic malignancies), and those receiving glucocorticoids, chemotherapeutic agents, and other immunosuppressive medications. Trimethoprim-sulfamethoxazole is an established first-line line agent for prevention and treatment of PJP. However, in some situations, this medication cannot be used and dapsone is considered a suitable cost-effective second line agent. However, information on potential interactions with drugs commonly used in immunosuppressed patients is lacking or contradictory. In this this article we review the metabolic pathway of dapsone with a focus on interactions and clinical significance particularly in patients with haematological malignancies. An understanding of this process should optimise the use of this agent.

Effects of H(2)-receptor antagonists on dapsone-induced methaemoglobinaemia in rats.

Dapsone (DDS) (4,4'diaminodiphenylsulfone), the drug of choice for the treatment of leprosy, frequently induces haemolytic anaemia and methaemoglobinaemia. N-hydroxylation, one of the major pathways of biotransformation, has been constantly related to the methaemoglobinaemia observed with the use of the drug. In order to determine the reversible inhibition of this toxicologic bioactivation pathway without changing the detoxification pathways of the drug or cytosolic acetylation, cimetidine (CIM), ranitidine and famotidine were administered in combination with DDS to male Wistar rats weighing 200-220 g. The animals were divided into nine groups of eight: group 1 received a single dose of 40 mg kg (-1) DDS in dimethylsulfoxide (DMSO) and groups 2-4 received the same treatment as group 1 but after the administration of a single dose of 100, 150 and 200 mg kg (-1) CIM, respectively, injected 2 h prior DDS administration. Groups 5-9 received the same treatment as group 2 but after the treatment of ranitidine (50 and 100 mg kg (-1) intraperitoneally (i.p.) in 200 microl DMSO) and famotidine (10, 50 and 100 mg kg (-1) i.p. in 200 microl DMSO), respectively. The animals were then anaesthetized with ether and blood was collected from the aorta for the determination of plasma DDS and monoacetyldapsone concentrations by HPLC and later for the determination of methaemoglobinaemia by spectrophotometry. CIM showed a higher affinity for cytochrome P-450 than famotidine and ranitidine. The results obtained showed the potentiality of the pharmacological effects of DDS with a low risk of adverse reactions, especially methaemoglobinaemia, which is dose dependent.

A dapsone-induced blood dyscrasia in the mouse: evidence for the role of an active metabolite.

In the female mouse, dapsone (50-500 mg kg-1, p.o.) caused a dose-related methaemoglobinaemia which peaked at 0.5-1 h with recovery to baseline values occurring by 4 h. Cimetidine (100 mg kg-1, p.o.), a known inhibitor of several hepatic P450 isozymes administered 1 h before dapsone, prevented the methaemoglobinaemia. In-vitro, dapsone required activation by mouse hepatic microsomes to cause methaemoglobin formation in mouse erythrocytes and cytotoxicity to human mononuclear leucocytes. In both instances, the toxic effects were markedly reduced by cimetidine. Daily dosing of mice with dapsone (50 mg kg-1, p.o.) for 3 weeks induced a blood dyscrasia, characterized by a fall of platelet and white blood cell counts, which was inhibited by cimetidine (100 mg kg-1, p.o. daily). It is concluded that an active metabolite of dapsone arising from a P450-dependent pathway is involved in the genesis not only of the methaemoglobinaemia but also the blood dyscrasia arising from repeated administration of the drug in this species.

Monoacetyldapsone inhibition of dapsone N-hydroxylation by human and rat liver microsomes.

Dapsone (DDS) is metabolized by N-hydroxylation and N-acetylation to DDS hydroxylamine (DDS-NOH) and monoacetyldapsone (MAD), respectively. The activities of these two alternative and independent reactions vary widely between individuals and show an inverse relationship during chronic DDS therapy. Toxicity observed during DDS therapy has been attributed to DDS-NOH. The observation of reduced toxicity in rapid acetylators, who are also poor hydroxylators, therefore, raised the possibility that MAD may be inhibiting DDS-NOH formation. This hypothesis was tested in human and rat liver microsomes. Human liver microsomes hydroxylated DDS with a lower affinity (KM 2-fold greater) and lower maximal catalytic activity (Vmax 12-fold lower) than that of the rat. The relative catalytic activity (Vmax/KM) was 22-fold higher in rat compared with human liver microsomes. Furthermore, MAD was a potent inhibitor of DDS N-hydroxylation by rat liver microsomes (52% inhibition at 0.01 mM MAD) compared with human liver microsomes (23% inhibition at 0.4 mM MAD). Human, but not rat, liver microsomes deacetylated MAD to DDS by an NADPH independent mechanism. These results show that substantial differences exist in DDS N-hydroxylase between rats and humans, with respect to substrate affinity, enzyme activity, and susceptibility to inhibition, such that information obtained from the rat should not be extrapolated to humans. We conclude that MAD is a potent inhibitor of DDS-NOH formation in rat liver microsomes. The degree of inhibition in human microsomes, however, suggest that MAD is unlikely to be a significant modulator of enzyme activity in vivo.

In vitro and in vivo results of brodimoprim and analogues alone and in combination against E. coli and mycobacteria.

New diaminodiphenylsulfone inhibitors of dihydropteroate synthase are described with increased inhibitory activity against mycobacteria and plasmodia, whereas their side effect of methemoglobin formation could be suppressed. The optimization of diaminobenzylpyrimidines, inhibitors of dihydrofolate reductase, led to derivatives with increased inhibitory effect against mycobacteria, especially M. leprae and plasmodia. Some of these derivatives show autosynergism. Finally the combination of brodimoprim (BDP) and dapsone (DDS) was developed for the treatment of leprosy. First clinical trials in Paraguay and Ethiopia show that combinations of BDP/DDS and BDP/DDS plus rifampicin were highly effective and may become an alternative multi-drug therapy for the treatment of leprosy. The tolerance of the regimens used was generally good.

Lack of effect of proguanil on the pharmacokinetics of dapsone in healthy volunteers.

The multiple-dose kinetics of dapsone (DDS) and its principal metabolite monoacetyldapsone (MADDS) were determined in 6 healthy volunteers after daily administration of low-dose dapsone (10 mg). Comparison with a previous study involving the same volunteers on a daily regimen of proguanil (200 mg) plus dapsone (10 mg) revealed no statistically significant differences in the maximum plasma concentrations, area under the plasma drug concentration curves and elimination half-lives of both DDS and MADDS in the presence of proguanil. Although these findings suggest that proguanil does not alter the pharmacokinetics of DDS and MADDS, the possibility that proguanil affects the disposition of hydroxylated metabolites of dapsone, which appear to mediate dapsone toxicity, cannot be excluded.

Superoxide is an antagonist of antiinflammatory drugs that inhibit hypochlorous acid production by myeloperoxidase.

Myeloperoxidase, the most abundant enzyme in neutrophils, catalyses the conversion of hydrogen peroxide and chloride to hypochlorous acid. This potent oxidant has the potential to cause considerable tissue damage in many inflammatory diseases. We have investigated the ability of dapsone, diclofenac, primaquine, sulfapyridine and benzocaine to inhibit hypochlorous acid production by stimulated human neutrophils. The drugs were also tested against purified myeloperoxidase using xanthine oxidase to generate hydrogen peroxide and superoxide. The inhibitory effects of the drugs on hypochlorous acid production, either by cells stimulated with phorbol myristate acetate or by myeloperoxidase and xanthine oxidase, were significantly less than those determined with myeloperoxidase and reagent hydrogen peroxide. Comparable potency was observed only when superoxide dismutase was present to remove superoxide. We also observed that with the xanthine oxidase system, inhibition of hypochlorous acid production by dapsone decreased markedly as the concentration of myeloperoxidase increased. Dapsone was a poor inhibitor of hypochlorous acid production by neutrophils stimulated with opsonized zymosan, regardless of the presence of superoxide dismutase. With this phagocytic stimulus, catalase inhibited hypochlorous acid formation by only 60%, which indicates that a substantial amount of the hypochlorous acid detected originated from within phagosomes. Thus, it is apparent that dapsone is unable to affect intraphagosomal conversion of hydrogen peroxide to hypochlorous acid. All the drugs inhibit myeloperoxidase reversibly by trapping it as its inactive redox intermediate, compound II. We propose that superoxide limits the potency of the drugs by reducing compound II back to the active enzyme. Furthermore, under conditions where the activity of myeloperoxidase exceeds that of the hydrogen peroxide-generating system, which is most likely to occur in phagosomes, partial inhibition of myeloperoxidase need not affect hypochlorous acid production. We conclude that drugs that inhibit myeloperoxidase by converting it to compound II are unlikely to be effective against hypochlorous acid-mediating tissue damage.

Inhibition of dapsone-induced methaemoglobinaemia by cimetidine in the rat during chronic dapsone administration.

Dapsone undergoes N-acetylation to monoacetyl dapsone as well as N-hydroxylation to a hydroxylamine which is responsible for the haemotoxicity (i.e. methaemoglobinaemia; Met Hb) of the drug. Since dapsone is always given chronically, we have investigated the ability of cimetidine to inhibit Met Hb formation caused by repeated dapsone administration. The drug was given (i.p.) to four groups (n = 6 per group) of male Wistar rats, 300-360 g. Group I received 10 mg kg-1 at 1, 24, 48 and 72 h. Group II received 10 mg kg-1 at 1, 8, 24, 32, 48, 56, 72 and 80 h. Groups III and IV received the drug as for groups I and II, respectively, as well as cimetidine (50 mg kg-1) 1 h before each dose of dapsone. Twice daily dapsone administration (Group II) resulted in a significantly greater (P less than 0.05) Met Hb AUC (757 +/- 135 vs 584 +/- 115% Met Hb h), dapsone AUC (140 +/- 17.5 vs 113 +/- 13.0 micrograms h mL-1) and monoacetyl dapsone AUC (48.2 +/- 18.3 vs 10.8 +/- 4.6 micrograms h mL-1) compared with a single daily dapsone dose (group I). The administration of cimetidine before the once daily dose of dapsone (group III) resulted in a significant (P less than 0.05) fall in Met Hb (302 +/- 179 vs 584 +/- 115% Met Hb h) and an increase in both the dapsone (151 +/- 22.2 vs 113 +/- 13.0 micrograms h mL-1) and monoacetyl dapsone AUC values (33.6 +/- 5.8 vs 10.8 +/- 4.0 micrograms h mL-1) compared with a single daily dose of dapsone (group I).(ABSTRACT TRUNCATED AT 250 WORDS)

Inhibition of dapsone-induced methaemoglobinaemia in the rat isolated perfused liver.

We have investigated the disposition of dapsone (DDS, 1 mg) in the rat isolated perfused liver in the absence and the presence of cimetidine (3 mg). After the addition of DDS alone to the liver there was a monoexponential decline of parent drug concentrations and rapid formation of DDS-NOH (within 10 min) which coincided with methaemoglobin formation (11.7 +/- 3.0%, mean +/- s.d.) which reached a maximum (22.6 +/- 9.2%) at 1 h. The appearance of monoacetyl DDS (MADDS) was not apparent until 30-45 min. Addition of cimetidine resulted in major changes in the pharmacokinetics of DDS and its metabolites. The AUC of DDS in the presence of cimetidine (1018.8 +/- 267.8 micrograms min mL-1) was almost three-fold higher than control (345.0 +/- 68.1 micrograms min mL-1, P less than 0.01). The half-life of DDS was also prolonged by cimetidine compared with control (117.0 +/- 48.2 min vs 51.2 +/- 22.9, P less than 0.05). The clearance of DDS (3.0 +/- 0.55 mL min-1) was greatly reduced in the presence of cimetidine (1.03 +/- 0.26 mL min-1 P less than 0.01). The AUC0-3h for DDS-NOH (28.3 +/- 21.2 micrograms min mL-1) was significantly reduced by cimetidine (8.1 +/- 3.40 micrograms min mL-1, P less than 0.01). In contrast, there was a marked increase in the AUC0-3h for MADDS (32.7 +/- 25.8 micrograms min mL-1) in the presence of cimetidine (166.0 +/- 26.5 micrograms min mL-1 P less than 0.01). The methaemoglobinaemia associated with DDS was reduced to below 5% by cimetidine. Hence, a shift in hepatic metabolism from bioactivation (N-hydroxylation) to detoxication (N-acetylation) caused by cimetidine, was associated with a fall in methaemoglobinaemia. These data suggest that the combination of DDS with a cytochrome P450 inhibitor might reduce the risk to benefit ratio of DDS.