INTRODUCTION
Orally administered itraconazole and sulfamethoxazole-trimethoprim1 are drugs recommended for the treatment of patients with paracoccidiodomycosis of low or moderate severity1. Previous studies have shown the clinical efficacy of itraconazole2 and sulfamides3 as well as ketoconazole4 and fluconazole5 for the control of systemic disease caused by Paracoccidioides brasiliensis. Comparative studies of antifungal agents for the treatment of paracoccidioidomycosis have detected similar efficacy of ketoconazale and amphotericin B6 and also of itraconazole, ketoconazole and sulfadiazine7. A more recent controlled open label comparative clinical investigation showed that itraconazole and voriconazole were equally successful in controlling this fungal disease after six to 12 months of treatment8.
Voriconazole is a second-generation triazole currently recommended for the treatment of patients with acute invasive aspergillosis. This drug has a broad spectrum of antifungal activity including dimorphic fungi, Cryptococcus spp., Candida spp., Trichosporon spp. and Fusarium spp.9 . Voriconazole shows good bioavailability, providing maximum plasma concentrations of 2 to 3µg/mL after the oral administration of a 200mg dose10. A minimum inhibitory concentration of < 0.03 to 2µg/mL voriconazole against P. brasiliensis has been observed11, suggesting that this triazole drug could also be a therapeutic alternative for the control of paracoccidioidomycosis.
Voriconazole has been studied in only a few patients with paracoccidioidomycosis8,12 and its in vivoefficacy against P. brasiliensis is still little known. Experimental models of infection can provide relevant data about the comparative efficacy of antifungal drugs against this dimorphic fungus13.
The objective of the present study was to assess the effectiveness of voriconazole, ketoconazole, itraconazole, fluconazole and sulfamethoxazole-trimethoprim in the control of experimental paracoccidioidomycosis of the rat.
METHODS
Animals
The study was conducted on female Wistar rats (Rattus novergicus) supplied by the University of São Paulo, Ribeirão Preto Campus, SP, Brazil. At the time of inoculation the animals weighed 45 to 60g and were housed 6 to 8 to a cage, with free access to food and water.
Paracoccidioides brasiliensis and inoculums
The rats were infected with the BOAS strain of P. brasiliensis isolated from a patient and kept in the laboratory by subculture in Sabouraud medium. The fungus was cultured at 35ºC for 7 to 15 days and yeast-like cells were collected in isotonic saline. The suspension was homogenized, adjusted to a content of 1×107 to 2×107 yeast cells/mL in a hemocytometer and 0.2mL amounts were injected into the lateral vein of the rat tail. Fungus viability was confirmed by culturing the yeast cell suspension on plates containing brain-heart-infusion agar supplemented with 4% horse serum and 5% of a culture filtrate of the B339 strain of P. brasiliensis.
Drugs and treatment
The following drugs, manufactured for clinical use, were employed to treat the animals: voriconazole (Vfend(r), Pfizer), ketoconazole (Nizoral(r), Jansen-Cilag), fluconazole (Zoltec(r), Pfizer), and sulfamethoxazole-trimethoprim (Bactrim(r), Roche). The tablets or capsules of each drug were weighed and macerated and added to a 3% aqueous solution of gelatin (Vetec(r), Brazil) immediately before being administered to the animals. The drugs suspended in gelatin were administered by gavage once a day (ketoconazole, fluconazole and itraconazole) or twice a day with an 8 hour interval between doses (voriconazole and sulfamethoxazole-trimethoprim). The antifungal treatment was started seven days after P. brasiliensis inoculation.
Paracoccidioides brasiliensis colony forming units
The rats were sacrificed 48 or 78 hours after the last dose of the antifungal agents. The lungs and spleen were placed in a solution of penicillin G (200U/mL) and gentamicin (48µg/mL) and triturated with a tissue homogenizer (Marconi(r), Brazil). Serial dilutions of the lung and spleen homogenates were cultured on plates containing brain-heart-infusion agar supplemented with 4% horse serum and 5% of a culture filtrate of the B339 strain of P. brasiliensis. The colony forming units (CFU) of P. brasiliensis were counted after 15 days of culture at 35ºC and the total fungal burden of each organ was estimated.
Experimental design
A) To assess the efficacy of different doses of voriconazole, 5, 10 or 20mg/kg weight/day of the drug was administered to the animals for three weeks. The number of P. brasiliensis CFU in the lungs and spleen of rats was compared between the treated animals and also to the CFU of rats that received only gelatin (controls); B) In two independent experiments, the CFU of P. brasiliensis in tissues were analyzed in rats respectively receiving for three weeks voriconazole (7 or 10mg/kg weight/day), ketoconazole (12 or 15mg/kg weight/day), fluconazole (6mg/kg weight/day), itraconazole (4mg/kg weight/day), sulfamethoxazole-trimethoprim (120 or 150mg/kg weight/day of sulfamethoxazole), or only 3% gelatin (control); C) For the survival study, the animal groups received the same drugs for 12 days and were followed up together with the controls for 36 days after inoculation with P. brasiliensis.
Statistical analysis
The mean numbers of P. brasiliensis CFU in the tissues of the different animal groups were compared by analysis of variance (ANOVA) with pairwise comparisons in according to Bonferroni correction and using the PROC GLM feature of the SAS software, version 9.0. Survival was evaluated by the Kaplan-Meier method using a parametric model based in log-normal distribution; this analysis employed the SAS software, version 9.0 (PROC LIFEREG) and R software with a library survival. Differences were considered to be significant when p<0.05.
RESULTS
Voriconazole significantly reduced the number of P. brasiliensis CFU in the lungs and spleen of infected animals. No difference in efficacy was observed between the doses of 5, 10 or 20 mg/kg weight/day, except for a tendency to a greater reduction of the number of CFU in the lungs with increasing doses (Table 1 ). In independent experiments, voriconazole 7 or 10mg/kg weight/day reduced the fungal burden in the spleen of rats, but only the higher dose reduced the pulmonary CFU ( Table 2 ).
CFU: colony forming units; SD: standard deviation;
CFU: colony forming units; SD: standard deviation; SMX-TMP: sulfamethoxazole-trimethoprim;
a NS: non significant (C xV) p- NS ;
(CxK) p=0.0187, (CxF) p<0.0001, (CxI) p<0.0001, (CxST) p=0.0003, (IxV) p<0.001, (F x V) p = 0.0002, (ST x V) p= 0.0115;
b(CxV) p=0.0291, (CxK) p=0.0897, (CxF) p<0.0001, (CxI) p<0.0001, (CxST) p=0.0020; (I x V) p<0.0001, (FxV) p= 0.0016;
c(CxV) p<0.0001, (CxK) p<0.0001, (CxF) p<0.0001, (CxI) p<0.0001, (CxST) p<0.001, (KxV) p= 0.0155, (FxV) p<0.0001 and (IxV) p< 0.0001;
d(CxV) p=0.0003, (CxK) p<0.0001, (CxF) p<0.0001, (CxI) p<0.0001, (CxST) p=0.0003, (FxV) p= 0.0233 and (IxV) p =0.0349.
Compared to other drugs, voriconazole 7mg/kg weight/day was less effective in reducing the pulmonary fungal burden than fluconazole, itraconazole and sulfamethoxazole-trimethoprim. At the dose of 10mg/kg weight/day it was also less efficient in reducing pulmonary CFU than ketoconazole, fluconazole and itraconazole ( Table 2 ). The last two drugs were also more effective than voriconazole in reducing the fungal burden of the spleen ( Table 2 ).
Figure 1 and Table 3 presents the survival data for infected animals treated with different antifungal agents for 12 days. Fluconazole, itraconazole and sulfamethoxazole-trimethoprim increased the mean survival of the animals, while the survival of animals treated with voriconazole and ketoconazole was similar to that of the untreated group.
DISCUSSION
This was the first study to assess simultaneously the antifungal efficacy of voriconazole, other azole drugs and sulfamethoxazole-trimethoprim in experimental paracoccidioidomycosis. We used a model of intravenous animal infection in which granulomas containing P. brasiliensis are formed, being more numerous in the lungs, spleen and liver. After the occurrence of established fungal infection in tissues, the animals were treated with medications available for clinical use and at doses similar to those recommended for patient treatment.
Voriconazole was effective against P. brasiliensis, as demonstrated by the reduction of the fungal burden in the lungs and spleen of the animals. Three separate experiments have suggested a dose-dependent action of voriconazole. The dose-dependent efficacy of this drug was detected in mice infected with Aspergillus spp. and treated with 10, 20 and 30mg/kg weight/day14. In the present study, doses of 5 and 7mg/kg weight/day had a lesser impact in the reduction of number of CFU in the lungs. In another preliminary experiment in which voriconazole 7mg/kg weight/day was administered in a single daily dose, again no reduction of the pulmonary fungal burden occurred (data not shown).
These results suggest that the dose of voriconazole recommended for clinical treatment (4 to 6mg/kg weight/twice daily)9 does not have a sufficient antifungal activity to control the paracoccidioidomycosis induced in rats. This is probably due to an accelerated metabolism self-induced by voriconazole in rats, leading to a half-life of about one to two hours15.
Voriconazole 10mg/kg weight/day significantly reduced the fungal burden in the lungs and spleen after three weeks of treatment. However, it failed to prolong the life of rats treated for 12 days, a period sufficient to demonstrate the anti-P. brasiliensis action of other drugs. The survival of mice infected with Blastomyces dermatitidis and treated for 23 days with voriconazole 1, 5 and 20mg/kg weight/day was increased compared to untreated animals16. Voriconazole 5 and 10mg/kg/day prolonged the survival of guinea pigs with invasive trichosporonosis and reduced the number of CFU in the liver and kidney of these animals17. Mice infected with Cryptococcus neoformans showed 100% survival after receiving voriconazole 60mg/kg weight/day, but doses of 10 and 40mg/ kg weight/day had a lower impact on animal survival18. In addition to the quantity of voriconazole administered daily, other variables may influence the results of different studies, such as time and route of administration of the drug, number of CFU inoculated, and the animal model employed.
Fluconazole and itraconazole showed the best performance in the comparative assessment of the drugs by both reducing the tissue fungal burden in a more intense manner and by prolonging animal survival. In the same model of rat paracoccidioidomycosis, fluconazole showed efficacy comparable to that of amphotericin B in reducing the number of CFU in the lungs13. This triazole is little used for the treatment of paracoccidioidomycosis, but has proved to be effective even in immunosuppressed patients5,19. Itraconazole reduced the fungal burden of tissues and increased the survival of mice and rats infected with P. brasiliensis 20,21 . In two controlled clinical studies, this triazole elicited high rates of a favorable response in patients with paracoccidioidomycosis7,8 . In the model of rat paracoccidioidomycosis, sulfamethoxazole-trimethoprim had a lower antifungal activity than itraconazole and fluconazole, but a higher activity than ketoconazole and voriconazole. This experimental result supports the use of sulfamethoxazole-trimethoprim for the treatment of patients with paracoccidioidomycosis but this drug combination is now recommended only as an alternative to itraconazole1. Among the drugs tested, ketoconazole showed the lowest antifungal activity, perhaps due to its lower bioavailability.
The application of the results of antifungal efficacy obtained in animal models of infection to human therapy should be considered with caution. The data of the present study validate the therapeutic use of itraconazole, fluconazole and sulfamethoxazole-trimethoprim in human paracoccidioidomycosis. Voriconazole showed in vivo activity against P. brasiliensis and has the potential for use in the treatment of some patients with paracoccidioidomycosis of low or moderate severity, especially those with damage to the central nervous system8. The rate of voriconazole metabolism varies from person to person22 and patients treated with the recommended doses usually present plasma levels of less than 1mg/mL23. It may be necessary to monitor the levels of this drug in order to establish an adequate dose, to achieve a favorable therapeutic response and to minimize adverse effects24,25 .