Evaluation of anti-sleeping-sickness drugs and topoisomerase inhibitors in combination on Trypanosoma brucei

Sir, The chemotherapy of human African trypanosomiasis (HAT) or sleeping sickness is unsatisfactory. For the treatment of this fatal disease, only four drugs, some of which were developed over 50 years ago, are available.1 In the light of declining investment in drug development for parasitic diseases, new strategies for the chemotherapy of HAT are urgently required. Drugs developed as potential anti-tumour agents could also be of use against HAT. For example, the ornithine decarboxylase inhibitor eflornithine (d,l-α-difluoromethylornithine; DFMO) introduced in 1990 for the treatment of West African sleeping sickness was initially developed as an anti-cancer drug.2 Other examples of anti-cancer drugs with anti-trypanosomal activity include DNA topoisomerase inhibitors,3 a number of which are in clinical use to treat cancer.4 It is well established that combination drug regimens often achieve greater therapeutic efficacy than monotherapies (e.g. in the treatment of cancer, AIDS and tuberculosis). Eflornithine/nifurtimox and melarsoprol/nifurtimox combination therapies are currently being evaluated in clinical trials for the treatment of HAT to replace monotherapies with melarsoprol or eflornithine.5–7 In the best case, drug combinations interact synergistically. However, additive effects are also desirable as they permit decreased dosage while maintaining efficacy.8 Thus, they may reduce the toxicity of monotherapy regimens and have the additional advantage of reducing risks of development of drug resistance.8 We therefore investigated the trypanocidal activities of combinations of anti-HAT drugs (eflornithine, melarsoprol, pentamidine and suramin), DNA topoisomerase I inhibitors [topotecan and 7-ethyl-10-hydroxycamptothecin (SN-38; the active compound of irinotecan)] and DNA topoisomerase II inhibitors (aclarubicin, daunorubicin, doxorubicin, epirubicin, idarubicin and mitoxantrone) against bloodstream forms of Trypanosoma brucei cultured in vitro. For evaluation of the interaction between drugs, the isobolographic method was selected.8 First, the 50% growth inhibition (GI50) value of each drug was determined. Then, the concentrations of the combinations were computed based on the GI50 values of each individual agent at a constant concentration ratio of 1:1. Bloodstream forms of the T. brucei clone 427-221a9 were then incubated with the calculated concentrations of the combination in a range from 4× GI50 to 0.24× GI50. After 48 h of incubation, live cells were counted using a Neubauer haemocytometer and the GI50 value of each drug in the presence of the other co-administered drugs was calculated. The combination index (CI) for each drug combination was determined using the equation (GI50,Acomb/GI50,Bcomb)+(GI50,Asingle/GI50,Bsingle), where GI50,Acomb and GI50,Bcomb are the concentrations of drugs A and B used in the combination to achieve GI50, and GI50,Asingle and GI50,Bsingle are the concentrations of the single drugs to achieve the same effect. A CI value of less than, equal to and more than 1 indicates synergy, additivity and antagonism, respectively. Any combination of anti-HAT drugs and DNA topoisomerase I inhibitors produced CI values >1 (1.11–1.95), indicating an antagonistic effect of these drug combinations. With the exception of the eflornithine/mitoxantrone combination, an antagonistic effect was also found for all combinations of anti-HAT drugs and DNA topoisomerase II inhibitors. The CI values for these combinations were between 1.11 and 2.34. The eflornithine/mitoxantrone combination showed an additive effect with a CI value of 0.94 ± 0.10. The GI50 of eflornithine dropped from 75.27 to 34.23 µM, and the GI50 of mitoxantrone dropped from 5.99 to 2.89 nM (Figure 1a). An antagonistic effect was also observed for most combinations of DNA topoisomerase I and II inhibitors with CI values ranging between 1.12 and 1.28. Only the combinations topotecan/mitoxantrone and SN-38/doxorubicin displayed additive effects with CI values of 1.01 ± 0.07 and 1.05 ± 0.03, respectively. The GI50 values of topotecan and mitoxantrone dropped from 2.29 to 1.26 µM and from 6.41 to 2.92 nM (Figure 1b), and the GI50 values of SN-38 and doxorubicin dropped from 0.34 to 0.19 µM and from 35.6 to 16.68 nM (Figure 1c), respectively. Isobolographic plot for the interaction between eflornithine and mitoxantrone (a), topotecan and mitoxantrone (b) and SN-38 and doxorubicin (c). The broken line that connects the GI50 points for the single drug treatments (filled squares) is the theoretical additive line. The GI50 values of the combinations are indicated by open circles. Each point represents the mean ± SD of three experiments. Next, the general cytotoxicity of the three drug combinations that act additively on trypanosomes was assayed with human myeloid leukaemia HL-60 cells. Each drug combination was tested at the concentration that inhibited the growth of trypanosomes by 50%. The combinations topotecan/mitoxantrone (1.26 µM/2.92 nM) and SN-38/doxorubicin (0.19 µM/16.68 nM) were much more toxic to HL-60 cells than to trypanosomes. They inhibited the growth of the human cells by 97% and 99%, respectively. Encouragingly, the combination eflornithine/mitoxantrone (34.23 µM/2.89 nM) was less toxic to HL-60 cells than to the parasite cells. The eflornithine/mitoxantrone combination inhibited the growth of HL-60 cells by only 20%. In this study, we tested 44 drug combinations of which 3 showed an additive effect on the growth inhibition of bloodstream forms of T. brucei. Of the three drug combinations with additive action, only the eflornithine/mitoxantrone combination was found to be less toxic to human HL-60 cells than to trypanosomes, indicating some degree of selectivity. This finding suggests that the combination of eflornithine and mitoxantrone could be a potential alternative treatment for HAT. As this was an in vitro study, further investigation should be carried out in order to establish the in vivo efficacy of the eflornithine/mitoxantrone combination. This study was supported by a research grant from the British Society for Antimicrobial Chemotherapy (reference GA725). None to declare. We thank Dr Kevin Tyler for critical reading of the manuscript.