Pharmacodynamics of Flubendazole for Cryptococcal 1 Meningoencephalitis: Repurposing and Reformulation of an Anti- 2 Parasitic Agent for a Neglected Fungal Disease 3

49 Current therapeutic options for cryptococcal meningitis are limited by toxicity, global supply and 50 emergence of resistance. There is an urgent need to develop additional antifungal agents that 51 are fungicidal within the central nervous system and preferably orally bioavailable. The 52 benzimidazoles have broad-spectrum anti-parasitic activity, but also have in vitro antifungal 53 activity that includes Cryptococcus neoformans . Flubendazole (a benzimidazole) has been 54 reformulated by Janssen Pharmaceutica as an amorphous solid drug nanodispersion to develop 55 an orally bioavailable medicine for the treatment of neglected tropical diseases such as 56 onchocerciasis. We investigated the in vitro activity, the structure-activity-relationships and both 57 in vitro and in vivo pharmacodynamics of flubendazole for cryptococcal meningitis. Flubendazole 58 has potent in vitro activity against Cryptococcus neoformans with a modal MIC of 0.125 mg/L 59 using European Committee for Antimicrobial Susceptibility Testing (EUCAST) methodology. 60 Computer models provided an insight into the residues responsible for the binding of 61 flubendazole to cryptococcal ß-tubulin. Rapid fungicidal activity was evident in a hollow fiber 62 infection model of cryptococcal meningitis. The solid drug nanodispersion was orally 63 bioavailable in mice with higher drug exposure in the cerebrum. The maximal dose of 64 flubendazole (12 mg/kg/day) orally resulted in a ~2 log 10 CFU/g reduction in fungal burden 65 compared with vehicle-treated controls. Flubendazole was orally bioavailable in rabbits, but 66 there were no quantifiable drug concentrations in the CSF or cerebrum and no antifungal activity 67 was demonstrated in either CSF or cerebrum. These studies provide evidence for the further 68 study and development of the benzimidazole scaffold for the treatment of cryptococcal 69 meningitis.


72
Cryptococcal meningoencephalitis (herein meningitis) is a common and lethal disease in 73 immunosuppressed patients (1, 2). This disease is predominately associated with advanced HIV 74 infection and has the highest incidence in low to middle income countries (1). The number of 75 effective agents is despairingly small (3). All available induction and maintenance regimens are 76 constructed with three antifungal agents: amphotericin B (AmB), flucytosine (5FC) and 77 fluconazole (4). Each of these compounds has significant adverse effects that include infusional 78 toxicity (AmB), nephrotoxicity (AmB (5)), bone marrow suppression (AmB and 5FC (5, 6)) and 79 hepatotoxicity (fluconazole and 5FC (7)). Moreover, there are significant inherent limitations 80 that include fungistatic effects (fluconazole; (8)) and the potential emergence of drug resistance 81 (fluconazole and 5FC; (9-11)). Thus, there is an urgent imperative to develop new agents. Orally  In vitro studies 107 Flubendazole displayed potent in vitro activity (MIC 0.06-0.25 mg/L; Table 1) against C. 108 neoformans. The MICs were comparable when EUCAST and CLSI methodology was used.

113
In vitro DMPK assessment of commercially available flubendazole powder confirmed a 114 favorable logD7.4 of 2.9. Plasma protein binding was 90.6% and there was low metabolic 115 turnover (Hu Mic Clint = 44 µl/min/mg and Rat Hep Clint = 39 µl/min/10 6 cells). However, 116 aqueous solubility was poor (0.8 µM), which is characteristic of the benzimidazoles. This in vitro 117 DMPK assessment was consistent with subsequent in vivo observations (see below). Poor 118 aqueous solubility limits absorption through the gut, but once in the bloodstream the drug has 119 favorable pharmacokinetic properties (e.g. ability to pass through cell membranes, low 120 metabolism, and high concentrations of free drug) that enable it to reach the effect site.  Rapid fungicidal activity was observed in the hollow fiber infection models. Controls 140 grew from an initial density of approximately log 10 CFU/mL 6 to log 10 CFU/mL 8-9. Following the 141 administration of flubendazole there was a progressive decline in the fungal density in the 142 hollow fibre in all arms. There was an exposure-dependent decline in fungal burden.    There was no demonstrable antifungal effect in rabbits receiving 6 mg/kg/day. There 184 may be some effect in rabbits receiving 22.5 mg/kg q24h, but if present the effect was small and 185 these assessments were limited by few animals. There were no statistically significant 186 differences in the area under the log 10 CFU/g-time curve for each regimen even though this may 187 be a relatively insensitive test of antifungal effect. Furthermore, there was no difference in the  Flubendazole has striking in vitro activity against Cryptococcus neoformans that was 208 evident in the MIC testing and the pharmacodynamic studies in the hollow fiber infection model.

209
There was modest antifungal activity in the murine model, which is not as prominent as that binding interactions with C. neoformans β-tubulin. This may provide the potential to exploit this 241 differential binding to establish a favorable therapeutic index. It is also worth emphasizing that 242 the benzimidazoles may have additional targets beyond β-tubulin that have the further potential 243 to provide differential activity between human and fungal proteins, but this requires further 244 investigation (21-24).

245
The potential utility of congeners of flubendazole now rests with medicinal chemistry 246 programs. Compounds must be synthesized that exhibit differential activity against cryptococcal 247 and human tubulin (if that is possible) so that there is an acceptable safety margin and toxicity 248 profile. Furthermore, the compound must be able to traverse the gut (compounds that are not 249 orally bioavailable will be less clinically valuable) and then the blood-brain-barrier to achieve 250 concentrations that are ideally fungicidal. The latter will be promoted by new molecules that 251 low molecular weight lipophilic compounds that are not substrates for active pumps such as P-252 glycoprotein. This will undoubtedly also require the use of novel formulation technologies to 253 ensure compounds that are poorly soluble to become useful agents for disseminated infections.  These isolates were identified to species level using standard microbiological techniques.

294
The porcine tubulin assay was performed according to the manufacturer's instructions. 295 Briefly, the 96-well assay plate was pre-warmed to 37°C prior to use. Five μL of test

305
All data points were acquired in triplicate and IC 50 values were calculated with GraphPad Prism.

306
The IC 50 value was defined as the drug concentration required to inhibit tubulin polymerization 307 by 50% compared with negative control.     The pharmacokinetic and pharmacodynamic datasets from mice were modelled using the 433 program Pmetrics (33) and the following five inhomogeneous differential equations: