Rate-of-Kill (RoK) assays to triage large compound sets for Chagas disease drug discovery: Application to GSK Chagas Box

Chagas disease (CD) is a human disease caused by Trypanosoma cruzi. Whilst endemic in Latin America, the disease is spread around the world due to migration flows, being estimated that 8 million people are infected worldwide and over 10,000 people die yearly of complications linked to CD. Current chemotherapeutics is restricted to only two drugs, i.e. benznidazole (BNZ) and nifurtimox (NIF), both being nitroaromatic compounds sharing mechanism of action and exerting suboptimal efficacy and serious adverse effects. Recent clinical trials conducted to reposition antifungal azoles have turned out disappointing due to poor efficacy outcomes despite their promising preclinical profile. This apparent lack of translation from bench models to the clinic raises the question of whether we are using the right in vitro tools for compound selection. We propose that speed of action and cidality, rather than potency, are properties that can differentiate those compounds with better prospect of success to show efficacy in animal models of CD. Here we investigate the use of in vitro assays looking at the kinetics of parasite kill as a valuable surrogate to tell apart slow- (i.e. azoles targeting CYP51) and fast-acting (i.e. nitroaromatic) compounds. Data analysis and experimental design have been optimised to make it amenable for high-throughput compound profiling. Automated data reduction of experimental kinetic points to tabulated curve descriptors in conjunction with PCA, k-means and hierarchical clustering provide drug discoverers with a roadmap to guide navigation from hit qualification of a screening campaign to compound optimisation programs and assessment of combo therapy potential. As an example, we have studied compounds belonging to the GSK Chagas Box stemmed from the HTS campaign run against the full GSK 1.8 million compounds collection [1].Author summary: One of the challenges in early drug discovery of small molecules is the improvement of the poor success rate in the translation from in vitro biological profile into efficacy in disease models, and ultimately in the clinic. Reductionist in vitro models on the bench may not properly recapitulate disease biology, thus overlooking critical properties of candidate compounds. Chagas Disease is a neglected tropical disease caused by Trypanosoma cruzi, a protozoan parasite with a complex life cycle. Despite the promising prospect based on in vitro and in vivo preclinical studies, efforts to reposition antifungal azoles turned out to be disappointing in clinical trials, with treatment failure in Chagas patients. This raises the question of whether we are using the right preclinical tools for decision-making about moving compounds forward for the treatment of this disease. We hypothesise that in vitro potency and efficacy values alone might be distorting the translational power of preclinical compounds, and we propose the use of rate-of-kill (RoK) assays in high-throughput mode. Herewith we disclose a simple, systematic, and automated methodology of analysis of the otherwise complex kinetic patterns, which provides drug discoverers with a navigation guide along a compound optimisation program or prioritisation of best exemplars across different chemical series.