Combinations of two or more drugs frequently offer therapeutic benefits due to improved potency and safety, and decreased possibility of developing resistance to therapy. However, it remains extremely challenging to identify potent drug combinations by trial and error because of the large number of possible candidate combinations that need to be tested. For example, even with the recent dramatic advances in screening technologies (aided, for example, by the establishment of the acoustic dispense platforms), it still remains impossible to screen a relatively modest set of a dozen drugs at half a dozen concentrations efficiently, as the total number of combinations already exceeds a billion. To tackle this problem, the team resorted to building a multiphase process (see figure) based on stochastic search algorithms, which used biased random walks of the input states (drug concentrations) to maximize the output (a chosen biological activity measured through an assay). First, a single-agent screen is conducted to determine the concentration range of interest for each drug. The second phase, called the screening test, uses a two-level orthogonal array experimental design to construct a first-order linear model of the relationship between drug–dose combinations and bioactivity whose application yields the drugs with the greatest impact on bioactivity. In the third phase, the selected drugs are tested in an iterative fashion to yield a quadratic model of the relationship between drug–dose combinations and bioactivity. In the last phase, in-depth analysis of the finalized second-order quadratic model is used to construct a model surface of the relationship between drug–dose combinations and bioactivity, ultimately yielding the final drug combinations and corresponding dose ratios. The team used an IPTG-inducible GFP expression system to evaluate the antimicrobial efficacy against Mycobacterium tuberculosis of drug combinations in an in vitro human macrophage culture system. The fluorescent readout reported on inhibition of Mtb metabolic activity and could be taken to a high throughput. Fourteen drugs were tested at five doses each, an experiment that if conducted using the traditional all-versus-all strategy would have required the setup of more than 6 billion tests. Using the present approach allowed the elimination of low-performing drugs very rapidly. Interestingly, both isoniazid (INH) and rifampicin (RIF) were eliminated from the optimized regimens because of a small effect on efficacy (INH) or antagonistic interactions with other drugs (RIF), in agreement with previous observations of antagonistic interactions of INH and RIF with other first-line tuberculosis drugs. While the method may seem complicated as described, automation of the steps involved offers the promise to make this platform accessible to a wide range of users. Contributed by Anton Simeonov.
* Abstract from Proc Natl Acad Sci U S A 2016, [Epub ahead of print]; DOI: 10/1073/pnas.1600812113.
Tuberculosis (TB) remains a major global public health problem, and improved treatments are needed to shorten duration of therapy, decrease disease burden, improve compliance, and combat emergence of drug resistance. Ideally, the most effective regimen would be identified by a systematic and comprehensive combinatorial search of large numbers of TB drugs. However, optimization of regimens by standard methods is challenging, especially as the number of drugs increases, because of the extremely large number of drug–dose combinations requiring testing. Herein, we used an optimization platform, feedback system control (FSC) methodology, to identify improved drug–dose combinations for TB treatment using a fluorescence-based human macrophage cell culture model of TB, in which macrophages are infected with isopropyl β-D-1-thiogalactopyranoside (IPTG)-inducible green fluorescent protein (GFP)-expressing Mycobacterium tuberculosis (Mtb). On the basis of only a single screening test and three iterations, we identified highly efficacious three- and four-drug combinations. To verify the efficacy of these combinations, we further evaluated them using a methodologically independent assay for intramacrophage killing of Mtb; the optimized combinations showed greater efficacy than the current standard TB drug regimen. Surprisingly, all top three- and four-drug optimized regimens included the third-line drug clofazimine, and none included the first-line drugs isoniazid and rifampin, which had insignificant or antagonistic impacts on efficacy. Because top regimens also did not include a fluoroquinolone or aminoglycoside, they are potentially of use for treating many cases of multidrug- and extensively drug-resistant TB. Our study shows the power of an FSC platform to identify promising previously unidentified drug–dose combinations for treatment of TB.