Pharmacy Distribution

To objectively identify compounds whose response distributions show exceptional positive response, that is, a relatively few highly responsive samples at the right tail of the drug response distribution, we calculated the sample skewness γ of the drugs' empirical response distribution over all the samples under analysis. The one-sided significance p-value of the observed positive skewness was assessed using the D'Agostino15 test in the R-package “moments” (version 0.13, http://cran.r-project.org/package=moments). This enables systematic detection of drug-sensitive patient sub-groups for a given compound, without visually going through all the drug response distributions. When comparing two sets of samples, such as highly responsive patient samples against the remaining samples for those compounds initially identified with positive skewness, we assessed the difference in the response levels between the two pre-defined sample groups with the Wilcoxon rank-sum test. We chose to use the non-parametric test because the response distributions cannot be assumed to be normally distributed.
The predictive accuracy of the DSS, IC₅₀ and AA metrics was assessed in terms of their capability to distinguish the active dose-response curves from the inactive ones using the receiver operating characteristic (ROC) analyses; ROC curves evaluate the relative trade-off between true positive rate (sensitivity) and false positive rate (1 – specificity) of the metric when ordering the dose-response curves according to the increasing value of the response metric16 (link). The overall accuracy of each response metric was summarized using the area under the ROC curve (AUROC) measure; for an ideal metric, AUROC = 1, whereas a random metric obtains an AUROC = 0.5 on average. Statistical significance of an observed AUROC, when compared to random classifier, was assessed using the roc.area function in the R-package “verification”. Statistical significance of an observed AUROC difference between two response metrics was assessed using the “pROC” package with the De Long's test17 (link).

PK properties such as absorption, distribution, metabolism, excretion and toxicity (ADMET) profiling of compounds were determined using the pkCSM ADMET descriptors algorithm protocol¹ and the Discover Studio 4.0 (DS4.0) software package (Accelrys Software, Inc., San Diego, CA, United States). Two important chemical descriptors correlate well with PK properties, the2D polar surface area (PSA_2D, a primary determinant of fractional absorption) and the lipophilicity levels in the form of atom-based LogP (AlogP98). The absorption of drugs depends on factors including membrane permeability [indicated by colon cancer cell line (Caco-2)], intestinal absorption, skin permeability levels, P-glycoprotein substrate or inhibitor. The distribution of drugs depends on factors that include the blood–brain barrier (logBB), CNS permeability, and the volume of distribution (VDss). Metabolism is predicted based on the CYP models for substrate or inhibition (CYP2D6, CYP3A4, CYP1A2, CYP2C19, CYP2C9, CYP2D6, and CYP3A4). Excretion is predicted based on the total clearance model and renal OCT2 substrate. The toxicity of drugs is predicted based on AMES toxicity, hERG inhibition, hepatotoxicity, and skin sensitization. These parameters were calculated and checked for compliance with their standard ranges.
The prediction of genotoxicity used the OECD QSAR toolbox 4.1 software package (Organization for Economic Co-operation and Development, Paris, France) and Toxtree, Version 2.6.13 (Ideaconsult, Ltd., Sofia, Bulgaria). Both software are open source freely available in silico programs that identify the chemical structural alerts (SA).

The clinician, or veterinarian, choosing an antimicrobial agent to treat humans or animals with a bacterial infection requires information that the antimicrobial selected is effective against the bacterial pathogen. Such information will be used, together with clinical details such as the site of infection, ability of the antimicrobial to reach the site of infection, formulations available and dosage regimes, when determining an appropriate therapeutic course of action. The in vitro susceptibility of the bacterial pathogen can be determined and CBPs used to ascertain whether the organism is likely to respond to treatment. CBPs will take into account the distribution of the drug in the tissues of the body following administration and assume that a clinical response will be obtained if the drug is given as recommended and there are no other adverse factors which affect the outcome. Conversely, if the CBP indicates resistance, then it is likely that treatment will be unsuccessful. Frequency of dosing is one factor that can affect the antimicrobial concentration achieved at the site of infection. Therefore, different dosing regimens can lead to the development of different CBPs, as occurs in some countries for certain antimicrobials where different therapeutic regimes are in place. Although the rationale for the selection of different CBPs may be clear, their use makes the interpretation of results from different countries in reports of this type problematic, as the results are not directly comparable between those different countries.