Application of dose-addition analyses to characterize the abuse-related effects of drug mixtures

Abstract

Polysubstance use makes up a majority of drug use, yet relatively few studies investigate the abuse-related effects of drug mixtures. Dose-addition analyses provide a rigorous and quantitative method to determine the nature of the interaction (i.e., supraadditive, additive, or subadditive) between two or more drugs. As briefly reviewed here, studies in rhesus monkeys have applied dose-addition analyses to group level data to characterize the nature of the interaction between the reinforcing effects of stimulants and opioids (e.g., mixtures of cocaine + heroin). Building upon these foundational studies, more recent work has applied dose-addition analyses to better understand the nature of the interaction between caffeine and illicit stimulants such as MDPV and methamphetamine in rats. In addition to utilizing a variety of operant procedures, including drug discrimination, drug self-administration, and drug-primed reinstatement, these studies have incorporated potency and effectiveness ratios as a method for both statistical analysis and visualization of departures from additivity at both the group and individual subject level. As such, dose-addition analyses represent a powerful and underutilized approach to quantify the nature of drug-drug interactions that can be applied to a variety of abuse-related endpoints in order to better understand the behavioral pharmacology of polysubstance use.

Keywords: dose-addition analyses; drug mixtures; self-administration.

Figure 1.

Idealized isobole (panels A-C) and corresponding predicted vs. observed dose-response curves (bottom row). Each column represents a different fixed dose mixture relative to the ED₅₀ of each constituent drug [3:1 drug A: drug B (left column), 1:1 drug A: drug B (center column), and 1:3 drug A: drug B (right column)] and shows different types of potential interactions (additive, supra-additive, and sub-additive interactions with respect to potency and/or effectiveness). The colored, dashed/dotted lines (blue, black, and red in panels A, B, and C) represent all possible ED₅₀s for 3:1, 1:1, and 1:3 mixtures, respectively. The observed ED₅₀s that fall on the line of additivity (panels A-C; grey line) are considered strictly additive interactions (panel B), whereas data points that fall to the left are supra-additive (panel A) or to the right are sub-additive (panel C). Similarly, compared to the predicted additive effect line (grey), the observed dose-response curves (colored lines; panels D-F) that fall to the left (panel D), overlap (panel E), or fall to the right (panel F) are considered supra-additive (panel D), additive (panel E), or sub-additive (panel F) with respect to potency, respectively. Not visualized using a traditional isobole, compared to the predicted additive effect line (grey), the observed dose-response curves (colored lines; panels D-F) that result in maximal effects that are equal to (panel D), greater than (panel E), or less than (panel F) are considered additive (panel D), supra-additive (panel E), or sub-additive (panel F), which respect to effectiveness.

Figure 2.

Dose-response curves for self-administration of binary mixtures of MDPV + caffeine (top row), methylone + caffeine (second row), cocaine + caffeine (third row), or methamphetamine + caffeine (bottom row) under a progressive ratio schedule of reinforcement. Mixtures were tested at three fixed dose ratios relative to the ED₅₀ of each constituent drug [3:1 (left column), 1:1 (center column), and 1:3 (right column)]. Experimentally-determined dose response curves (black lines and filled symbols) represent the mean ± S.E.M. for 12 (MDPV + caffeine and methylone + caffeine), 10 (cocaine + caffeine) or 7 rats (methamphetamine + caffeine). Predicted, additive dose-response curves (grey lines, no symbols) represent the mean ± S.E.M. for the rats that contributed to the experimentally-determined dose-response curves. Abscissa: doses refer to total dose equivalents of the illicit stimulant available for infusion, expressed as mg/kg/infusion on a log scale. Ordinate: percent of maximal effect, normalized to the maximal responding maintained by the illicit stimulant (100%) and saline (0%). MDPV + caffeine and methylone + caffeine mixture dose-response curves are replotted from Gannon, Galindo, et al. (2018).

Figure 3.

Potency (predicted ED₅₀ / observed ED₅₀; left column) and effectiveness (predicted E_max / observed E_max; right column) ratios in rats with behavior maintained by mixtures of MDPV + caffeine (top row), methylone + caffeine (second row), cocaine + caffeine (third row), or methamphetamine + caffeine (bottom row). Group means and 95% confidence intervals are depicted by the red line and error bars. Individual subjects are represented by the same symbol shape and color in the panels displaying potency and effectiveness ratios. Abscissa: potency or effectiveness ratio. Ordinate: the three fixed dose ratios (3:1, 1:1, or 1:3). MDPV + caffeine and methylone + caffeine potency and effectiveness ratios are replotted from Gannon, Galindo, et al. (2018).