Combined effects and variability

Calling a combination additive, potentiating or synergistic should be a measurement, not an impression. The classical method is the isobologram. Each drug is tested alone and a dose–effect curve is drawn; then a fixed combination is tested, and the pair of doses that produces a chosen effect level is plotted as a point on a graph whose axes are the doses of the two drugs. If the point lies on the straight line joining the two individual effective doses, the combination is additive. If it lies below and to the left of that line, a smaller total dose than expected was needed, which is synergism; if it lies above and to the right, more drug was needed than the parts predicted, which is antagonism or sub-additivity. A related calculation uses the combination index, which expresses how far the experimental point deviates from the additive line, and is widely used in anticancer drug development to find pairs worth taking forward. Two cautions belong with the method. First, the answer depends on the effect level chosen: a pair can be synergistic at low effect levels and additive at high ones, because the curves have different shapes. Second, synergy for the desired effect does not imply safety; the same pair may be synergistic for toxicity, and a combination that is more potent must be examined for its margin as well as its power. In the clinic the tool is rarely a formal isobologram; it is the same logic applied plainly, by asking whether the second drug added something the first could not achieve, or merely a second version of the same effect.

It helps to separate variability into two kinds, because they call for different responses. Pharmacokinetic variability changes the concentration of drug that reaches the receptors: differences in absorption, in the volume of distribution, in metabolism and in excretion. A patient with renal failure accumulates a renally cleared drug; a patient with a larger fat mass stores more of a lipophilic drug; a fast or slow metaboliser, determined by inherited enzyme variants, clears the same dose at very different rates. Here the response can often be predicted from a measurement — a plasma concentration, a creatinine value, a body weight — which is why therapeutic drug monitoring exists for drugs with a narrow index and a clear relationship between concentration and effect. Pharmacodynamic variability changes how strongly the tissue answers to the same concentration: differences in receptor number or sensitivity, in the efficiency of the signal that follows receptor activation, in age-related change and in the presence of disease. This kind is harder to see because there may be no concentration to measure, which is why clinicians monitor the effect itself — blood pressure, blood glucose, pain score — instead. Two further sources straddle the two categories. Tolerance develops with repeated exposure and shifts the curve to the right, so the same concentration produces less effect. Adherence changes the average concentration actually achieved, sometimes in a way that looks like resistance to treatment. Recognising which kind of variability is at work matters because it decides the corrective action: if the concentration is wrong, adjust the dose; if the tissue response is wrong, the drug or the target must change.