Drug combinations, offering increased therapeutic efficacy and reduced toxicity, play an important role in treating multiple complex diseases. Yet, our ability to identify and validate effective combinations is limited by a combinatorial explosion, driven by both the large number of drug pairs as well as dosage combinations. Here we propose a network-based methodology to identify clinically efficacious drug combinations for specific diseases. By quantifying the network-based relationship between drug targets and disease proteins in the human protein-protein interactome, we show the existence of six distinct classes of drug-drug-disease combinations. Relying on approved drug combinations for hypertension and cancer, we find that only one of the six classes correlates with therapeutic effects: if the targets of the drugs both hit disease module, but target separate neighborhoods. This finding allows us to identify and validate antihypertensive combinations, offering a generic, powerful network methodology to identify efficacious combination therapies in drug development.

/pdf/network-based-prediction-of-drug-combinations-34ao5xywzf.pdf

Network-based prediction of drug combinations.

All beta-adrenoreceptor blocking drugs seem to be fairly rapidly and completely absorbed from the gastro-intestinal tract. The rate of absorption, however, appears to be lower in elderly patients and possibly also in patients with renal failure than in younger patients. The extent of bioavailability varies considerably between different beta-blockers. Some of these drugs(e.g. alprenolol and propranolol) have a low extent of bioavailability due to a high first-pass elimination effect, while pindolol and practolol for example are in influenced very little by this effect. However, as some beta-blockers from active metabolites, the bioavailability calculated as the ratio between the area under the plasma concentration time curve of unchanged drug after oral and intravenous administration does not give an accurate estimation of the fraction of the biologically active dose reaching the systemic circulation. The beta-blockers so far studied are rapidly distributed in the body. The t1/2 of distribution ranges between 5 to 30 minutes. The apparent volume of distribution varies 3- to 4-fold between the compounds but in all cases the apparent volume of distribution exceeds the physiological body space. In patients with impaired liver function an increase of the volume of distrubution of propranolol has been found. The beta-blockers are relatively rapidly eliminated from the body and most of them have an elimination half-life between 2 to 4 hours. For atenolol, practolol and sotalol higher values have been reported. The most lipophilic beta-blockers are almost completely metabolised in the liver, wheras those of lower lipophilicity are mainly excreted via the kidneys. Impraired liver and kidney function have been found to significantly influence the rate of elimination of those beta-blockers eliminated via the insufficient organ of elimination. Numerous investigators have shown that the beta-blocking effect is linearly related to the logarithm of the plasma concentration. In spite of this relationship, it is difficult from mean data to predict the individual plasma concentration which is necessary for a certain degree of beta-blockade. This might be due to variations in the quantitative formation of active metablolites, individual differences in the plasma protein binding and rather flat plasma level-response curves. Also with respect to the therapeutic effect, the plasma levels vary considerably between individuals. This limits the value of determination of plasma concentrations in order to adjust the therapeutic dose. Our recommendation is that these facilities should be utilised in selected patient groups, eg. those who have a poor therapeutic response to a beta-blocker although the dose is high, and those patient with impaired renal or liver function. The duration of beta-blockade is dose-dependent since the pharmacological effect declines with a constant rate (zero-order kinetics) within relatively wide dosage intervals...

Clinical pharmacokinetics of beta-adrenoreceptor blocking drugs.

The significance of studying the kinetics of drug partitioning into red blood cells (RBCsb) in animals and humans is not fully appreciated, although the importance of routine determination of rate and extent of partitioning of investigational drugs has been stressed ([Lee et al., 1981b][1]; [

Red Blood Cells: A Neglected Compartment in Pharmacokinetics and Pharmacodynamics

14C-hydrochlorothiazide (hct) was administered orally (n=4) and iv (n = 2 to healthy subjects. The gastrointestinal absorption ranged between 60% and 80%, most of it took place in the duodenum and the upper jejunum. The radioactivity was eliminated mainly in the urine, while no sigificant biliary excretion was observed. Chromatographic analysis of the urinary radioactivity demonstrated that greater than 95% of the absorbed or injected 14C-hct was excreted unchanged. The radioactivity in plasma during the first 10 hr after oral administration declined with a fast phase but the levels of label thereafter suggested a slow phase. The existence of such a phase was verified in 1 subject given 75 mg hct orally. His plasma levels of hct (determined with gas-liquid chromatography) declined according to a 2-compartment model, the half-lives of the alpha-and beta-phases being 1.7 and 13.1 hr, respectively. Hct accumulated in the blood cells and the ratio between the radioactivity in cells and that in plasma averaged 3.5. The fate of a single dose of 14C-hct in 2 hypertensive patients treated with the drug chronically was similar to that in the healthy subjects. A third patient, who had slightly elevated serum creatinine, eliminated hct more slowly than the others. Like the healthy subjects, the patients eliminated hct to greater than 95% in unchanged form.

Absorption, metabolism, and excretion of hydrochlorothiazide.

Hydrochlorothiazide (hct) was administered orally in four different doses (12.5, 25, 50 and 75 mg), to eight healthy volunteers. Plasma and urine concentrations of hct were determined by GLC. Maximal plasma levels were found at 1.5–5 h, and averaged 70, 142, 260 and 376 ng × ml−1 respectively. The peak plasma levels and AUC0→9h of hct were highly correlated (p<0.001) with the dose. The decline in the plasma curve was biphasic in those experiments in which the plasma levels of hct could be determined for at least 24 h. The half life of the slower phase lay between 5.6 and 14.8 h. The urinary recovery of hct, which represented the gastrointestinal absorption, averaged 65 to 72 per cent of the dose. The mean renal plasma clearance did not vary with the dose and averaged 319 to 345 ml × min−1. The diuresis during the 10 h after hct 12.5 mg exceeded that after placebo by a mean of 800 ml. The diureses was not increased further after higher doses of hct. The maximal natriuretic effect (+ 100 mmol), too, was found after the 12.5 mg dose. The excretion of potassium, however, rose with increasing doses; the maximal increment, after 75 mg hct, averaged 25 mmol. The excretion of calcium was significantly increased after 50 mg hct (+ 0.6 mmol). The maximal effect on magnesium excretion occurred after 25 mg hct (+ 0.5 mmol). In healthy volunteers there was no correlation between peak plasma level of hct or AUC0→9h and the renal excretion of water and electrolytes.

Pharmacokinetics of hydrochlorothiazide in man.

When 14C-chlorthalidone was administered orally to 2 healthy volunteers, the total recovery of radioactivity in urine (about 75 percent) and feces was close to 100 percent. Most of the label recovered in the blood was bound to the blood cells. When the procedure was repeated while the 2 subjects were receiving acetazolamide, the excretion of labeled material in urine and feces was essentially unchanged, but the blood cells contained less and the plasma more of the blood radioactivity. The half-life of the radioactivity in plasma and blood cells had decreased by about 65 percent. Intravenous administration of acetazolamide (single dose) to 2 other subjects who had received 14C-chlorthalidone orally resulted in a marked drop in the blood cell radioactivity, whereas that in plasma increased. The affinity of chlorthalidone for red blood cells was further evidenced on incubation of 14C-chlorthalidone with human blood. Of the incubated radioactivity, 94 percent to 99 percent was recovered in the erythrocytes. Preincubation of the blood samples with acetazolamide prior to the addition of 14C-chlorthalidone, as well as incubation of acetozolamide in blood samples previously incubated with 14C-chlorthalidone, demonstrated that acetazolamide is able to inhibit and to displace chlorthalidone from blood cells. There are several lines of evidence indicating that chlorthalidone is transported attached to the erythrocyte carbonic anhydrase.

Binding-site interaction of chlorthalidone and acetazolamide, two drugs transported by red blood cells.

Propantheline, 60 mg., labeled with 14C in its xanthene carboxylic moiety, was given orally or intrajejunally to 5 healthy subjects. The absorption of label did not reflect the uptake of the intact drug. Samples of intestinal aspirates obtained a few minutes after administration of the test solution contained label with chromatographic properties the same as those of the free xanthene carboxylic acid. Following incubations in duodenal bile, the 14C-propantheline showed marked modification. According to the autoradiogram of the thin-layer chromatogram, most of the label was localized in two spots. The ma;or one had an Rf value identical to that of xanthene carboxylic acid, while the other one was unidentified. The modification of propantheline in the incubates was more pronounced in alkaline solutions but decreased markedly when the drug was incubated with samples of duodenal aspirates that had been brought to boil and then cooled. The absorption of intact drug after oral administration was estimated to be less than 50 per cent.

Anders Rosén

Papers

Absorption, metabolism, and excretion of hydrochlorothiazide.

Binding-site interaction of chlorthalidone and acetazolamide, two drugs transported by red blood cells.

On the metabolism of propantheline in man.