Summary We have assessed the pharmacokinetics, pharmacological and anti-tumour effects of the specific steroidal anti-oestrogen ICI 182780 in 19 patients with advanced breast cancer resistant to tamoxifen. The agent was administered as a monthly depot intramuscular injection. Peak levels of ICI 182780 occurred a median of 8-9 days after dosing and then declined but were above the projected therapeutic threshold at day 28. Cmax during the first month was 10.5 ng/ml-' and during the sixth month was 12.6 ng ml-l. The AUCs were 140.5 and 206.8 ng day ml-' on the first and sixth month of dosing respectively, suggesting some drug accumulation. Luteinising hormone (LH) and follicle-stimulating hormone (FSH) levels rose after withdrawal of tamoxifen and then plateaued, suggesting no effect of ICI 182780 on the pituitary-hypothalamic axis. There were no significant changes in serum levels of prolactin, sex hormone-binding globulin (SHBG) or lipids. Sideeffects were infrequent. Hot-flushes and sweats were not induced and there was no apparent effect of treatment upon the endometrium or vagina. Thirteen (69%) patients responded (seven had partial responses and six showed 'no change' responses) to ICI 182780, after progression on tamoxifen, for a median duration of 25 months. Thus ICI 182780, given by monthly depot injection, and at the drug levels described, is an active second-line anti-oestrogen without apparent negative effects on the liver, brain or genital tract and warrants further evaluation in patients with advanced breast cancer.

1. The effect of variability of gastric emptying and oro‐caecal transit on the absorption of a multicomponent solution of frusemide, atenolol, hydrochlorthiazide and salicylic acid has been studied in six healthy subjects. Each subject was studied on five separate occasions: three times under basal conditions, once following metoclopramide and once following codeine pretreatment in an attempt to speed and slow transit respectively. 2. Inter‐subject variability of gastric emptying, oro‐ caecal transit and the rate and extent of drug absorption was considerable. 3. The absorption of salicylic acid appeared rate‐limited by gastric emptying but the rate and extent of frusemide, atenolol and hydrochlorthiazide absorption were unrelated to measures of gastric emptying or oro‐caecal transit. 4. Codeine phosphate caused a two‐fold delay in oro‐caecal transit but did not influence gastric emptying while metoclopramide had no significant effect on either function. 5. Metoclopramide and codeine had no significant effect on the rate or extent of absorption of any of the study drugs. 6. Within the limits of this experiment, oro‐caecal transit time did not appear to be an important determinant of frusemide, atenolol, hydrochlorothiazide or salicylic acid absorption. Other factors must account for the observed variability in drug absorption.

1 We have studied the effects of a non-absorbable osmotic load on the absorption of a multicomponent solution of frusemide, atenolol, hydrochlorothiazide and salicylic acid in six healthy volunteers. 2 Each subject was studied on up to four separate occasions. The drugs were administered in one of four solutions: a) a mannitol/electrolyte solution, b) a double-strength mannitol/electrolyte solution, c) a glucose/electrolyte solution and d) water. Lactulose or sulphasalazine were added as oro-caecal transit markers. Lactulose was included in the mannitol-and glucose-based solutions, adding a further non-absorbable osmotic load, and sulphasalazine was added to the water, adding little osmotic load. 3 The absorption of atenolol and hydrochlorothiazide was two-to three-times less from all lactulose-containing solutions than from the sulphasalazine-containing solution.The absorption of frusemide and salicylic acid was similar from all four solutions. 4 The largest non-absorbable osmotic load impaired the absorption of atenolol and hydrochlorothiazide most and the incorporation of glucose only partly restored absorption. 5 These results suggest that transmucosal water movement is an important determinant of atenolol and hydrochlorothiazide absorption but is less relevant for the absorption of frusemide and salicylic acid. Furthermore, these data demonstrate a previously unrecognised interaction between a commonly prescribed laxative -lactulose, and atenolol and hydrochlorothiazide.

Summary This open-label, non-randomized, parallel-group trial investigated the pharmacokinetics of raltitrexed (Tomudex, formerly ZD1 694) after a single intravenous dose of 3.0 mg m-2, comparing eight cancer patients with mild to moderate renal impairment (creatinine clearance 25-65 ml min-') with eight cancer patients with normal renal function (creatinine clearance >65 ml min-'). The primary end points were area under the plasma raltitrexed concentration-vime curve from the start of the infusion to the last determined concentration (AUCo_) and AUC to infinity (AUCo); secondary end points were peak corncentrations of ralftitrexed (C,n,,) and elimination half-life (t,>). The groups were compared statistically using analysis of covariance. The AUCs were greater for patients with renal impairment than for patients with normal renal function (2452.2 compared with 1247.3 ng h ml-' for AUCo._. (ratio 1.97; 95% Cl 1.36-2.84); 2961.5 compared with 1457.0 ng h ml-' for AUCo_ (ratio 2.03; 1.25-3.29). These differences were statistically significant (P = 0.002 and P = 0.008 for AUC,, and AUCo_ respectively. 50-lOO h (Clarke et al. 1996). This prolonged t1,, may represent hydrolysis of the polvalutamated

1 The effects of various drugs on the pharmacokinetics of tolbutamide have been examined in the rat.2 Phenobarbitone pretreatment caused a significant decrease in half life and area under the curve (AUC) and a significant increase in clearance and volume of distribution (Vd). 3 Acute administration of primaquine significantly increased half life and AUC and decreased clearance. In contrast, the related animoquinolone chloroquine, was without effect. 4 Acute administration of cimetidine produced similar changes to primaquine but of lesser magnitude. 5 Formation of the major metabolite hydroxytolbutamide, was markedly enhanced by phenobarbitone and reduced by primaquine and cimetidine. 6 We conclude that due to its single pathway of metabolism, tolbutamide is a good substrate to use when examining pharmacokinetic interactions involving hepatic enzyme induction and inhibition.