In order to determine the dose regimen of new antihypertensive compounds, between-patient trials are usually performed. However, the combined use of a crossover design and a precise methodology to measure blood pressure (BP) and biological effects can provide relevant data with a minimal number of patients, if there is no carryover effect which invalidates the experiment. Such goals were successfully achieved with just 25 hypertensive patients who were randomly allocated in double-blind fashion every 2 weeks to a new angiotensin converting enzyme (ACE) inhibitor, benazepril [10 mg once a day (o.d.), 20 mg o.d., 10 mg twice a day (b.i.d.) and 20 mg b.i.d.], or a placebo. The mean BP fall [systolic (SBP)/diastolic (DBP), measured in mmHg] just before drug intake was significantly greater with benazepril: -14/-9 (10 mg o.d.); -15/-8.5 (20 mg o.d.); -22.5/-14 (10 mg b.i.d.), and -21/-13 (20 mg b.i.d.) in comparison with placebo (-3/-3). Mean active plasma renin (measured in pg/ml), assessed by an immunoradiometric assay based on two monoclonal antibodies, increased significantly in a dose-dependent manner, by +0.7 (placebo), +15.0 (10 mg o.d.), +23.4 (20 mg o.d.), +44.4 (10 mg b.i.d.) and +78.8 (20 mg b.i.d.), whereas plasma ACE decreased (by 67 and 78% after 10 and 20 mg o.d., respectively, and by 91-92% after 10 and 20 mg b.i.d.). In the clinical development of an antihypertensive drug, the earlier use of such within-patient studies, with the random insertion of one placebo period between the active periods, should help in the dose-response curve search.(ABSTRACT TRUNCATED AT 250 WORDS)