CysLT(1) antagonists are effective for a subset of patients with asthma; however, there has been no good way to predict a given patient's response. We examined the interaction between the clinical response to a cysLT(1) antagonist, pranlukast, and DNA sequence variant A(-444)C in leukotriene C(4) synthase (LTC(4) S) gene in Japanese patients with moderate asthma. The frequency of LTC(4) S C(-444) allele was 21.6% in the Japanese general population (n = 171) and 19.4% in the asthmatic subjects ( n= 349). A 4-week prospective, open trial of pranlukast (225 mg twice daily) was performed in 50 patients with moderate asthma who had been well controlled with inhaled corticosteroid (beclomethasone 400-800 microg/day or fluticasone 200-400 microg/day). The C(-444) allele carriers (n = 16) responded better to pranlukast compared to the A(-444) allele homozygotes ( n= 31) [14.3 5.3% vs. 3.1 2.4% improvement of forced expiratory volume in one second (FEV(1) ), 0.01], while LTC(4) S genotype-stratified response to inhaled beta-agonist salbutamol (200 microg) was not observed (17.5 2.1% vs. 18.7 2.2% improvement of FEV(1) ). Univariate analysis demonstrated that the better response to pranlukast (more than 10% improvement of FEV(1) ) was correlated with LTC(4) S genotype (P < 0.01) and pretreatment airway reversibility to salbutamol (P < 0.01), but not with sex, age, atopic status, urinary leukotriene E(4) excretion rate, or daily dose of inhaled corticosteroid. Furthermore, multivariate regression analysis suggested that LTC(4) S genotype and the bronchodilatory effect of salbutamol were independent variables to predict the clinical response to pranlukast (P < 0.05). We conclude that LTC(4) S genotype is predictive of the clinical response to a cysLT(1) antagonist, pranlukast, in Japanese patients with moderate asthma.