To investigate whether a pharmacokinetic drug-drug interaction exists between theophylline (THEO), a CYP1A2 substrate with a narrow therapeutic index, and the concomitant substrate roflumilast (ROF), a novel selective PDE4 inhibitor partially metabolized by CYP1A2. In an open-label, 2-period, crossover study, Treatment A (oral ROF 500 μg q.d. on Days 6 - 10 in addition to oral THEO 375 mg b.i.d. on Days 1 - 10) and treatment B (oral ROF 500 μg q.d. on Days 1 - 5) were administered consecutively in random order to each of 24 healthy adult subjects. Both periods were separated by a wash-out phase of at least 10 days. Plasma samples for pharmacokinetic evaluation (AUC, Cmax, t1/2, tmax) including percent peak-trough fluctuation (%PTF) of THEO were taken. Point estimates and the 90% confidence interval of the geometric mean ratios were calculated for AUC and Cmax and descriptive statistics for other pharmacokinetic parameters. Concomitant administration of ROF did not alter pharmacokinetics of THEO. With coadministered THEO, only steady-state total exposure to ROF (AUC) was increased by 28% whereas other pharmacokinetic parameters (t1/2, Cmax, tmax) of ROF and of the active metabolite roflumilast-N-oxide (R-NO), its main contributor to the pharmacodynamic effects, remained unchanged. Neither ROF nor its main metabolite had any impact on the metabolism of the concomitant CYP1A2 substrate THEO in humans. Though co-administration of THEO resulted in a minor increase (28%) in total ROF exposure, no safety or tolerability concerns and no altered total PDE4 inhibition of both ROF and R-NO, were observed.