Methods: Enrolled in this study were 31 patients with different advanced carcinoma: 9 patients were given GEM (1 h i.v. infusion, 1000 mg/m2) followed after 5 min by VNR (10 min i.v. infusion, 25 mg/m2) (GEM-VNR sequence); 17 patients received VNR followed by GEM (VNR-GEM sequence), at the same doses and with the same infusion period; as a control group (GEM), 5 patients were given only single-agent gemcitabine (1 h i.v. infusion, 1000 mg/m2).Results: GEM serum levels showed higher Cmax and AUCtot in the VNR-GEM protocol than in the GEM-VNR and GEM groups. The GEM pharmacokinetic profile in both schedules showed biphasic elimination, as in monotherapy. VNR concentration/time curves showed rapid plasma clearance and wide interpatient variability in both sequences. VNR Cmax was higher in the VNR-GEM group than in GEM-VNR, while VNR AUCtot and Cltot did not differ significantly in the two sequences.Conclusions: Some pk-values were altered for both GEM and VNR, following the two alternate protocols. A possible rationale for this behaviour is that VNR and GEM may influence each other during liver elimination extraction and metabolism. [ABSTRACT FROM AUTHOR]