SLOG regimen:

S-1 20-40 mg/m2/b.i.d., day 1-7;Leucovorin 20 mg/m2/b.i.d., day 1-7;Oxaliplatin 85 mg/m2 in 250 mL of D5W,given as 2-hour intra-venous infusion, day 1;Gemcitabine 800 mg/m2 in 250 mL of normal saline, given as fixed dose-rate infusion,day 1;After the administration of gemcitabine, the infusion line should be flushed with 20 ml of normal saline and then 50 ml of 5% glucose solution before the administration of oxaliplatin; Every 14 days, as one cycle.Prophylactic G-CSF or GM-CSF will not be allowed in this study. In case of grade 4 or complication neutropenia, patients may receive G-CSF according to the regulation of National Insurance Bureru treated with appropriate antibiotics. Therapeutic G-CSF may be used at the discretion of attending physicians.

Biweekly S-1, Leucovorin, Oxaliplatin and Gemcitabine (SLOG) in metastatic pancreatic adenocarcinoma

Patients will be entered in cohorts of three. In any dose level of S-1, if none develops DLT as listed after the first two doses of SLOG chemotherapy, another cohort of 3 patients will be accrued to the next dose level. If at any time two or more patients develop DLT at the same dose level, the dose escalation will be terminated, and the prior dosage level will be considered the MTD.

9.2.1 Simon's optimal two-stage design will be used to determine the target patient number for the phase II part of this study. Using the approach, we test a null hypothesis that the true-response probability is less than an uninteresting level (p0) of 25% against the alternative hypothesis that the true response probability is at least as great as a target level (p1) of 40%. Response probabilities less than 25% will be considered inactive while response probabilities greater than 40% will be called highly effective.