Purpose

Erlotinib marginally improves survival when administered continuously with gemcitabine to patients with advanced pancreatic cancer; however, preclinical data suggest that there is antagonism between chemotherapy and epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors when these are delivered concurrently. We tested a pharmacodynamic separation approach for erlotinib plus gemcitabine and interrogated EGFR signaling intermediates as potential surrogates for the efficacy of this strategy.

Methods

Patients with measurable, previously untreated locally advanced unresectable or metastatic pancreatic cancer were treated with gemcitabine 1000 mg/m(2) as an intravenous infusion over 30-min on days 1, 8, 15 and erlotinib 150 mg/day on days 2-5, 9-12, 16-26 of each 28-day cycle. The primary endpoint was progression-free survival (PFS); secondary endpoints included RECIST objective response rate (ORR) and safety. The study was terminated after thirty patients due to funding considerations.

Results

The median PFS was 2.07 months (95% CI; 1.87-5.50 months) and the ORR was 11%. No unexpected safety signals were seen: the most common grade 3 or higher adverse events were neutropenia (23%), lymphopenia (23%), and fatigue (13%). Patients with mutant plasma Kirsten rat sarcoma virus (KRAS) had significantly lower median PFS (1.8 vs. 4.6 months, p = 0.014) and overall survival (3.0 vs. 10.5 months, p = 0.003) than those without detected plasma KRAS mutations.

Conclusions

Although pharmacodynamically separated erlotinib and gemcitabine were feasible and tolerable in patients with advanced pancreatic cancer, no signal for increased efficacy was seen in this molecularly unselected cohort. Detection of a KRAS mutation in circulating cell-free DNA was a strong predictor of survival.