Phase II trial of erlotinib in advanced pancreatic cancer (PC)
4609 Background: The epidermal growth factor receptor (EGFR) is a potentially important target in PC. Benefit from erlotinib (Tarceva), an oral EGFR tyrosine kinase inhibitor has been associated with the presence of a skin rash. The purpose of this study was to determine the efficacy of erlotinib, dosed to achieve a rash, in patients (pts) with PC. Methods: Erlotinib was given at an initial dose of 150 mg/day to eligible pts with locally advanced (LA) or metastatic PC who had progressed or were unable to tolerate gemcitabine-based chemotherapy. The dose of erlotinib was increased by 50mg every 2 weeks (maximum 300 mg/day) until > grade 1 rash (CTCAE v 3.0) or other dose-limiting toxicities occurred. Erlotinib pharmacokinetic (PK) studies were performed. Baseline tumor tissue was collected for analysis of Kras mutations, EGFR by IHC and FISH. The primary endpoint of this two- stage phase II trial was prolonged disease control (PR + SD > 8 wks) with a rate of >20% assumed to be significant . Results: Fifty pts were accrued (median age 61, M:F = 25:25, ECOG 0:1:2 = 5:41:4, LA:Metastatic = 5:45, prior gemcitabine none:adjuvant:palliative = 2:16:35). 47 and 40 pts were evaluable for toxicity and response, respectively. Dose-escalation to 200–300 mg of erlotinib was possible in 9 pts. Most common treatment adverse events (TAEs) of any grade were: rash (35 pts, 74.5%), diarrhea (18 pts, 38.3 %), and fatigue (8 pts, 17%). Grade 3+ TAEs were rash in 2 pts and diarrhea in 2 pts. Best response was SD in 14 pts, 0.35 (95% CI: 0.2–0.5). Prolonged disease control (SD > 8 wks) was observed in 10/40 evaluable pts, 0.25 (95% CI: 0.12–0.38), which met the primary study endpoint. Median TTP was 1.6 mo (95% CI:1.6–2.1), mOS 4.1 mo (95% CI:3.2–7.3), and 6 mo OS rate was 39% (95%CI: 24–61%). PK and correlative data are being analyzed and will be presented. Conclusions: Erlotinib is associated with prolonged stable disease in a subset of pts with advanced refractory PC. Dose escalation in the absence of toxicity is feasible and safe. [Table: see text]