In summary, a single randomized double-blind study presented in this application (PA.3) demonstrated a statistically significant increase in overall survival for the combination of 100 mg erlotinib/gemcitabine, only when cox proportional hazards ratio were adjusted for PS and extent of disease. However, the very small increase in median overall survival (12.8 days) and median PFS (11 days) raises the question whether the difference is clinically important. In addition, there is lack of differential response rate , along with a significant increase in toxicity associated with toxic death and SAEs (e.g. > 6 fold increase in strokes) leading to drug discontinuation and worse quality of life, suggesting that the benefit with erlotinib/gemcitabine combination in the treatment of advanced or metastatic pancreatic carcinoma may not outweigh the risks associated with this therapy. Although diarrhea and skin rash are the most frequent EG toxicities in relation to placebo/gemcitabine, there are several serious toxicities that, while low in frequency, are more frequent in the EG arm. These include stroke, cardiac ischemia/infarction, stent occlusion, infections, ILD-like events and GI bleeding. A confirmatory second well-controlled and well-conducted trial may help to discern whether erlotinib adds a clinically significant improvement to gemcitabine with an acceptable toxicity profile in the therapy of locally advanced or metastatic pancreatic adenocarcinoma.