TORONTO, Canada-In a phase III international study of patients with advanced pancreatic cancer, addition of erlotinib (Tarceva) to gemcitabine(Drug information on gemcitabine) (Gemzar) increased 1- year survival by 40%, marking the trial as "the first to show a survival benefit for a targeted therapy in advanced pancreatic cancer", reported lead investigator Malcolm Moore, MD, director of the New Drug Development Program at Princess Margaret Hospital in Toronto (abstract 1). Study Protocol The double-blind placebo-controlled phase III study was conducted by the National Cancer Institute of Canada Clinical Trials Group at Queens University, with investigators from OSI Pharmaceuticals. A total of 569 patients with locally advanced or metastatic pancreatic cancer were enrolled. Treatment groups were well balanced in terms of performance status, locally advanced vs metastatic dis-ease, and presence of pain. Most of the patients had measurable disease. Patients were randomized to receive gemcitabine at 1,000 mg/m2 IV weekly for 7 of 8 weeks and then 3 of 4 weeks plus daily oral placebo (n = 284) or erlotinib (n = 285). Erlotinib was given at a dose of 100 mg during most of the trial, however an additional, smaller cohort of patients receiving 150 mg was later added to the study. The analyses reported are based on the entire study group, Dr. Moore said. Survival Improvement The study met its primary endpoint of overall survival. "The hazard ratio of survival is 0.81, which translates into a 24% improvement in survival from the use of erlotinib with gemcitabine," Dr. Moore said. Median survival time for patients in the erlotinib arm was 6.4 months, vs 5.9 months in the placebo arm. "The 1- year survival was improved from 17% to 24%, which is an approximate improvement of 40%," he told ONI. Median progression-free survival time for patients treated with erlotinib was 3.75 months, vs 3.55 months for patients randomized to placebo. Tumor response occurred in 57.5% of patients receiving erlotinib, vs 49.2% of patients receiving placebo. Rash Linked to Better Response In all patient subsets, Dr. Moore said, the hazard ratio was always less than 1, indicating a positive effect in favor of erlotinib. This effect of erlotinib appeared to be pronounced in some patient subsets, he added, particularly patients at performance status 2 and those who had metastatic disease. Nevertheless, he emphasized, because these are subset analyses, "they are really just hypotheses-generating." Expected adverse events-rash, diarrhea, infection, and stomatitis- were generally tolerable and manageable, he reported. Similar to other studies with erlotinib, patients whodeveloped a rash responded better to treatment. Grade 3 and 4 toxicities "were relatively low in both arms," Dr. Moore said, and there were "no differences in hematologic, in renal, or in hepatic toxicities between the arms." Quality-of-life assessments, he added, showed "no detrimental effects on quality of life from the addition of erlotinib to gemcitabine." Consider Therapeutic Index "The small improvement in survival afforded to erlotinib-treated patients was accompanied by increased toxicity, challenging us to think caretinibfully about the therapeutic index, when an unselected population of patients with pancreatic cancer is exposed to the combination of gemcitabine and erlotinib," commented discussant James L. Abbruzzese, MD, of The University of Texas M.D. Anderson Cancer Center, Houston. An improved understanding of the relationship between the skin rash induced by erlo-tinib and improved patient outcome, he said, "will be a critical first step" in identifying which patients will be likely to benefit from EGFR antagonists. He also advocated "a detailed pharmacoeconomic analysis to include the costs of drug treatment as well as the costs of managing incremental toxicity encountered with this combination." Refine Treatment Population "Based on these considerations, at this time, I do not feel that the results clearly alter the standard of care for patients treated with advanced pancreatic cancer, and continued refinement of the patient population most likely to benefit from this combination will be needed rather than to generally adopt this regimen for all patients," Dr. Abbruzzese said. "In the end," he added, "I think this trial's greatest contribution may be to focus us, the oncology community, on increasing our efforts to therapeutically exploit the molecular biology of complex malignancies like pancreatic cancer, to look for opportunities to intervene earlier in the process of pancreatic carcinogenesis, and finally to continue the hard work of identifying which patients are likely to benefit from targeted therapy."