Abstract
Clinical outcomes of chemotherapy for patients with advanced pancreatic adenocarcinoma in a real-world setting might differ from outcomes in randomized clinical trials (RCTs). Here we show in a single-institution cohort of 595 patients that median overall survival (OS) of patients who received gemcitabine alone (n = 185; 6.6 months (95% CI; 5.5-7.7)) was the same as in pivotal RCTs. Gemcitabine/capecitabine (n = 60; 10.6 months (95% CI; 7.8-13.3)) and gemcitabine/nab-paclitaxel (n = 66; 9.8 months (95% CI; 7.9-11.8)) resulted in a longer median OS and fluorouracil/oxaliplatin/irinotecan (n = 31, 9.9 months (95% CI; 8.1-11.7)) resulted in a shorter median OS than previously reported. Fluorouracil/oxaliplatin (n = 35, 5.8 months (95% CI; 4.5-7)) and best supportive care (n = 206, 1.8 months (95% CI; 1.5-2.1)) could not be benchmarked against any RCTs. The degree of protocol adherence explained differences between real-world outcomes and the respective RCTs, while exposure to second-line treatments did not.