Targeted agent use in cancer patients at the end of life

Context: The use of targeted therapy at the end of life has not been well characterized. Objectives: To determine the frequency and predictors of targeted therapy use in the last days of life. Methods: All adult patients residing in the Houston area who died of advanced cancer between September 1, 2009 and February 28, 2010 and had contact with our institution within the last three months of life were included. We collected baseline demographics and data on chemotherapy and targeted agents. Results: Eight hundred sixteen patients were included: average age 62 years (range 21-97), female 48% and white 61%. The median interval between the last treatment and death was 47 (interquartile range [IQR] 21-97) days for targeted agents and 57 (IQR 26-118) days for chemotherapeutic agents. Within the last 30 days of life, 116 (14%) patients received targeted agents and 147 (18%) received chemotherapy. Regimens given in the last 30 days of life included a median of one (IQR 1-2) chemotherapeutic or targeted agent and 43 (5%) patients receiving targeted agents had concurrent chemotherapy. The most common targeted agents in the last 30 days of life were erlotinib (n = 25), bevacizumab (n = 20), rituximab (n = 11), gemtuzumab (n = 8), and temsirolimus (n = 8). On multivariate analysis, younger age (odds ratio [OR] 0.98 per year, P = 0.01), hematologic malignancy (OR = 6.1, P < 0.001), and lung malignancy (OR = 2.6, P = 0.05) were associated with increased targeted agent use in the last 30 days of life. Conclusion: Targeted agents were used as often as chemotherapy at the end of life, particularly among younger patients and those with hematologic malignancies. Guidelines on targeted therapy use at the end of life are needed.