581 Background: In contrast to trials investigating treatment efficacy, patients perspective on given chemotherapy is less studied. Known surveys indicate that patients accept therapy for very limited benefit. Methods: Patients who had received 3 or more months of systemic, palliative therapy for colorectal (CRC) or non-colorectal (upper GI, NSCLC, SCCHN) cancer were eligible. They completed a questionnaire including their retrospective report on symptomatic toxicity (according to NCI-CTC) and using numeric ranking scales their burden from the toxicity and the comparison to the informed consent. Depression and anxiety was measured by HADS. Finally, the patients were asked for the minimal gain in life expectancy they would feel it worth to repeat the therapy for. Results: Between August 2008 and December 2009, 134 patients were enrolled. Median age was 63 (32-86); 71% of patients were male. Colorectal cancer patients would repeat their therapy for a minimal gain of life expectancy of 36.0 mo (14.0-57.0 mo), non-CRC patients for 15.9 [12.3-19.5] months (p = 0.01). Considering additional time with best supportive care, the minimal expectation from treatment trended to be longer than the actual survival of the same patients (39.3 [33.6-45.0] and 19.9 [16.3- 23.5] months for CRC and non-CRC patients, respectively), p = 0.08. Depression, diarrhea and tumor entity (CRC vs. non-CRC) were an independent factors for the minimal expected life expectancy in a multivariate analysis. Most frequent self reported toxicity (grade 3/4) was fatigue (9.0%), nausea (8.1%), diarrhea (8.5%) and stomatitis (8.3%), severe burden was felt from fatigue (14.3%) followed by neuropathy (12.0%), diarrhea (11.1%), nausea (9.8%) and stomatitis (8.8%). Fatigue was felt ′a little bit more′ or ′much more′ than in the patient information in 29.9%, followed by neuropathy (20.5%), acne (18.4%) and diarrhea (17.6%). Conclusions: Patients' expectations of life-prolonging systemic therapy are higher than expected from previous studies or the median survival in phase III studies. Compared to their actual survival it may be regarded as realistic. Fatigue might deserve more attention when toxicity of treatment and symptoms of disease are explained. [Table: see text]