Purpose: Although patients with incurable disease and Eastern Cooperative Oncology Group performance status (ECOG-PS ≥ 2) are underrepresented in clinical trials, they are frequently offered palliative chemotherapy (pCT) in daily clinical practice in order to improve symptoms and quality of life. In this case-control retrospective analysis, our goal was to identify factors associated with poorer survival and lack of benefit of pCT in this population.

Patients and methods: We evaluated 2,514 patients who died between August 2011 and July 2012 in an academic cancer care institution and its hospice. A total of 301 patients with solid tumours and ECOG-PS ≥ 2 at prescription of pCT were selected for this case-control retrospective analysis. Cases were defined as patients who survived less than 90 days after the first cycle of first line pCT, and controls were those who had a longer survival.

Results: 142 cases and 159 controls were included. Cases were more likely to experience grade ≥ 3 toxicity (43% versus 28%; p = 0.005), die of toxicity (16% versus 6%; p < 0.001) and not be offered best supportive care (BSC) only (47% versus 71%; p < 0.001). Median overall survival was 204 among controls and 34 days in cases (hazard ratio = 0.177; 95%, confidence interval = 0.015–0.033, p < 0.001). Logistic regression analysis identified ECOG-PS > 2 (odds ratio (OR) = 2.3, p = 0.044) and serum creatinine (sCr) > 1 mg/dL (OR = 11.2, p < 0.001)
as independent predictors of 90-day mortality.

Conclusions: The independent predictors of short survival (less than 3 months) after initiation of pCT in this population were ECOG-PS > 2 and elevated sCr. Therefore, patient selection is crucial, as pCT may be deleterious in ECOG-PS ≥ 2 pts.