Abstract only e21502 Background: Limited data are available on how the setting and timing of palliative care (PC) referral can affect end-of-life (EoL) care and survival in cancer patients (pts). The aim of our study was to evaluate the impact of PC referral (hospice and/or home-care services) versus family physician care (FP-C), in terms of both EoL quality of care and survival in cancer pts. Methods: The study included 337 cancer pts who died between January 2015 and February 2016. We retrieved data on type of EoL care and on poor quality of care indicators in the last 30 days of life. Palliative Care Survival (PCS) was defined as the interval between timing of PC referral and death. The survival after treatment (Treatment-Free Survival, TFS) was defined as the interval between the last administration of anticancer therapy and death. Results: FP-C court included 89 pts and PC services court 248 pts. The 248 PC pts were divided in three groups: 99 assisted by both PC services, 58 only by home-care and 91 only by hospice care. The median PCS was 24 days. TFS was significantly longer for the three groups assisted by PC (105, 95, and 82 days respectively) than for that assisted only by FP-C (49 days, p < 0.0001). Compared to PC services, FP-C was associated with higher frequency of poor quality of care indicators, such as emergency room visits (p < 0.0001), hospitalizations (p < 0.0001), hospital death (p < 0.0001) and chemotherapy administration (p < 0.0001). In addition, earlier PC referral (30-60 and > 60 days before death) versus late referral ( < 30 days before death) was associated with a lower frequency of poor quality indicators. Variables such as age, sex and primary tumor were not associated with a different quality of care. Conclusions: Pts referred to PC services, compared to pts referred only to FP-C, had improved EoL survival and quality of care. A better definition of PC referral timing can affect the quality of EoL care in cancer pts.