Background: Sleep disturbance (SD) is one of the most distressing symptoms in patients with advanced cancer. There is limited data on the effectiveness of treatment and predictors of response. Methods: We reviewed 442 consecutive patients with advanced cancer presenting in the Supportive care clinic for SD. All patients underwent screening for SD (‘+’ defined as ≥3/10) and interdisciplinary assessment and treatment including drug review, counseling, sleep hygiene, and drug therapy (PC). Response was defined as ≥30% improvement from baseline to follow up visit. Baseline patient characteristics, cancer related symptoms (Edmonton symptom assessment scale), delirium (Memorial delirium assessment scale), use of opioids, steroids and anxiolytics/hypnotics, and CAGE score, were analyzed to determine their association with response to PC. Results: The median age was 58 years; males were 53%. The most common cancer types were H&N and lung and GI. (55%). 330/ 442 pts had SD (75 %). Median, mean (SDev) for SD (0=best sleep, 10- worst sleep imaginable) were 5 and 5(3). Sedative use (r-0.16, p=0.002), pain (r - 0.31), fatigue (r-0.31, <0.001), Depression(r-0.25, p<0.001), anxiety (r-0.28, p<0.001), drowsiness (r-0.30, p<0.001) and feeling of wellbeing (r-0.33, p=0.02) are associated with SD at initial consult. The median time between visits was 15 days. SD response was seen in 143/330 (43%) [95% CI 38% to 49%]. Use of medications such as anxiolytics/hypnotics, steroids, opioid dose, pain, depression, anxiety, fatigue and delirium were not predictive of SD response. Higher SD score (OR=1.14 per pt; p=0.01) was associated with response in patients with baseline SD. Conclusions: Frequency and severity of SD is high. Response to interdisciplinary supportive care is low. Further research is greatly needed.