148 Background: Cancer patients have an elevated risk of suicide given the prevalent psychological distress, treatment side effects, and uncontrolled pain. This study aims to examine temporal trends in cancer-related suicide in the US during the past two decades in view of progress in psycho-oncology care and symptom control for cancer patients in the country. Methods: Cancer-related suicide was defined as deaths with suicide as the underlying cause and cancer as a contributing cause in the US Multiple Cause of Death Data. We first calculated annual age-standardized cancer-related suicide rates and overall suicide rates from 1999 to 2015. Then, we fitted the log-transformed age-standardized rates to Joinpoint regression models to calculate the annual percentage change (APC) by demographic factors and overall. Results: Among 599,786 suicides in the US from 1999 to 2015, 5,559 were cancer-related. Seventy-three percent of the cancer-related suicides were committed by firearm and 14% were by poisoning. The age-standardized cancer-related suicide rate per 100,000 persons decreased from 0.17 in 1999 to 0.12 in 2015, with an APC of -2.5% (95% CI: -3.3%, -1.7%). In contrast, the age-standardized overall suicide rate per 10,000 persons increased from 13.22 in 1999 to 16.73 in 2015, with APC of 0.9% (95% CI: 0.4%, 1.3%) from 1999 to 2006 and 2.0% (95% CI: 1.7%, 2.2%) from 2006 to 2015. The decline in cancer-related suicide rate was largest among male (APC = -3.1%, 95% CI: -3.9%, -2.3%), 65-74-year-olds (APC = -3.1%, 95% CI: -4.2%, -2.0%), in the South (APC = -4.0%, 95% CI: -5.8%, -2.3%), and in urban area (APC = -3.0%, 95% CI: -4.2%, -1.7%). Conclusions: Despite the continuous increasing overall suicide rates in the US during the past two decades, cancer-related suicide has been decreasing, suggesting an evolving role of psycho-oncology care and symptom control during this period. Well-designed prospective studies are warranted to identify cancer patients at high-risk of suicide and to develop effective care intervention.