Outcomes Of Advanced Gastrointestinal (Gi) Cancer Patients In Relationship To Opioid Use: An Individual Patient Data Pooled Analysis From Eight Clinical Trials

687 Background: The current study aims at assessing the patterns of opioid use, and evaluating the impact of opioid use on survival outcomes among patients with advanced GI cancers who were included in eight clinical trials. Methods: De-identified datasets of eight clinical trials evaluating first-line systemic treatment for advanced GI cancers (NCT01124786; NCT00844649; NCT00290966; NCT00678535; NCT00699374; NCT00272051; NCT00305188; NCT00384176) were accessed from the Project Data Sphere platform. These trials evaluated patients with pancreatic, gastric, hepatocellular and colorectal carcinoma. Multivariable logistic regression analysis was used to evaluate factors predicting the use of opioids. Kaplan-Meier survival estimates were used to compare survival outcomes in each disease entity among patients who did or did not receive opioid treatment. Multivariable Cox regression analysis was used to assess the impact of opioid use on survival outcomes in each disease entity. Results: A total of 3441 participants were included in the current analysis. The following factors predicted a higher probability of opioid use within logistic regression analysis: younger age (P = 0.004), non-white race (P = 0.010), higher ECOG score (P < 0.001) and pancreatic primary site (P < 0.001). Use of opioids was consistently associated with worse overall survival in Kaplan-Meier survival estimates of each disease entity (for pancreatic cancer: P = 0.008; for gastric cancer: P < 0.001; for hepatocellular carcinoma: P < 0.001 and for colorectal cancer: P < 0.001). Within multivariable Cox regression analysis, opioid use was associated with worse overall survival among patients with pancreatic cancer (HR = 1.245; 95% CI: 1.063-1.459; P = 0.007), gastric cancer (HR = 1.725; 95% CI: 1.403-2.122; P < 0.001), hepatocellular carcinoma (HR = 1.841; 95 CI: 1.480-2.290; P < 0.001) and colorectal cancer (HR = 1.651; 95% CI: 1.380-1.975; P < 0.001). Conclusions: Opioid use is consistently associated with worse overall survival among patients with different GI cancers. Further studies are needed to evaluate the underlying mechanisms of this observation.