e14087 Background: Gastrointestinal (GI) cancers are among the most common cancers in adult patients. The prognosis of GI cancers differs widely between specific subsites, with some cancers having excellent prognoses, and others (such as pancreatic and hepatobiliary cancers) having an extremely poor prognosis. We sought to characterize the landscape of GI oncologic clinical trials in the last two decades. Methods: We searched ClinicalTrials.gov for phase III RCTs between 2003 and 2018. Randomized multi-arm trials assessing a therapeutic intervention in adult cancer patients were included. GI cancer trials were identified and grouped into specific disease subsites. Differences in trial factors between GI and non-GI trials and between different GI subsites were analyzed. Results: Out of the 600 RCTs identified, 76 trials assessed gastrointestinal cancers (12.7%). By disease subsite, we identified 21 (27.6%) gastro-esophageal cancer trials, 12 (15.8%) pancreatic cancer trials, 11 (14.5%) hepatobiliary cancer trials, and 32 (42.1%) colorectal cancer trials. GI cancer trials were more likely to be industry funded when compared to non-GI cancer trials (88.2% vs 76.3% respectively, p = 0.02). Furthermore, only 18.4% of GI trials had cooperative group sponsorship (vs 33.3% for non-GI trials, p = 0.01). Most GI cancer trials assessed systemic therapy interventions (85.5%); only 6 trials (7.9%) addressed supportive care interventions (vs 20.2% for non-GI trials, p = 0.01). Around half of GI cancer trials (51.3%) met their primary endpoints, but only 8 trials (10.5%) led to subsequent FDA approval. All FDA approvals were for colorectal cancer treatments, and none for gastroesophageal, pancreatic or hepatobiliary cancers (p = 0.01). Most patients (97.1%) accrued on GI cancer trials had a good performance status (Eastern Cooperative Oncology Group of 0 to 1). Specifically, 95.9% of patients in pancreatic trials and 98.6% of patients in hepatobiliary trials had a good performance status. Conclusions: While many GI cancers have a poor prognosis, very few trials assess supportive care interventions in this populations. Moreover, most patients included in GI cancer trials have a good performance status. In that regard, future GI cancer trials addressing supportive care measures with a wider cancer population are still warranted.
