Abstract
OBJECTIVES: Early-phase clinical trials play a pivotal role in drug development. However, limited data are available on outcomes of gastrointestinal (GI) cancer patients enrolled in phase I clinical trials. Here, we evaluated the characteristics associated with survival in GI cancer patients participating in phase I clinical trials and attempted to validate previously established prognostic models. MATERIALS AND METHODS: All consecutive patients with advanced GI tumors who participated in phase I clinical trials at our institution from January 2007 to December 2013 and received at least 1 dose of the study drug were included. Cox regression models were used to estimate multivariable-adjusted hazard ratio (HR) and 95% confidence interval. RESULTS: In 243 study patients (median age, 62 y [range, 26 to 82 y]; 55% male), treatment included chemotherapy only (14%), targeted therapy (41%), chemotherapy+targeted therapy (42%), and others (2%) for the following disease types: pancreatic (42%), colorectal (34%), gastroesophageal (10%), hepatobiliary (13%), and others (2%). Response rate was 4%, with 38% achieving stable disease and 42% having progressive disease. Median survival was 5.8 months (range, 0.2 to 52.4 mo). Our multivariable Cox regression analyses included the following as predictors of survival: Eastern Cooperative Oncology Group performance score ≥1 (HR=1.76), prior systemic therapies ≥2 (HR=1.63), lactate dehydrogenase >618 IU/L (HR=1.85), sodium >135 mmol/L (HR=0.46), and white blood count >6×10/L (HR=1.5). Our data set was consistent with previous prognostic scores. CONCLUSIONS: This is the largest study to assess clinical outcomes in this patient population. Phase I trials provide clinical benefit to patients with advanced GI malignancies and should be recommended as a treatment option in appropriate patients.