Background: Gemcitabine plus cisplatin (GC) is currently the standard regimen for advanced biliary tract cancers (BTC) based on the outcomes in ABC-02 trial. Multiple factors can affect outcomes in these patients. This retrospective review evaluates the University of Cincinnati experience with GC in advanced intrahepatic (IHC)/ extrahepatic cholangiocarcinoma (EHC) and gall bladder carcinoma (GBC). Methods: In this study approved by University of Cincinnati IRB, retrospective analysis of advanced BTC patients seen between 01/2008 and 01/2015 was done. Kaplan Meyer method was used to calculate progression free survival (PFS), and overall survival (OS). Cox model was used to test the association between baseline variables and OS/PFS, adjusting for gender and age at diagnosis. Patients were identified using ICD code for BT tumors, 26 patients met inclusion criteria: histologically proven advanced BTC that received GC as their initial chemotherapy. GC was given as per ABC-02 protocol with appropriate modifications until disease progression or unacceptable toxicities. Results: Median age at diagnosis was 62 years (range, 31-81 years). Eighteen (69%) were IHC, 5 EHC, 3 GBC, 61% male, 73% whites. Performance status (PS): 0-1: 69%, PS 2: 31%. Baseline CA19-9 data was available for 21 patients, (range 1 to 69,543), and abnormal CA19-9 was seen in 14 patients (54%). PFS was 4.5 months (95% CI: 3.1-8.9 months) and OS was 10.5 months (95 % CI: 7.9-18.8 months). OS at 6 and 12 months was 69% (18/26) and 42% (11/26). Thirty-eight percent (10/26) received 2nd line chemotherapy, of these 9/10 received 5FU based chemotherapy. Eleven percent (3/26) received 3rd line chemotherapy. Increase in baseline carcinoembryonic antigen (CEA), alanine aminotransferase, alkaline phosphatase (ALP) and total bilirubin was associated with increased risk of death while increase in baseline CEA and ALP was associated with increased risk of progression (P valve <0.05). In the group of patients who had all three major risk factors (PS ≥2, CEA >3, and stage IVb), the median survival was 2.9 months (95% CI: 2.6-9.3 months), which was significantly worse compared to rest of the population [median 18 months (95% CI: 5.4-19.5 months), P<0.01]. Conclusions: Our data supports the use of GC as a first line regimen for advance BTC in a non-clinical trial setting. Results are comparable to those reported in ABC-02 trial, despite inclusion of PS 2 patients whom constituted 31% of our population. In the patient population studied, baseline CEA and liver function test appeared able to predict response to GC in advanced BTC. Patients with all three high risk factors (PS ≥2, CEA >3, and stage IVb) did poorly and may need careful selection prior to initiating chemotherapy.