Background: Genetic alterations (GAs) in the fibroblast growth factor receptor (FGFR) pathway are emerging as promising therapeutic targets in CCA. The clinical and molecular features of patients (pts) with CCA harboring FGFR GAs are reported here. Methods: A retrospective chart review was performed in pts with CCA who were found to have an FGFR GA on tumor molecular profiling as part of routine care. Data on demographics, risk factors, pathology, systemic therapy, radiographical response, progression free survival (PFS), and overall survival (OS) were collected. Results: Among 65 pts, the median age at diagnosis was 55 years old (range = 27-92), and 38 (58%) pts were female, 63 (97%) had intrahepatic CCA, and 5 (11%) had chronic HBV. At presentation, 37% of pts had resectable disease. Of 47 pts with a known CA 19-9 at the time of initial diagnosis, 21 (45%) had a value < 35U/mL. FGFR2 fusions were the most common FGFR GA (78%), followed by FGFR2 mutations (14%), FGFR3 mutations (4%), FGFR3 fusion (2%) and FGFR1 amplification (2%). The most common fusion partners were BICC1 (20%), POC1B (6%), SORBS1 (6%), DBP (4%), and TACC2 (4%). The most common co-alterations were in ARID1A, CDKN2A/B, TP53, BAP1, IDH1, HER2, BRCA2, and PTEN. The median lines of palliative systemic therapies received was 3 (range = 0-8), and 9/65 (14%) pts had > 1 FGFR inhibitor (FGFRi). For the 30 (46%) pts with FGFR2 fusions who received gemcitabine/platinum as first line palliative systemic therapy, the median PFS was 4.7 months (95% CI: 2.1-6.0). In the overall population, the median OS from time of initial diagnosis was 35.8 months (95% CI:29.7-52.7). Among 46 pts who received an FGFRi on a clinical trial and had ≥ 1 follow-up scan, the overall response rate (ORR) by RECIST v1.1 in pts with FGFR2 fusions, was 35.8% (14/39) on their first FGFRi; ORR was 16.7% (1/6) for pts with FGFR2 mutations. Conclusions: Pts with CCA harboring FGFR GAs were found to have a high rate of normal CA 19-9 and short median PFS on first line gemcitabine/platinum compared to historical controls but additional comparative studies are necessary to evaluate these findings.