Tigecycline, the first in a new class of glycylcyclines, has been approved for the treatment of complicated skin and skin structure and intraabdominal infections in adults. However, clinical data on its safety and effectiveness in cancer patients are lacking. We reviewed the records of all cancer patients treated with tigecycline for more than 48 hours between June 2005 and September 2006 at our institution and identified 110 consecutive cases (median age, 58 yr; range, 18-81 yr). We collected data on demographics, cancer type, tigecycline indication, microbiologic characteristics, side effects, and outcome. Sixty-four (58%) patients had hematologic malignancies; 27 patients had undergone hematopoietic stem cell transplantation. Thirty-one (28%) patients had neutropenia, and 62 (56%) were in the intensive care unit at the start of therapy. Most patients (106 [96%]) received tigecycline as a second-line agent (after not responding to other broad-spectrum antibiotics), and 101 (92%) received it in combination with an antipseudomonal drug. The mean duration of therapy was 11 days (range, 3-35 d). Sixty-six (60%) patients received tigecycline for refractory pneumonia, 19 (17%) had bacteremia, 9 (8%) had intraabdominal infections, and 7 (6%) had complicated skin and soft tissue infections. Fifty (45%) patients had microbiologically documented infections, and the remaining patients had negative cultures at the start of therapy.An overall clinical response was noted in 70 (64%) patients. More clinical responses were seen in patients with bacteremia than in those with pneumonia (79% vs. 51%; p = 0.029). Patients with microbiologically documented infections had significantly higher clinical response rates than patients with non-microbiologically documented infections (73% vs. 55%; p = 0.047). Forty (36%) patients did not respond to treatment; 36 of these patients died of active infection during tigecycline therapy. Patients with pneumonia had a significantly higher mortality rate than patients with bacteremia (44% vs. 16%; p = 0.026). During the 60 days of follow-up from the date of clinical response, patients with pneumonia had significantly shorter survival durations than patients with other infections. Of the 42 patients who were not on antiemetics or ventilator support at the start of tigecycline therapy, 2 (5%) experienced mild nausea, and 1 (2%) experienced nausea and vomiting. Only 4 (4%) patients overall experienced diarrhea during tigecycline therapy, all of whose stools were negative for Clostridium difficile toxin. No serious adverse events related to tigecycline use were identified. The combination of tigecycline and an antipseudomonal drug may be appropriate for treating refractory infections and multidrug-resistant organisms in cancer patients, including hematopoietic stem cell transplant recipients. Patients with refractory pneumonia had a relatively low clinical response rate in our study.