Infections in Patients with Chronic Lymphocytic Leukemia Treated with Chemoimmunotherapy.

Chronic lymphocytic leukemia (CLL) is a clonal B cell malignancy characterized by defects in both cellular and humoral immunity. Each immune defect can be linked to common infectious complications affecting these patients. Predisposition to infections can be compounded by immunosuppression from cytotoxic chemotherapy. The purine analog fludarabine has been associated with a spectrum of infections caused by prolonged T lymphocyte depletion. The combination of fludarabine, cyclophosphamide, and rituximab (FCR) was developed based on complementary activity of the combination. FCR has significant activity in both chemotherapy-naïve (ChN) and previously treated (PrT) patients with CLL. We analyzed risk factors for infection and describe the infections experienced by both ChN and PrT patients during FCR treatment.

We performed a retrospective review of medical records of 224 ChN and 177 PrT patients with CLL treated with the FCR regimen at MD Anderson Cancer Center from 7/99-8/02. Review was until relapse, death, or end of treatment. We focused on infectious events that occurred during and within 30 days of the last cycle of FCR treatment. Prophylaxis for herpes virus (valacyclovir) was recommended for all patients. Prophylaxis for P carinii (trimethoprim-sulfa) was recommended for all PrT patients and was done at the discretion of the treating physician for ChN patients.

Infectious events were experienced by 96/224 (43.2%) ChN and 99/177 (56.2%) PrT patients. Causative agents were identified in 23% of events for ChN and 17% of events for the PrT patients (Table 1). There was a higher incidence of any event (p=0.012) and major infectious (MI) events (p<0.001) in PrT versus ChN patients. Major infections included pneumonia, sepsis and fever of unknown origin.

Among ChN patients, univariate analysis identified low serum IgG (<700 mg/dL) (p=0.04), increased WBC count (p=0.049), and >2 chemotherapy courses (p=0.044) as risk factors for MI. Multivariate analysis identified low serum IgG (p=0.03) as the only predictor for MI. Among PrT patients, univariate analysis identified female gender (p=0.015) and baseline hemoglobin level (p=0.031) as risk factors for MI. Multivariate analysis identified increased WBC count (p=0.026) and high (>4.0 mg/dL) beta-2 microglobulin (ß2M) (p=0.027) as risk factors for MI in PrT patients.

Among ChN and PrT patients none of the variables were found to be risk factor for minor infections (mI) in univariate analysis; multivariate analysis identified high ß2M (p=0.034) and splenomegaly (p=0.033) as significant predictors for occurrence of mI among PrT patients.

For ChN patients, the nadir neutrophil count was not significantly associated with occurrence of mI or MI over the 6 courses of treatment. On the contrary, it was associated with occurrence of minor, major and any infection in the PrT patients. FCR is a higly effective and well-tolerated regimen. IgG <700 mg/dL was the main predictive variable for MI in ChN patients; baseline WBC count and ß2M >4.0 mg/L were the main predictive variables for MI in PrT patients.