Abstract
Autoimmune cytopenias (AIC), particularly autoimmune hemolytic anemia (AIHA), pure red cell aplasia (PRCA) and immune-thromobcytopenic purpura (ITP) are important complications of fludarabine (F) treatment of chronic lymphocytic leukemia (CLL). Rituximab (R), a chimeric anti-CD20 monoclonal antibody, is an effective treatment for these autoimmune complications. The introduction of rituximab into fludarabine containing combination regimens has been expected to decrease the incidence of AIC. The experience in some studies (CALBG 9712, 2% incidence of AIC with FR; Byrd, 2003) but not others (6.5% incidence in 300 patients treated with FCR; Borthakur, 2007) supported this view. We compared the incidence of AIC during treatment in 2 CLL cohorts, one treated with F (1998–2004, n=21), the other with FR (2005–2008, n=17, rituximab 375mg/m2 q4 weeks). Five of 21 patients (24%) in the F group developed AIC (3 ITP, 1 PRCA, 2 AIHA), while 3 of 17 patients (18%) in the FR group developed AIC (2 AIHA, 1 autoimmune neutropenia). Two patients with ITP in the first cohort and all three patients in the second were successfully treated for their AIC with 4 infusions of rituximab over 2–4 weeks. Interestingly, although rituximab did not prevent the occurrence of AIC it remained effective in treating the complication, indicating that serum levels of rituximab could be too low to prevent the onset of AIC. We therefore measured rituximab serum levels in 7 patients (average ALC 134 K/mL, range 10–423) at 6, 24 and 120 hours from the start of the infusion. The average peak serum level at 6 hours, typically shortly after the end of the infusion, was 93μg/mL (32–155). By 24 hours, the average level declined to 66μg/mL (39–90), and by 120 hours to 11μg/mL (0–30, undetectable in 2). Based on these measurements, we estimate the half life of rituximab during this first cycle at 1–2 days. This extremely short half life is consistent with enhanced clearance of cell bound rituximab either during phagocytosis proper or during a form of abortive phagocytosis, also called trogocytosis, during which rituximab/CD20 complexes are pulled off the CLL cell surface and ingested by phagocytic cells (Beum, 2006). Consistent with the expected correlation of high tumor burden with rituximab clearance, we found higher rituximab serum levels in subsequent cycles: in cycle 2, 3 of 4 patients had an average increase in serum concentration of 27% at 24 hours, and in cycle 3 one patient had a 72% higher level than in the first cycle. Thus, at least in the first cycles, rituximab is cleared within a week, giving rise to a prolonged period of essentially single agent fludarabine effects. This could contribute to the persistently high incidence of AIC despite the use of rituximab in fludarabine combination regimens. These data indicate that, at least in the first cycles, additional infusions of rituximab may be necessary to prevent the onset of AIC in CLL patients treated with fludarabine-containing regimens.
Disclosures: No relevant conflicts of interest to declare.
Author notes
Corresponding author