CLL Cells From Subjects Treated with Rituximab and Alemtuzumab Can Lose Target Antigen and Develop Resistance to Mab-Mediated Complement Dependent Cytotoxicity

Abstract 3585

Relapsed/refractory CLL is often resistant to antibody-based therapy. Understanding the dynamics of interaction between complement and cell-based antibody-antigen complexes that promote complement dependent cytotoxicity (CDC) of targeted cells could lead to improved therapies. We are conducting a clinical trial of chemoimmunotherapy for relapsed/refractory CLL using pentostatin, alemtuzumab (ALEM), and low-dose rituximab (RTX). Correlative studies were designed to determine if circulating CLL cells developed resistance to CDC.

Peripheral blood CLL cells were obtained from 12 patients (median age 66 years, 11 male, 17p13- in 4) including 7 (58%) who responded (4 CR, 3 PR) to treatment. Blood samples were obtained before initiating therapy (day 1), 48 hr after the 1^st RTX dose (20 mg/m² IV MWF), one hour after the 2^nd dose of RTX and 1^st dose of ALEM (3 mg SC) (day 3), and 72 hr after the 3^rd doses of RTX (cumulative dose 60 mg/m²) and ALEM (cumulative dose 43 mg) (day 8). All blood samples were taken prior to the 1^st dose of pentostatin. B cells were examined for expression of several markers including complement control proteins (CCP), CD20 and CD52.

By treatment day 8, ALC decreased by 70 % or more for 7 patients. The other 5 patients had ALC that were 50 – 100% of pre-treatment levels. In agreement with previous studies from our laboratories, low-dose RTX rapidly decreased CD20 levels on circulating CLL cells; on day 3 after the 2^nd RTX dose (and 1^st ALEM dose), CD20 was reduced by > 75% in 10 subjects. CD20 levels recovered by day 8 in 6 of these subjects. One hour after completion of mAb infusions on day 3, fragments of complement breakdown product C3dg were demonstrable on circulating CLL cells in 7 subjects, suggesting that infusion of RTX had promoted complement activation on circulating cells. After the RTX and ALEM infusions on day 3, circulating CLL cells with bound C3dg fragments had considerably reduced levels of CD20 but were not cleared from the circulation, suggesting that RTX had bound to the cells, activated complement and promoted shaving (CD20 loss) without inducing bloodstream clearance of these cells. CD52 levels on circulating CLL cells showed modest decreases by day 8 (~ 50% of pre-treatment values) in 4 subjects.

To study CDC in vitro, washed blood samples were reconstituted with ABO-matched normal human serum, ALEM was added and complement activation (C3 fragment deposition) and CDC of targeted CLL cells were measured by flow cytometry. Alternatively, mononuclear cells were isolated by ficoll-hypaque centrifugation and subjected to the same analyses, with very similar results. ALEM-mediated in vitro CDC of CLL cells varied from 10 to 90 % for pre-treatment blood samples. Measures of ALEM-mediated CDC and C3b fragment deposition in vitro correlated with the density of CD52 on CLL cells (P<0.001). The initial burden of circulating cells was not predictive of a drop in ALC, and pre-therapy levels of CDC were not correlated with cell clearance after therapy. On day 8, ALEM-mediated in vitro CDC of circulating CLL cells had decreased to <50% of the values measured on pre-treatment samples in 6 subjects, suggesting that cells of these patients might be developing resistance to ALEM-mediated CDC. This decrease in CDC was not correlated with decreases in levels of CD52 expression on patients’ CLL cells or with increases in expression of CCP (CD35, CD46, CD55, CD59). Our findings demonstrate that even after treatment with both RTX and ALEM, CLL cells can persist in the bloodstream in a subset of patients, suggesting that some patients have circulating CLL cells that are resistant to mAb-mediated clearance and CDC due to mechanism(s) that may include loss of antigen and possibly intrinsic resistance.