Figure 3. C3b and C5b-9 Expression and Complement Dependent Cytotoxicity.

C5b-9 levels were measured on CLL cells from a subset of 10 patients including all 5 patients with < 75% CDC on initial treatment with mAb and 10% normal human serum (NHS). A: C5b-9 binding was measured by flow cytometry in viable (propidium iodide (PI) negative) and intact dead (PI positive) CLL cells. The median C5b-9 binding (dMFI) for viable cells (open bars) and intact dead (black bars) is shown with error bars representing the standard deviation. C5b-9 binding was significantly higher in intact dead cells compared to viable cells for all cases (p=0.002). The level of binding of C5b-9 was higher in cells treated with ALM compared to OFA (live and intact dead cells p≤0.006). Addition of OFA (p=0.02) and RTX (p=0.01) to ALM significantly increased C5b-9 binding in intact dead cells but not viable cells (p>0.1). B: The binding of C3b and C5b-9 was compared in viable (open circles) and intact dead (black squares) CLL cells treated, in 10% NHS, with OFA, ALM, ALM + RTX, and ALM + OFA. There was a significant linear relationship between binding levels of C3b and C5b-9 in intact dead cells (black squares, r²=0.77, p<.0001) but not live cells (open circles r²=0.19, p=0.23). There was also considerable overlap in C5b-9 levels between viable and intact dead cells for both of these measurements. C: CLL cells from a representative patient were treated with 10%NHS and either ALM or OFA. This histogram shows C5b-9 binding to viable cells treated with ALM (grey) and intact dead cells (black) with OFA.