CD38 is a transmembrane glycoprotein expressed on the surface of leukemic cells in a significant percentage of patients with Bcell chronic lymphocytic leukemia (B-CLL). A recent study suggested that CD38 expression has prognostic value in CLL. Peripheral blood samples from 218 patients with B-CLL were analyzed by flow cytometry for CD38 expression on CD5/19 ؉ leukemic cells. Various patient characteristics were studied including age, sex, Rai and Binet stages, splenomegaly, hepatomegaly, hemoglobin (Hgb) level, ␤-2 microglobulin (␤2M) level in the serum, number of nodal sites involved with disease, and length of survival. The Kaplan-Meier method was used to construct survival curves, and the log-rank statistic was used to compare these curves. CD38 was expressed in 20% or more of leukemic cells in 43% of the patients. Patients with high CD38 expression (20% or more) had significantly shorter survival times (P ‫؍‬ .00005). Multivariate analyses showed that CD38 expression is an important prognostic factor associated with high incidence of lymph node involvement (P ‫؍‬ .004), lower hemoglobin level (P ‫؍‬ .001), hepatomegaly (P ‫؍‬ .05), and high ␤2M level (P ‫؍‬ .00005).

CD20 is a 33-to 36-kDa transmembrane phosphoprotein involved in the activation, proliferation, and differentiation of B lymphocytes. The predicted amino acid sequence of the CD20 suggests 4 transmembrane-spanning regions with both Nand C-termini located in the cytoplasm. We demonstrate herein that significant levels of circulating CD20 (cCD20) can be detected in the plasma of patients with chronic lymphocytic leukemia (CLL) and that cCD20 interferes with the binding of rituximab, a humanized anti-CD20 monoclonal antibody, to CLL cells. An enzymelinked immunosorbent assay (ELISA) was developed to measure circulating cCD20 levels in the plasma. We measured cCD20 levels in the plasma of 180 patients with CLL and correlated these levels with clinical characteristics and outcome. Circulating CD20 levels correlated positively with ␤ 2 -microglobulin level (p ‫؍‬ .006) and percentage of CD38 ؉ cells (p ‫؍‬ .03) and negatively with platelet count (p ‫؍‬ .004) and hemoglobin level (p ‫؍‬ .02). Patients with advanced Rai (III/IV) or Binet (C) stage disease had significantly higher levels of cCD20 than did patients with earlier-stage disease (P ‫؍‬ .01 and P ‫؍‬ .006, respectively). There was no correlation between cCD20 level and age, lymphocyte count, or white blood cell count. Using a recursive classification method, we found that patients with a cCD20 level more than 1875 nM/L had significantly shorter survival than those with cCD20 1875 nM/L or below (P ‫؍‬ .01). The prognostic value of cCD20 was independent of Rai staging or hemoglobin level. Prospective evaluation is indicated to establish whether rituximab dosing should be adjusted according to cCD20 levels. (Blood. 2003;101: 2507-2513)

Lymphoid cells in most patients with chronic lymphocytic leukemia (CLL), when treated with rituximab, become CD20 ؊ . This is thought to be due to masking of CD20 by rituximab. We used specific antimouse immunoglobulin antibodies to detect rituximab on the surface of CLL lymphocytes and we demonstrate that rituximab is rarely detectable after therapy. Only 3 of 65 patients with CLL had rituximab detectable on their lymphocytes after rituximab therapy despite the fact that most had no detectable CD20 expression. In vitro mixing of CLL or Raji cells with rituximab demonstrated that rituximab was detectable on the surface of cells due to its binding to CD20. However, the addition of plasma led to the down-modulation of CD20 expression, and the rituximab became undetectable. This down-modulation of CD20 protein expression was associated with a downmodulation of CD20 mRNA. CLL cells that lost their CD20 expression regained CD20 expression after 24 hours in culture. These data suggest that rituximab therapy leads to a substantial but transient downmodulation of CD20 expression and that negativity for CD20 in cells from patients treated with rituximab is not necessarily due to CD20 masking.

We measured the concentration of CD33 antigen on the surface of cells in 315 bone marrow (BM) samples and 114 corresponding peripheral blood (PB) samples from patients with various leukemias (acute myeloid leukemia [AML], chronic myelogenous leukemia [CML], myeloproliferative disorder [MPD] other than CML, myelodysplastic syndrome [MDS]) and from control subjects. Overall CD33 intensity in total CD33+ cells was significantly higher in BM than in PB. CD33 intensity in total BM CD33+ cells differed significantly with the type of disease. The median number of CD33 molecules per cell was highest in AML, followed by MDS, CML, and control subjects and lowest in MPD. When only CD34+/CD33+ cells were examined, CD33 molecules per cell were highest in CD34+ cells in AML and lowest in MPD (P = .027). Patients with AML or MDS younger than 60 years had significantly higher intensity of CD33 expression on CD34+ cells than patients 60 years or older. Levels of CD33 intensity did not correlate with cytogenetics in patients with AML or MDS. There was no correlation between CD33 intensity and response to therapy or overall survival in 35 patients treated with protocols including Mylotarg. These data demonstrate variation in CD33 intensity between various leukemias.