Quantitative Measurement of CD52 Expression and Alemtuzumab Binding in Adult Acute Lymphoblastic Leukemia (ALL): Correlation with Immunophenotype and Cytogenetics in Patients (Pts) Enrolled on a Phase I/II Trial from the Cancer and Leukemia Group B (CALGB 10102).

The feasibility of incorporating Alemtuzumab (A) into frontline therapy of adult ALL pts was evaluated in a recently completed Phase I/II clinical trial, CALGB 10102. Pts were considered CD52+ and eligible to receive A during post-remission therapy if at least 10% of lymphoblasts had detectable expression of CD52 as measured by qualitative flow cytometry analysis in a CALGB reference laboratory. Using this qualitative assay, we have determined that 212 of 302 (70%) pts were CD52+ and eligible to receive A. Interestingly, 40/40(100%) pts with t(9;22) were CD52+. To better characterize CD52 expression level on lymphoblasts at the time of diagnosis and to obtain insights into subset specific therapies in adults with ALL, we developed a quantitative flow cytometry method to measure CD52 antigen expression and A binding using custom conjugated clinical grade A antibody. Results of this quantitative assay are expressed in arbitrary units of Antibody Bound per Cell (ABC). To determine whether ABC correlated with immunophenotype and specific cytogenetic subsets of adult ALL, we assessed ABC in a randomly selected subset of 88 untreated pts entered onto CALGB 10102. Seventy-two (81%) pts had precursor-B cell (pre-B) ALL and 16 (19%) had precursor T-cell (pre-T) ALL. The median ABC for pre-B cases was 37,178 (range: 1,545–228,247) and was significantly higher than the median ABC of 15,585 (range: 5,231–53,409) for pre-T cases (p=0.01). Interestingly, ABC on both pre-B and pre-T lymphoblasts was significantly lower than the median ABC on residual normal B-lymphocytes of 135,047 (range: 30,277–1,214,141), and on residual normal T -lymphocytes with a median of 70,139 (range: 20,621–557,859) that were present in the diagnostic bone marrow specimen (p<0.001). ABC by cytogenetic subset was also evaluated with results tabulated below for some of the commonly recurring abnormalities:

ABC on lymphoblasts differed significantly across the cytogenetic subsets examined (p<0.001). The highest ABC levels were present in pts with del 9p and the lowest ABC levels were detected in pts with t(4;11). In conclusion, CD52 expression and A binding are significantly higher in pts with pre-B ALL than those with pre-T ALL. However, overall CD52 expression and A binding is significantly lower on lymphoblasts in comparison with normal residual lymphocytes present in the diagnostic bone marrows. We also demonstrate for the first time that differential CD52 expression and A binding occurs in specific cytogenetic subsets of adults with ALL. Quantitative PCR analysis of minimal residual disease following A treatment is being performed to determine whether pretreatment ABC levels identify pts who benefit from A therapy.