Background: CD33 is variably expressed on leukemia blasts in most patients with acute myeloid leukemia (AML). Efforts to target CD33 therapeutically have focused on gemtuzumab ozogamicin (GO), an antibody-drug conjugate delivering a DNA-damaging calicheamicin derivative. Qualification for GO therapy has been determined by expression of CD33 by multidimensional flow cytometry (MDF); patients with positive CD33 expression receive GO. Previous studies from the AAML0531 protocol demonstrated that the cell surface intensity expression of CD33, determined by MDF, predicts the response to treatment with GO, and that in part this expression is regulated by a pair of CD33 single nucleotide polymorphisms (SNPs) in linkage disequilibrium that also, independently, predict response to therapy. Patients with CC genotype for rs12459419 have lower relapse risk and improved disease-free survival, compared to CT and TT genotypes. Because GO therapy is associated with treatment-related toxicity, it is important to identify biologic variables associated with benefits and risks. To date, there has been no report associating SNP status among CD33 positive versus CD33 negative patients that could assess how the combination of these biomarkers for determining administration of GO therapy could improve patient outcomes.

Objective: In an effort to determine which children would benefit most from GO treatment in future prospective pediatric AML protocols, we aimed to elucidate if CD33 SNP genotype status should be combined with quantitative CD33 cell surface antigen expression on the diagnostic leukemia cells (CD33+ versus CD33-) to determine GO eligibility, with a retrospective analysis of children enrolled in Children's Oncology Group AAML0531.

Methods: Of 1022 newly diagnosed pediatric patients with de novo AML enrolled on protocol AAML0531, 666 satisfied two criteria for this study: (1) submission of a blood or bone marrow sample for MDF at diagnosis with corresponding CD33 SNP genotype data available, and (2) proper consent for specimen testing.

CD33 Expression Levels on Leukemic Blasts

The diagnostic AML leukemia population was identified by gating on CD45 versus log-side scatter and CD33 expression levels were determined by measuring the mean fluorescence intensity (MFI) by flow cytometry. For a patient to be considered CD33+ two criteria were required: intensity of CD33 on blasts was at minimum four times the MFI of its correspondent autofluorescent control, and at least 80% of the total blasts were greater than this minimum.

Genotyping CD33 SNPs

CD33-coding SNP rs12459419-Ala14Val and linked promoter SNP rs3865444 were genotyped using the Sequenome (San Diego, CA) platform at the Biomedical Genomics Center, University of Minnesota. Both SNPs had a call rate of 0.98 and were in Hardy-Weinberg equilibrium (P=0.05).

Results:Association of Genotype and cell surface expression of CD33 for AAML0531 patients Of 666 patients, 84% (560/666) were CD33+. CC genotype was observed in 54.5% (305/560) of CD33+ cases, 37.5% (210/560) had CT genotype and 8% (45/560) TT genotype. Of the 16% (106/666) of patients who were CD33 negative, 33% (35/106) had CC genotype, 47% (50/106) CT genotype, and 20% (21/106) had TT genotype. Comprehensively, 340/666 (51%) had CC genotype (51%) and 10% were CD33 negative (35/340). Conversely, out of a total of 66 patients with TT genotype, 45 (68%) were CD33+ and 32% (21/66) were CD33 negative.

Conclusions: These results clarify the relationship between the amount of CD33 expressed on AML at diagnosis as measured by MDF and CD33 SNP genotype status. While correlation with clinical outcome analysis is ongoing, these data support inclusion of CD33 SNP genotyping for eligibility of GO therapy. Therefore, the current recommendation for future COG AML clinical protocols is that GO will be administered to patients with CD33 expression as determined by MDF and CC genotype patients regardless of CD33 expression.

Disclosures

Pardo:Hematologics, Inc: Employment. Eidenschink Brodersen:Hematologics, Inc: Employment. Paine:Hematologics, Inc: Employment. Loken:Hematologics, Inc: Employment, Equity Ownership.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution