University of Florida
Project Term: July 1, 2019 - June 30, 2022
Acute myeloid leukemia remains a refractory disease with dismal outcome. Recent approval of CD33-directed gemtuzumab ozogamicin (GO) is a significant step in AML treatment. We recently identified a genetic variant in CD33 resulting loss if the domain that is recognized by GO. Presence of this variant significantly impacts patient’s GO response. The proposed research seeks to perform mechanistic evaluation of CD33 variant to improve our ability to personalize CD33-directed immunotherapy in AML.
Acute myeloid leukemia (AML) is the second most common childhood leukemia with dismal outcome. It is a heterogeneous disease with significant proportion of patients in spite of achieving remission after induction therapy experience relapse. CD33 is expressed on leukemic cells of majority of AML patients and thus is one of the key targets that are explored for development of various immunotherapies to treat AML. Gemtuzumab Ozogamicin (GO) is a CD33-antibody that is conjugated to cytotoxin calicheamicin and promising results from several adult and pediatric AML trials resulted in its recent approval by for treatment of CD33+ve acute myeloid leukemia (AML). Given the success of gemtuzumab numerous other CD33-directed agents are currently under different phases of development and clinical trials. This recent expansion of agents targeting CD33 in AML has opened up opportunities for identifying biomarkers predictive of response to CD33-directed agents such as GO. Since GO targets CD33, any inter-patient variation in CD33 will have a profound clinical impact on therapeutic efficacy. Precious results have shown variation in CD33 expression in leukemic cells, which in turn impacts response to GO. Therefore, there is enormous clinical benefit from preemptively identifying patients most likely to benefit from GO and individualizing therapy to obtain the maximum clinical benefit with reduced toxicity in patients who would not respond to GO. Our group was first to investigate genetic variation in CD33 in relation to GO response in pediatric AML and our recent results reported a genetic variant known as rs12459419 in exon 2 of CD33 that results in loss of the IgV domain in certain patients. This is clinically relevant, as GO as well as other CD33-directed agents in development recognize IgV domain. Our results from pediatric AML clinical trial (AAML0531) show significant association of this variant leukemic cell surface CD33 expression as well as GO response. These exciting results holds potential to personalize GO treatment by preemptively identifying patients based on the rs12459419 variant testing, who are likely to benefit from GO. Of special clinical relevance are patients who are carriers of this variant and thus do not benefit from CD33-directed GO and thus should not be considered to receive GO-based therapies. Given our promising results so far, the objective of the current study is to expand on these findings and provide a comprehensive molecular and functional characterization CD33 genetic and splice variants for impact on clinical response to CD33-directed agents as well as to establish the diagnostic application of a CD33 SNP-score based model in prospective clinical trials. We are confident that this highly translational project will have immediate impact on the treatment of AML, and our findings can be readily extrapolated to other CD33-directed therapies in cancer.