Abstract Acute myeloid leukemia (AML) therapy requires very intensive chemotherapy/stem cell transplant that cures only 60% of children and 25% of adults with AML. The development of chimeric antigen receptor (CAR) T cells for AML has been hampered by the lack of optimal surface antigens. We hypothesized that unbiased identification of AML cell surface proteins will enable creation of novel CAR T cell immunotherapies and that integration of AML surfaceome data with AML RNA and DNA sequencing data will facilitate identification of new therapeutic targets. We also hypothesized that pediatric and adult surfaceomes are overlapping but distinguishable. Sucrose gradient ultracentrifugation was used for surfaceome enrichment of patient leukemia and normal bone marrow samples without chemical labeling or affinity purification followed by GeLC-MS. The MOLM-14 human AML cell line served as internal control. Mass spectrometry data were processed with MaxQuant, and 9 GO terms were used to define putative cell surface proteins. Samples with at least 6 of 12 known AML surface markers detected were further analyzed. RNA and DNA were extracted for a subset of samples and sequenced. To date, we have surfaceome data for 25 pediatric AML, 12 pediatric normal bone marrow, 9 adult AML, and 9 MOLM-14 samples. A median of 2050 ± 319 surface proteins were detected in the MOLM-14 samples. The mean Pearson correlation coefficient for these replicates was 0.82, confirming the robustness of our workflow. In addition, we found a high representation of known AML cell surface proteins among the captured surface proteins, such as CD33, CD123 and FLT3. A median 1269 ± 474 and 1994 ± 480 surface proteins were found in the pediatric primary AML and normal control samples, respectively and 1790 ± 553 proteins were identified in adult primary AML samples. The average correlation coefficients for the pediatric AML, pediatric control, and adult AML samples were 0.63, 0.75, and 0.63, respectively. Comparison of the pediatric AML and control samples revealed 208 unique and 19 differentially expressed (>2-fold, p<0.05) surface proteins in the patient samples. Similarly, comparison of pediatric and adult AML proteomes identified 210 unique and 23 differentially expressed surface proteins in the pediatric samples. RNA and DNA sequencing has also been performed on a subset of the pediatric and adult primary AML samples, and proteogenomic integration of the different datasets is in progress. We demonstrate that the surfaceome of primary pediatric and adult AML patient samples can be surveyed at depth for the discovery of new potential immunotherapeutic targets. The presence of unique or differentially expressed surface proteins increases our optimism that targeting AML with potent immunotherapies may become clinically feasible. A proteogenomic integration of these AML surfaceome and RNA and DNA sequencing datasets may illuminate AML biology as well as present novel therapeutic targets. Citation Format: Tina Glisovic-Aplenc, Kevin Nestler, Congcong Lu, Lusha Cao, Joseph Cesare, Hyoungjoo Lee, Mitchelle Matesva, Terzah Horton, Sarah Tasian, John Shern, Saar Gill, Benjamin Garcia, Richard Aplenc. Surveying the AML surfaceome for novel immunotherapeutic targets [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5120.
Read full abstract