Abstract B003: Characterization of CD20 transcriptional diversity using short- and long-read sequencing in pediatric acute lymphoblastic leukemia: potential implications for rituximab plus chemotherapy treatment

Abstract Randomized trials in adult patients have shown that rituximab, a monoclonal antibody targeting CD20 (MS4A1 gene), can improve treatment response in de novo Acute Lymphoblastic Leukemia (ALL). These findings have led to clinical trials incorporating rituximab into first-line chemotherapy for pediatric patients with high- or intermediate-risk CD20+ ALL. While immunotherapy in this context might improve survival, resistance to antibody-based therapies remains a major challenge, and its interaction with chemotherapy has not been studied in large cohorts of childhood ALL (cALL). Given the potential toxicities of immunotherapies, identifying predictive markers is critical for precision medicine. We hypothesized that the expression and relative abundance of MS4A1 isoforms influence the response of leukemic cells to rituximab in combination with standard chemotherapy. To test this, we studied MS4A1 transcriptional variability at cALL diagnosis using short-read (Illumina, n=31) and long-read (Oxford Nanopore Technologies ONT, n=2) RNA-seq. We obtained bone marrow aspirates of patients enrolled in the ALLIC-GATLA 2010 clinical protocol. ONT RNA-seq was performed at the Genomics Center of Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, while Illumina RNA-seq was conducted at Macrogen, Korea. We used Salmon, FLAIR and IsoQuant to annotate and quantify known and novel MS4A1 transcripts. Our results showed high heterogeneity among patients, both in transcriptional diversity and total transcript abundance. Notably, the MS4A1-204 isoform, which lacks the rituximab-binding domain, was expressed at diagnosis in >50% of patients. Isoform quantification differed significantly between ONT and Illumina, despite similar relative sequencing coverage of exons. Illumina predicted similar proportions of isoforms 201, 212, and 206, whereas ONT predicted dominance of the 201 isoform. Nevertheless, despite ONT sequencing depth was 6 to 11-fold lower than Illumina, molecular subtype predictions were concordant between both platforms. The discovery of novel MS4A1 transcripts using FLAIR and IsoQuant in ONT-sequenced samples showed differences in transcript identities. Structural analysis of two novel isoforms suggested the coding of putative proteins lacking extracellular or transmembrane domains, which are important for rituximab binding or may interfere with protein function. In conclusion, we successfully implemented ONT transcriptome sequencing in Argentina for ALL molecular subtyping, isoform annotation, quantification and novel transcript discovery. However, we detected high variability between sequencing platforms and bioinformatic tools, highlighting the need for further studies to improve reproducibility. These results are crucial for designing prospective studies evaluating patients treated with rituximab plus chemotherapy, currently underway in a multicentric clinical trial in Argentina. Additionally, this analysis could be extended to whole-transcriptome isoform diversity, complementing the characterization of MS4A1. Citation Format: Maria Sol Ruiz, Lucila Viappiani, Daniel Avendaño, Ignacio Gomez Mercado, Marina L Ingravidi, Geraldine Gueron, Javier Cotignola. Characterization of CD20 transcriptional diversity using short- and long-read sequencing in pediatric acute lymphoblastic leukemia: potential implications for rituximab plus chemotherapy treatment [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Discovery and Innovation in Pediatric Cancer— From Biology to Breakthrough Therapies; 2025 Sep 25-28; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2025;85(18_Suppl_2):Abstract nr B003.