Lineage Tracing Sheds Light on Factors Affecting CAR-T Expansion in B-cell Malignancies

A chimeric antigen receptor T-cell (CAR-T) atlas created from data derived from patients with large B-cell Lymphoma (LBCL) or acute lymphoblastic leukemia (ALL) treated with CAR-T therapies has revealed factors affecting CAR-T expansion. All told, the knowledge gained from the atlas may help to improve treatment response durability in LBCL and ALL and carries implications for CAR-T therapy manufacturing.

The data were presented by Zinaida Good, PhD, an instructor at Stanford Institutes of Medicine, at the American Association for Cancer Research (AACR) Annual Meeting 2023, held April 14-19, 2023, in Orlando, Florida. The unified T-cell atlas was based on data from 22 patients who received treatment with either the commercial CAR-T axicabtagene ciloleucel (axi-cel; Yescarta) (n=13) or CD19/22-CAR (n=9), an investigational bispecific CAR-T product currently being evaluated in an investigator-initiated trial (NCT03233854). 

To Good and her colleagues’ knowledge, it was the largest CAR-T atlas ever created, with more than 800,000 cells in total and more than 200,000 unique T-cell receptor (TCR) clonotypes. The goal of the research was to gather information that could potentially be useful for improving durable responses in LBCL and ALL; Good noted that although more than 80% of patients achieve a complete response (CR) with CAR-T treatments for LBCL, currently only about 40% of patients treated with CAR-T therapies for LBCL have CRs that remain durable for 6 months or more.

“Importantly, we detected over 17,000 TCR clonotypes in CAR-positive cells at 2 or more timepoints, enabling lineage tracing at scale for the first time” Good said during her presentation. Good and colleagues found that T-cells in manufactured CAR-T product (infusion T-cells) had increased clonal diversity compared withthe T-cells initially leukapheresed from the patients before manufacturing (source T-cells). Clonal diversity ultimately decreased after infusion.

Furthermore, Good and colleagues sought to determine the phenotype, based on gene expression, of source T-cells with clonotypes that were expanded 2-fold or more in infusion T-cells. They determined that clonally expanded source T-cells had features of naïve, regulatory, and proliferating T-cells. On the other hand, clonally contracted source T-cells had features of effector T-cells. Naïve, regulatory, and proliferating T-cells were also found to clonally expand from source T-cells to infusion T-cells, whereas effector T-cells were contracted from source T-cells to infusion T-cells.

“The observation of regulatory T-cells expanding during CAR T-cell manufacturing was especially interesting given our recent finding that CAR-positive T-regulatory cells in blood postinfusion are associated with disease progression following CD19 CART-cell therapy,” Good stated. “Further, our work was copublished with a second study led by Marcella Maus, MD, PhD, [an associate professor of medicine at Harvard Medical School and director of the Cellular Immunotherapy Program at Massachusetts General Hospital,] that independently identified CAR-positive T-regulatory cells in infusion product as associated with disease progression. However, it was not known whether these CAR T-regulatory cells derived from preexisting T-regulatory cells or are induced during manufacturing.”

Good noted that the data she presented point to the likely source being preexisting T-regulatory cells. She and her colleagues further confirmed this with T-regulatory cell-specific demethylated region (TSDR) analysis on blood from 6 of the patients treated with axi-cel who had the highest CAR T-regulatory cell content at 7 days postinfusion. It was found that T-regulatory cells had low TSDR methylation indicating the T-regulatory cells likely derived from preexisting thymic derived T-regulatory cells. This conclusion was additionally supported by lineage tracing. Good stated that these findings have “important implications for future CAR T-cell manufacturing.” In response to an audience question, Good added that she and her colleagues will conduct further experiments to evaluate whether depleting T-regulatory cells in source T-cells could be beneficial for CAR-T efficacy.

In addition, Good and colleagues applied “grouping of lymphocyte interactions by paratope hotspots”(GLIPH) algorithms to their dataset in collaboration with Shin-Heng Chiou, PhD, an assistant professor at Rutgers Cancer Institute of New Jersey. Flu-specific and Epstein Barr virus-specific clones were found to be enriched in expanded in source T-cells. Cytomegalovirus-specific clones were found to be enriched to a lesser extent in expanded source T-cells. On the other hand, Dengue-specific T-cells were not enriched in expanded T-cell clones. Good pointed out in response to an audience question that this area of the research will need to be investigated further before the specifics of the work’s implications for manufacturing of CAR-T therapies can be drawn.

Click here for more coverage of AACR 2023.

REFERENCE
Good Z. Lineage tracing of CAR T cells in patients with B cell malignancies. Presented at: AACR Annual Meeting. April 14-19, 2023; Orlando, FL. Abstract 1128.