What platelets tell us about CARs Free

In this issue of Blood, Rejeski et al introduce a new grading system, termed T-ICAHT, as an important tool for assessment and quantification of immune effector cell–associated hematotoxicity (ICAHT).¹ 

Chimeric antigen receptor T-cell therapy (CART) has become a cornerstone of salvage treatment of B-cell malignancies, such as B-lineage acute lymphoblastic leukemia, most B-cell lymphomas, and multiple myeloma. However, the high efficacy of CARTs comes at the price of some significant toxicities, including hematotoxicity, which is probably the most important driver of nonrelapse mortality (NRM) associated with CART use.² The most prominent feature of CART-associated hematotoxicity is neutropenia, which can be categorized using the N-ICAHT grading system.³ The N-ICAHT score helps to standardize post-CART neutropenia management and research efforts.³ Moreover, N-ICAHT grading can predict CART outcomes, including infections, NRM, and survival.⁴ 

In addition to neutropenia, acute and prolonged thrombocytopenias frequently occur after CART therapy.^5,6 However, a systematic approach to the categorization and quantification of post-CART platelet deficiency and its implications for clinical outcomes has been lacking to date. The aim of the present study by Rejeski et al was to close this gap by providing T-ICAHT as a grading tool to complement N-ICAHT. Using a large training sample consisting of 744 recipients of commercial CD19-CARTs for predominantly aggressive B-cell lymphoma, they developed a grading system closely aligned to the N-ICAHT structure: T-ICAHT grades were assigned on the basis of depth and duration of thrombocytopenia within the first 30 days after CART administration (early T-ICAHT), and between day 31 and 100 after CART (late T-ICAHT). Higher grade early T-ICAHT was observed in 23% of the patients of the training sample and was associated with adverse baseline factors, such poor performance status, extensive pretreatment, elevated lactate dehydrogenase (LDH), low platelet counts, use of bridging therapies, and high HEMATOTOX and EASIX (endothelial activation and stress index) scores. Although T-ICAHT and N-ICAHT were partially correlated, remarkable discrepancies were noted, as 40% of all patients with grade 3 to 4 early T-ICAHT had only grade 1 to 2 early N-ICAHT, and 4% had no early N-ICAHT at all. Similar relations were found for late T-ICAHT and N-ICAHT, suggesting that there are different biological drivers behind the 2 forms of ICAHT.

Notably, and independent of N-ICAHT, both early T-ICAHT and late T-ICAHT predicted overall and progression-free survival (PFS) measured from the day 30 and day 100 landmarks. Intriguingly, the modest PFS disadvantage of grade 1 to 2 early and late T-ICAHT was largely due to increased NRM, whereas the much more pronounced PFS disadvantage of grade 3 to 4 early and late T-ICAHT was predominantly driven by a markedly increased relapse risk (16% and 20% at 2 years for early and late T-ICAHT, respectively; see figure). Furthermore, the T-ICAHT grading system was validated in 3 additional cohorts, including patients with B-cell lymphoma receiving CD19 CARTs produced by academic centers, patients with multiple myeloma treated with commercial anti–B-cell maturation antigen CARTs, and adult and pediatric patients with B-lineage acute lymphoblastic leukemia (ALL) infused with commercial or investigational CD19 CARTs. Although the proportion of patients experiencing severe early T-ICAHT varied substantially between these 3 cohorts (from 18% in the cohort with lymphoma to 53% in the cohort with ALL), the 2-year survival difference between the patients without early T-ICAHT and those acquiring grade 3 to 4 early T-ICAHT was ≈40% in all 3 samples, confirming the strong adverse impact of severe early T-ICAHT independent of underlying condition and CART product used.

Collectively, these data show that T-ICAHT is a practical system for characterization and quantification of CART-associated thrombocytopenia. T-ICAHT promises to become a valuable tool for standardizing CART-related toxicities in both research and real-world settings. More importantly, it seems to be a powerful, albeit time-dependent, predictor of CART therapy outcomes. Unlike N-ICAHT, which is predominantly associated with NRM, the predictive capacity of T-ICAHT appears to be largely driven by relapse-related mortality.

For the time being, however, T-ICAHT is just a descriptive instrument. Although clearly correlated with several baseline indicators of hematopoietic stress, it does not provide an explanation for the mechanisms behind CART-associated thrombocytopenia. Also, the reasons for the correlation between severe T-ICAHT and increased relapse risk remain unclear. The fact that it retains its adverse effect on PFS even after adjustment for confounders, such as performance status, elevated LDH, and underlying lymphoma, suggests that T-ICAHT is not just a surrogate for poor-risk disease. As speculated by the authors, 1 mechanistic link between thrombocytopenia and impaired survival outcomes could be endothelial distress induced by treatment toxicity and/or lymphoma activity. This hypothesis is supported by the strong correlation between early T-ICAHT and the EASIX score obtained at the time of lymphodepletion, even though this relationship is obscured by the fact that the EASIX score is directly driven by the platelet count.⁷ 

Thus, T-ICAHT raises novel research questions and simultaneously promises to become an important tool to help answering them. Provided it can be further validated in additional prospective and real-world samples, T-ICAHT has the potential—complementary to N-ICAHT—to play an important role in the field of complication assessment and outcome prediction of CART therapies.

Conflict-of-interest disclosure: P.D. declares consultancy for AbbVie, AstraZeneca, BeiGene, Bristol Myers Squibb, Gilead, and Miltenyi (all honoraria to institution); speaker honoraria from AbbVie, AstraZeneca, BeiGene, Bristol Myers Squibb, Gilead, Riemser, and Roche (all honoraria to institution); meeting attendance support from BeiGene and Gilead; participation on a data safety monitoring board for Novartis; and is current chairman of the German Working Group for Hematopoietic Stem Cell Transplantation and Cellular Therapy.