A road map for navigating CAR T hematotoxicity

In this issue of Blood, Rejeski et al, representing an international panel of experts from the European Hematology Association (EHA) and the European Society for Blood and Marrow Transplantation (EBMT), review available data on chimeric antigen receptor (CAR) T cell–associated cytopenias, which they define as immune effector cell–associated hematotoxicity (ICAHT), and assemble recommendations on grading, predicting, preventing, evaluating, and managing ICAHT.¹ 

This is the first such guideline by a major organization and is a much-needed development for the management of this important CAR T cell–associated toxicity. Neutropenia is the most common toxicity observed after CAR T therapy. Infections are the most common cause of nonrelapse mortality in CAR T recipients, more frequently observed than cytokine release syndrome (CRS) or deaths related to immune effector cell–associated neurotoxicity (ICANS), and they are often observed in the setting of severe and prolonged neutropenia.^2,3 Whereas other CAR T cell–associated toxicities such as CRS and ICANS have been heavily studied with robust guidelines, post–CAR T therapy cytopenias are less understood, despite their clinical significance.

It is important to note that the cytopenias seen after CAR T therapy are distinct from standard chemotherapy-associated cytopenias. A neutropenia-predominant, biphasic presentation is commonly seen, with early cytopenias within the first 2 weeks of CAR T therapy and a late recurrence often more than 30 days after treatment.^4,5 A small subset of patients suffer persistent, severe neutropenia, with associated high risk of severe infection. Although lymphodepleting chemotherapy and acute inflammation related to cytokine release provide ready explanations for early cytopenias, the pathophysiology of late hematotoxicity remains poorly understood. There are likely a combination of factors at play in late ICAHT, including limited marrow reserve, immune-driven suppression, and marrow microenvironmental changes, along with other potential factors in subsets of patients such as hemophagocytic lymphohistiocystosis/macrophage activation syndrome, clonal hematopoiesis of indeterminate potential, and secondary malignancies.⁶ 

Perhaps most importantly, this work defines the problem specifically and provides a novel grading system that can be implemented in studies and clinical trials (see figure). The system is feasible, relying only on absolute neutrophil count and timing. The grading scale distinguishes “early” ICAHT within 30 days of CAR T administration and “late” ICAHT, occurring more than 30 days after CAR T therapy. These classifications better capture the biphasic physiology of post–CAR T therapy cytopenias, which is often observed, and offers an improvement over conventional grading scales such as the Common Terminology Criteria for Adverse Events,⁷ which only captures the depth of neutropenia without regard for timing or duration.

The authors also highlight the use of the CAR-HEMATOTOX prognostic score, which uses easily measurable lab values obtained prior to lymphodepletion to estimate a patient’s risk of experiencing prolonged neutropenia.^3,5 The CAR-HEMATOTOX score has been validated in large B-cell lymphoma, mantle cell lymphoma, and multiple myeloma, although further study is needed to confirm whether it is prognostic in indolent non-Hodgkin lymphoma or acute lymphoblastic leukemia. The scoring system has shown high sensitivity but lower specificity for predicting patients with severe and prolonged neutropenia after CAR T therapy; nearly all patients with severe, prolonged neutropenia were identified as being at high risk, but many patients identified as high risk did not have prolonged neutropenia, whereas low-risk patients have minimal risk of prolonged neutropenia. Further study is needed to determine how best to leverage this score and whether preventative strategies, such as earlier colony-stimulating factor use, more aggressive antibiotic prophylaxis, immunomodulating therapies, or stem cell collection prior to CAR T therapy for subsequent stem cell boost, are effective approaches.

Important approaches to the management of ICAHT are covered, including the use of prophylactic antibiotics, colony-stimulating factors, and transfusions. One of the most challenging scenarios in ICAHT remains that of late, aplastic disease, with persistent cytopenias sometimes lasting months after treatment. A reflection of this is their stated "ultima ratio,” their last resort: the use of allogeneic transplant. The authors recommend initiation of a donor search if grade 4 ICAHT persists beyond day 30, with a time frame between months 3 to 6 post–CAR T therapy deemed reasonable to consider transplant. They point out that this is a highly individualized decision based on donor availability, patient fitness, comorbidities, and disease status. We agree that this is a difficult decision, given the high risks of mortality associated with allogeneic transplant; our practice has generally been to defer transplant to 6 or more months post–CAR T therapy, based on our own experience and data suggesting that late recoveries between 3 and 6 months still commonly occur.⁸ 

While providing comprehensive recommendations for ICAHT, the EHA/EBMT guidelines also highlight important gaps in our current knowledge of ICAHT, which are significant. This is not the destination but an early road map, offering standardized approaches to ICAHT, which will hopefully lead to subsequent preclinical and clinical research to further our understanding and improve outcomes for CAR T recipients.

Conflict-of-interest disclosure: C.J. has served as a consultant for Kite/Gilead, Novartis, BMS/Celgene, Instil Bio, Abintus Bio, Caribou Bio, ImmPACT Bio, Miltenyi, Ipsen, Daiichi-Sankyo, Morphosys, AbbVie, ADC Therapeutics, and Synthekine and has received research funding from Kite/Gilead and Novartis. D.Q. declares no competing financial interests.