In this issue of Blood Advances, Wudhikarn et al1 report on the incidence, types, risk factors, and outcomes of clinically significant infections (CSIs) within the first 100 days after the first infusion of CD19 chimeric antigen receptor (CAR) T-cell therapy (eg, axicabtagene ciloleucel or tisagenlecleucel) for the treatment of relapsed/refractory (R/R) large B-cell lymphoma (LBCL). This large retrospective observational study that spanned almost 5 years (December 2017 to June 2022) revealed a concerningly high incidence of at least 1 CSI (24.9%; n = 834) with a 100-day infection–related mortality of 1.6% (95% confidence interval, 1.2-2.0). Furthermore, infection within the first 100 days after receiving CD19 CAR T-cell therapy was identified as an independent risk factor for reduced overall longer-term survival (ie, beyond 100 days of CAR T-cell therapy).1
CAR T cells are genetically modified T lymphocytes that are engineered to express synthetic receptors that allow the cells to target and eliminate specific tumor cells.2 Although CAR T cells have revolutionized the treatment of LBCL by enhancing long-term survival over standard of care, this therapeutic modality is associated with appreciable side effects.2,3 Cytokine release syndrome (CRS) and immune effector cell–associated neurotoxicity syndrome (ICANS) have long been recognized as acute life-threatening side effects of CAR T-cell therapy; consequently, much effort has gone into their early detection and treatment. As their management has become standardized, it has become obvious to clinicians working in the field that infections represent the next great challenge.4 As a result, many centers have begun to report their infection rates; however, these reports have produced conflicting results largely related to small patient numbers.5 The advantage of the current study by Wudhikarn et al is the large number of included patients (n = 3350), drawn from 121 centers that manage patients undergoing CAR T-cell therapy across the United States, Australia, Belgium, Canada, and Israel. The Center for International Blood and Marrow Transplant Research (CIBMTR) has collected data from >700 000 patients undergoing hematopoietic stem cell transplantation or other cellular therapies.6 Data contributed to the CIBMTR database are audited for accuracy, making it a powerful tool. Therefore, as clinicians, we can have a degree of confidence that the findings of Wudhikarn et al can be taken into consideration when developing infection preventive strategies for patients receiving CD19 CAR T-cell therapy for R/R LBCL.
However, the disadvantage of registry data is that it generally lacks granularity, including precise details of prior treatments for lymphoma (eg, dose and duration of prior corticosteroids and dose and number of cycles of other prior treatments) or prior infections (eg, number, timing, site, severity, organism, and treatment). Moreover, prior infections in this CIBMTR study were recorded only as a composite of documented infection, fever of unknown origin, and infection requiring continued antimicrobial therapy beyond day 0.1 The heterogeneity of this variable is large. Detailed data on the treatment of infections that occurred in the first 100 days post-CD19 CAR T-cell therapy are also not available.1 These issues likely confound the findings on the risk and outcome of infections after CD19 CAR T-cell therapy in the study by Wudhikarn et al. Given this information, clinicians should be encouraged to collect detailed data on infections, both before and after the administration of CAR T cells, at their center to supplement these findings of Wudhikarn et al. A detailed analysis regarding the relationship and sequence between CRS, ICANS, and infections needs to be included as well. In addition, the use of a core, standardized yet detailed data set7 will facilitate the pooling of data across multiple centers. Such an approach can strengthen data accuracy to inform the development of center-specific strategies to prevent infection in this patient population.
Wudhikarn et al report that 38 (1.1%) patients had fungal infection of the respiratory tract. In the upper respiratory tract, Candida and Candida-like species were the most commonly detected fungal organisms and were the second most common fungi, after Aspergillus, in the lower respiratory tract.1,Candida and Candida-like species rarely cause invasive infection in the respiratory tract, and their detection in this study most likely represents contamination from the mouth or colonization.8 This finding further emphasizes the importance of detailed data collection to determine the import of any isolated organisms and place the findings correctly in the clinical context.9 This finding, however, may still influence prophylaxis guidelines in the sense that topical oral antifungal agents against Candida and Candida-like species may have utility in patients receiving CAR T-cell therapy.
As the use of CAR T-cell therapy expands to other hematological malignancies (eg, multiple myeloma) and to earlier stages of disease (ie, as first-line treatment),10 ongoing national and international infection surveillance is critical to detect changes in infection epidemiology, risk factors, and outcomes. Furthermore, any implemented infection preventive strategies, including vaccination and immunoglobulin replacement, will change the epidemiology, risk, and outcomes of infections after CAR T-cell therapy over time. This adds further strength to the critical importance of ongoing surveillance to update infection preventive strategies on an ongoing basis.
Further research in this domain requires a detailed analysis of baseline factors that place patients at high risk of developing infections after CAR T-cell administration. This will facilitate the implementation of strategies to prevent infections early (ie, at the time of CAR T-cell infusions), which is likely to further reduce the incidence of infections and improve outcomes.
In conclusion, even though the study of Wudhikarn et al has limitations, to date it still provides clinicians with the best real-world data on infectious complications in patients receiving CD19 CAR T-cell therapy for R/R LBCL. The authors are to be applauded for their efforts and publication. The data generated by Wudhikarn et al can be used by CAR T-cell centers globally to inform the development of infection preventive strategies aimed at improving the long-term survival of these vulnerable patients.
Conflict-of-interest disclosure: The author declares no competing financial interests.
