Search for a command to run...
Venous thromboembolism (VTE) is a well-recognized complication of multiple myeloma (MM), particularly in patients with relapsed or refractory disease treated with immunomodulatory drug–based regimens. In this setting, reported VTE incidence ranges from approximately 2% to 8% in the absence of thromboprophylaxis and is reduced to lower rates, generally around 3%–5%, following the introduction of low-molecular weight heparin (LMWH), underscoring the importance of disease phase, treatment exposure and prophylaxis strategy in determining thrombotic risk.1 Against this background, the thrombotic risk associated with B-cell maturation antigen (BCMA)–directed chimeric antigen receptor (CAR) T-cell therapy remains incompletely defined, despite the rapid incorporation of these therapies into routine clinical practice for relapsed or refractory MM due to their high rate of response.2 Patients undergoing CAR T-cell therapy typically represent a heavily pretreated population with advanced disease, frequent exposure to prior thrombogenic regimens and accumulation of patient-related and treatment-related risk factors for thrombosis, including advanced age, central venous access, infections, immobilization and prolonged cytopenias.3 In addition, CAR T-cell therapy is associated with unique inflammatory toxicities, most notably cytokine release syndrome (CRS), which is characterized by systemic inflammation, endothelial activation and coagulation abnormalities. Experimental and clinical observations have linked CRS to transient disturbances in haemostatic balance, raising concern that CAR T-cell therapy might exacerbate the underlying thromboinflammatory state already present in MM.4 However, the extent to which these mechanisms translate into clinically overt thrombotic events has not been systematically established. To address this question, we conducted a systematic review of published clinical trials and real-world studies evaluating the two approved BCMA-directed CAR T-cell products, idecabtagene vicleucel and ciltacabtagene autoleucel, with a specific focus on the reporting of VTE events. This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and was prospectively registered in the PROSPERO database (CRD420251105194). Searches were performed in PubMed, Scopus and Embase databases. Across 18 eligible studies comprising 2543 patients treated with CAR T-cell therapy, only 10 VTE events were explicitly reported (Table 1). A full table with all the 18 studies and their incidence of VTE is available in Supporting Information. At face value, this corresponds to an overall incidence of 0.43%. The median follow-up across the included studies ranged from approximately 8 to 12 months in single-arm trials and real-world cohorts, extending to up to 18–19 months in the phase III randomized controlled trials. This reported incidence is substantially lower than that historically observed with other anti-myeloma therapies in relapsed or refractory disease. However, closer examination of the source and context of these events reveals important limitations that substantially affect interpretation. 1 PE 1 DVT 3DVT 1PE 4 PE 2 DVT Notably, all reported VTE events occurred in only five studies, all of which were prospective, industry-sponsored clinical trials with predefined safety monitoring and structured adverse event reporting. These trials included the pivotal KarMMa and CARTITUDE studies5-8 in which adverse events were systematically captured and reported through formal trial databases and regulatory submissions. In contrast, the remaining 13 studies, predominantly retrospective or real-world cohorts, did not report a single VTE event among more than 1800 CAR T-cell–treated patients. Given the established thrombotic risk associated with MM and the advanced disease characteristics of these populations, the complete absence of reported events in these cohorts is highly unlikely to reflect a true absence of thrombosis and instead strongly suggests under-ascertainment or incomplete reporting. This distinction between absence of reported events and absence of events is central to the interpretation of the available literature. Many real-world studies were designed to evaluate efficacy, durability of response or feasibility of CAR T-cell delivery rather than thrombotic outcomes. In most cases, VTE was not listed as a prespecified end-point, adjudication procedures were not described and follow-up relied on retrospective chart review rather than systematic event surveillance. Consequently, thrombotic complications may have gone unrecognized, undocumented or unreported, particularly if they occurred outside the immediate post-infusion hospitalization or were managed at external institutions. When analysis is restricted to prospective clinical trials with systematic adverse event capture, a different picture emerges. In these studies, VTE incidence ranged from approximately 1% to 3%, depending on product and follow-up duration (Table 2). Although numerically low, these rates are consistent with contemporary estimates for relapsed or refractory MM patients receiving pharmacological thromboprophylaxis, particularly LMWH, as recommended by international guidelines.9 Importantly, most CAR T-cell trials did not mandate CAR T-cell–specific thromboprophylaxis protocols. When anti-coagulation was recommended, it was generally driven by prior or concurrent exposure to immunomodulatory drugs rather than by the CAR T-cell phase itself. As such, the observed VTE rates should be interpreted within the context of background prophylaxis practices rather than as evidence of a CAR T-cell–specific protective effect. The relevance of appropriate contextualization is further underscored by historical data on VTE risk in MM. Previous data indicate that, even in advanced disease, effective thromboprophylaxis can substantially mitigate thrombotic risk.10 Therefore, the low VTE incidence reported in prospective CAR T-cell trials appears biologically plausible when viewed against this prophylaxis-modulated baseline, rather than representing a departure from established MM-associated risk. More recent real-world evidence further supports this interpretation. A contemporary cohort of 108 patients with MM and AL amyloidosis treated with an academic BCMA-directed CAR T-cell product11 reported a 2.8% incidence of VTE within 3 months of infusion. Importantly, this study included a particularly high-risk population, with a substantial proportion of patients not meeting eligibility criteria for pivotal CAR T-cell trials. Thrombotic events were actively captured, timing relative to CAR T-cell infusion was documented and thromboprophylaxis strategies were explicitly reported. These findings provide a valuable counterpoint to larger real-world cohorts reporting zero events and illustrate that, when thrombotic outcomes are systematically assessed, VTE incidence remains low but clearly non-zero. Taken together, these observations suggest that the currently available literature likely underestimates the true incidence of VTE following BCMA-directed CAR T-cell therapy, particularly outside the setting of prospective clinical trials. The apparent discrepancy between prospective and retrospective studies highlights the limitations of relying on passive or non-standardized reporting for relatively infrequent but clinically meaningful complications. Importantly, the data do not support the conclusion that CAR T-cell therapy eliminates thrombotic risk nor do they establish that standard thromboprophylaxis is universally sufficient in this setting. Rather, they indicate that observed VTE rates are compatible with those expected in a heavily pretreated MM population receiving background prophylaxis, while emphasizing substantial gaps in reporting quality. These limitations have important implications as CAR T-cell therapies move earlier in the treatment course of MM12 and are increasingly administered to broader patient populations, including those with higher baseline thrombotic risk. Without standardized definitions, prospective adjudication and explicit reporting of thromboprophylaxis strategies, it will remain difficult to accurately quantify VTE risk or to identify patient- or treatment-specific modifiers. Future studies should, therefore, incorporate predefined thrombotic end-points, systematic event capture and transparent documentation of anti-coagulation practices to enable meaningful comparisons across trials and real-world cohorts. In conclusion, reported VTE incidence following BCMA-directed CAR T-cell therapy in MM is low, but likely underestimated, particularly in real-world studies. When thrombotic events are actively sought and documented, VTE rates appear comparable to those expected in relapsed or refractory MM patients receiving standard thromboprophylaxis. These findings highlight the need for improved methodological rigor in the assessment and reporting of thrombotic complications in CAR T-cell therapy, rather than supporting definitive conclusions regarding thrombotic safety or prophylaxis adequacy. Pedro Henrique Fernandes do Carmo Las Casas: Conceptualization, investigation, writing, original draft, methodology, writing, review & editing, formal analysis, project administration and supervision. Javier Marco-Ayala: Investigation, writing—original draft, writing—review & editing, validation, methodology and project administration. Elmina-Eleftheria Lefkou: Visualization and writing—review & editing. Thamiris Silva Soares: Writing—original draft and Investigation. Mohammed A. Baghdadi: Investigation. Prakasha Kempaiah and Jawed Fareed: Supervision. Patrick Vandreden: Supervision and writing—review & editing. Vania Hungria: Writing—review & editing, methodology, formal analysis and project administration. Marina Marchetti and Anna Falanga: Writing—review & editing and formal analysis. Laurent Garderet: Formal analysis and supervision. Grigoris Gerotziafas: Supervision. The authors would like to thank the Saint-Antoine Research Center (CRSA) for their assistance and all collaborating centres for their data contributions. The authors declare that they have no conflicts of interest regarding this paper. All data supporting the findings of this study are available from the corresponding author upon reasonable request. Data S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.