PDF
Citation
Print
Introduction: CAR T‑cell therapy and emerging infectious challenges
Chimeric antigen receptor (CAR) T-cell therapy has transformed the treatment landscape of hematological malignancies and is now widely used for several high-risk and relapsed diseases. It was first approved for patients with relapsed or refractory large B-cell lymphoma (LBCL) and B-cell acute lymphoblastic leukemia (B-ALL) and has since been expanded to include multiple myeloma (MM). In heavily pretreated patients, CAR T-cell therapy has substantially improved response and survival rates [1–7]. As indications continue to expand and eligibility criteria are broadened to include older and more comorbid patients, the global use of CAR T-cell therapy continues to grow [8–10].
With improved survival, non-relapse mortality (NRM) has emerged as a critical outcome [11, 12]. Although early attention was focused on disease control and acute toxicities, such as cytokine release syndrome (CRS) and immune effector cell–associated neurotoxicity syndrome (ICANS), longer follow-up has clarified that infections are a major contributor to NRM [11–13]. A recent metaanalysis reported that infections accounted for approximately half of non-relapse deaths after CAR T-cell therapy, exceeding mortality attributable to CRS, ICANS, and immune effector cell–associated hemophagocytic syndrome (IEC-HS) [11]. Treatment-related mortality in clinical trials generally ranges from 0 to 6%, whereas real-world NRM rates may reach 5% to 10%, highlighting infection as a key determinant of outcomes after CAR T-cell therapy [3, 11, 12]. Importantly, infectious complications occur across distinct temporal phases that reflect evolving immune dysfunction resulting from lymphodepletion, CAR T-cell activity, and immunosuppressive therapies [3, 14, 15].
Burden of infection after CAR T‑cell therapy: incidence, timing and clinical impact
Infectious complications are common after CAR T-cell therapy and remain important contributors to morbidity and NRM. Across trials and real-world cohorts, reported infection rates range from approximately 15% to 50%, reflecting differences in patient populations, CAR T-cell constructs, follow-up duration, and supportive care practices [14, 15, 18, 19]. Infection-related mortality accounts for approximately 2–4% of deaths and is a significant component of NRM [11, 18, 19]. NRM also varies according to disease type, with higher rates reported in mantle cell lymphoma (10.6%) and MM (8.0%) than in LBCL (6.1%) or indolent lymphoma (5.7%) [11].
The infection risk is time-dependent and closely linked to the evolving immune dysfunction that follows CAR T-cell therapy. Three clinical phases are commonly described: early (< 30 days), prolonged (30–90 days), and late (> 90 days) [14–16]. This temporal framework can help guide strategies for infection surveillance, antimicrobial prophylaxis, and vaccination.
During the early phase (days 0–30), patients experienced a hematologic nadir following lymphodepleting chemotherapy and CAR T-cell infusion. Profound neutropenia and lymphopenia are nearly universal and represent the major drivers of the risk of early infection [3, 16]. Neutrophil recovery typically begins within 2–4 weeks, although recovery kinetics vary widely [20–22]. Some patients demonstrate a prompt and sustained recovery, whereas others develop delayed or biphasic cytopenia characterized by secondary decline after initial improvement [23]. These patterns may reflect bone marrow suppression, inflammatory signaling associated with CAR T-cell expansion, or limited marrow reserves from previous therapies [20, 21, 24]. Prolonged neutropenia is strongly associated with an increased risk of bacterial and invasive fungal infections (IFI) [14, 16].
Clinically, bacterial infections predominate in the early phases and are the leading cause of infectious morbidity [3, 15]. Gram-negative bacteremia, including Escherichia coli and Pseudomonas aeruginosa, as well as gram-positive infections, such as Staphylococcus and Enterococcus species, are commonly reported. These infections are often present as bloodstream pneumonia or catheterassociated infections [19]. Frequent exposure to antibiotics and gastrointestinal toxicity also increase the probability of Clostridioides difficile (C. difficile) infection, which should be considered in patients with similar symptoms. Early fungal infections are less common but may occur in patients with prolonged neutropenia or prior extensive immunosuppression, with Candida species being the most frequently reported [3, 25]. Viral infections are less prominent in the early phase but may emerge in patients who require high-dose corticosteroids or cytokine-directed therapy for severe CRS or ICANS [26, 27]. Reactivation of cytomegalovirus (CMV) infection has been described in selected high-risk populations within the first two to three weeks of infusion [26].
Several clinical factors contribute to delayed hematologic recovery and increased risk of infection after CAR T-cell therapy. High disease burden at infusion, extensive exposure to prior chemotherapy, previous autologous or allogeneic hematopoietic stem cell transplantation (HSCT), and baseline cytopenia are consistently associated with a diminished bone marrow reserve [20, 22, 28]. Severe CRS has also been associated with prolonged cytopenias, likely mediated by cytokinedriven marrow suppression and endothelial activation [20, 29]. Additionally, elevated levels of inflammatory markers such as ferritin and CRP further support the idea that systemic inflammation contributes to bone marrow dysfunction and delayed hematologic recovery after CAR T-cell therapy [21]. Persistent anemia and thrombocytopenia are often accompanied by neutropenia, which contributes to transfusion dependence, recurrent healthcare encounters, and cumulative immunological stress.
During the prolonged phase (days 30–90), neutrophil counts frequently recover but adaptive immune dysfunction persists. B-cell aplasia, hypogammaglobulinemia, and delayed CD4+ T-cell reconstitution continue to impair host defenses [3, 30, 31]. Consequently, viral and opportunistic infections have become increasingly prevalent [15, 16]. Respiratory viral infections, including influenza, respiratory syncytial virus, parainfluenza, and SARS-CoV-2, are the main sources of morbidity during this period and may require hospitalization [3, 15]. Herpes simplex virus (HSV) and varicella-zoster virus (VZV) infections can occur in the absence of antiviral prophylaxis [27]. CMV infections may persist or develop during this phase, particularly in patients receiving prolonged immunosuppressive therapy [26].
Fungal infections during the prolonged phase are less frequent than bacterial infections seen earlier after infusion, but cause substantial mortality when they occur [25]. Invasive mold infections, including Aspergillus species, are most commonly observed in patients with prolonged cytopenia, prior allogeneic hematopoietic cell transplantation, or exposure to high-dose corticosteroids [3, 25]. Pneumocystis jirovecii (PJP) has also emerged as a clinically relevant opportunistic infection during this window, particularly in patients who have not received adequate prophylaxis or with delayed T-cell recovery [16, 32].
Beyond 90 days after CAR T-cell therapy, the infection risk decreases but does not resolve completely [14, 30]. Late infections are largely driven by persistent humoral and cellular immune dysfunction rather than cytopenia [30, 33]. Persistent B-cell aplasia and hypogammaglobulinemia predispose patients to recurrent sinopulmonary infections, often caused by encapsulated bacteria [17, 30]. These infections may be repetitive and tend to contribute more to chronic morbidity than to acute mortality [30]. Viral susceptibility also persists in the late phase with an ongoing risk of respiratory viral infections and herpes virus reactivation [3, 30]. Vaccine responses are frequently suboptimal in this population, further complicating infection prevention efforts [30]. Opportunistic infections such as PJP or invasive fungal disease are rare, but may occur in patients with sustained immunosuppression or delayed immune reconstitution [30].
Together, these temporal patterns highlight that the risk of infection after CAR T-cell therapy is dynamic and closely related to the pace of immune recovery [14, 15]. Therefore, a time-adapted framework for infection surveillance and prophylaxis is essential to minimize NRM and improve long-term outcomes in CAR T-cell recipients [14]. These phases, their associated immune deficits, and infection risks are summarized in Fig. 1.
Pathophysiology of infection risk after CAR T‑cell therapy
Immune dysregulation
Susceptibility to infection after CAR T cell therapy reflects dynamic immune dysregulation rather than a single immunological defect. Lymphodepleting chemotherapy, CAR T-cell–mediated cytotoxicity, and immunosuppressive therapies used to manage therapy-related toxicities collectively shape the timing and spectrum of infections [3, 14, 15].
Early risk is primarily driven by profound cytopenia induced by lymphodepleting regimens. Nearly all patients develop neutropenia and lymphopenia, which alter innate and adaptive immune responses [14, 15]. Delayed or biphasic cytopenia may extend the vulnerability to the prolonged phase.
Delayed CD4+ T-cell reconstitution, including prolonged CD4+ lymphopenia and functional impairment, further increases susceptibility to viral and opportunistic infections, particularly beyond the early post-infusion phase [15, 34].
Immunosuppressive management of complications associated with immune effector cells further increases this risk. Corticosteroids and cytokine-targeted therapies (i.e., tocilizumab, anakinra) impair innate and adaptive immune function in a dose-dependent manner [3, 26, 33]. Patients requiring prolonged or multi-agent immunosuppression represent a particularly high-risk subgroup [11, 26, 32, 33].
Differences in the risk of infection based on CAR T‑cell construct and disease entity
The risk of infection after CAR T-cell therapy is not uniform across all products or disease settings. Increasing clinical experience has highlighted meaningful differences in both the incidence and nature of infections based on the CAR T-cell construct and the underlying malignancy being treated [14, 34]. Among these, recipients of BCMA-directed CAR T-cell therapy appear to carry a particularly high infection burden compared with those treated with CD19-directed CAR T-cell products [11, 34, 35].
Lineage-specific immune defects contribute to susceptibility to infections. CD19-directed CAR T-cell therapy induces prolonged B-cell aplasia, leading to hypogammaglobulinemia and impaired humoral immunity [3, 30, 31, 36]. In contrast, BCMA-directed CAR T-cell therapy directly depletes plasma cells, resulting in a more profound and sustained antibody deficiency [30]. A central mechanistic driver of the increased risk of infection with BCMA-directed CAR T cell therapy is profound humoral immunodeficiency. In plasma cell–driven diseases, such as MM, this effect is further compounded by pre-existing immune dysfunction and cumulative prior therapies, resulting in severe and prolonged hypogammaglobulinemia [37–40]. Comparative studies suggest that although early bacterial infection rates are broadly similar between CD19- and BCMA-directed CAR T-cell recipients, the pattern of infections diverges over time. In BCMAdirected CAR T cell therapy, subsequent infections, particularly respiratory viral and recurrent sinopulmonary bacterial infections, are more frequent and persistent. Opportunistic infections, including PJP and IFI, are also more frequently reported, particularly in patients who require prolonged corticosteroid exposure or additional immunosuppressive therapies [14, 16, 32, 34, 35].
In contrast, CD19-directed CAR T-cell therapy is more commonly associated with B-cell aplasia than with complete plasma cell depletion [3, 41]. Although hypogammaglobulinemia is still frequent, immunoglobulin levels may recover over time, and the infection risk may decline accordingly [14, 41]. Nevertheless, patients with delayed B-cell recovery remain vulnerable to late infections, reflecting the heterogeneity within CD19-directed treatment populations [17, 41].
In addition to the target antigen–related differences, the costimulatory domain may further modulate the risk of infection. CD28-based CAR T-cell constructs are associated with rapid expansion and higher rates of severe CRS and ICANS, often requiring immunosuppressive therapy and contributing to an increased risk of early infection. In contrast, 4-1BB-based constructs exhibit prolonged persistence, which can lead to sustained B-cell aplasia and a greater burden of late infections [42–45].
Disease-specific factors further increase the risk of infection in patients with MM. Patients with myeloma often enter CAR T-cell therapy with advanced age, extensive prior exposure to immunosuppressive agents, baseline cytopenia, renal dysfunction, and compromised marrow reserve. The prior use of proteasome inhibitors, immunomodulatory drugs, monoclonal antibodies, and corticosteroids contributes to cumulative immune exhaustion [32, 39]. In addition, myeloma-related immune paresis and impaired antigen presentation persist even after disease response, limiting immune recovery despite effective tumor control [39, 40].
These overlapping factors help explain why infectionrelated morbidity and mortality appear disproportionately higher in patients with MM undergoing CAR T-cell therapy [11, 35]. Our findings have several practical implications. They support earlier considerations of immunoglobulin replacement, heightened vigilance for viral infections, and more aggressive risk-adapted prophylactic strategies in BCMA-directed CAR-T cell recipients [17, 32]. As newer CAR T-cell constructs and bispecific therapies continue to emerge, understanding construct- and disease-specific infection risks is essential to refine prevention strategies and minimize NRM [46, 47]. The mechanistic differences in humoral immune depletion with CD19- versus BCMA-directed CAR-T cell therapy are illustrated in Fig. 2.
A similar pattern of disease-related vulnerability has been observed in other lymphoma subtypes, particularly mantle cell lymphoma (MCL). Patients with MCL receiving CAR T-cell therapy often represent an older and heavily pretreated population with limited bone marrow reserve and baseline immune dysfunction. Although product-related factors, including the toxicity profile of brexucabtagene autoleucel, may contribute to the higher NRM observed in the MCL group, it is more likely driven by cumulative patient vulnerability and infection-related complications, which ultimately translate into an increased burden of severe and sometimes fatal infections [48].
Hematologic recovery and infection susceptibility
Hematologic recovery after CAR T-cell therapy is heterogeneous and plays a central role in determining susceptibility to infection across all post-infusion phases. Although early cytopenias are nearly universal and largely attributable to lymphodepleting chemotherapy, a substantial proportion of patients experience delayed, prolonged, or biphasic cytopenias that extend vulnerability to infection beyond the first month after treatment [21, 22, 49].
Neutrophil recovery following CAR T-cell infusion typically begins within 2–4 weeks; however, recovery kinetics vary widely [22, 49]. Some patients demonstrate rapid and sustained neutrophil recovery, whereas others develop delayed recovery or secondary decline after the initial improvement [50]. These biphasic patterns are increasingly recognized and may reflect inflammatory signaling associated with CAR T-cell expansion or limited bone marrow reserve from prior therapies [21, 24, 49]. In patients with prolonged or unexplained cytopenia, a bone marrow evaluation may be required to assess the marrow reserve and exclude alternative etiologies [49]. Importantly, prolonged neutropenia is strongly associated with an increased risk of bacterial and IFI [14, 16].
Several clinical factors have been associated with delayed or prolonged cytopenia after CAR T-cell therapy, including a high disease burden at infusion, extensive prior chemotherapy exposure, prior autologous or allogeneic HSCT, and baseline cytopenia [21, 28]. Severe CRS has also been associated with prolonged cytopenias, likely mediated through cytokine-driven marrow suppression and endothelial activation [29, 49]. Elevated levels of inflammatory markers such as ferritin and CRP are similarly associated with delayed hematologic recovery [22].
These observations have led to the development of risk stratification tools to better predict hematological toxicity after CAR T-cell therapy. One such model, the CAR-HEMATOTOX score, integrates baseline hematological and inflammatory parameters to estimate the risk of immune effector cell–associated hematotoxicity. Higher CAR-Hematotox scores are associated with prolonged cytopenia, severe infections, and increased NRM. Prospective validation studies have further supported its ability to identify patients at high risk of delayed hematologic recovery and infection-related complications [32, 34, 51].
The clinical consequences of prolonged cytopenia extend beyond those of neutropenia. Persistent anemia and thrombocytopenia contribute to transfusion dependence and increased healthcare utilization, whereas recurrent hospitalization and prolonged antimicrobial exposure may further increase the risk of resistant infections. These factors may collectively increase the risk of infection during the prolonged and late post-infusion phases [22, 49].
Granulocyte colony-stimulating factor (G-CSF) is frequently used to support neutrophil recovery after CAR-T cell therapy, although its optimal timing remains debatable. Early concerns suggested that G-CSF may exacerbate CRS or ICANS by enhancing myeloid activation. However, emerging clinical experience suggests that delayed administration, typically after resolution of CRS and ICANS, does not significantly increase toxicity and may shorten the duration of neutropenia [50, 52]. Consequently, many centers avoid G-CSF during the first 5 days after infusion and initiate G-CSF treatment approximately 7–14 days (or later) in patients with persistent grade 3–4 neutropenia and high infection risk [17, 28, 53].
Despite evolving experience, no universally accepted strategy has been proposed for managing post–CAR T-cell cytopenia. Therefore, clinical decisions regarding the use of growth factors, transfusion support, and infection prophylaxis are often individualized, based on patient-specific risk factors and institutional practices [33, 49]Understanding cytopenia kinetics and marrow reserve provides essential context to interpret risk of infection across post-infusion phases and informs subsequent discussions on targeted prophylaxis and monitoring strategies [21, 22].
Antimicrobial prophylaxis and vaccination: evidence‑based and risk‑adapted prevention
Antimicrobial prophylaxis is a central component of infection prevention after CAR T-cell therapy, but must balance efficacy with toxicity, antimicrobial resistance, and risk of overtreatment [16, 33]. Current preventive strategies are largely extrapolated from HSCT and intensive chemotherapy settings, although specific evidence relative to CAR cells and guideline recommendations are emerging [54, 55]. Prevention of infection after CAR T-cell therapy generally follows a phaseadapted and risk-stratified approach. Key international guideline recommendations for antibacterial, antiviral, and antifungal prophylaxis from the European Society for Blood and Marrow Transplantation (EBMT) and the American Society for Transplantation and Cellular Therapy (ASTCT) are summarized and compared in Table 2. Practical antimicrobial prophylaxis and vaccination strategies after CAR T-cell therapy are summarized in Table 3.
Antibacterial prophylaxis
Fluoroquinolone prophylaxis is selective, rather than universal. The National Comprehensive Cancer Network (NCCN) and the American Society of Clinical Oncology (ASCO)/Infectious Diseases Society of America (IDSA) recommend fluoroquinolone prophylaxis use during severe neutropenia, considering institutional resistance patterns and Clostridioides difficile infection [58]. In CAR T-cell recipients, antibacterial prophylaxis is typically reserved for patients with an absolute neutrophil count (ANC) < 500/μL and is discontinued upon neutrophil recovery [16, 54, 59]. Empiric broad-spectrum antimicrobial therapy should not be delayed in febrile patients even when CRS is suspected [22].
Antiviral prophylaxis
HSV and VZV prophylaxis with acyclovir or valacyclovir is universally recommended, typically initiated with lymphodepleting chemotherapy and continued for at least 1 year [32, 54]. This recommendation reflects prolonged T-cell dysfunction beyond the neutropenic period [16, 27]. Routine prophylaxis for cytomegalovirus (CMV) is not recommended; instead, a risk-adapted surveillance approach is generally employed in selected high-risk patients. Further details regarding CMV reactivation patterns and management are discussed in Sect. "CMV Reactivation" [26, 60–63].
Antifungal prophylaxis
Fluconazole is commonly used to prevent candidiasis [16, 32]. Mold-active prophylaxis is generally reserved for patients with established risk factors, including prolonged neutropenia, prior invasive fungal infection (IFI), prior allo-HSCT, or prolonged or high-dose corticosteroid exposure [16, 51, 64]. In addition to the EBMT and ASTCT guidelines, the updated ECIL-10 recommendations provide a more structured, risk-adapted framework for antifungal prophylaxis in CAR-T cell recipients. Notably, ECIL-10 distinguishes between pre-infusion factors (e.g., prior therapies, baseline cytopenia, and prior IFI) and postinfusion factors (e.g., CRS requiring corticosteroids, prolonged neutropenia, or the use of additional immunosuppressive agents) and suggests that mold-active prophylaxis should be considered when such risk factors are present. This approach extends previous guideline recommendations by incorporating dynamic treatmentrelated risk factors specific to CAR T-cell therapy [51].
Vaccination after CAR T‑cell therapy
Current guidelines and expert consensus statements suggest that vaccination after CAR T cell therapy should be considered in the context of delayed and variable immune reconstitution. Inactivated vaccines are generally considered safe and are typically initiated approximately 3–6 months after infusion, whereas live vaccines should be deferred following more complete immune recovery, often 12–24 months post-therapy [65–70].
Vaccine responsiveness is frequently reduced after CAR T-cell therapy, largely reflecting impaired humoral immunity [67, 71, 72]. Although responses may improve with B-cell recovery, serological measures alone do not consistently predict protective immunity [73]. Vaccination is generally used as a complementary strategy along with antimicrobial prophylaxis and intravenous immunoglobulin (IVIG) in high-risk patients [65, 66]. Vaccination of household contacts is also commonly recommended to improve indirect protection [16, 65, 74]. Given the limited prospective data in this setting, current practice often relies on a pragmatic, time-based approach with selective consideration of immune recovery and the clinical context [65, 67].
Role of intravenous immunoglobulin
Hypogammaglobulinemia is a common complication after CD19- and BCMA-directed CAR T-cell therapy and contributes to intermediate and late risk of infection [17, 75]. IVIG provides passive antibody replacement and may enhance opsonization and immune modulation [75, 76]. Current practice favors selective administration [17, 75]. Most guidelines recommend IVIG in patients with serum IgG levels < 400–600 mg/dL accompanied by recurrent or severe infections. The typical dose is 400–500 mg/kg of IVIG every 3–4 weeks, with the practical goal of maintaining serum IgG at > 400–600 mg/dL. BCMA-directed CAR T-cell therapy, particularly for patients with MM, may warrant earlier consideration because of plasma cell depletion and sustained antibody deficiency [17, 32, 40]. Routine prophylactic IVIG is not recommended for any recipients (Table 4).
Infection screening and monitoring
Baseline screening
Pre–CAR T evaluation includes HIV testing, hepatitis B and C serologies, and evaluation of CMV status, with antiviral prophylaxis for patients at risk of hepatitis B reactivation [16, 17, 78, 79]. Screening for tuberculosis, Strongyloides, or endemic fungi infections should be guided by epidemiologic exposure risk [54, 80]. Active infections should be identified and controlled before administering lymphodepleting chemotherapy.
Post–CAR T surveillance
Routine surveillance culture is not recommended for asymptomatic patients. Instead, post-CAR T cell monitoring is guided by clinical status and individual risk factors [32]. CMV polymerase chain reaction (PCR) monitoring is generally performed in high-risk patients. A detailed discussion of CMV reactivation patterns, risk factors, and management strategies is provided in Sect. "CMV Reactivation" [26, 32].
Distinguishing infection from CRS/ICANS
Fever, hypotension, and neurological symptoms overlap between infections and immune effector cell-associated toxicity [3, 16, 17]. Therefore, clinicians should maintain a low threshold for diagnostic evaluation and empirical antimicrobial therapy when an infection is suspected. Biomarkers such as CRP and ferritin are nonspecific and should be interpreted with caution in this setting [17, 33]. Interleukin-6 (IL-6) levels are also of limited diagnostic value, particularly in patients who have received anti-IL-6 therapy, as circulating IL-6 levels may remain elevated due to receptor blockade. In patients with persistent fever after CAR T-cell infusion, careful evaluation of atypical infections, including viral, fungal, and opportunistic pathogens, is essential. At the same time, clinicians should consider the possibility of IEC-HS and perform appropriate laboratory assessments, including liver function tests and fibrinogen levels, to allow timely recognition and management, including anti-IL-1–directed therapy when indicated.
Therefore, infection screening and monitoring after CAR T-cell therapy requires careful baseline evaluation and risk-adapted surveillance during follow-up [36, 80]. Early recognition of infection is essential for guiding diagnostic evaluation and timely antimicrobial therapy in this highly immunocompromised population.
Special clinical scenarios
COVID‑19
CAR T recipients may experience prolonged viral shedding and impaired vaccine responses due to B cell depletion [81–84]. Infection that occurs early after infusion or during periods of immunosuppression carries an increased risk of severe disease [60, 81]. Thus, early antiviral therapy and repeated diagnostic testing may be required in patients with persistent symptoms [61, 84].
The management of SARS-CoV-2 infection before CAR T-cell therapy should be guided by symptom severity. In symptomatic patients, CAR T-cell infusion is generally delayed for approximately 7–14 days until clinical improvement is achieved. Repeated SARS-CoV-2 testing after symptom resolution is not always required. Antiviral therapies, such as remdesivir or nirmatrelvir/ritonavir, should be considered when clinically indicated.
In patients with asymptomatic or mild infections, CAR T-cell therapy may be considered in selected cases, although delayed treatment is generally preferred when feasible. Clinical decision-making should consider symptom severity, evidence of lower respiratory tract involvement, inflammatory markers, and the urgency of treating the underlying malignancy [62].
Invasive fungal infections
Although IFI is relatively rare after CAR T-cell therapy, it is associated with substantial mortality and occurs in a setting of cumulative and evolving immunosuppression. This risk reflects the interplay of multiple baseline and treatment-related factors rather than a single dominant determinant [3, 25]. Recent guideline updates, including ECIL-10, emphasize risk stratification by integrating both pre-infusion (e.g., prior therapies, baseline cytopenia, or prior IFI) and post-infusion factors (e.g., CRS requiring corticosteroids, prolonged neutropenia, or additional immunosuppressive therapy) [51]. The NCCN guidelines for MM further define high-risk scenarios based on treatment-related immunosuppression, including corticosteroid exposure (≥ 10 mg dexamethasone daily for > 3 days within a 7-day period), receipt of ≥ 1 g methylprednisolone per day, more than one dose of tocilizumab, or the use of additional immunosuppressive agents such as anakinra or siltuximab [32]. In this context, persistent respiratory symptoms or unexplained fever should prompt early imaging and evaluation of IFI [63, 85].
CMV reactivation
CMV reactivation is increasingly recognized after CAR T-cell therapy, typically occurring within the first month after infusion, with a reported incidence ranging from approximately 17% to 56% and clinically significant infection in a smaller subset [86–89].
In contrast to allo-HSCT, current guidelines do not support universal CMV surveillance in CAR T-cell recipients [16, 56, 57]. The ECIL-10 recommendations suggest a risk-adapted approach, in which monitoring is considered primarily in CMV-seropositive patients with additional risk factors such as severe CRS, prolonged corticosteroid exposure, use of multiple immunosuppressive agents, or persistent lymphocytopenia. In these patients, CMV PCR monitoring may be performed during the early post-infusion period, typically between 2 and 6 weeks after CAR T-cell therapy, with the frequency adjusted according to the clinical course and immunosuppressive burden [26].
Preemptive therapy is generally initiated in the setting of high or rapidly increasing CMV DNAemia. However, ECIL-10 emphasizes that no universal viral load threshold can be defined and treatment decisions should be based on center-specific cut-offs, considering assay performance and calibration using WHO International Standard Units (IU/mL). From a clinical perspective, it is important to distinguish asymptomatic CMV DNAemia from clinically significant CMV infection and CMV end-organ disease because these entities have different management implications. When treatment is indicated, antiviral options, such as valganciclovir, ganciclovir, and foscarnet, may be considered. In the CAR T-cell setting, drug selection should account for toxicity profiles, particularly myelosuppression associated with ganciclovir-based therapies, which may limit their use in patients with ongoing cytopenia.
Given the absence of prospective CAR-T cell-specific data, current CMV monitoring and management strategies are largely extrapolated from the HSCT experience, which may limit their direct applicability in this setting.
Conclusions
Current infection prevention strategies after CAR T-cell therapy are largely extrapolated from experience with intensive chemotherapy and HSCT and are primarily based on observed infection patterns rather than on evidence from randomized controlled trials. As CAR T-cell therapy continues to expand, more individualized approaches are needed. In this context, integrated patient-specific risk assessments may play an increasingly important role in guiding monitoring and preventive strategies. Biomarkerguided strategies that incorporate dynamic CD4+ T cell counts, B cell recovery, and immunoglobulin levels may help refine risk-adapted prevention, although prospective validation remains necessary. In addition, the infection risk appears to vary across CAR T constructs, particularly with a more profound and sustained humoral deficiency observed in BCMA-directed therapy. As next-generation cellular therapies, including newer targets and dual-target constructs, are introduced into clinical practice, defining their long-term infectious profiles will be an important priority for future research.
XML Download