Invasive fungal disease is rare in patients with relapsed/refractory multiple myeloma treated with BCMA CAR T-cell therapy Open Access

B-cell maturation antigen (BCMA)–directed therapies including chimeric antigen receptor (CAR) T-cell therapy have increasingly been applied to patients with relapsed/refractory multiple myeloma with excellent efficacy.^1-3 Despite improved management of toxicities, including cytokine release syndrome (CRS), immune effector cell (IEC)–associated neurotoxicity syndrome (ICANS), IEC-associated hematotoxicity, and hypogammaglobulinemia, infections remain an important concern and the leading cause of nonrelapse mortality, with more than half of the nonrelapse deaths in recipients of BCMA CAR T-cell therapy attributed to infection.^4-7 Unfortunately, the lack of standardized infection reporting and absence of information on antimicrobial prophylaxis in many clinical trials have limited understanding of infection risks associated with these novel immunologic therapies.^8,9 Postmarketing studies that are principally retrospective and single-center have provided additional information on the epidemiology of infections in recipients of CAR T-cell therapy.^10-13 

Fungal infections are relatively rare after CAR T-cell therapy, but the incidence and characteristics of invasive fungal disease (IFD) in recipients of BCMA CAR T-cell therapy remain less well-elucidated, with fewer studies, small numbers of patients (<100), and relatively short follow-up times compared with recipients of CD19 CAR T-cell therapy.^11,14-18 Notably, the incidence of IFD in many published studies may be overestimated in the absence of the use of consensus definitions of IFD, and with the inclusion of noninvasive oral/mucocutaneous candidiasis.^19-22 Nonetheless, fungal infections have been associated with excess mortality, emphasizing the importance of improved understanding of the epidemiology of IFD following BCMA CAR T-cell therapy.^4,5,19,20 

Here we sought to characterize the incidence of IFD in a retrospective cohort of 234 patients with relapsed/refractory multiple myeloma treated with BCMA CAR T-cell therapy at 2 institutions from 1 November 2016 through 18 October 2023 (investigational autologous BCMA CAR T-cell products at the recommended phase II dose; investigational/commercial idecabtagene vicleucel; and investigational/commercial ciltacabtagene autoleucel). CRS and ICANS were graded according to the American Society for Transplantation and Cellular Therapy (ASTCT) criteria.²³ IEC hemophagocytic lymphohistiocytosis-like syndrome (IEC-HS) was defined according to ASTCT criteria.²⁴ Supplemental Table 1 summarizes institutional protocols for antifungal prophylaxis for IEC therapy (institution 1 used no routine antiyeast prophylaxis, and institution 2 used antiyeast prophylaxis with fluconazole/micafungin for patients with CRS/ICANS treated with corticosteroids or prolonged neutropenia >10 days; no azole antimold prophylaxis was used at either institution). Both institutions utilized febrile neutropenia protocols with micafungin initiated for persistent or recurrent neutropenic fever >4 days after initiation of antibacterial therapy through neutrophil recovery. Proven/probable IFD was recorded from day 0 through last follow-up, day 365, or death according to the EORTC/MSGERC consensus criteria.²¹ All autopsies were reviewed for the presence of IFD. This study was approved by the Dana-Farber Cancer Institute Office of Human Research Studies and Institutional Review Board, and research was conducted in accordance with the Declaration of Helsinki.

Among 234 recipients of BCMA CAR T-cell therapy, the median age was 66 years, and 57% were male (Table 1). Patients had a median of 6 prior therapy lines, and 68% had undergone autologous stem cell transplant. Patients received idecabtagene vicleucel (59%), ciltacabtagene autoleucel (33%), or an investigational BCMA-directed CAR T-cell product (8%). Only 12% of patients qualified for fluconazole or micafungin prophylaxis per institutional guidelines. Acute IEC toxicities included CRS (85%), ICANS (15%), atypical neurotoxicity (4%), and IEC-HS (2%). Among those who developed IEC toxicities, 98 (49%) received treatment with corticosteroids including 18 (9%) who received cumulative doses of >1000 mg prednisone equivalent, 132 (67%) received tocilizumab, and 21 (11%) received anakinra. Within 1 year, 44% developed disease relapse and 19% died.

The 1-year cumulative incidence of proven/probable IFD following BCMA CAR T-cell therapy accounting for the competing risks of infection-free death or receipt of additional antineoplastic therapies was 1.72% (95% confidence interval, 0.6-4.5). There were no possible IFD cases: 2 patients received empiric antifungal therapy but did not meet criteria for possible IFD (1 treated for positive 1-3 β-d-glucan without clinical criteria; 1 with alternate diagnosis). Four proven/probable cases were identified, including 1 case of Candida albicans peritonitis and 3 cases of invasive aspergillosis (Table 2). The timing of diagnosis ranged from 6 to 104 days after infusion. All 4 patients had significant IEC toxicity with CRS/ICANS, as well as 1 patient who developed atypical neurotoxicity with parkinsonism, and 1 patient who developed IEC-HS. In addition, all received high doses of corticosteroids (cumulative doses ranging from 200 mg to 6790 mg prednisone equivalent), as well as tocilizumab, anakinra, or both. CAR-Hematotox (CAR-HT) scores were high in 3 of 4 patients who developed IFD. Two patients had complete response to antifungal therapy and were alive at the 1-year censoring time point, and 2 patients died with IFD. One case of aspergillosis was identified on autopsy after the patient died of progressive parkinsonism related to CAR T-cell therapy and Stenotrophomonas sepsis. The second death occurred in a patient with Staphylococcus aureus ventilator-associated pneumonia, progressive IEC-HS, and newly positive galactomannan with associated pulmonary infiltrates.

Here we demonstrate a low incidence of IFD despite infrequent use of antiyeast prophylaxis and no antimold prophylaxis in a large cohort of recipients of BCMA CAR T-cell therapy. All 4 patients who developed IFD in this study had CRS/ICANS requiring tocilizumab or anakinra and multiple doses of corticosteroids, and 2 patients had prolonged atypical CAR-associated toxicities (IEC-HS, parkinsonism) requiring extended immunosuppression and ultimately contributing to their deaths. There were no cases of IFD in patients without CRS/ICANS or those with grade 1 CRS, and IFD occurred primarily in the first 100 days after cell infusion, emphasizing that the toxicities of CAR T-cell therapy may have a greater impact on risk for IFD than the direct immunologic effects of the therapy. Taken together, these findings suggest universal antifungal prophylaxis is likely not needed in patients with multiple myeloma (MM) undergoing CAR T-cell therapy. Tailored approaches based on host risk and employing principles of antifungal stewardship have been applied with success for CD19 CAR T-cell therapy, and might be extended to the BCMA CAR T-cell setting, particularly focusing on the first 100 days in patients with significant IEC-associated toxicities.^22,25,26 

Formal assessment of risk factors for IFD in patients receiving BCMA CAR T-cell therapy was not possible given the low number of events: a recurring challenge across multiple studies of IFD after CAR T-cell therapy.^19,20 One recent study of infections in a combined cohort of recipients of BCMA and CD19 CAR T-cell therapy identified hematopoietic cell transplantation, CRS, and hypogammaglobulinemia as risk factors for IFD, whereas idecabtagene vicleucel was associated with a lower risk of IFD compared with CD19-directed products. However, this study used codes from the International Classification of Diseases, Tenth Revision to identify cases of IFD rather than consensus definitions, which the authors acknowledge were likely to include a substantial number of mucocutaneous candidiasis cases, limiting the applicability of these findings.^13,21 CAR-HT scores have been utilized to predict patients at high risk of severe bacterial infections following CD19 and BCMA CAR T-cell therapy, but have not been validated for independent prediction of IFD.^6,27 Here, 3 of the 4 cases had high CAR-HT scores prior to lymphodepletion, raising the question of whether this might be investigated as a predictor of IFD as well in the future.

Differences in rates of IFD across disease groups and CAR T-cell products are not well established. Although this study indicates that rates of IFD remain low among patients with MM, larger comparative studies utilizing rigorous definitions of IFD may help to clarify the specific risks within unique CAR T-cell therapy populations (lymphoma vs leukemia vs MM). Differential risk in disease groups related to intrinsic states of immunosuppression and prior therapies, as well as specific CAR T-cell products, remains an important consideration for management as these therapies are applied in novel nononcologic settings such as autoimmune disease.²⁸ 

Despite the relative rarity of IFD after CAR T-cell therapy, the associated morbidity and mortality remain high.^12,29 In this study, all autopsies and patient deaths were reviewed for evidence of the presence of fungal infection, with 1 case identified by autopsy alone without prior clinical or microbiologic evidence, emphasizing the importance of maintaining a high clinical index of suspicion in patients at risk, and pursuing early diagnostics and treatment when appropriate. In addition, patients at risk for IFD may have other serious comorbidities impacting their outcomes, as demonstrated by the 2 patients in this study who died with IFD: both with severe IEC-associated toxicities and concomitant bacterial infections that contributed significantly to their deaths.

Contribution: J.S.L., A.M.P., D.D.C., and S.P.H. designed the study; A.M.P., E.B.K., J.S.L., D.D.C., and S.P.H. performed data collection; A.M.P., J.S.L., and S.P.H. performed the analysis; J.S.L. wrote the manuscript; A.M.P., E.B.K., A.J.Y., O.N., S.M., A.S.S., N.C.M., N.R., M.J.F., D.D.C., and S.P.H. edited the manuscript.

Conflict-of-interest disclosure: J.S.L. reports research funding from Merck and consulting for Basilea. A.J.Y. reports consulting for AbbVie, Adaptive Biotechnologies, Amgen, Bristol Myers Squibb, Celgene, GSK, Johnson & Johnson, Karyopharm, Oncopeptides, Pfizer, Prothena, Regeneron, Sanofi, Sebia, and Takeda; and receives research funding from Amgen, Bristol Myers Squibb, GSK, Johnson & Johnson, and Sanofi. O.N. reports consulting or advisory role with Bristol Myers Squibb, Janssen, Takeda, Sanofi, Pfizer, and GPCR Therapeutics Inc.; research funding from Bristol Myers Squibb, Janssen, and Takeda; and honoraria from Pfizer. S.M. reports consulting for Janssen and Pfizer; honoraria for Pfizer; and stock ownership with AbbVie. A.S.S. reports consulting for Novartis. N.C.M. reports consulting for Bristol Myers Squibb, Janssen, Oncopep, Amgen, Karyopharm, Legened, AbbVie, Takeda, and GSK; and serves on the board of directors of, and has stock options in, Oncopep. N.R. reports consulting for AbbVie, Amgen, Bristol Myers Squibb, Janssen, Pfizer, Immuneel, GSK, K36 Therapeutics, Sanofi, and AstraZeneca; and research funding from Pfizer. M.J.F. reports consulting for Novartis, Kite/Gilead, Bristol Myers Squibb, JnJ/Legend, and SOBI; and receives research support from SOBI. S.P.H. reports consulting or advisory board roles for Pfizer, Roche, Melinta, Seres therapeutics, Takeda, and Treeline biosciences; and research funding from GSK, Scynexis, F2G, Cidara, and Mundipharma. The remaining authors declare no competing financial interests.

Correspondence: Jessica S. Little, Brigham and Women’s Hospital, 75 Francis St, PBB-A4, Boston, MA 02115; email: jlittle@bwh.harvard.edu.