Coordinated antiviral immune response in a patient with myeloma and systemic adenovirus infection post-BCMA CAR T cells

TO THE EDITOR:

Infections pose a serious problem after chimeric antigen receptor (CAR) T-cell therapies¹ but only a few cases of systemic adenovirus (Adv) infection after CAR T-cell therapy have been described.² In patients who undergo hematopoietic stem cell transplantation (HSCT), the incidence of AdV disease is 3% to 47%,³ with mortality for pneumonia/disseminated disease of 50% and 80%, respectively.^4,5 In allogeneic HSCT, the main risk factors for AdV infections are T-cell depletion in the graft, patient, or haploidentical/mismatched donor. Although these factors point to an important role for T cells in the control of adenoviral infections, AdV-specific T-cell immunity in CAR T-cell patients is not well characterized, and it is unclear how to enforce T-cell responses in these patients. This question is particularly important because treatment with BCMA (B-cell maturation antigen) CAR T-cells depletes plasma cells, which may affect the induction of virus-specific T cells by reducing (1) viral uptake by antigen-presenting cells after opsonization and (2) formation of immune complexes between antibodies and viral antigens.

Our patient is a 67-year-old female diagnosed with immunoglobulin G (IgG) kappa multiple myeloma 4.5 years before CAR T-cell therapy. She received multiple prior lines of treatment (supplemental Table 1), and upon another relapse underwent lymphodepleting chemotherapy followed by ciltacabtagene autoleucel CAR T-cell therapy in November 2022. After CAR T-cell therapy, she developed grade 1 cytokine release syndrome and a marked expansion of BCMA-targeted CAR T cells (Figure 1B), followed by neutrophil recovery (Figure 1C). As reported before,⁶ shortly after CAR T-cell therapy, the patient developed bilateral cranial nerve VII palsy requiring steroids.

Two months after CAR T-cell therapy, the patient was readmitted for bloody diarrhea, and was initially treated for Clostridioides difficile colitis with fidaxomicin and budesonide for possible immune-mediated colitis. One week after admission, the patient was found to have a positive AdV polymerase chain reaction (PCR) in blood (91 700 copies per mL). At the time of diagnosis, the patient showed humoral/cellular immunosuppression with an IgG level of 523 mg/dL and a lymphocyte count of 0.52x10³/μL. Despite receiving 2 doses of cidofovir, she developed acute respiratory failure and a dramatic increase in liver enzymes and viremia (Figure 2A). On hospital day 23, she was transferred to the intensive care unit (ICU) with multiorgan failure (pneumonitis, hepatitis, and pancytopenia) due to disseminated adenoviral infection requiring oxygen via a high-flow nasal cannula. A third dose of cidofovir was given in the ICU (Figure 2A). The patient had received 1 dose of 20 g of IV immunoglobulin before admission and 7 additional doses during her hospital stay (Figure 2B-C), subsequent to the diagnosis of disseminated Adv infection, with AdV PCR in the blood peaking at >10 000 000 copies per mL (Figure 2A). Upon marked clinical improvement 5 days into the ICU stay, she was transferred back and continued to improve. She was discharged after 42 days from the hospital. Her PCR, measuring expression of the AdV hexon gene, decreased to 59 600 copies per mL at discharge but she continued to be viremic until 4 months later (Figure 2A).

Blood samples were collected under the institutional review board (IRB)–approved protocol 2043GCCC (IRB HP-00091736). Our analyses suggested that the first IV immunoglobulin infusion led to a modest increase in her anti-AdV and anti-influenza IgG antibody titers (Figure 2B) with subsequent doses, resulting in a substantial increase in serum levels of total IgG/IgM and a marked further increase in titers against both viral proteins (Figure 2B,D). Hypothesizing that the administration of IV immunoglobulin was the driving factor behind the development of strong humoral anti-Adv immunity, we performed serial analyses of antibody responses against tetanus toxoid, rhinovirus, varicella-zoster virus, and respiratory syncytial virus. We found that IgG antibody titers against all these microbial agents, including vaccine-induced anti-tetanus toxoid titers, increased only after IV immunoglobulin (Figure 2C). After the cessation of IV immunoglobulin administration, all antibody titers decreased markedly over the following days/week, comparable with the anti-Adv titers. Combined with the absence of IgM anti-Adv antibodies, these findings strongly suggest that it was indeed the IV immunoglobulin, and not an endogenous antibody response, that caused a marked increase in the anti-Adv humoral immunity in our patient.

The patient did not show pre-existing CD4⁺/CD8⁺ T cells against AdV hexon protein at baseline, at 2 weeks or 2 months after receiving her CAR T-cell therapy (Figure 1A,D). However, after being diagnosed with Adv and receiving multiple IV immunoglobulin infusions, the patient developed substantial numbers of both anti-AdV CD4⁺ and CD8⁺ T cells (Figure 1A,D). Ten days after transfer from the ICU, she showed a 120-fold expansion in anti-AdV CD8⁺ and a 40-fold expansion in AdV-specific CD4⁺ T cells (Figure 1A,D; supplemental Figure 1). At that time, the total CAR T-cell levels had already decreased to <5% (Figure 1B). We found that lymphodepleting chemotherapy before CAR T cells depleted all pre-existing influenza A–specific CD4⁺ and CD8⁺ T cells (Figure 1E). In contrast, Adv-reactive T cells remained detectable for months after she was discharged and became undetectable once she cleared the Adv (Figure 1A,D).

The antiviral agent cidofovir is commonly used in disseminated AdV infections, despite its unclear efficacy and known nephrotoxicity.^5,7 The mainstay of the management of AdV infections consists of the reduction of immunosuppression in transplant patients. The application of IV immunoglobulin is a well-supported prophylactic strategy for primary immunodeficiencies.⁸ However, its role in acute infections in hematologic malignancies has remained controversial¹; therefore, more research is needed to identify patients who will benefit from IV immunoglobulin.¹ 

Our patient received 3 doses of cidofovir and multiple IV immunoglobulin doses, which led to the complete resolution of clinical symptoms and a marked increase in antibody titers against the AdV hexon protein. The AdV capsid consists of 3 major structural proteins: hexon, penton, and fiber.⁹ It has been shown that neutralizing antibodies to AdV are directed primarily against the hexon protein.^10-12 Accordingly, neutralizing antibody responses to hexon were dominant in vaccinated mice and vaccinated/naturally exposed humans.¹³ Finally, therapeutic monoclonal antibodies with neutralizing activity against AdV preferentially target the hexon protein.^14,15 

We consider it possible that, in addition to their immediate effect on viral entry, our patient’s hexon-specific polyclonal antibodies also contributed to the development of hexon-specific T cells. It has been shown before that antiviral antibodies can enhance the Fc receptor–mediated uptake/cross-presentation of viral proteins through immune complex formation, resulting in T-cell responses against the same antigen.^16,17 Our group and others have shown that a similar mechanism occurs in the setting of antitumor immunity.^17-20 Such a mechanism would explain the development of pronounced hexon-specific T-cell responses in our patient, which only became undetectable once the patient cleared the virus. In line with this observation, the occurrence of AdV hexon–specific T cells has been linked closely to viral control in allogeneic HSCT.²¹ 

In conclusion, we describe here a patient with multiple myeloma who developed a disseminated, life-threatening AdV infection after CAR T-cell therapy. The patient received multiple IV immunoglobulin doses leading to an enhanced humoral antiviral immunity and the development of a pronounced Adv-specific T-cell response. We propose that the passive transfer of donor-derived polyclonal AdV-specific antibodies contributed to the development of antiviral T cells which in combination resulted in antibody-mediated inhibition of viral entry and elimination of infected cells and viral clearance. Future studies should examine this proposed mechanism in more detail and larger cohorts of patients after CAR T-cell therapy.

Acknowledgments: This study was funded by grants from the Kahlert Foundation (D.A.), the Maryland Department of Health's Cigarette Restitution Fund Program (D.A. and X.F.), and National Cancer Institute Cancer Center Support grant P30CA134274.

Contribution: D.A. designed the study, performed the experiments, analyzed the data, made the figures, and wrote the manuscript; P.L. and M.Y. collected and processed the patient samples and clinical data; D.O., E.G., R.M., D.Y., and T.I. processed patient samples and performed the experiments; M.H.K., J.T.B., J.B., X.F., J.M.B., K.A.D., K.G.H., J.A.Y., A.B., A.P.R., and N.M.H. analyzed the data and wrote the manuscript; and T.L. analyzed the data, prepared the figures, and wrote the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Djordje Atanackovic, Fannie Angelos Cellular Therapeutics Good Manufacturing Practice Laboratory, University of Maryland Greenebaum Comprehensive Cancer Center, Bressler Research Building, Room 9-011, 655 W Baltimore St, Baltimore, MD 21201; email: datanackovic@som.umaryland.edu.