Human herpesvirus 6 and CAR T-cell toxicity

In this issue of Blood, Kampouri et al report their findings on incidence of human herpesvirus 6B (HHV-6B) reactivation in patients receiving chimeric antigen receptor (CAR) T-cell therapies, both in a cohort of 89 patients monitored prospectively for up to 12 weeks following infusion and retrospectively in a larger group of 626 patients who had symptom-driven screening of plasma (n = 24) or cerebrospinal fluid (n = 34), concluding that HHV-6B reactivation and disease are infrequent in this setting and that routine monitoring is not warranted.¹ 

Treatment-emergent infections are recognized as a significant challenge following CAR T-cell therapies. The underlying immune deficiency in these patients is multifactorial, relating not only to the specific, often profound, and in some cases prolonged B-cell aplasia that results directly from most CAR T-cell targets, but also in varying degrees to aspects of the underlying malignancy, prior therapies, prolonged cytopenias (particularly neutropenia), and immune suppressive drugs that are often required to manage CAR T-cell toxicities (most notably immune effector cell–associated neurotoxicity syndrome [ICANS]). Many of these contribute to enhanced risk of bacterial or fungal infections. Many viral infections, including those in the extended human herpesvirus group, are more heavily dependent on T-cell function for control and containment. T-cell dysfunction following CAR T-cell therapy is likely less pervasive than following allogeneic stem cell transplantation (SCT). Most of the 9 herpesviruses recognized to primarily infect humans are highly prevalent among human populations, with >90% of adults having been infected with at least 1 of these viruses. Herpesviruses cause chronic infections, such that most adults can be considered infected. Viral “latency” following primary infection is variably compromised following allogeneic SCT, resulting in detection of viral replication and in some instances end organ disease. HHV-6B infects most individuals within the first 3 years of life, often manifesting as roseola.² Unlike other human herpesviruses, HHV-6B can integrate into chromosomes as a mechanism of latency, which may result in inheritance of chromosomally integrated HHV-6, which can complicate interpretation of diagnostic testing, as indicated by the authors. Systemic reactivation of HHV-6B, as detected by HHV-6B DNA in blood, is frequent after SCT, occurring in up to 40% of patients within the first few months, dependent on the transplant platform and donor type.³ Of particular relevance to the field of CAR T-cell therapy, HHV-6B is the most frequent infectious cause of encephalitis after SCT.⁴ Although HHV-6B encephalitis is a relatively uncommon event in this setting, occurring in ≈1% of SCTs, the higher incidence and severity of neurologic complications after CAR T-cell therapies, and overlapping manifestations with ICANS, highlights the importance of understanding any potentially contributory role of HHV-6B, including guidance regarding the need for either routine or targeted surveillance.

The study of Kampouri et al concludes that HHV-6B reactivation and disease are infrequent after CAR T-cell therapy and that routine monitoring is not warranted. In their prospective surveillance cohort, the cumulative incidence of reactivation was 6% (95% confidence interval [CI], 2.2%-12.5%) but importantly detection events were at low viral titer levels and self-limiting. In a larger cohort with more targeted symptom-driven sampling, a retrospective analysis suggested a cumulative incidence of HHV-6B encephalitis of 0.17% (95% CI, 0.02%-0.94%), noting also that the single case identified improved without treatment. This study does not definitively answer all the questions relevant to the field. The general recommendation regarding routine surveillance does seem justified, although confirmation by other groups in similarly sized cohorts would be welcomed. Patient numbers are relatively small when considering specific subgroups that might be considered at heightened risk of reactivation, and any introduction of new biologics into the management algorithms for ICANS may warrant reevaluation of these issues and the question of whether more targeted surveillance has any role. Furthermore, some elements of risk are likely CAR T-cell target specific rather than platform specific, and further evaluation of patients receiving CAR T cells with alternative target specificity are warranted. Nevertheless, the study does provide clinically relevant and tractable information that will be useful for the increasing number of clinicians and treatment centers facing the challenges of managing CAR T-cell–associated ICANS.

Conflict-of-interest disclosure: The author declares no competing financial interests.