Letermovir for cytomegalovirus (CMV) prophylaxis in allogeneic haematopoietic cell transplant (HCT) recipients has decreased anti-CMV therapy use. Contrary to letermovir, anti-CMV antivirals are also active against human herpesvirus-6 (HHV-6). We assessed changes in HHV-6 epidemiology in the post-letermovir era.
We conducted a retrospective cohort study of CMV-seropositive allogeneic HCT recipients comparing time periods before and after routine use of prophylactic letermovir. HHV-6 testing was at the discretion of clinicians. We computed the cumulative incidence of broad-spectrum antiviral initiation (foscarnet, (val)ganciclovir, and/or cidofovir), HHV-6 testing, and HHV-6 detection in blood and cerebrospinal fluid within 100 days after HCT. We used Cox proportional-hazards models with stabilized inverse probability of treatment weights to compare outcomes between cohorts balanced for baseline factors.
We analysed 738 patients, 376 in the pre-letermovir and 362 in the post-letermovir cohort. Broad-spectrum antiviral initiation incidence decreased from 65% (95% CI, 60-70%) pre-letermovir to 21% (95% CI, 17-25%) post-letermovir. The cumulative incidence of HHV-6 testing (17% [95% CI, 13-21%] pre-letermovir versus 13% [95% CI, 10-16%] post-letermovir), detection (3% [95% CI, 1-5%] in both cohorts), and HHV-6 encephalitis (0.5% [95% CI, 0.1-1.8%] pre-letermovir and 0.6% [95% CI, 0.1-1.9%] post-letermovir) were similar between cohorts. First HHV-6 detection occurred at a median of 37 days (interquartile range, 18-58) in the pre-letermovir cohort and 27 (interquartile range, 25-34) in the post-letermovir cohort. In a weighted model, there was no association between the pre-versus post-letermovir cohort and HHV-6 detection (adjusted hazard ratio, 1.08; 95% CI, 0.44-2.62).
Despite a large decrease in broad-spectrum antivirals after the introduction of letermovir prophylaxis in CMV-seropositive allogeneic HCT recipients, there was no evidence for increased clinically detected HHV-6 reactivation and disease.