HHV-6 Viremia Is Frequent but the Incidence of Encephalitis Is Low in Double-Unit Cord Blood Transplant Recipients Transplanted without ATG

While cord blood transplantation is a known risk factor for Human Herpesvirus-6 (HHV-6) reactivation, recent studies have yielded inconsistent results in regard to the level of viremia which is associated with a high risk of HHV-6 encephalitis. Moreover, the association with graft failure or transplant-related mortality is controversial.

We conducted a retrospective analysis of 125 patients who underwent double-unit cord blood transplantation (DCBT) for the treatment of hematologic malignancies from 2/2006-3/2012 who were monitored for HHV-6 reactivation to examine the incidence and severity of HHV-6 viremia, the incidence of encephalitis, and the association with DCBT outcome. HHV-6 viremia was measured by quantitive PCR of HHV-6 DNA from plasma (lower limit of detection 100 DNA copies/ml).

Of the 125 patients (median age 42, range 1–69), 93 (74%) received myeloablative conditioning and 32 (26%) received a non-myeloablative regimen followed by 4–6/6 HLA-A, B antigen, DRB1 allele matched DCBT for the treatment of AML (N = 43, 34%), ALL (N = 24, 19%), MDS/CML/other leukemia (N = 12, 10%), or lymphoma/CLL (N = 46, 37%). No patient received anti-thymocyte globulin (ATG). Of 125 monitored patients, 117 (94%) reactivated HHV-6 to a median peak of 7,600 (range 100–160,000) copies/ml at a median onset of 20 days (range 10–59) post-DCBT. The median time to peak viremia was 23 days (range 12–62) with a median viremia duration of 10 days (range 1–60 days). Fifty-one patients (41% of total, 44% of viremic patients) developed HHV-6 > 10,000 copies/ml (median peak viremia 31,200 copies/ml at 20 days, range 12–57). Only 6 patients (5% of total, 5% of viremic patients) developed HHV-6 > 100,000 copies/ml (median peak 130,000 copies/ml at 19 days, range 14–29). HHV-6 encephalitis occurred in 2 patients (1.6%, peak viremias 13,000 and 118,000, respectively). One patient died from encephalitis and the other had a complete recovery following therapy. Four additional viremic patients had HHV-6 isolated from bronchoalveolar lavage but did not meet criteria for HHV-6 pneumonia. Using a high level viremia definition of > 10,000 copies/ml from days 14–60 (a level reportedly associated with end-organ disease and decreased survival, Dulery, BBMT 2011), viremia was not associated with diagnosis or conditioning regimen. Engrafting unit-recipient HLA-match and TNC, CD34⁺ and CD3⁺ cell dose were not associated with high level viremia. Treating viremia as a time-dependent covariate in Cox regression analysis, no association was found between viremia and neutrophil or platelet engraftment. Specifically, there were two graft failures in patients with no or low level viremia and two in patients with high level viremia. There was also no association between viremia and CMV reactivation, day 100 grade II-IV aGVHD, day 100 TRM, relapse or overall survival. A second analysis was performed to examine effects of high level viremia defined as a level > 25,000 copies/ml (N = 31 patients, the highest peak viremia quartile during days 14–60) and no associations with DCBT outcomes were detected.

In our DCBT population, nearly all patients have reactivated HHV-6. However, the incidence of end-organ disease is relatively low. We postulate the difference between our findings and other studies reporting higher rates of HHV-6 encephalitis in DCBT recipients could be due to our exclusion of ATG from the conditioning regimen. At this time our understanding of the significance of HHV-6 after CBT is incomplete. We are currently evaluating anti-viral treatment responses, and ultimately a prospective trial is needed to better define the causality between HHV-6 viremia and transplantation outcomes, and to investigate the risk-benefits of pre-emptive therapy.

No relevant conflicts of interest to declare.