Prophylactic Foscarnet and Suppression of Pre-Engraftment Immune Reaction Are Essential to Overcome the Development of HHV-6 Encephalitis From the Experience of 496 Adult Cord Blood Transplants in Toranomon Hospital

Human herpes virus 6 (HHV-6) encephalitis is a growing concern after cord blood transplantation (CBT), which leads to permanent neurological sequelae that do not allow patients to get back into society. The pathogenic mechanism remains unknown and appropriate preventative strategies have not been established.

To evaluate incidences, risk factors and outcomes of HHV-6 encephalitis after CBT, we reviewed the medial records of 496 adult patients who underwent CBT for the first time at the Toranomon Hospital between 2002 and 2011, and who survived more than 7 days. Over the entire period, routine prophylaxis with acyclovir against reactivation of herpesviruses and pre-emptive therapy with ganciclovir or foscarnet (FCV) against cytomegalovirus disease were performed. From January 2006, prophylactic administration of FCV with monitoring of serum HHV-6 viral load against HHV-6 disease was introduced, the indication of which was decided at a physician's discretion. Plasma HHV-6 DNA copy numbers were measured using real-time PCR method, and the detection limit was 200 copies/mL. Immune reaction prior to neutrophil engraftment characterized by unexplained fever in the absence of documented infection with skin eruption, peripheral edema and body-weight gain was defined as pre-engraftment immune reaction (PIR), which was categorized as mild and severe, according to the degree of concomitant organ dysfunctions. HHV-6 encephalitis was defined as the neurological symptoms with positive PCR results for HHV-6 in cerebrospinal fluid (CSF) and the absence of other identified etiologies of encephalitis.

Forty-two patients developed HHV-6 encephalitis with a cumulative incidence of 8.5%. The incidence was significantly higher in the period from 2006 to 2011 compared to the period from 2002 to 2005 in spite of the introduction of prophylactic FCV after 2006 (11.1% vs. 4.3%, P=0.01), probably owing to more vigorous investigation. Among the 308 patients who received CBT after 2006, engraftment was achieved in 239 at a median of 20 (10–66) days after CBT, and PIR occurred in 133 patients at a median of 10 (4–24) days, the severity of which was classified into mild in 114 and severe in 19. 132 patients developed grade II-IV acute graft-versus-host disease (GVHD) at a median of 31 (9–91) days, and 182 received high-dose corticosteroids at >= 0.5mg/kg with a median starting point of 19.5 (0–1363) days. Serum HHV-6 PCR measurement was performed in 280 patients, 122 of who showed positive test results at a median of 20 (8–193) days. Prophylactic FCV was given in 234 patients from a median of 11 (0–35) days, with a median duration of 23 (2–79) days. HHV-6 encephalitis occurred in 34 of the 308 patients at a median of 22 (11–49) days, with a cumulative incidence of 11.2%. The median peak level of HHV-6 DNA was 20,000 (200–400,000) copies/mL in CSF, in contrast with 200 (0–89,300) copies/mL in plasma. PIR was identified as an independent significant risk factor for HHV-6 encephalitis (HR 3.13(1.52–6.44), P=0.002), which occurred more frequent in patients with severe PIR compared to those with mild one (HR 2.82(1.11–7.18), P=0.029). Another significant risk factor was high-level of HHV-6 DNA in plasma at >=10,000 copies/mL (HR 3.29(1.17–9.30), P=0.024), which was, however, observed in only 7 of the 34 patients with encephalitis. Neither grade II-IV acute GVHD nor the use of high-dose steroids influenced an incidence of the encephalitis (HR 2.2(0.90–5.37), P=0.083, and HR 0.98(0.47–2.03), P=0.95). Prophylactic FCV at >=60mg/kg showed a significant benefit in preventing HHV-6 encephalitis (HR 0.27(0.09–0.77), P=0.015), which substantially reduced the incidence to 1.6% if PIR was completely suppressed, whereas the incidence reached 23.4% in the presence of PIR if the FCV was not given (P<0.001). Although no patient directly died of HHV-6 encephalitis, the encephalitis tended to affect non-relapse mortality after CBT (HR 1.75(0.99–3.07), P=0.053).

PIR seems to bring at least similar or more risk for HHV-6 encephalitis compared to high-level HHV-6 viral load. Prophylaxis with FCV at >=60mg/kg combined with the optimal suppression of PIR might provide a further reduction of HHV-6 encephalitis after CBT.

No relevant conflicts of interest to declare.