Prospective Study of High Dose Foscarnet Prophylaxis for HHV-6 Related CNS Diseases in Cord Blood Transplantation for Adult Patients

Background: Umbilical cord blood transplant (UCBT) recipients showed higher incidence of HHV-6 related CNS diseases (HHV-6 RCD) than those of bone marrow (BM) or peripheral blood (PB). It causes short-term memory impairment or disorientation, which sometimes become irreversible, and convulsions in severe cases, resulted in high non-relapse mortality. We reported previously that in the group given higher doses of prophylactic Foscarnet (FCV) (≥60mg/kg), the incidence of HHV-6 RCD was reduced (Asano-Mori Y, et al. ASH 2013). We thus conducted a prospective study to assess efficacy of higher dose (120mg/kg) of prophylactic FCV to further reduce the incidence and to improve outcome of HHV-6 RCD.

Patients and Methods: The study was started since May 2014 and was finished on September 2015. The protocol was approved by local IRB and UCBT recipients who gave written informed consent were included. FCV 120 mg/kg was administered prophylactically from day 7 to day 50 post UCBT.. FCV dose was adjusted according to renal function. The incidence of HHV-6 RCD, types of manifestation and severity of the symptoms, treatment response, and prognosis were evaluated.

Results: Among 113 who were included in this study, 9 failed to start FCV from day 7 due to early death or renal impairment, 37 were switched to other drugs at some point before day 50 for reasons such as renal impairment or viral infections other than HHV-6 (23 changed to GCV, 10 to ACV, and 4 to CDV), and 67 received planned dose of FCV. Fifteen developed HHV-6 RCD (12 HHV-6 limbic encephalitis (LE) and 3 myelitis). The cumulative incidence of HHV-6 RCD was 11.5%. No significant risk factors before transplant associated with the development of HHV-6 RCD were identified. Six out of 22 who developed severe pre-engraftment immune reactions (PIR) developed HHV-6 RCD. PIR was shown to be a significant risk factor for HHV-6 RCD development when it was treated as a time-dependent covariate. The median onset day was day 40, and 4 out of 15 patients developed after the cessation of FCV. Only one of 12 encephalitis patients developed convulsions. The median number of copies of HHV-6 in the cerebrospinal fluid at onset was 40,400/µL. HHV-6 viral load in peripheral blood was examined on the same day of CSF examination in 13 and was not detected in 7. Six patients were treated with FCV and GCV in combination, 4 received increased dose of FCV, and 1 was switched from FCV to GCV. Two out of 4 patients who developed after the prophylactic administration of FCV resumed FCV, and the remaining 2 resumed GCV. The virus became undetectable in all 13 evaluable patients by follow-up CSF examination after treatment. The median day of viral disappearance from CSF was 19 days (8-34 days) after diagnosis. All but 2 who were unevaluable due to other complications, had their symptoms improved without sequalae. The median time from treatment intervention to symptomatic improvement was 28 days. The OS after 1 year from HHV6 encephalitis myelitis group was 46.2%, and there was no significant difference in the prognosis as 47.3% OS after 1 year from non-onset group when time-dependent co-analysis was performed (HR 1.88, 95%CI;0.49-2.84, p=0.69). Analysis of OS for 67 patients who were able to undergo preventative FCV administration and other group showed 53.2% and 38.3% respectively in 1 year, revealing a trend in which OS appeared to become higher as a result of continued prophylaxis.

Conclusion: There was no difference in the incidence of HHV-6 RCD compared to the previous report even after prophylactic administration of high dose 120 mg/kg of FCV. However, there were few severely ill individuals, most of their symptoms improved due to antiviral treatment without significant sequalae.