Preemptive Therapy with Rituximab for Epstein-Barr Virus Reactivation in Children after T-Cell-Replete Haploidentical Hematopoietic Stem Cell Transplantation with Low-Dose Anti-Thymocyte Globulin

[Introduction] EBV-PTLD is a serious complication in hematopoietic stem cell transplantation (HSCT) recipients. We conducted a retrospective study to investigate the incidence and potential risk factors for EBV reactivation and to assess the efficacy of the management of EBV reactivation with preemptive rituximab in children who had T-cell-replete haploidentical HSCT (TCR-haplo-SCT) with low-dose anti-thymocyte globulin (ATG).

[Methods] A total of 38 patients who underwent haplo-SCT in a single institution (Fukushima Medical University Hospital, Fukushima, Japan) between October 2009 and December 2014 were included in this retrospective study. Those patients were tested for EBV-DNA levels in PB at least once after HSCT when they showed suspicious symptoms of EBV reactivation. During this study period, haplo-SCT with low-dose ATG was performed on those patients with high-risk childhood acute leukemia or solid tumor with the expectation of a strong GVL/GVT effect. On the other hand, a patient with immunodeficiency syndrome underwent haplo-SCT because of the lack of an HLA-matched related or unrelated donor. When the EBV-DNA level in PB increased to >1,000 copies/10⁶ peripheral blood mononuclear cells (PBMCs), patients were preemptively treated with rituximab (375 mg/m²/dose). For patients with a decrease in EBV viremia titer following one dose of rituximab, viral loads were monitored weekly to follow their response to rituximab. For patients with persistent viremia, an additional weekly dose of rituximab was administered until the resolution of the viremia. Whenever possible, immunosuppressants were tapered.

[Results] All HSCTs were T-cell-replete, and all patients received ATG at a dose of 2.5 mg/kg during conditioning. Thirty-eight patients were included in this study with a median follow up of 625 (range, 59-2084) days post-transplantation. Among them, 32 (84%) patients became EBV-DNA-positive, with 19 of 38 (50% of the total population) patients presenting with EBV reactivation and requiring preemptive treatment. In detail, 13 patients with transient, self-limiting EBV-DNA positivity had a mean peak level of 140 cp/10⁶ PBMCs (range: 17-880), while the 19 patients who presented with EBV reactivation showed a mean EBV-DNA level of 2900 cp/10⁶ PBMCs (range: 1000-650000). The median time from transplantation to EBV reactivation was 56 (range, 17-270) days and the median viral load at initiation of therapy was 2900 cp/10⁶ PBMCs (range: 1000-650000). Preemptive therapy was started after a median of 2 (range, 0-7) days of qPCR result. The median number of weekly treatment cycles was 2 (range, 1-3). Fever was noted after rituximab infusion in one of the 19 patients, although no other side effects were observed. Seventeen of those 19 (89%) patients sustained viral loads of <1000 cp/10⁶ PBMCs after the therapy. Two patients died from veno-occlusive disease and an underlying disease before achieving resolution of the viremia. None of the patients developed PTLD or other EBV-associated diseases. Twenty-three patients (61%) were still alive and the OS rate was 63% at 24 months. In all, 9 patients died owing to disease progression and 6 patients died because of transplant-related complications. Although the numbers were small, the OS rate was lower in patients who showed EBV reactivation than in patients who did not; the 24-month OS rates were 54% versus 72% and 28%, but the difference was not statistically significant (OS rate: log rank test, p=0.09). Although there was also no statistically significant difference, TRM and relapse/disease progression rates tended to be higher in patients who experienced EBV reactivation than in those who did not (TRM rate: Gray test, p=0.23, relapse/disease progression rate: Gray test, p=0.09). Univariate analysis for risk factors associated with EBV reactivation is shown in Table 2. Although no risk factors were significantly different between subgroups with and without EBV reactivation, there was a tendency for patients aged 13 years and above to develop EBV reactivation and patients with high-level EBV reactivation tended to have a higher incidence of acute GVHD grades II-IV. Because of the relatively small number of events, multivariate analysis was not performed.

[Conclusion] We conclude that preemptive rituximab therapy is effective for EBV-PTLD in TCR-haplo-SCT recipients with low-dose ATG.