Background. Chronic active Epstein-Barr virus (EBV) infection is characterized by recurrent infectious mononucleosis–like symptoms, and infected patients have high viral loads in their peripheral blood. Standard therapy for the disease has not yet been established. Recently, hematopoietic stem cell transplantation (HSCT) has been introduced and has the potential to become a standard treatment, although guidelines for HSCT to treat chronic active EBV infection have not yet been proposed.
Methods. Fifteen patients were retrospectively analyzed, both clinically and virologically, to investigate the factors associated with prognosis of chronic active EBV infection treated with HSCT.
Results. After HSCT, 7 patients died after survival periods that ranged from 1 to 16 months (mean duration of survival after HSCT, 5 months). Three patients were considered to have died of transplantation-related complications. The duration between infection onset and diagnosis was significantly longer in patients who died than in those who survived. Five of the 7 patients who died experienced ⩾3 life-threatening complications. The plasma concentrations of interferon-γ, interleukin-10, thrombomodulin, and soluble E-selectin did not differ significantly between the groups of patients. With regard to sequence variations in the EBV latent membrane protein 1 gene, no specific patterns were found in the patients who died. Importantly, the plasma EBV load at diagnosis was significantly higher in patients who died than in living patients. Moreover, plasma viral load was shown to be an important factor to monitor during follow-up for patients after HSCT.
Conclusions. The number of life-threatening complications and plasma viral load are indicative of the stage of disease progression and may be useful factors for predicting the outcome of HSCT.
Epstein-Barr virus (EBV) is a ubiquitous virus that infects most humans during early adulthood. Primary EBV infection is usually asymptomatic, but sometimes it results in infectious mononucleosis, which resolves spontaneously after the emergence of EBV-specific immunity [1]. EBV can cause chronic infections in apparently immunocompetent persons. Chronic active EBV (CAEBV) infection is characterized by chronic or recurrent infectious mononucleosis-like symptoms, such as fever, swelling of lymph nodes, and hepatosplenomegaly. CAEBV infection is associated with high mortality and morbidity, with various life-threatening complications, such as virus-associated hemophagocytic syndrome, interstitial pneumonia, lymphoma, coronary artery aneurysm, and CNS involvement [2–4]. Patients with CAEBV infection also have an unusual pattern of EBV-related antibodies and high viral loads in their peripheral blood [5–8]. Recent studies have indicated that clonal expansion of EBV-infected T cells and natural killer (NK) cells plays a central role in the pathogenesis of CAEBV infection [7, 9–12], although CAEBV infection in Western countries may not always be associated with the expansion of EBV-infected T or NK cells [13].
Standard therapy for CAEBV infection has not yet been established. Antiviral, anticancer, and immunomodulating agents have been administered, although they seem to have only a transient effect and are unable to control CAEBV infection [13]. Recently, hematopoietic stem cell transplantation (HSCT) has been introduced as a curative therapy for CAEBV infection. Bone marrow [14], cord blood [15–17], and peripheral blood [18–20] have been reported as stem cell sources. HSCT is not a standard therapy, even after successful cases have been reported, because patients with CAEBV infection are at high risk for treatment-related toxicity attributable to multiple organ dysfunction [12, 13].
In the present study, 15 patients with CAEBV infection treated with HSCT were investigated to establish which factors affect treatment outcome. Clinical characteristics were compared between living patients and patients who died. Viral loads in PBMCs and plasma, the plasma concentrations of cytokines and vascular endothelial cell–associated molecules, and sequence variations in EBV isolated from samples from patients with CAEBV infection were compared.
Patients. Fifteen patients treated with allogeneic HSCT during the period 1999–2006 were enrolled in this study. These patients were referred to the Nagoya University School of Medicine (Nagoya, Japan) with CAEBV infection. Informed consent was obtained from all patients or their parents. All patients fulfilled the following diagnostic criteria [7]: (1) EBV-related illness or symptoms for >6 months, including fever, persistent hepatitis, extensive lymphadenopathy, hepatosplenomegaly, pancytopenia, uveitis, interstitial pneumonia, hydroa vacciniforme, or hypersensitivity to mosquito bites; (2) increased quantity of EBV in either affected tissues or peripheral blood (the amount of EBV was defined as increased when ⩾1 of the following criteria were met: EBV DNA detected in tissue or peripheral blood samples by Southern blot hybridization; EBV-encoded small RNA1–positive cells detected in tissue or peripheral blood samples; or an EBV DNA level >102.5 copies/µg of DNA detected in PBMCs [21]); and (3) no evidence of any prior immunologic abnormalities or of any other recent infection that might explain the condition. The time of diagnosis was defined as the time when the patient was found to meet the above criteria. Serial blood samples were obtained before and after transplantation. Clinical assessments, such as standard cardiac and neurological tests, were performed to evaluate life-threatening complications.
Determination of EBV-infected cells. To determine which cells harbored EBV, PBMCs were fractionated into CD3+, CD4+, CD8+, CD16+, CD19+, and CD56+ cells with use of Dynabeads (Invitrogen). Patients were defined as having T cell–type infection when CD3+ cells were the major group of cells that harbored EBV. Patients were defined as having NK cell–type infection when their CD16+ or CD56+ cells were the major group of cells infected with EBV [7, 22].
Plasma concentrations of cytokines and vascular endothelial cell–associated molecules. Plasma concentrations of IFN-γ, IL-6, IL-10, thrombomodulin, and soluble E-selectin were measured using ELISA kits (for IFN-γ and IL-10, BioSource Europe; for IL-6, Fuji Rebio; for thrombomodulin, Daiichi Fine Chemicals; for soluble E-selectin, R&D Systems), according to the manufacturers' instructions.
Quantification of EBV DNA. DNA was extracted from either 1×106 PBMCs or 200 µL of plasma. A real-time quantitative PCR assay was performed, as described elsewhere [21]. The amount of EBV DNA was calculated as the number of virus copies per microgram of PBMC DNA or per milliliter of plasma. The lower limits of detection in this assay were ∼10 copies/µg of DNA for PBMCs and 50 copies/mL for plasma.
Expression of the EBV latent membrane protein 1 (LMP1) gene in PBMCs. To detect expression of the LMP1 gene, RNA was extracted from the PBMCs of 12 patients at diagnosis and was used for a nested RT–PCR assay, as described elsewhere [22].
Sequence analysis of the C-terminal region of the LMP1 gene. DNA was extracted from PBMCs at diagnosis and was used for sequencing. Sequences of the C-terminal region of the LMP1 gene were obtained using the direct sequencing method as described elsewhere [23]. PCR products were not obtained from the PBMCs of some patients. In these cases, a seminested PCR was performed using the primers described elsewhere [24]. The sequence data obtained were checked for homology in BLAST [25]. The sequences of the C-terminal region of the LMP1 genes of patients 3, 5, 8, and 11 have been reported elsewhere [26].
Statistical analysis. Statistical analysis was conducted using StatView software, version 5.0 (SAS Institute). Either Fisher's exact test or the χ2 test was used for the comparison of clinical data. The Mann–Whitney U test was used to compare viral loads and plasma concentrations of IFN-γ, IL-6, IL-10, thrombomodulin, and soluble E-selectin. Multivariate analysis was also performed using a multiple regression model to analyze factors identified as statistically significant. P<.05 was considered to be statistically significant.
Clinical characteristics of patients with CAEBV infection treated with HSCT. The clinical features and follow-up results of all 15 patients are shown in table 1, and the life-threatening complications encountered by each patient are listed in table 2. There were 6 male and 9 female patients, and the age at onset of disease ranged from 1 to 26 years (median age as disease onset, 4.2 years). Seven patients were defined as having T cell–type infection, and 7 patients had NK cells predominantly infected with EBV. Seven patients died after survival periods that ranged from 1 to 16 months (median period of survival, 3 months) after HSCT. Each cause of death is shown in table 1. Three patients who experienced reappearance of clinical symptoms and increased viral load after undergoing HSCT were considered to have experienced relapse. Three patients (patients 5, 6, and 12) died of transplantation-related complications. One patient (patient 14) died of acute disseminated encephalomyelitis. Transplantation-related complications were defined as those that occurred within 60 days after transplantation and that were considered to be associated with transplantation rather than the disease itself. Eight patients were alive after a follow-up period of 9–67 months (median follow-up period, 40 months). One of the 8 patients was excluded from our analysis because of graft rejection.
Comparison of viral loads in PBMCs or plasma samples from living patients with chronic active Epstein-Barr virus (EBV) infection and patients with such infection who died. EBV load was measured using real-time PCR. A, EBV load at diagnosis. B, EBV load before hematopoietic stem cell transplantation. Samples were collected 7–50 days (median, 10 days) before conditioning for transplantation. Dashed lines represent detection limits. The Mann-Whitney U test was used to compare viral loads.
Dynamics of viral load in patients with chronic active Epstein-Barr virus (EBV) infection treated with hematopoietic stem cell transplantation (HSCT). A, EBV loads in PBMCs and plasma samples from living patients (patients 1–3, 8–10, and 15). The number of plasma samples from patient 10 was insufficient to establish the time course of plasma viral load measurements. B, EBV loads in patients who died without experiencing relapse (patients 5, 6, 12, and 14). Patients 5, 6, and 12 died of transplantation-related complications. Patient 14 died of acute disseminated encephalomyelitis. C, EBV loads in patients who died after experiencing relapse (patients 4 and 11). EBV load was measured using real-time PCR. Dotted lines represent detection limits. †, Death.
Alignment of the DNA sequences of the C-terminal regions of the LMP1 gene from patients (Pts) with chronic active Epstein-Barr virus infection before hematopoietic stem cell transplantation. DNA sequence differences from the B95-8 strain are shown for each sequence pattern. Pt numbers are the same as those listed in table 1. Dots indicate that the nucleotide sequence is the same as the above sequence. Dashes represent deletions.
Clinical characteristics of 15 patients with chronic active Epstein-Barr virus (EBV) infection treated with stem cell transplantation.
Life-threatening complications for 15 patients with chronic active Epstein-Barr virus infection treated with stem cell transplantation.
Fourteen patients achieved myeloid engraftment (absolute neutrophil count, >500 neutrophils/µL) in a median period of 18 days (range, 12–26 days). Persistent complete donor chimerism within 70 days after transplantation was achieved in 9 of 12 patients. The number of recipient cells increased after complete donor chimerism was achieved in 1 patient who experienced relapse (patient 4). Two patients (patients 7 and 14) did not achieve complete donor chimerism. Chimerism was not evaluated in 3 patients who died of transplantation-related complications. We did not evaluate the relationship between the EBV status of the donors and patient outcome after transplantation because of the limited availability of information.
Comparison of clinical features between living patients and patients who died. Fourteen patients were divided into living patients and patients who died (7 patients each), and the clinical manifestations of these groups were compared (table 3). One of 8 living patients (patient 7), who had been alive for 5 years after transplantation, was excluded from this comparison. This patient had recently experienced full relapse of disease and received a second transplantation. The age at diagnosis was significantly older for patients who died than for living patients (table 3). The interval between disease onset and diagnosis was also significantly longer for patients who died than for living patients. Ten patients underwent nonmyeloablative HSCT to avoid transplantation-related complications. Six of 7 living patients were treated with a nonmyeloablative regimen. There were no statistically significant differences between patients who lived and patients who died with regard to the other factors listed (table 3).
The relationship between individual outcome and the number of life-threatening complications before HSCT is shown in table 2. Five of 7 patients who died experienced ⩾3 life-threatening complications, whereas only 1 of the 7 surviving patients experienced ⩾3 life-threatening complications (P=.051). Five of 6 patients with complicated hematologic disorders, such as malignant lymphoma (n=3) and hemophagocytic syndrome (n=4), died.
Plasma concentrations of IFN-γ, IL-6, IL-10, thrombomodulin, and soluble E-selectin. To investigate whether the levels of cytokines known to be elevated in patients with CAEBV infection [22, 27–29] affect the mortality rate, the plasma concentrations of IFN-γ, IL-6, and IL-10 at a time when patients were treatment-free before HSCT were compared between living patients and patients who died. Moreover, thrombomodulin and soluble E-selectin concentrations were measured, because some patients with CAEBV infection develop vascular and cardiac complications [12]. The concentrations of IFN-γ, IL-6, and IL-10 in plasma specimens from one-half of the patients with CAEBV infection increased, compared with normal levels in healthy persons, although there were no statistically significant differences between the 2 groups (table 4). The mean levels of thrombomodulin and soluble E-selectin were not elevated in either group of patients.
Plasma concentrations of IFN-γ, IL-6, IL-10, thrombomodulin, and soluble E-selectin in patients who lived and patients who died after hematopoietic stem cell transplantation.
Viral load in the peripheral blood. Viral load was measured by real-time quantitative PCR in consecutive samples obtained from each patient. First, the amounts of EBV DNA in PBMCs and plasma were compared between living patients and patients who died (figure 1). Interestingly, the number of copies of EBV DNA in the plasma at diagnosis was significantly higher in patients who died than in living patients. Of note, all of the patients who had a plasma viral load ⩾103.8 copies/mL at diagnosis died after undergoing HSCT (figure 1). On the other hand, the plasma viral load before HSCT was similar between living patients and patients who died. The plasma EBV load at diagnosis and 2 other clinical factors that differed significantly (age at diagnosis and the interval between disease onset and diagnosis) (table 3) were analyzed using a multivariate analysis. Only plasma viral load at diagnosis was confirmed to be significantly higher in patients who died than in living patients (P=.007).
Second, viral load was measured longitudinally after HSCT. Figure 2 shows the time course of EBV DNA load in 12 of the patients. Two patients, for whom the number of samples was not sufficient to see the time course, and patient 7, who experienced graft rejection, were excluded from the longitudinal study. In living patients, viral load decreased in both PBMCs and plasma after HSCT (figure 2A). None of these patients showed any symptoms associated with CAEBV infection after HSCT. There was good agreement between the disappearance of clinical symptoms and decreased viral load after HSCT. Moreover, viral DNA was not detected in plasma 70 days after HSCT. Viral load also decreased after HSCT in 3 patients who died of transplantation-related complications and in 1 patient who died of a neurological complication (figure 2B). In contrast, the viral load increased or did not decrease in patients who experienced relapse. Importantly, plasma EBV reappeared in these patients after they underwent HSCT (figure 2C).
Expression of LMP1. in PBMCs and sequence variation in the LMP1 gene. Expression of the LMPFigure 1. Comparison of viral loads in PBMCs or plasma samples from living patients with chronic active Epstein-Barr virus (EBV) infection and patients with such infection who died. EBV load was measured using real-time PCR. A, EBV load at diagnosis. B, EBV load before hematopoietic stem cell transplantation. Samples were collected 7–50 days (median, 10 days) before conditioning for transplantation. Dashed lines represent detection limits. The Mann-Whitney U test was used to compare viral loads.
We thank the following for their contributions to this study: Yoshitoyo Kagami (Aichi Cancer Center); Kazushi Tanimoto (Ehime Prefectural Central Hospital); Masaki Yasukawa (Ehime University); Masaki Ito, Mitsuaki Hosoya, and Atsushi Kikuta (Fukushima Medical University); Hitoshi Kiyoi, Tomohiro Kinoshita, and Tomoki Naoe (Nagoya University); Motoko Koyama, Nobuharu Fujii, and Takanori Teshima (Okayama University); Misako Ueda and Yasunobu Takeoka (Osaka City University); Hirokazu Kanegane (Toyama University); Tsuyoshi Ito (Toyohashi City Hospital); Masahiro Tsuchida (Ibaraki Children's Hospital); and Chikako Kanazawa (Yamagata University). We thank F. Ando for secretarial assistance.
Financial support. Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (18591212 to Y.I. and 19591247 to H.K.).
Potential conflicts of interest. All authors: no conflicts.
IDSA Members: For your free access to this journal, log in via the IDSA members area.
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy