Analysis of the efficacy and survival of allogeneic hematopoietic stem cell transplantation and nontransplantation treatment in chronic active Epstein

Abstract

This retrospective study evaluated 75 patients with chronic active Epstein–Barr virus infection (CAEBV) to compare the efficacy and survival outcomes of allogeneic hematopoietic stem cell transplantation (HSCT) and programmed death-1 (PD-1) blockade therapy. Patients were classified into HSCT and non-HSCT groups. The primary endpoints were overall response rate (ORR), overall survival (OS), and event-free survival (EFS). HSCT significantly improved ORR, 3-year OS, and 3-year EFS compared to non-HSCT treatment. Subgroup analysis showed that PD-1 blockade achieved outcomes comparable to HSCT in a subset of patients; however, HSCT remained superior, overall, particularly in patients without hemophagocytic lymphohistiocytosis (HLH). The presence of HLH was identified as an independent risk factor for inferior survival. In conclusion, allogeneic HSCT remains the preferred curative strategy for CAEBV, whereas PD-1 blockade represents a promising alternative for carefully selected patients. Early recognition and management of HLH are crucial for improving prognosis.

Graphical Abstract

Keywords

chronic active Epstein–Barr virus infection allogeneic hematopoietic stem cell transplantation hemophagocytic lymphohistiocytosis programmed death-1 blockade

Introduction

Chronic active Epstein–Barr virus infection (CAEBV) is a subtype of Epstein–Barr virus-associated T/NK-cell lymphoproliferative disease (EBV-T/NK-LPD). Its clinical manifestations result from systemic inflammation and the proliferation of EBV-infected T/NK lymphocytes, leading to recurrent inflammatory flares and progressive organ dysfunction¹. During the active phase, CAEBV presents with infectious mononucleosis (IM)-like symptoms, including fever, hepatosplenomegaly, lymphadenopathy, and hepatic dysfunction, often accompanied by pancytopenia or progressive cutaneous lesions, together with elevated EBVDNA levels in the peripheral blood^2–4. In contrast to EBV infection of B lymphocytes, CAEBV is not self-limiting and may rapidly deteriorate when complicated by uncontrollable hemophagocytic lymphohistiocytosis (HLH), cytokine release syndrome (CRS), multiple organ dysfunction syndrome (MODS), or progression to fatal hematologic malignancies⁵.

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is currently regarded as the only curative treatment for CAEBV because it can reconstitute the hematopoietic and immune systems^6–9. Bollard and Cohen¹⁰ proposed that allo-HSCT eradicates EBV⁺-T/NK lymphocytes through immune reconstitution mediated by donor-derived immunity and recommended initiating the transplantation as early as possible after CAEBV diagnosed. However, not all patients with CAEBV are eligible for or have access to allo-HSCT, and transplant-related mortality substantially influences treatment decisions. Non-transplant therapy, including combination chemotherapy, programmed death-1 (PD-1) blockade, CD20 monoclonal antibody, and cellular immunotherapy, have shown partial efficacy in clinical studies, by rapidly alleviating patient symptoms, reducing EBV⁺-T/NK lymphocytes, and slowing disease progression¹¹.

Given advances in basic and translational research, more clinical data are required to clarify whether allo-HSCT should be prioritized at diagnosis or whether non-transplant approaches can serve as alternatives in selected patients. These questions remain central to CAEBV management. In this context, we conducted a retrospective study at our center to compare outcomes of CAEBV patients treated with or without allo-HSCT. We evaluated overall response rate (ORR), overall survival (OS), and event-free survival (EFS) and explored prognostic risk and protective factors based on OS. Subgroup analyses were performed for patients without HLH during the disease course and patients who underwent HSCT.

Patients and methods

Patients

Patients diagnosed with CAEBV between January 2010 and December 2021 in the Department of Hematology, Tongji Hospital, Wuhan, China, were included in this retrospective study. CAEBV was diagnosed according to the criteria of the Japanese Ministry of Health and the 2016 World Health Organization (WHO) classification, and required all of the following: (1) persistent (>3 months) or recurrent IM-like symptoms, such as fever, pharyngotonsillitis, lymphadenopathy, and hepatosplenomegaly, with or without additional manifestations including vasculitis, uveitis, gastroenteritis, neuritis, or cutaneous lesions; (2) quantitatively elevated EBV-DNA levels in peripheral blood or t affected tissues, defined as >1 × 10^2.5 copies/µg DNA; (3) evidence of EBV infection of T or NK cells in tissues or peripheral blood; and (4) exclusion of other possible diseases including primary EBV infection (IM), autoimmune diseases, congenital immunodeficiencies, human immunodeficiency virus (HIV) infection, and other conditions requiring or associated with immunosuppressive therapy.

Exclusion criteria were as follows: (1) patients not fulfilling all diagnostic criteria for CAEBV; patients not fulfilling all diagnostic criteria for hydroa vacciniforme-like lymphoproliferative disorder (HV-LPD) and severe mosquito bite allergy (sMBA) are regarded as cutaneous forms of EBV-positive lymphoproliferative disease and, in accordance with the 2016 WHO classification, were not diagnosed as CAEBV; (2) patients with EBV-associated or other malignant lymphomas, including Hodgkin’s lymphoma (HL), extranodal NK/T-cell lymphoma (ENKTL), angioimmunoblastic T-cell lymphoma (AITL), peripheral T-cell lymphoma (PTCL), or aggressive NK-cell leukemia (ANKL), were not diagnosed. Patients with extranodal NK/T-cell lymphoma (ENKTL), angioimmunoblastic T-cell lymphoma (AITL), peripheral T-cell lymphoma (PTCL), and aggressive NK-cell leukemia (ANKL) are not eligible for a diagnosis of CAEBV. When CAEBV has been established and patients subsequently developed HLH or progressed to EBV-positive T/NK-cell leukemia or lymphoma during disease course, the primary diagnosis of CAEBV was retained.

Treatment

As this was a retrospective study, treatment allocation was not randomized. In accordance with international recommendations, all patients diagnosed with CAEBV were informed that allogeneic HSCT is currently considered the only potentially curative option. After initial conservative management for symptom control, treating physicians recommended HSCT to clinically eligible candidates. However, the final decision to proceed with HSCT versus continued non-transplant therapy was made by patients and their families based on preference and informed consent, reflecting real-world practice constraints.

HSCT group

Before transplantation, HLA typing was recommended for all patients with potential family donors and for the search for unrelated donor. The minimum required HLA compatibility was ≥5/10 for related donors and 10/10 for unrelated donors.

A conditioning regimen was administered to transplantation. The main regimens were as follows:

Fu/Bu/Cy/ATG: fludarabine (Fu), busulfan (Bu), cyclophosphamide (CTX), and antithymocyte globulin (ATG);

VP-16/Bu/Cy/ATG: Etoposide (VP-16), leucovorin, CTX, and ATG;

Ara-c/VP-16/Bu/Cy/ATG: Cytarabine (Ara-C), etoposide, busulfan, CTX, and ATG;

TBI/VP-16/Cy: Total body irradiation (TBI), etoposide and CTX.

Before conditioning and HSCT, all transplanted patients had received various conservative therapies, including but not limited to antiviral and HLH-directed therapy, combination chemotherapy, CD20 monoclonal antibody, and PD-1 blockade, tailored to individual clinical status. Detailed pretreatment information for each HSCT recipient is provided in Supplementary Table 1.

Peripheral blood and/or bone marrow hematopoietic stem cells from donor were collected to achieve a dose of at least 2.0 ×10⁶ CD34⁺ cells/kg recipient body weight. In patients with CAEBV undergoing HSCT, neutrophil engraftment was defined as an absolute neutrophil count ≥0.5 × 10⁹/L for at least 3 consecutive days. Platelet engraftment was defined as a platelet count ≥20 × 10⁹/L for at least 7 consecutive days without platelet transfusion. Successful engraftment was defined as both neutrophil and platelet engraftment within 28 days after allogeneic stem cell infusion. Neutrophil or platelet engraftment occurring after 28 days was defined as poor engraftment, and failure to achieve engraftment by 180 days was considered primary graft failure¹².

Non-HSCT group

Patients in the non-HSCT group received antiviral and HLH-directed therapy, combination chemotherapy, PD-1 blockade, anti-CD20 monoclonal antibody, or cellular immunotherapy. Treatment allocation reflected real-world practice, specific modalities were chosen at physician discretion according to disease activity, organ involvement, prior treatment response, and comorbidities, together with patient preference after informed counseling. Detailed variant distributions are summarized in Supplementary Table 2.

Antiviral (ART) and HLH-directed therapy

Patients received ganciclovir or penciclovir, together with HLH-directed therapy. In the presence of concomitant HLH, first-line immunochemotherapy included HLH-94 or HLH-04-based protocols (etoposide plus dexamethasone, with or without cyclosporine A), the DEP regimen (liposomal doxorubicin, etoposide, and methylprednisolone), and, in selected refractory cases, ruxolitinib or anti-CD25 monoclonal antibody for induction of remission.

Chemotherapy

Cytoreductive combination chemotherapy was used for overt lymphoproliferation, progressive organ infiltration, or failure of purely immunosuppressive regimens. Protocols included SMILE (methotrexate, dexamethasone, ifosfamide, leucovorin, and etoposide), P-GemOx (pegaspargase, gemcitabine, and oxaliplatin), MESA (methotrexate, etoposide, dexamethasone, and asparaginase), GAD-M (gemcitabine, asparaginase, dexamethasone, and methotrexate), CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone), and DA-EPOCH regimen (etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin).

PD-1 blockade

Anti-PD-1 therapy (sindilizumab or tirilizumab, 100–200 mg every 3 weeks) was reserved for patients with refractory inflammatory manifestations and/or progressive disease with PD-L1 expression, particularly when conventional chemotherapy or immunosuppression had failed or was contraindicated. The use of PD-1 inhibitors took into account the potential risk of cytokine release syndrome.

CD20 monoclonal antibody

Rituximab was employed in patients with EBV-driven HLH or B-cell lymphoproliferation, where rapid B-cell depletion was expected to reduce EBV burden and attenuate cytokine storm, providing a window for HLH control. A 200 mg single-dose regimen was used as an empiric, risk-adapted strategy distinct from the standard 375 mg/m² lymphoma dose. Given that CAEBV predominantly involves T/NK-cell, rituximab has limited activity against the primary pathogenic clone; thus, deep and sustained B-cell aplasia would disproportionately increase opportunistic infections risk in patients with baseline T-cell dysfunction and cytopenia. The reduced 200 mg dose achieves transient (approximately 2–4 weeks) B-cell suppression sufficient to dampen acute inflammatory activity (e.g. high fever, HLH) while preserving residual humoral immunity, prioritizing short-term stabilization and bridging rather than intensive, long-term immunosuppression.

Cellular immunotherapy

Selected patients with persistent or relapsing disease despite antiviral and immunochemotherapy received Epstein–Barr virus-specific cytotoxic T lymphocyte (EBV-CTL) therapy or T-cell receptor-like chimeric antigen receptor T-cell (TCR-like CART) therapy as salvage or bridging approaches.

Treatment protocols (choice of regimen, dose calculation, and schedule) were applied similarly in pediatric (<18 years) and adult (≥18 years) patients; all cytotoxic drugs were dosed by body surface area, and no statistically significant differences in treatment patterns between age groups were observed (Supplementary Table 4).

Genetic testing in this cohort covered STAT3, DDX3X, SH2D1A, XIAP, PRF1, UNC13D, STX11, STXBP2, RAB27A, AP3B1, LYST, and ITK. Variants were frequent and often monoallelic, consistent with polygenic susceptibility rather than classic monogenic HLH. Heterozygous variants in cytolytic-pathway genes (e.g. PRF1, UNC13D, STXBP2, STX11, RAB27A) and X-linked genes (SH2D1A, XIAP, including potential female carriers with skewed X-inactivation) were considered clinically relevant, prompting lower thresholds for initiating etoposide-based immunochemotherapy, intensifying immunosuppression, and expediting HSCT referral. This management-oriented framework acknowledges that even monoallelic or carrier states may modify CAEBV severity and therapeutic responsiveness, independent of strict hereditary HLH classification. Detailed variant distributions are summarized in Supplementary Table 2.

Endpoints

This retrospective study was conducted by review of medical records and telephone follow-up, with follow-up censored on 1 February 2022. OS in the HSCT group was defined as the interval from stem cell infusion to death from any cause or last follow-up; OS in the non-HSCT group was defined as the time from the first receipt of the treatment to death or follow-up cutoff time. EFS in the HSCT group was defined as the interval from HSCT infusion to relapse, progression, death, or last follow-up, and in the non-HSCT group as the time from first treatment to relapse, progression, death, or last follow-up.

Treatment response was assessed at 3 months. Virological complete response (vCR) defined as complete resolution of clinical symptoms with peripheral blood EBV-DNA <500 copies/ml. Complete response (CR) was defined as complete symptom resolution with EBV-DNA ≥500 copies/ml. Partial response (PR) was defined as partial improvement of clinical symptoms. Stable disease (SD) was defined as absence of symptoms remission but without uncontrollable complications or evolution to T/NK-cell neoplasia. Progressive disease (PD) was defined as difficult-to-control complications, transformation to T/NK-cell neoplasia, or death. ORR was defined as vCR + CR + PR.

HLH complicating the course of CAEBV was diagnosed when at least five standard HLH-2004 criteria were present before initiation of systemic therapy^13,14: including fever ≥38.5°C, hepatosplenomegaly, cytopenias in ≥2 lineages, hypertriglyceridemia and/or hypofibrinogenemia, hemophagocytosis without malignancy, reduced/absent NK-cell activity, serum ferritin ≥500 µg/L, and soluble interleukin-2 receptor (sIL- 2R) ≥2400 U/ml.

Progressing to T/NK-cell neoplasia was defined by peripheral or immunophenotypic detection of T/NK-cell lymphoma or leukemia in peripheral blood or tissue specimens.

Statistical analyses

All statistical analyses were performed using GraphPad Prism (version 9.5) and SPSS (version 27). Categorical variables were compared using the χ² test or Fisher’s exact test, as appropriate, and continuous variables using the Student’s t test or Kruskal–Wallis test, according to distributional assumptions. Distributions for OS and EFS were estimated with the Kaplan–Meier method and compared by the log-rank test. Multivariable analyses of prognostic factors were conducted using Cox proportional hazards regression models. Variables with P < 0.05 in univariable analyses were entered into stepwise multivariate Cox regression model. Two-sided P < 0.05 were considered statistically significant.

Results

Baseline characteristics of the cohort

Baseline variables collected included age, gender, dominant EBV-infected subset (T or NK), elevated EBV-DNA level (defined as ≥1×10⁴ copies/ml), multilineage cytopenia (≥2 lineages), hepatic dysfunction (LT ≥123 U/L), hyperpyrexia (≥39°C), hepatosplenomegaly, lymphadenopathy, occurrence of HLH during the disease course, and HLH-related gene mutations (STAT3, DDX3X, XIAP, PRF1, UNC13D, STX11, STXBP2, RAB27A, AP3B1, LYST, and ITK).

Seventy-five patients with CAEBV were included. Median age was 30 years (range, 2–73), and 58.7% (44/75) were male. NK-cell-type CAEBV accounted for 70.7% (53/75), and T-cell-type for 29.3% (22/75). Among 50 patients with pre-treatment plasma EBV-DNA testing, 50.0% (25/50) had elevated EBV-DNA. Multilineage cytopenia was present in 53.5% (40/75), hepatic dysfunction in 42.7% (32/75), and hyperpyrexia in 66.7% (50/75). Among 72 patients with organ imaging, hepatosplenomegaly occurred in 70.7% (53/72) and lymphadenopathy in 79.2% (57/72). HLH developed during the disease course in 49.3% (37/75). Of 65 patients undergoing genetic testing, 42.2% (27/65) carried at least one HLH-related gene variant.

Patients were classified into HSCT (n = 15) and non-HSCT (n = 60) groups. Apart from less frequent hyperpyrexia at the onset in the HSCT group (26.7% vs 76.7%, P < 0.001), baseline features were comparable between HSCT and non-HSCT groups, including age, gender, CAEBV subtype, plasma EBV-DNA elevation, multilineage cytopenia, hepatic dysfunction, hepatosplenomegaly, lymphadenopathy, HLH during the disease course, and presence of HLH-related gene variants (all P >0.05; Table 1). Pediatric (<18 years) and adult (≥18 years) patients showed no significant differences in baseline characteristics; a nonsignificant trend toward lower HLH incidence was observed in pediatric patients (Supplementary Table 3).

Table 1.

Baseline characteristics of CAEBV patients.

Characteristic	Whole series	HSCT group	Non-HSCT group	P value
Number	75	15	60
Age, median (range)	30 (2–73)	30 (12–55)	30 (2–73)	0.5887
Gender, male	44 (58.7%)	9 (60.0%)	35 (58.3%)	0.2521
Dominant EBV-infected subset, NK-type	53 (70.7%)	8 (53.3%)	45 (75.0%)	0.1198
Elevated plasma EBV-DNA level^a	25 (50%)	4 (26.7%)	21 (60.0%)^b	0.0876
Multilineage cytopenia	40 (53.3%)	9 (60.0%)	31 (51.7%)	0.7732
Hepatic dysfunction^c	32 (42.7%)	4 (26.7%)	28 (46.7%)	0.398
Hyperpyrexia^d	50 (66.7%)	4 (26.7%)	46 (76.7%)	<0.0001
Hepatosplenomegaly	53 (70.7%)	12 (80.0%)	41 (71.9%)^e	0.5302
Lymphadenopathy	57 (79.2%)	12 (80.0%)	45 (78.9%)^f	>0.9999
Concomitant HLH	37 (49.3%)	7 (46.7%)	30 (50.0%)	0.3861
Gene mutations^g	27 (42.2%)	6 (40.0%)	21 (42.9%)^h	0.7158

Elevated plasma EBV-DNA level (defined as ≥1×10⁴ copies/ml).

40 of 65 patients in the non-HSCT group were tested.

Elevated plasma EBV-DNA level (defined as ≥1×10⁴ copies/ml).

Hyperpyrexia (≥39°C).

57 of 65 patients in the non-HSCT group were tested.

Gene mutations: STAT3, DDX3X, SH2D1A, XIAP, PRF1, UNC13D, STX11, STXBP2, RAB27A, AP3B1, LYST, ITK.

49 of 65 patients in the non-HSCT group were tested.

Of the 75 patients, 15 (20.0%) underwent HSCT, including 5 with HLH and 10 without HLH. Six patients (80.0%) received non-HSCT treatments: PD-1 blockade (10 cases), CD20 monoclonal antibody (11), ART and HLH-directed therapy (24), chemotherapy (13), and EBV-specific CTL therapy (2). Among patients receiving PD-1 blockade, 1 had HLH and 9 did not (Fig. 1).

Figure 1.

Flow chart of CAEBV patients in this study.

Treatment characteristics

CAEBV treated with HSCT or non-HSCT

Among the 15 CAEBV patients in the HSCT group, treatment response at 3 months was evaluated as vCR in 6 patients (40.0%), CR in 6 patients (40.0%), PD in 3 patients (20.0%), and no patients achieved PR or SD. In the 60 CAEBV patients in the non-HSCT group, treatment response was vCR in 4 (6.7%), CR in 15 (25.0%), PR in 2 (3.3%), SD in 10 (16.7%), and PD in 29 (48.3%). The ORR was therefore 80.0% in the HSCT group and 35.0% in the non-HSCT group, indicating a significantly higher ORR in transplanted patients (P = 0.003; Fig. 2a).

Figure 2.

ORR, OS, and EFS in HSCT group and non-HSCT group in CAEBV ((a), (b), and (c)). OS in CAEBV with and without HLH (d). OS in CAEBV using and not using PD-1 blockade (e).

At the data cutoff, the median follow-up was 14.7 months (range, 0.9–64.9) in the HSCT group and 3.6 months (range, 0.2–59.9) in the non-HSCT group. In the HSCT group, 4 of 15 patients had died and 11 were alive. In the non-HSCT group, 42 of 60 patients had died and 18 patients alive. The 3-year OS rate was 70.7% in the HSCT group and 28.5% in the non-HSCT group, demonstrating superior survival in transplanted patients (P = 0.003; Fig. 2b). The relatively short median follow-up in the non-HSCT cohort reflects high early mortality (3-year OS 28.5%) rather than insufficient observation time.

With respect to events, by the end of follow-up, 4 HSCT recipients had died, 1 had relapsed, and 10 remained event-free. In contrast, among non-HSCT patients, 2 had died without relapse, 42 patients relapsed, and 16 remained event-free. The 3-year EFS rate was 70.7% in the HSCT group and 22.6% in the non-HSCT group (P < 0.001; Fig. 2c).

Analysis of independent factors affecting OS

Cox univariate and multivariate regression analyses were performed to identify factors associated with OS in 75 CAEBV patients. In univariate analysis, concomitant HLH during the disease course (P = 0.001, HR: 6.539, 95% CI: 0.221–19.25), antiviral and HLH-directed therapy (P = 0.011, HR: 4.808, 95% CI: 1.440–16.254), and chemotherapy-containing regimens (P = 0.033, HR: 9.882, 95% CI: 1.210–80.375) were associated with inferior OS, whereas PD-1 blockade (P = 0.006, HR: 0.049, 95% CI: 0.006–0.566) and HSCT (P = 0.004, HR: 0.156, 95% CI: 0.044–0.555) were associated with improved OS.

Variables entering the multivariate model included age ≤14 years, gender type of EBV-infected cell, marked elevation of plasma EBV-DNA, multilineage cytopenia, hepatic insufficiency, presence of gene mutations, use of CD20 monoclonal antibody, antiviral and HLH-directed therapy, chemotherapy, and cellular immunotherapy. In multivariate analysis, allo-HSCT (P = 0.027, HR: 0.295, 95% CI: 0.100–0.872) and PD-1 blockade (P = 0.021, HR: 0.091, 95% CI: 0.012–0.694) remained independent protective factors for OS, whereas HLH during the disease course was an independent adverse prognostic factor (P = 0.017, HR: 2.145, 95% CI: 1.149–4.005) (Table 2).

Table 2.

Cox univariate and multivariate regression analyses in CAEBV.

	Univariate analysis			Multivariate analysis
	P	HR	95% CI	P	HR	95% CI
Age (≤14 years)	0.147	0.360	0.093-1.427
Gender (male)	0.995	1.003	0.390-2.577
Dominant EBV-infected subset (T-type)	0.073	0.394	0.142-1.089
Elevated plasma EBV-DNA level	0.420	1.598	0.512-4.993
Elevated MNC EBV-DNA	0.391	0.375	0.400-3.533
Multilineage cytopenia	0.110	3.562	1.341-9.462
Hepatic dysfunction	0.076	2.424	0.913-6.437
Gene mutations	0.279	2.207	0.527-9.249
Concomitant HLH	0.001	6.539	0.221-19.25	0.017	2.145	1.149-4.005
PD-1 blockade	0.006	0.049	0.006-0.566	0.021	0.091	0.012-0.694
CD20 monoclonal antibody	0.865	1.122	0.298-4.229
Antiviral and HLH-directed therapy	0.011	4.808	1.440-16.254	0.074	1.766	0.946-3.297
Chemotherapy	0.033	9.882	1.210-80.735	0.335	1.536	0.642-3.675
Cellular Immunotherapy	0.999	1.000	0.000
Allo-HSCT	0.004	0.156	0.044-0.555	0.027	0.295	0.1-0.872

P < 0.05 for multifactorial inclusion.

Subgroup analyses

CAEBV within the HLH subgroup

Based on the presence of HLH during the disease, the 75 CAEBV patients were categorized into an HLH (n = 37) and non-HLH group (n = 38). By the data cutoff, the median follow-up was 2.7 months (range, 0.2–44.7) in the HLH group and 8.5 months (range, 0.9–64.9) in the non-HLH group. In the HLH group, 1 patient was lost to follow-up, 28 had died, and 8 patients were alive; in the non-HLH group, 6 were lost, 16 had died, and 16 were alive. The 3-year OS rate was 18.8% in patients with CAEBV complicated by HLH and 52.5% in those without HLH (P < 0.001; Fig. 2d).

CAEBV patients treated with or without PD-1 blockade

According to whether PD-1 blockade was used as primary therapy, patients were divided into a PD-1 group (n = 10) and a non-PD-1 group (n = 65). The median follow-up was 9.0 months (range, 3.0–44.0) in the PD-1 group, and 3.1 months (range, 0.2–64.9) in the non-PD-1 group. At the cutoff, 1 of 10 patients in the PD-1 group had died and 9 were alive, whereas 45 of 65 patients in the non-PD-1 group had died and 19 were alive. The 3-year OS rate was 90.0% in patients receiving PD-1 blockade and 27.2% in those not receiving PD-1 blockade (P = 0.001; Fig. 2e).

CAEBV treated with HSCT or PD-1 blockade

Among the 75 CAEBV patients, 10 received PD-1 blockade (PD-1 group) and 15 patients allo-HSCT (HSCT group). In the PD-1 group, best responses were vCR in 1 patients (10.0%), CR in 5 (50.0%), SD in 3 (30.0%), PD in 1 (10.0%); no patient achieved PR. In the HSCT group, 6 patients (40.0%) achieved vCR, 6 (40.0) achieved CR, and 3 (20.0%) had PD; no PR or SD was observed. The ORR was 60.0% in the PD-1 group and 80.0% in the HSCT group, with no statistically significant difference between the two group (P = 0.378; Fig. 3a).

Figure 3.

ORR, OS, and EFS in PD-1 group and HSCT group in CAEBV ((a), (b), and (c)). ORR, OS, and EFS in PD-1 group and HSCT group in CAEBV without HLH ((d), (e), and (f)). ORR, OS, and EFS in HSCT group in CAEBV with or without HLH ((g), (h), and (i)).

By the cutoff, the median follow-up was 9.0 months (range, 3.0–44.0) in the PD-1 group (1 death, 9 alive) and 14.7 months (range, 0.9–64.9) in the HSCT group (4 deaths, 11 alive). The 3-year OS rate was 90.0% in PD-1 group and 70.7% in HSCT group (P = 0.258; Fig. 3b). Regarding events, 1 patient in the PD-1 group had died, 4 remained event-free, and 5 relapsed; in the HSCT group, 4 had died, 10 remained event-free, and 1 relapsed. The 3-year EFS rate was 63.5% in the PD-1 group and 70.7% in the HSCT group, with no statistically significant (P = 0.818; Fig. 3c).

CAEBV without HLH treated with HSCT or PD-1 blockade

Among patients without HLH, 9 received PD-1 blockade and 10 received allo-HSCT. In the 9 evaluable PD-1-treated patients, responses were vCR in 1 (11.1%), CR in 5 (55.5%), SD in 2 (22.2%), and PD in 1 (11.1%); none achieved PR. In the 10 non-HLH patients undergoing HSCT, 4 (40.0%) achieved vCR and 6 (60.0%) achieved CR, with no PR, SD, or PD. The ORR was 66.7% in PD-1 group and 80.0% in HSCT group (P = 0.378; Fig. 3d).

By the cutoff, among non-HLH patients, the median follow-up was 18.5 months (range, 3.0–44.0) in the PD-1 (1 death, 8 alive) group and 15.1 (range, 0.9–64.9) in the HSCT group (1 death, 9 alive). The 3-year OS was 88.9% in patients treated with PD-1 blockade and 87.5% in those treated with HSCT (P > 0.999; Fig. 3e). For EFS, among the 9 PD-1 blockade-treated patients, 3 were event-free and 6 relapsed. Among the 10 HSCT-treated patient, 1 died, 8 were event-free, and 1 relapsed. The 3-year EFS was 58.3% in the PD-1 group and 87.5% in the HSCT group; numerically higher EFS with allo-HSCT did not reach statistical significance (P = 0.437; Fig. 3f).

CAEBV with or without HLH treated with HSCT

Of the 15 CAEBV patients who underwent allo-HSCT, 5 had concomitant HLH (transplanted HLH group) and 10 did not (transplanted non-HLH group). In the transplanted HLH group, two patients (40.0%) achieved vCR and three (60.0%) achieved CR; no PR SD, or PD occurred. In the transplanted non-HLH group, five patients (50.0%) achieved vCR, four (40.0%) achieved CR, and one (10.0%) had PD; no PR or SD occurred. The ORR was 100% in the transplanted HLH group and 90% in the transplanted non-HLH group (P > 0.999; Fig. 3g).

By the end of follow-up, the median follow-up duration was 13.0 months (range, 1.5–40.8) in the transplanted HLH group (3 deaths, 2 alive) and 15.1 months (range, 0.9-64.9) in the transplanted non-HLH group (1 death, 9 alive). The 3-year OS was 40.0% in the transplanted HLH group and 87.5% in the transplanted non-HLH group (P = 0.034; Fig. 3h). For EFS, 3 of 5 HLH patients had died and 2 remained event-free, whereas among the 10 non-HLH patients, 1 had died, 8 remained event-free, and 1 relapsed. The 3-year EFS was 40.0% in the transplanted HLH group and 87.5% in the transplanted non-HLH group (P = 0.034; Fig. 3i).

In the transplanted HLH subgroup, all three deaths were attributable to transplant-related mortality (TRM). Patient 11 died on day +46 from infectious complications, including Pneumocystis jirovecii pneumonia and Stenotrophomonas maltophilia sepsis, accompanied by concurrent CMV viremia and post-transplant lymphoproliferative disorder (PTLD). Patient 12 died on day +65 due to transplant-associated thrombotic microangiopathy (TA-TMA). Patient 14 died on day +86 from graft failure complicated by systemic infection and hepatic failure before hematopoietic recovery. In the transplanted non-HLH subgroup, the single death (Patient 3) resulted from TA-TMA at 11 months post-transplantation. No deaths in either subgroup were attributed to disease recurrence.

Discussion

In this retrospective cohort of 75 patients with CAEBV, we systematically compared allo-HSCT, PD-1 blockade, and other non-transplant modalities and further evaluated the prognostic impact of HLH complicating the disease course. Consistent with previous multicenter series, allo-HSCT remained the only modality associated with durable disease control in the majority of patients, while PD-1 blockade emerged as a promising non-transplant strategy, particularly in patients without HLH. Importantly, multivariable analysis identified the occurrence of HLH during the disease course as an independent adverse prognostic factor, whereas both allo-HSCT and PD-1 blockade were independent protective factors for overall survival.

In our cohort, the HSCT group achieved significantly higher ORR (80.0% vs 35.0%), 3-year OS (70.7% vs 28.5%), and 3-year EFS (70.7% vs 22.6%) compared with the non-HSCT group. These data align with large Japanese and international series reporting 3-year OS rates of approximately 65%–82% after allo-HSCT, versus dismal survival with chemotherapy or immunotherapy alone¹⁵. Taken together, these results support the current view that allo-HSCT remains the cornerstone and only potentially curative treatment for CAEBV^10,16.

PD-1 blockade, as an immune checkpoint inhibitor, plays an important role in the management of EBV-associated diseases¹⁷. In a prospective study of CAEBV, combination therapy with PD-1 blockade and lenalidomide was both effective and well tolerated, achieving an overall response rate of 54.2%, significantly decreasing EBV-DNA copies in peripheral blood, and increasing the proportion of CD8⁺ T lymphocytes¹⁸. In our study, use of PD-1 inhibitors was an independent protective factor for OS. When we directly compared patients treated with PD-1 blockade versus those undergoing allo-HSCT, the ORR, 3-year OS, and 3-year EFS were 60.0% versus 80.0%, 90.0% versus 70.7%, and 63.5% versus 70.7%, respectively, without statistically significant differences. However, these results should be interpreted with caution because baseline disease severity differed between groups: only 10% of patients in the PD-1 cohort had concomitant HLH, compared with 33.3% in the HSCT cohort.

To reduce confounding by HLH, we performed a subgroup analysis restricted to patients without HLH during the disease course. After excluding this independent risk factor, patients treated with PD-1 blockade still tended to have a lower ORR than those receiving allo-HSCT (66.7% vs 100.0%, P = 0.087), and their 3-year EFS was numerically inferior (58.3% vs 87.5%, P = 0.437), despite similar 3-year OS (88.9% vs 87.5%, P > 0.999). Thus, our data do not support a conclusion that PD-1 blockade is therapeutically equivalent to allo-HSCT in CAEBV; rather, they indicate that PD-1 blockade has considerable therapeutic potential and may provide survival benefits comparable to allo-HSCT in carefully selected patients, especially those without HLH^19,20.

Consistent with previous reports, HLH was common and strongly associated with poor outcomes^13,16,21,22. In a prospective cohort of 121 patients, the cumulative incidence of HLH reached 24.9%, 47.3%, 55.1%, and 85.2% at 1, 3, 5, and 10 years after diagnosis, respectively²³.

In our cohort, 37 of 75 patients developed HLH. The 3-year OS was only 18.8% in patients with HLH versus 52.5% in those without HLH (P < 0.001), and multivariable analysis confirmed HLH during the disease course as an independent risk factor for OS. These findings are in line with prior long-term observational studies showing a high cumulative incidence of HLH in CAEBV and its association with aggressive disease, cytokine storm, and multi-organ failure.

We further examined the impact of HLH among patients who underwent allo-HSCT. CAEBV patients receiving transplantation were divided into a transplanted HLH group (n = 5) and a transplanted non-HLH group (n = 10). Although ORR was high in both groups (100.0% vs 90.0%; P > 0.999), patients with HLH had significantly worse long-term outcomes, with 3-year OS and EFS of 40.0% versus 87.5% (both P = 0.034). Importantly, all three deaths in the transplanted HLH subgroup were due to transplant-related mortality rather than disease recurrence. One patient died on day +46 from infectious complications with concurrent CMV viremia and post-transplant lymphoproliferative disorder, another on day +65 from TA-TMA, and the third on day +86 from graft failure with systemic infection and hepatic failure prior to hematopoietic recovery. In the transplanted non-HLH group, the single death was also attributable to TA-TMA at 11 months post-transplant. These patterns suggest that although allo-HSCT improves prognosis in CAEBV overall, patients with HLH at any point during the disease course remain at high risk of transplant-related complications and cannot achieve outcomes comparable to those without HLH. This observation supports an early-transplant strategy in proceeding to allo-HSCT as soon as feasible after CAEBV is diagnosed and before irreversible immune dysregulation and HLH-related organ damage develop.

In our study, one patient who received TCR-like CAR T cells died from HLH-related complications during therapy. EBV-CTLs likewise produced only short-lived responses: the patient treated in our cohort experienced temporary clinical improvement but relapsed within 1 month. EBV-CTL is probably not suitable for CAEBV. EBV-CTLs predominantly clear type III latently infected B cells and thus fail to control CAEBV, in which the infected T/NK cells exhibit type I/II latency. Consistent with this biology, patients in our study showed only transient symptom relief after EBV-CTL, with relapse within 1 month, and previously reported CAEBV cases also failed to achieve durable clinical benefit^24–26.

Conclusion

This study has several limitations, including its retrospective single-center design, limited sample size, particularly in the PD-1 and transplanted-HLH subgroups, and short follow-up for some patients. Treatment allocation was not randomized and was influenced by disease severity, comorbidities, and patient preference, which may introduce selection bias, especially in comparisons between PD-1 blockade and HSCT. Furthermore, comprehensive virological and immunological correlative studies were not available for all patients. Despite these limitations, our findings provide important real-world evidence that HLH during the disease course is a major determinant of poor prognosis in CAEBV, that allo-HSCT significantly improves long-term outcomes but does not fully offset the adverse impact of HLH, and that PD-1 blockade represents a promising non-transplant option and an independent protective factor for OS, particularly in patients without HLH.

Clinically, these findings support early consideration of allo-HSCT at the time of CAEBV diagnosis, careful monitoring and aggressive management of HLH, and the incorporation of PD-1 blockade either as a bridge to transplant or as an alternative in patients who are not HSCT candidates. Future prospective, multicenter studies with larger cohorts are needed to validate the role of PD-1 inhibitors, optimize transplant timing and conditioning, and clarify the place of emerging cellular therapies in the treatment algorithm for CAEBV.

Supplemental Material

sj-docx-1-cll-10.1177_09636897261443609 – Supplemental material for Analysis of the efficacy and survival of allogeneic hematopoietic stem cell transplantation and nontransplantation treatment in chronic active Epstein–Barr virus infection

Supplemental material, sj-docx-1-cll-10.1177_09636897261443609 for Analysis of the efficacy and survival of allogeneic hematopoietic stem cell transplantation and nontransplantation treatment in chronic active Epstein–Barr virus infection by Xi Ming, Xiaoying Zhang, Jiaying Wu, Wanying Liu, Qi Zhang, Delian Zhou, Yicheng Zhang, Xiaojian Zhu and Yi Xiao in Cell Transplantation

Supplemental Material

sj-docx-2-cll-10.1177_09636897261443609 – Supplemental material for Analysis of the efficacy and survival of allogeneic hematopoietic stem cell transplantation and nontransplantation treatment in chronic active Epstein–Barr virus infection

Supplemental material, sj-docx-2-cll-10.1177_09636897261443609 for Analysis of the efficacy and survival of allogeneic hematopoietic stem cell transplantation and nontransplantation treatment in chronic active Epstein–Barr virus infection by Xi Ming, Xiaoying Zhang, Jiaying Wu, Wanying Liu, Qi Zhang, Delian Zhou, Yicheng Zhang, Xiaojian Zhu and Yi Xiao in Cell Transplantation

Supplemental Material

sj-docx-3-cll-10.1177_09636897261443609 – Supplemental material for Analysis of the efficacy and survival of allogeneic hematopoietic stem cell transplantation and nontransplantation treatment in chronic active Epstein–Barr virus infection

Supplemental material, sj-docx-3-cll-10.1177_09636897261443609 for Analysis of the efficacy and survival of allogeneic hematopoietic stem cell transplantation and nontransplantation treatment in chronic active Epstein–Barr virus infection by Xi Ming, Xiaoying Zhang, Jiaying Wu, Wanying Liu, Qi Zhang, Delian Zhou, Yicheng Zhang, Xiaojian Zhu and Yi Xiao in Cell Transplantation

Supplemental Material

sj-docx-4-cll-10.1177_09636897261443609 – Supplemental material for Analysis of the efficacy and survival of allogeneic hematopoietic stem cell transplantation and nontransplantation treatment in chronic active Epstein–Barr virus infection

Supplemental material, sj-docx-4-cll-10.1177_09636897261443609 for Analysis of the efficacy and survival of allogeneic hematopoietic stem cell transplantation and nontransplantation treatment in chronic active Epstein–Barr virus infection by Xi Ming, Xiaoying Zhang, Jiaying Wu, Wanying Liu, Qi Zhang, Delian Zhou, Yicheng Zhang, Xiaojian Zhu and Yi Xiao in Cell Transplantation

Footnotes

Acknowledgements

The authors sincerely thank all colleagues in Department of Hematology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, for their help in this study.

ORCID iDs

Xi Ming

Jiaying Wu

Wanying Liu

Qi Zhang

Yi Xiao

Ethical considerations

This retrospective study involving human participants was approved by the Institutional Review Board of Tongji Hospital on 16 October 2025 (Approval No. TJ-IRB20250042).

Consent to participate

The requirement for written informed consent was waived by the IRB due to the retrospective nature of the study.

Author contributions

X.M. collected, analyzed the data, and wrote the main manuscript text; X.Z., J.W., W.L. Q.Z., and D.Z. helped to collected the data and gave critical suggestions; Y.Z., Y.X., and X.Z. conceived and designed the study and revised the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China (Grant Nos. 82070213 and 82370196).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

All data has been outlined in TABLEs. Raw data can be obtained by contacting the corresponding author.

Supplemental material

Supplemental material for this article is available online.

Statement of human and animal rights

This article does not contain any studies with human or animal subjects.

Statement of informed consent

There are no human subjects in this article and informed consent is not applicable.

References

Luo

Liu

Jin

Zhang

Zheng

Xiao

Liu

Zhou

, et al. Clinical and genetic characterization of Epstein-Barr virus-associated T/NK-cell lymphoproliferative diseases. J Allergy Clin Immunol. 2023;151(4):1096–1109.

Arai

. Advances in the study of chronic active Epstein-Barr virus infection: clinical features under the 2016 WHO classification and mechanisms of development. Front Pediatr. 2019;7:14.

Dojcinov

Fend

Quintanilla-Martinez

. EBV-positive lymphoproliferations of B- T- and NK-cell derivation in non-immunocompromised hosts. Pathogens. 2018;7(1):28.

Hue

Oon

Wang

Tan

. Epstein-Barr virus-associated T- and NK-cell lymphoproliferative diseases: an update and diagnostic approach. Pathology. 2020;52(1):111–27.

Gotoh

Ito

Shibata-Watanabe

Kawada

Takahashi

Yagasaki

Kojima

Nishiyama

Kimura

. Clinical and virological characteristics of 15 patients with chronic active Epstein-Barr virus infection treated with hematopoietic stem cell transplantation. Clin Infect Dis. 2008;46(10):1525–34.

Sawada

Inoue

. Hematopoietic stem cell transplantation for the treatment of Epstein-Barr virus-associated T- or NK-cell lymphoproliferative diseases and associated disorders. Front Pediatr. 2018;6:334.

Sato

. Challenges in managing EBV-associated T- and NK-cell lymphoproliferative diseases. Front Pediatr. 2018;6:320.

Cohen

Jaffe

Dale

Pittaluga

Heslop

Rooney

Gottschalk

Bollard

Rao

Marques

Burbelo

, et al. Characterization and treatment of chronic active Epstein-Barr virus disease: a 28-year experience in the United States. Blood. 2011;117(22):5835–49.

Yamamoto

Sato

Onishi

Sasahara

Sano

Masuko

Nakamae

Matsuoka

K-I

Ara

Washio

Onizuka

, et al. Registry data analysis of hematopoietic stem cell transplantation on systemic chronic active Epstein-Barr virus infection patients in Japan. Am J Hematol. 2022;97(6):780–90.

10.

Bollard

Cohen

. How I treat T-cell chronic active Epstein-Barr virus disease. Blood. 2018;131(26):2899–2905.

11.

Zhang

Bao

Luo

Wang

Huang

Zheng

. Outcomes of programmed death protein-1 inhibitors treatment of chronic active Epstein Barr virus infection: a single center retrospective analysis. Front Immunol. 2023;14:1093719.

12.

Shi

Kong

Song

Sun

Wang

Zhang

L-P

Liu

Huang

X-J

. Atorvastatin enhances endothelial cell function in posttransplant poor graft function. Blood. 2016;128(25):2988–99.

13.

Jordan

Allen

Weitzman

Filipovich

McClain

. How I treat hemophagocytic lymphohistiocytosis. Blood. 2011;118(15):4041–52.

14.

Schram

Berliner

. How I treat hemophagocytic lymphohistiocytosis in the adult patient. Blood. 2015;125(19):2908–14.

15.

Yonese

Sakashita

Imadome

Kobayashi

Yamamoto

Sawada

Ito

Fukuhara

Hirose

Takeda

Makita

, et al. Nationwide survey of systemic chronic active EBV infection in Japan in accordance with the new WHO classification. Blood Adv. 2020;4(13):2918–26.

16.

Sawada

Inoue

Kawa

. How we treat chronic active Epstein-Barr virus infection. Int J Hematol. 2017;105(4):406–18.

17.

Sasaki

Nishikawa

Sakai

Iizasa

Yoshiyama

Yanagihara

Shuto

Shimokuri

Kanda

Suehiro

Yamasaki

, et al. EBV-associated gastric cancer evades T-cell immunity by PD-1/PD-L1 interactions. Gastric Cancer. 2019;22(3):486–96.

18.

Song

Wang

You

Song

Chen

Meng

Cui

Wang

, et al. PD-1 blockade and lenalidomide combination therapy for chronic active Epstein-Barr virus infection. Clin Microbiol Infect. 2023;29(6):796.e7–96.e13.

19.

Wang

Xiong

Jiang

Zhu

Wang

Gong

Zhang

, et al. Epstein-Barr virus-encoded circular RNA CircBART2.2 promotes immune escape of nasopharyngeal carcinoma by regulating PD-L1. Cancer Res. 2021;81(19):5074–88.

20.

Xie

Huang

Qian

. Prognostic and clinicopathological significance of PD-1/PD-L1 expression in the tumor microenvironment and neoplastic cells for lymphoma. Int Immunopharmacol. 2019;77:105999.

21.

Liu

Pan

Chen

Niu

Shuai

Wang

Chen

Liu

Xie

Liu

, et al. Nivolumab treatment of relapsed/refractory Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis in adults. Blood. 2020;135(11):826–33.

22.

Sonke Gabe

Ludwig

van Oosten

Baars Joke

Meijer

Kater Arnon

de Jong

. Poor outcomes of chronic active Epstein-Barr virus infection and hemophagocytic lymphohistiocytosis in non-Japanese adult patients. Clin Infect Dis. 2008;47(1):105–108.

23.

Wang

Song

Wang

. Development of a nomogram to predict the risk of chronic active Epstein-Barr Virus infection progressing to hemophagocytic lymphohistiocytosis. Front Med (Lausanne). 2022;9:826080.

24.

McAulay

Haque

Urquhart

Bellamy

Guiretti

Crawford

. Epitope specificity and clonality of EBV-specific CTLs used to treat posttransplant lymphoproliferative disease. J Immunol. 2009;182(6):3892–3901.

25.

Meedt

Weber

Bonifacius

Eiz-Vesper

Maecker-Kolhoff

Delecluse

H-J

Lorenz

Schwarz

Meedt

Braess

, et al. Chronic active Epstein-Barr Virus (EBV) infection controlled by allogeneic stem cell transplantation and EBV-specific T cells. Clin Infect Dis. 2023;76(12):2200–2202.

26.

Kimura

Hoshino

Hara

Sugaya

Kawada

J-I

Shibata

Kojima

Nagasaka

Kuzushima

Morishima

. Differences between T cell-type and natural killer cell-type chronic active Epstein-Barr virus infection. J Infect Dis. 2005;191(4):531–39.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.04 MB

0.02 MB

0.03 MB

0.04 MB

0.00 MB