Introduction
Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a potentially curative therapy for hematological malignancies. However, infection, relapse and graft-versus-host disease (GvHD) remain the major therapeutic challenges and affect the transplant outcomes. The quality of reconstitution of the donor-derived immune system in the recipient is of utmost importance for long-term survival after allo-HSCT [
1‐
3]. Immune reconstitution (IR) would be completed in 2–5 years after allo-HSCT in a continuous and stepwise pattern, and this process would be affected by various factors [
4]. Different kinds of immune cell subsets recover with different dynamics. After intensive conditioning, transplant recipients were in the “pre-engraftment phase” with prolonged neutropenia. Previous studies have confirmed that ATG_based prophylaxis are related to superior CD8
+ T, γδ T, natural killer (NK) and NKT cell reconstitution, while the reconstitution of CD4
+ T, regulatory T and B cell are faster in PTCy-based GvHD prophylaxis [
5,
6] Neutrophil recovers within 14–30 days after graft infusion which defends against bacterial and fungal infections, which may be delayed after PTCy treatment, resulting more infections before day 100 [
6]. The initial 100 days after transplantation are characterized by cellular immunodeficiencies due to a paucity of NK and T cells [
7]. The compromised T cell reconstitution is primarily responsible for deleterious viral reactivations, including cytomegalovirus (CMV) and Epstein–Barr virus (EBV), as well as related viral end-organ diseases, rendering major reasons for morbidity and mortality for allo-HSCT [
8].
Successful IR was defined as CD4
+ T cells > 50 cells/μL in two consecutive measurements within 100 days after allo-HSCT [
9,
10], which was associated with increased overall survival (OS) [
9] and reduced non-relapse mortality (NRM) [
10]. Besides, patients who received ATG treatment showed slower lymphocytes [
11], CD4
+ T, CD4
+ CD25
+ CD127
− T (Treg) and CD4
+ CD25
− CD127
+ T (Tconv) cell reconstitution [
12] and higher rates of EBV reactivation [
6], which indicated a potential association of EBV reactivation and CD4
+ T subsets. However, the predictive value of CD4
+ T cell recovery for the risk of viral reactivation is still weak [
9,
10]. About 90% people in developing countries had primary EBV and CMV infections during childhood and adolescence [
13]. When allo-HSCT was performed, the majority of EBV and CMV reactivations were usually observed within 3 months due to deficiency of normal cytotoxic T-cell monitoring [
14,
15]. Persistent EBV reactivation is the most important risk factor for EBV-related post-transplant lymphoproliferative disorders (PTLD), resulting in dismal prognosis [
16].
Furthermore, T-cell includes a panel of compartments, and CD4
+ T-cell subset is not sufficient to recapitulate the status of cellular IR. Several lines of studies demonstrated the predictive value of other compartments of T-cells for transplant-related complications. Khandelwal et al. reported that CD38
bright CD8
+ TEM > 35 cells/μL could predict acute GvHD (aGvHD) at a median of 8 days ahead of aGvHD onset [
17]. Recent studies found that reduced numbers of total CD4
+ T cells and naive CD4
+ T cells at day 28 were significantly correlated with more infections [
18]. Camargo et al. proposed that the absolute number of IL2
−IFNγ
+TNFα
−MIP1β
+CD8
+T cells at a median of 30 days after allo-HSCT provided robust predictive value for risk of CMV reactivation [
19]. Interestingly, Itzykson R et al. revealed that the CMV serostatus in recipient was positively correlated with the proportions of HLA-DR
+ activated (CD8
+HLA-DR
+) and of late effector memory CD8
+ (CD8
+CD45RA
+CCR7
−) T cells [
20]. Besides, successful EBV-specific immune responses are characterized by effective cytotoxic CD8
+ T cells and NK cells [
21,
22]. Higher proportion of CD8
+ T cells had been observed in patients with EBV reactivation [
15,
23]. After EBV reactivation, a sustained low proportion of CD4
+ T cells was persistent within one year [
24]. These findings suggested that numerous kinds of CD4
+ and CD8
+ cells participated in immune responses after allo-HSCT. However, detailed studies concerning the IR pattern at early stage after allo-HSCT, and its underlying significance for transplantation-related complications are still lacking.
In the present study, we retrospectively investigated the quality of IR at day 30 and day 100 after allo-HSCT, respectively, and constructed a predictive model with proportions of CD8+ and CD4+CD45RO+ populations at day 30 to predict EBV reactivation. Our data provided a systematic characterization of early T-cell reconstitution and analyzed its relationship with EBV reactivation.
Discussion
In this study, we retrospectively analyzed the association of immune cell recovery at day 30 and IR quality of 122 patients who received myeloablative allo-HSCT. As previous reported, CD4
+ T cells reconstitution could predict OS in pediatric HSCT patients [
9,
34], and better IR was associated with less viral reactivation [
9]. In the present study, we confirmed that early recovery of T cell subsets at day 30 reflected IR quality in a cohort with more than 70% HID HSCT. And successful reconstitution of CD4
+ T cells within 100 days was associated with lower CI of EBV reactivation. Several variables contribute to the increased risk of EBV reactivation after allo-HSCT. HLA mismatch is the most important risk factor for EBV reactivation after allo-HSCT [
35‐
37]. Therefore, patients with HID HSCT would have a high risk for EBV reactivation. To note, we set the EBV-DNAemia (> 1000 IU/ml) as the diagnostic criterion for EBV reactivation according to the previous study [
38]. As EBV-DNAemia accumulated, the risk of EBV-PTLD rose rapidly [
38]. 27.40% (20/73) of patients with EBV reactivation in our cohort required rituximab intervention. Accurate prediction and close monitoring for EBV activation are warranted. Thus, we evaluated the T cell reconstitution at day 30 after allo-HSCT, and found that the lymphocyte reconstitution at day 30 effectively predicted EBV reactivation, which would be ahead of conventional IR evaluation timepoint (100 days post-transplantation).
Adaptive immunity is the core determinant for EBV prevention [
39‐
41]. Several factors have a role in affecting immune recovery after allo-HSCT, spanning donor age, donor gender, intensive chemotherapy, GvHD prophylaxis, and conditioning regimens [
42‐
45]. T cell recovery after allo-HSCT differs across individuals, and immune-monitoring might help to predict the risk of EBV reactivation shortly after allo-HSCT. ROC analysis revealed that the proportion of CD4
+ and its subpopulation could distinguish GIR from PIR, while PIR patients are associated with higher incidence of EBV reactivation and PTLD, suggested that early CD4
+ reconstitution may predict EBV infection post-transplantation. Our findings confirmed that impaired CD4
+ T cell recovery was correlated with EBV reactivation and PTLD, which is consistent with previous studies [
46,
47].
During primary EBV infection in healthy individuals, T cell numbers in peripheral blood were increased dramatically [
39]. EBV-specific immune cells differentiated into memory CD4
+ (including CD45RA
−CD45RO
+CCR7
− effector memory [EM] and CD45RA
−CD45RO
+CCR7
+ central memory [CM] ~ 0.1%) and CD8
+ (2–5%) T cells after infection [
48]. For allo-HSCT recipients, early recovery of donor-derived EBV-specific T cells within 60 days provided prophylactic effects against EBV-related diseases [
49]. The early recovery of the T cells relies on peripheral expansion of memory T cells, and CD8
+ T cells reconstitute earlier than CD4
+ T cells in early T-cell reconstitution. In the present study, CD8
+ and CD4
+CD45RO
+ T cells within 30 days was reversely associated with EBV reactivation. Thus, our data suggested that CD8
+ and CD4
+CD45RO
+ T cells after allo-HSCT provided a protection against EBV reactivation, possibly by driving early recovery of EBV-specific T cells. Furthermore, in the clinical setting, the criterion of the invention against EBV reactivation after allo-HSCT is not well established. Our data provided a promising method for risk-stratification of EBV reactivation, which might assist the judgment for early intervention.
Tregs is a subpopulation of CD4
+ T cells with the function of suppressing immune responses and maintaining self-tolerance [
50]. In our data, Tregs (CD4
+CD25
+ and CD4
+CD25
+CD127
low T cells) at day 30 after allo-HSCT were independent protective factors of EBV reactivation, which is consistent with previous studies that poor CD4
+CD25
+ T cell recovery at day 30 after allo-HSCT was associated with prolonged CMV and EBV duration [
51,
52]. Taking the high inflammatory status of the early period after allo-HSCT into consideration, Tregs could have a compensatory increase and reflect high cytotoxic activity of effector T cells [
53]. These findings suggested that a more careful evaluation of Tregs function in CMV/EBV reactivation is necessary, especially in the early period after allo-HSCT. Besides, there are many studies focus on the differential impact of CMV on outcome/immune reconstitution depending on CMV kinetics and higher CMV load is related to poor IR and clinical outcomes [
5,
54,
55]. However, CMV reactivation after allo-HSCT is significantly limited in our cohort (data not shown) in the letermovir era, which is the reason that we're focused on predicting EBV activation with IR.
Our study has several limitations due to its retrospective design, small sample size, and short follow-up duration. Ongoing follow-up observation of survival in all patients is needed to verify the long-term effects of IR in HSCT patients. And the predictive model should be further evaluated by external cohorts. Thus, larger multicenter retrospective studies or prospective research endeavors are warranted.
In summary, our data suggested that early lymphocytes recovery, especially the CD4+ T cell and its subsets, were correlated with the quality of IR. We developed a prognostic nomogram for EBV reactivation based on the proportion of CD4+CD45RO+ and CD8+ T cells at day 30 after allo-HSCT, which may help to surveil the risk of EBV reactivation in early stages and intervene promptly.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.