Epstein–Barr virus‐associated post‐transplant lymphoproliferative disorders in pediatric transplantation: A prospective multicenter study in the United States

Epstein–Barr virus (EBV)‐associated post‐transplant lymphoproliferative disorders (PTLD) is the most common malignancy in children after transplant; however, difficulties for early detection may worsen the prognosis.

The World Health Organization (WHO) has developed a classification of PTLD subtypes, [14][15][16] which can also be classified into two categories: hyperplasias and aggressive PTLD, such as polymorphic, monomorphic, and classical Hodgkin lymphoma PTLDs. 16gressive PTLD is classically neoplastic, whereas hyperplasias do not disrupt the tissue architecture and it may regress with modulation of immunosuppressive agents. 2,16[19][20][21][22] Some data on patient survival after PTLD showed improved survival compared to older eras 4,23 ; however, aggressive PTLD is still considered to have a poor prognosis and requires early detection and treatment for favorable outcomes.In addition, the oncogenic EBV drives abnormal lymphocyte proliferation in 50%-80% of PTLD, especially in early-onset PTLD. 1 EBV+ PTLD is more common in children and tends to develop earlier post-transplant than in adults. 9,13Although the consensus guidelines for EBV viral load and other biomarker monitoring for PTLD after solid organ transplantation have recently been published, 24 it remains essential to continue to accumulate evidence, especially for early detection of aggressive PTLD in pediatric recipients, who are frequently EBV-naïve at the time of transplant.

The National Institutes of Health Clinical Trials in Organ
Transplantation in Children 06 (CTOTC-06) was an NIH/NIAIDsponsored prospective, multicenter study intended to identify viral and immune biomarkers of EBV-associated PTLD.The primary mechanistic endpoints were: Identify gain of function mutations in EBV latent membrane protein 1 (LMP1) detected by DNA sequencing and pathogenic changes in B cell clonotype development as assessed using high-throughput sequencing.The results from these studies were published elsewhere. 25Other mechanistic end-points were to determine whether unique microRNA signatures are associated with PTLD and LMP1 mutations.The results of this work were also the subject of a publication. 26Secondary end-points were clinical observations detected during the follow-up of the pediatric transplant recipients enrolled in the trial.The aim of this study was to investigate clinical risk factors for developing PTLD, and further elucidate the characteristics that may contribute to the early detection of aggressive PTLD using the database of a prospective, multicenter study from seven pediatric transplant centers in the United States.Enhancing clinical knowledge of risk factors for aggressive PTLD may lead to lower morbidity and mortality in such a vulnerable pediatric population.
factor for PTLD development (sub-hazard ratio: 2.79 [1.34-5.78],p = .006)and EBV DNAemia (2.65 [1.72-4.09],p < .001).Patients with D+R− were significantly more associated with monomorphic/polymorphic PTLD than those with the other combinations (p = .02).Patients with monomorphic/polymorphic PTLD (n = 21) had significantly more EBV DNAemia than non-PTLD patients (p < .001)and an earlier clinical presentation of PTLD than patients with hyperplasias (p < .001),within 6-month post-transplant.Among non-liver transplant recipients, monomorphic/polymorphic PTLD were significantly more frequent than hyperplasias in patients ≥5 years of age at transplant (p = .01).(Clini calTr ials.gov Identifier: NCT02182986).Of these, 872 received liver, kidney, heart, intestinal, or multivisceral transplants, and participants were followed for a minimum of 12 months and up to 4 years based on the time of enrollment.In total, 34 pediatric EBV+ PTLD (3.9%) were identified by biopsy (Figure S1A). 14,15,25The entire transplanted cohort included patients enrolled post-transplant (n = 588, 67%), who were enrolled within 3 years of transplant; however, complete data in the early post-transplant period were not available for this cohort.Therefore, in the current study, we also created a separate cohort (n = 470) that included patients enrolled before transplant (n = 284) and within a month after transplant (n = 186, Figure S11B).These two cohorts were used separately depending on the outcome setting as described in the section of statistical analysis.Immunosuppression and clinical treatment were determined by the standard of care at each participating transplant center.The study was approved by the Institutional Review Board at each institute and conducted in accordance with the relevant guidelines and regulations.All participants provided written informed consent prior to inclusion in the study.For participants under the age of 18 years, informed consent was obtained from a parent and/or legal guardian.

| Variables
The clinical variables included sex, age at transplant, age at PTLD diagnosis, predominant race, ethnicity, type of transplanted organs, pre-transplant EBV serology of donor and recipient, use of induction immunosuppression, presence of extranodal lesions, response to chemotherapy and rituximab, and histopathological diagnosis.

| EBV Monitoring
6][27] In brief, EBV serology of donor and recipient was examined at the time of transplantation.EBV serologies were diagnosed by the combination of the results of viral capsid antigen

| Statistical analysis
Data were presented as the median with interquartile ranges (IQRs) for continuous variables and counts for categorical variables.
Comparisons of continuous variables and categorical variables were performed using the Mann-Whitney U-test and Chi-squared test, respectively.The entire transplanted cohort (Figure S1A) was applied when incidence of PTLD was set as the outcome.The separate cohort that included patients enrolled before transplant and within a month after transplant (Figure S1B) was used when EBV DNAemia and rejection were defined as outcomes since the relevant data/timepoints were available but not for the entire cohort.
Cumulative incidence of PTLD, EBV DNAemia, and rejection was compared, with patient death as a competing risk.Risk factors for PTLD, EBV DNAemia, and rejection were analyzed by univariable and multivariable Fine-Gray sub-distribution hazard competing risk regression models.Variables with p < .10 in the univariable analysis were included in the multivariable analysis.Patient survivals were estimated by the Kaplan-Meier method, followed by a log-rank test.
All analyses were two-sided, and p < .05 was considered statistically significant.All statistical analyses were performed using JMP Pro16 (SAS Institute Inc., Cary, NC, USA) and Stata MP 17 (Stata Corp., College Station, TX, USA).

| Patient characteristics
Patient characteristics were summarized in Table 1.The number of male and female patients was 470 (54%) and 402 (46%), respectively.

| Risk factors for PTLD
The entire transplanted cohort was used for the analysis (Figure S1A).In univariable analysis, EBV-naïve recipient

| Risk factors for EBV DNAemia
The cohort enrolled pre-transplant and within 1 month of transplant was applied for the analysis (Figure S1B).In univariable EBV DNAemia (Table 3).

| Risk factors for rejection
The cohort enrolled pre-transplant and within 1 month of transplant was employed for the analysis (Figure S1B).In univariable analysis, significant risk factors for rejection within a year after transplant were EBV-naïve recipient (SHR:   S3).In a 2:2 comparison of D+R− and the other combinations vs. monomorphic/polymorphic PTLD and non-PTLD patients, there were significantly more patients with monomorphic/polymorphic PTLD in D+R− than in the other combinations (p = .004,Table S3).

| Characteristics of PTLD patients
Characteristics of PTLD patients were presented in Table S4.
According to the WHO classification, 14,15 13, 15, and six patients were histopathologically diagnosed with early lesion (hyperplasias), monomorphic type, and polymorphic type, respectively.The number of transplanted organs was 12 livers, eight hearts, seven kidneys, and seven intestines.There were significantly more extranodal lesions in monomorphic/polymorphic PTLD than in hyperplasia groups (p < .001).

| Relationship between age and monomorphic/ polymorphic PTLD by organ
The distribution of monomorphic/polymorphic PTLD and hyperplasias by age for each organ was shown in Figure S3.Patients ≥5 years at transplant had significantly more monomorphic/ polymorphic PTLD than hyperplasias (p = .047,Figure S3A).
Additionally, considering the skewed ages at transplant for each organ (Figure S3B-E), non-liver transplant recipients were combined into a single group (Figure S3F).Among non-liver transplant recipients, monomorphic/polymorphic PTLD was significantly more frequent than hyperplasias in patients ≥5 years of age at transplant (p = .01,Figure S3F).TA B L E 3 Risk factors for EBV DNAemia within a year after transplant.
risk factor for PTLD development, 2,4,28,29 whereas the relationship between EBV-naïve recipients and rejection and the association of D+R− with monomorphic/polymorphic PTLD remain unclear.
The cumulative incidence of EBV DNAemia showed that patients with D+R+ and D+R− were significantly associated with EBV DNAemia compared to recipients of EBV-naïve organs (Figure 1C,D).
EBV reactivation in D+R+ patients and primary EBV infection in D+R− patients may be responsible for EBV DNAemia after transplant.
Additionally, recipients of EBV-naïve organs developed significantly less EBV DNAemia after transplant than D+R+ patients, suggesting  Both donor and recipient serostatus affected the incidence of EBV DNAemia, with an additive effect of D+ and R− combination.
In contrast, EBV-naïve recipients developed significantly more rejection than EBV-seropositive recipients while donor serostatus did not affect the occurrence of rejection.Infection can contribute to allograft rejection both directly and indirectly. 30Infection may induce highly inflammatory and cytotoxic cellular responses, which can directly increase the risk of rejection.In contrast, post-transplant infections are commonly treated by reduction of immunosuppressants, which can indirectly increase the risk of rejection. 30Cytomegalovirus (CMV) infection is known to be significantly associated with a higher incidence of rejection. 31like CMV, there is no clear evidence that EBV is associated with rejection, probably because over 95% of adults are EBVseropositive. 32As shown in this study, we need to focus on pedi- ).The 1-year, 2-year, and 3-year survival rates of patients after PTLD diagnosis were 94%, 90%, and 90%, respectively.(C) The prognosis of patients with monomorphic/polymorphic PTLD was likely to be worse than that of patients with hyperplasias after PTLD diagnosis (p = .08).The 1-year, 2-year, and 3-year survival rates of patients with monomorphic/polymorphic PTLD after PTLD diagnosis were 90%, 83%, and 83%, respectively while those of patients with hyperplasias were 100%, 100%, and 100%, respectively.PTLD, post-transplant lymphoproliferative disorders.
Patients with D+R− were found to be significantly associated with the development of monomorphic/polymorphic PTLD compared to patients with the other combinations.In addition, patients with monomorphic/polymorphic PTLD showed a significantly higher incidence of EBV DNAemia than non-PTLD patients within 6 months after transplant.Additionally, all PTLD that developed within 6 months of transplantation were monomorphic/ polymorphic PTLD, which were significantly more frequent than hyperplasias.These findings suggest that intensive follow-up of EBV viral load within 6-month post-transplant may help detect monomorphic/polymorphic PTLD early, especially in patients D+R− serostatus.EBV viral load was routinely checked at transplant and every 3 months after transplant in this study.It would be recommended that EBV PCR is performed every week before discharge, followed by biweekly to monthly after discharge for the first 6 months after transplant.
The age at transplant in patients with monomorphic/polymorphic PTLD was significantly higher than patients with hyperplasias.
The median age for each transplanted organ in the entire cohort significant difference in the incidence between monomorphic/polymorphic PTLD and hyperplasias by age in liver transplant recipients.
Accordingly, close follow-up may be indicated for early detection of monomorphic/polymorphic PTLD in non-liver transplant patients ≥5 years of age.
Since most organs are from EBV-seropositive donors, intensive follow-up is important to reduce EBV-related complications in EBVseronegative recipients.Identification of transplant recipients at high-risk for EBV-associated complications would be beneficial.In a recent study, we identified two mutations EBV LMP1 that carry a nearly 12-fold increased risk of development of EBV+ PTLD. 25 Conversely, transplant recipients without these mutations are at low risk of PTLD which can be informative in stratifying patients for risk of EBV+ PTLD.Currently, there is no approved vaccine for EBV; however, new promising vaccines are being developed [33][34][35] and may be useful to prevent high-risk patients from developing EBV-related complications in the future.
This study has several limitations that need to be considered when interpreting the results.First, each center may have different clinical practices, including surgical procedures, perioperative management, and preference for immunosuppressive drugs, making direct comparisons difficult.However, the multicenter study allowed us to collect data and clinical samples on 872 pediatric recipients during the 5-year study period.Second, the analysis was performed in a cohort that integrated four different types of transplanted organs.Investigating the cohorts of each organ individually was considered because of their different characteristics; however, the number of events and patients would be reduced, which could result in lower statistical power.The analysis in this study was thus conducted using the integrated cohort.Third, although we were able to prospectively identify 34 cases of PTLD, clinical studies with more PTLD patients could provide additional insights or validate our findings.Further studies with larger data sets are also required to elucidate the characteristics of monomorphic/polymorphic PTLD.
Fourth, the impact of the potential presence of passive EBV antibodies on D/R risk analysis could not be performed due to insufficient data information.
In conclusion, our study supports previous retrospective studies that D+R− serostatus is a significant risk factor for both EBV

CO N FLI C T O F I NTER E S T S TATEM ENT
Author Brian Armstrong was employed by Rho.The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

: 2 . 1 |
D+R− is a risk factor for PTLD and EBV DNAemia and associated with the incidence of monomorphic/polymorphic PTLD.Intensive follow-up of EBV viral load within 6-month post-transplant, especially for patients with D+R− and/or nonliver transplant recipients ≥5 years of age at transplant, may help detect monomorphic/polymorphic PTLD early in pediatric transplant.K E Y W O R D S Epstein-Barr virus, pediatric transplantation, PTLD 2 | PATIENTS AND ME THODS Study design and patients This is a sub-analysis of the NIAID-sponsored CTOTC-06 "Biomarkers for Post-Transplant Lymphoproliferative Disorders in Children."This study enrolled 944 children (≤21 years of age) at seven pediatric transplant centers in the US between August 2014 and May 2019

(
VCA)-IgM, VCA-IgG, EBV nuclear antigen-1 (EBNA-1) IgG, and EBV early antigen diffuse component (EA-D) IgG.Blood samples were prospectively collected (n = 4753) at multiple timepoints: enrollment or transplant, every 3 months during the first 2 years, twice yearly thereafter, at the time of primary EBV infection, at PTLD diagnosis and during the intense monitoring post-PTLD diagnosis.EBV PCR was performed at each center's local laboratory and EBV DNAemia was determined by the cutoff value established by each laboratory.

F I G U R E 1
Cumulative incidence of PTLD, EBV DNAemia, and rejection according to the D/R combinations.(A) Patients with D+R− developed significantly more PTLD than those with D+R+ (p = .007).(B) Patients with D+R− had significantly more incidence of PTLD than those with the other combinations (p = .006).(C) D+R− patients had significantly more EBV DNAemia than those with D+R+, D−R−, and D− R+ patients (p = .002,.003,<.001, respectively).(D) Patients with D+R− developed significantly more EBV DNAemia than those with D+R+ and D− (p = .002,<.001, respectively).Patients with D+R+ had significantly more incidence of EBV DNAemia than recipients of EBV-naïve organs (p < .001).(E) There was no significant difference in the incidence of rejection between the four groups of D+R−, D+R+, D−R−, and D−R+.(F) EBV-naïve recipients developed significantly more rejection than EBV-seropositive recipients (p = .03).D, donor; EBV, Epstein-Barr virus; PTLD, post-transplant lymphoproliferative disorders; R, recipient.

TA B L E 2
Risk factors for PTLD within 5 years after transplant.This study was conducted at seven pediatric transplant centers in the United States and included 872 transplant patients over a recent 5year period, 2014-2019, of which 34 patients with EBV+ PTLD (3.9%) were identified.In a prospective observational study with a control group, risk factors for PTLD, EBV DNAemia, and rejection were analyzed by a competing risk model.Patients with D+R− serostatus or EBV-naïve recipients were the consistent risk factors for PTLD, EBV DNAemia, or rejection.Furthermore, patients with D+R− were significantly associated with the development of monomorphic/polymorphic PTLD.The D+R− combination is well known as a representative

TA B L E 4
Risk factors for rejection within a year after transplant.that transplant of EBV+ organs may promote EBV DNAemia in not only D+R− patients but also D+R+ patients.The fact that D+R+ patients, the most common population of transplant recipients, are more likely to develop EBV DNAemia than recipients of EBV-naïve organs should be emphasized in terms of post-transplant follow-up.
patients to elucidate the relationship between EBV infection and rejection.The direct association between EBV DNAemia and rejection was not evident in this cohort, presumably due to insufficient sample size (data not shown).Further accumulation of evidence from both basic and clinical research is required to elucidate the mechanism.F I G U R E 2 Incidence of PTLD and patient survival after transplant and PTLD diagnosis.(A) Of 21 patients with monomorphic/ polymorphic PTLD, 13 patients (62%) developed PTLD within a year after transplant.The median time from transplant to PTLD diagnosis was 5.3 months (IQR: 3.8-23.5).(B) Patients with monomorphic/polymorphic PTLD showed significantly worse prognosis than non-PTLD patients after transplant (p < .001

was 2 .
3 years (IQR: 0.9-8.8) in liver, 3.2 years (1.4-8.8) in intestine/ multi-organs, 6.8 years (1.2-14.1) in heart, 11.4 years (5.6-16.4) in kidney.The incidence of PTLD was evaluated by ages at transplant and PTLD for each organ considering the different age distribution in each transplanted organ (Figure S3).Patients ≥5 years of age at transplant had significantly more monomorphic/polymorphic PTLD than hyperplasias in non-liver transplants.In contrast, there was no F I G U R E 3 EBV DNAemia and PTLD development within 6-month post-transplant.(A) Patients with monomorphic/polymorphic PTLD showed significantly higher incidence of EBV DNAemia within 6-month post-transplant than non-PTLD patients (p < .001).Patients with monomorphic/polymorphic PTLD tended to have more incidence of EBV DNAemia than patients with hyperplasias within 6-month posttransplant (p = .09).(B) Of 13 patients who had monomorphic/polymorphic PTLD within a year of transplant, 11 patients (85%) developed monomorphic/polymorphic PTLD within 6 months of transplant while no patient developed hyperplasias within 6-month post-transplant.(C) Patients with monomorphic/polymorphic PTLD had an earlier clinical presentation of PTLD than patients with hyperplasias within 6 months of transplant (p = .002).EBV, Epstein-Barr virus; PTLD, post-transplant lymphoproliferative disorders.
DNAemia and PTLD and significantly associated with the development of monomorphic/polymorphic PTLD.Patients with monomorphic/polymorphic PTLD had significantly more EBV DNAemia than non-PTLD patients and earlier clinical presentation of PTLD than patients with hyperplasias, within 6-month post-transplant.Intensive follow-up of EBV viral load early post-transplant, especially for patients with D+R− serostatus and/or non-liver transplant recipients ≥5 years of age at transplant, may facilitate early detection of monomorphic/polymorphic PTLD in pediatric transplant.ACK N OWLED G M ENTS This study was funded by NIH UO1 AI104342 (C.O.E.), UM2AI117870 (Rho Federal Systems), the Stanford Transplant and Tissue Engineering Center of Excellence fellowship (T.T.).