More than 90% of adults worldwide have been infected

Transcription

1 ORIGINAL ARTICLE Identifying Predictive Factors for Posttransplant Lymphoproliferative Disease in Pediatric Solid Organ Transplant Recipients With Epstein-Barr Virus Viremia Lauren Weintraub, MD,* Chana Weiner, MD,w Tamir Miloh, MD,z Juli Tomaino, MD,y Umesh Joashi, MD,8 Corinne Benchimol, MD,z James Strauchen, MD,# Michael Roth, MD,* and Birte Wistinghausen, MD** Summary: Epstein-Barr virus (EBV) viremia (EV) in pediatric solid organ transplant (SOT) recipients is a significant risk factor for posttransplant lymphoproliferative disease (PTLD) but not all patients with EV develop PTLD. We identify predictive factors for PTLD in patients with EV. We conducted a retrospective chart review of all pediatric SOT recipients (0 to 21 y) at a single institution between 2001 and A total of 350 pediatric patients received a SOT and 90 (25.7%) developed EV. Of EV patients, 28 (31%) developed PTLD. The median age at transplant was 11.5 months in the PTLD group and 21.5 months in the EV-only group (P = 0.003). Twenty-three (37%) EV-only patients had immunosuppression increased before EV, compared with 28 (100%) of PTLD patients (P < 0.001). The median peak EBV level was 3212 EBV copies/10 5 lymphocytes for EV-only and EBV copies/ 10 5 lymphocytes for PTLD (P = 0.005). All patients who developed PTLD had Z1 clinical symptoms. Younger age at transplant, increased immunosuppression before EV, higher peak EBV level, and presence of clinical symptoms have predictive value in the development of PTLD in SOT patients with EV. Key Words: posttransplant lymphoproliferative disease, Epstein- Barr virus, immunosuppression (J Pediatr Hematol Oncol 2014;36:e481 e486) More than 90% of adults worldwide have been infected with Epstein-Barr virus (EBV), a ubiquitous human herpesvirus. 1 Initial infection occurs during childhood, when EBV invades B lymphocytes and persists as a latent infection throughout life. 2 In the immunocompetent host, antibodies effectively neutralize virus infectivity and CD4- positive and CD8-positive T cells control the proliferation of EBV-infected cells. 3 However, in patients receiving T-cell targeted immunosuppressive therapy, EBV-infected B cells may proliferate uncontrollably and result in Received for publication July 10, 2013; accepted April 25, From the *Division of Pediatric Hematology/Oncology, Children s Hospital at Montefiore, Bronx; wdivision of Pediatric Hematology/ Oncology, Morgan Stanley Children s Hospital, New York; Divisions of zpediatric Nephrology; #Pathology; **Pediatric Hematology/Oncology, Mount Sinai Medical Center, New York, NY; zdivision of Pediatric Gastroenterology/Hepatology, Phoenix Children s Hospital, Phoenix, AZ; ydivision of Pediatric Gastroenterology, Ann and Robert H. Lurie Children s Hospital of Chicago, Chicago, IL; and 8Division of Pediatric Critical Care, North Carolina Children s Hospital, Chapel Hill, NC. The authors declare no conflict of interest. Reprints: Lauren Weintraub, MD, Division of Pediatric Hematology/ Oncology, Children s Hospital at Montefiore, 3415 Bainbridge Ave, Rosenthal 303, Bronx, NY ( lauren.weintraub@ gmail.com). Copyright r 2014 by Lippincott Williams & Wilkins symptomatic illness ranging from a mononucleosis-like syndrome to malignancy. 4 Posttransplant lymphoproliferative disease (PTLD) is a clinically and histologically heterogenous group of lymphocyte proliferations occurring in transplant recipients while being on immunosuppression. 4 Eighty percent of PTLDs overall, and the vast majority of pediatric PTLDs, are EBV-positive B-cell proliferations, making EBV infection a significant risk factor for the development of PTLD. 5 This group of disorders is a major cause of morbidity and mortality in the pediatric transplant population. In pediatric patients, PTLD has been reported to affect 5% to 15% of cardiac and liver transplants and an even higher percentage of patients who have undergone intestinal transplant. 5 9 The incidence of PTLD was thought to be relatively low in renal transplantation, ranging from 1% to 3%; however, more recent studies report a rate as high as 6.9% The clinical presentation of PTLD varies and includes fever, an infectious mononucleosis-like illness, localized lymphadenopathy, or disseminated disease. It may also involve extranodal sites such as the central nervous system, gastrointestinal tract, or other abdominal organs, bone marrow or allograft. The work-up typically includes imaging to evaluate for lymphadenopathy and extranodal disease, a tissue biopsy, bone marrow biopsy, and lumbar puncture. Histologically, PTLD is classified as early lesions, polymorphic PTLD, monomorphic PTLD, and other rare forms such as Hodgkin disease like lesions. 4,10,13,14 Most transplant centers monitor for the presence of EBV viremia (EV) by PCR in solid organ transplant (SOT) patients at least for the first year posttransplant and reduce immunosuppression to a varying degree if viremia occurs. However, there are no clinical guidelines about the work-up necessary in patients who are at risk for PTLD. In this retrospective chart review, our goal was to identify risk factors for the development of PTLD in patients with EV. METHODS We conducted a retrospective analysis of all pediatric SOT recipients, age 0 to 21 years, at our institution between 2001 and Patients were followed for a minimum of 1 year and maximum of 9 years. Data were obtained from patient charts and electronic patient databases. Inclusion criteria consisted of EV after SOT. EV was defined as >1000 EBV copies/10 5 lymphocytes in peripheral blood. As no patients with EBV levels <1000 copies/10 5 lymphocytes developed PTLD, we excluded this population J Pediatr Hematol Oncol Volume 36, Number 8, November e481

2 Weintraub et al J Pediatr Hematol Oncol Volume 36, Number 8, November 2014 with low-level EV, as per our laboratory value range. Exclusion criteria also included EBV-negative PTLD. Data collected included onset of EV, presence of PTLD and pathologic grading, patients sex, age at transplant and organ transplanted, indication for transplant, induction and maintenance immunosuppression, EBV serostatus at time of transplant, steroid intensification or increased tacrolimus level for rejection before EV, time from transplant to onset of EV, tacrolimus levels at time of EV onset and 8 averaged levels before and after onset of EV, development of CMV viremia by PCR, pattern of EBV PCR spikes, peak EBV value, treatment of EV (including antivirals and reduction of immunosuppression), ability to stay on immunosuppression, number and treatment of rejections (diagnosed histologically after a graft biopsy) after EV. EBV serostatus at time of transplant was determined by IgG positivity. All children under the age of 12 months were determined to be negative. EBV PCR was checked monthly in the year posttransplant. The pattern of EBVpositive PCRs was also noted. EV was recorded as being primarily <1000 EBV copies/10 5 if the EBV would peak >1000 and then return to <1000 for the majority of the time, intermittently >1000 EBV copies/10 5 if the EBV was sustained >1000 for <3 months at a time, or chronically active if the levels were >1000 EBV copies/10 5 lymphocytes for >3 months at a time. PTLD was diagnosed histologically after biopsy and was staged using the grading provided by Harris et al. 14 All patients received induction and maintenance immunosuppression based on organ-specific guidelines. The immunosuppression protocol for liver transplant patients included induction with mycophenolate mofetil (MMF), methylprednisolone, and daclizumab (humanized monoclonal anti-cd25 antibody), followed by a prednisone taper over 6 months and maintenance tacrolimus. The protocol for kidney transplants was induction with methylprednisolone and antithymocyte globulin (ATG), with the exception of a few transplant recipients who were induced with basiliximab (chimeric murine/human monoclonal anti IL-2R antibody) instead of ATG. Tacrolimus and MMF were started immediately postoperatively and continued as maintenance therapy and prednisone was tapered to 0.5 mg/kg by 3 months and continued as part of the maintenance regimen. Heart transplant patients were induced with MMF, methylprednisolone, and basiliximab. Prednisone was tapered by 6 months and patients were maintained on double immunotherapy with tacrolimus and MMF, with the exception of some patients for whom sirolimus was substituted for tacrolimus. Small bowel transplant patients received induction with methylprednisolone, MMF, and basiliximab, followed by a prednisone taper over 6 months and either monotherapy with maintenance tacrolimus or double immunotherapy with tacrolimus and MMF. For analysis, patients were divided into EV without evidence of PTLD (EV-only) or EV with PTLD (PTLD). Categorical data were compared using a Pearson w 2 analysis. The square root of the median peak EBV values was used to transform that data to meet the assumptions of the statistical model. P-values for continuous, nonparametric data were calculated using Mann-Whitney statistical analysis. A multivariate analysis using logistic regression was performed. Variables were selected for the multivariate analysis if they achieved statistical significance in the univariate analysis and were deemed clinically relevant. The number of variables was limited by the sample size of the population. e482 A P-value of <0.05 was considered statistically significant. All data analysis was performed using SPSS. RESULTS Ninety of 348 transplant patients met inclusion criteria and developed EV. Of those transplants, 59/175 (33.7%) were liver, 16/119 (13.4%) kidney, 8/32 (25%) heart, 3/12 (25%) small bowel, 3/9 (33.3%) liver/kidney, and 1/1 (100%) small bowel/kidney. Indication for transplant in each patient was recorded (Table 1). Twenty-eight of 90 (31%) EV patients developed PTLD: 18/59 (31%) liver, 3/16 (19%) kidney, 5/8 (63%) heart, and 2/3 (66%) small bowel (Table 2). All cases were EBV-positive, B-cell PTLD. Of the 258 patients that did not develop EV, none developed PTLD. The clinical variables analyzed in a univariate analysis are recorded in Table 3. There was no statistical difference in incidence of PTLD based on sex. The median age at transplant was 11.5 months (range, 1 to 245 mo) in the PTLD group and 21.5 months (range, 2.5 to 206 mo) in the EV-only group (P = 0.003). The median time from transplant to EV was 6 months (range, 2.4 to 15.3 mo) in the EV-only group and 9.5 months (range, 3 to 26.3 mo) in the PTLD group (P = 0.206). The median age at onset of EV was 73 months (range, 20.3 to mo) in the EV-only group and 30.3 months (range, 15 to 77.5 mo) in the PTLD group TABLE 1. Indication for Transplant PTLD (N = 28) EV-Only (N = 62) Liver (n = 59) N = 18 N = 41 Extrahepatic biliary atresia Primary sclerosing cholangitis 1 2 Hemangioendothelioma 1 1 Congenital biliary malformation 1 1 Metabolic 2 5 Idiopathic liver failure 0 6 Wilsons 0 2 Hepatoblastoma 0 1 Other 0 2 Kidney (n = 16) N = 3 N = 13 Autosomal recessive polycystic 1 1 kidney disease Focal segmental 0 4 glomerulosclerosis IgA nephropathy 0 2 Minimal change 0 1 Genitourinary malformation 1 1 Idiopathic renal failure 0 2 Medullary cystic kidney 1 1 Other 0 1 Heart (n = 8) N = 5 N = 3 Dilated cardiomyopathy 4 2 Ebstein anomaly 0 1 Hypoplastic heart 1 0 Small bowel N = 2 N = 1 Short gut syndrome 2 1 Liver/kidney (n = 3) N = 0 N = 3 Metabolic 0 3 Small bowel/kidney N = 0 N = 1 Smooth muscle dysfunction/ Tacrolimus toxicity 0 1 EV indicates EBV viremia; PTLD, posttransplant lymphoproliferative disease.

3 J Pediatr Hematol Oncol Volume 36, Number 8, November 2014 Predictive Factors for PTLD TABLE 2. Incidence of PTLD by Organ PTLD/EV (N = 90) PTLD by Organ/All PTLD (N = 28) Organ transplanted Liver 18/59 (31) 18/28 (64.3) Kidney 3/16 (19) 3 (10.7) Heart 5/8 (18) 5 (17.9) Small bowel 2/3 (66) 2 (7.1) Liver/kidney 0/ Small bowel/ kidney 0/ EV indicates EBV viremia; PTLD, posttransplant lymphoproliferative disease. (P = 0.026). The median peak EBV level was 3212 EBV copies/10 5 lymphocytes for EV-only and EBV copies/10 5 lymphocytes for PTLD (P = 0.005). There was no statistical significance found for pretransplant EBV-positive recipient serostatus, treatment with antivirals, or the incidence of CMV viremia. Whether the EBV PCR was primarily <1000 EBV copies/10 5 lymphocytes versus intermittently or chronically >1000 EBV copies/10 5 lymphocytes was statistically significant, with a higher percentage of PTLD patients having intermittently (46.4% compared with 35.5% in EV group) or chronically (25% compared with 11.3%) elevated PCRs, and a lower percentage of PTLD patients having only transient elevations (28.6% compared with 53.2%) (P = 0.039). With respect to immunosuppression, the response to EV was physician and transplant dependent and was heterogenous. The responses included complete cessation of immunosuppression, immunosuppression reduction with objective evidence (tacrolimus levels lower, documentation that steroids were stopped), reduced immunosuppression documented without objective evidence, and no change in immunosuppression. Thirteen (21%) patients in the EV-only group were induced with ATG, compared with 1 (3.6%) patient in the PTLD group (P = 0.035). Five (8.1%) of the EV-only patients and 9 (32.1%) of the PTLD patients were induced with basiliximab (P = 0.004). Forty one (66.1%) EV-only patients were induced with daclizumab, compared with 18 (64.3%) PTLD patients (P = 0.865). With respect to maintenance immunosuppression, there was no statistical difference found for the type of immunosuppression or for patients treated with monotherapy, double, or triple immunotherapy. There was no statistical significance found in tacrolimus levels, compared with pre-ev, at onset of EV, and after onset of EV. Of significance, 23 (37%) EV-only patients had immunosuppression increased (steroid intensification or increased tacrolimus level) before EV, whereas 28 (100%) of the PTLD patients had increased immunosuppression (P < 0.001). Sixteen (25.8%) EV-only patients had steroid intensification before EV compared with 24 (85.7%) PTLD patients (P < 0.001). After diagnosis of EV, 54 (87.1%) EV-only patients had a reduction or cessation of their immunosuppression, compared with 28 (100%) PTLD patients (P = 0.083). All patients who developed PTLD had Z1 clinical symptoms including fever, lymphadenopathy, or other gastrointestinal symptoms, most frequently diarrhea. P With respect to rejection post-ev, there was no statistical difference between groups in the number of rejection episodes (mean number of episodes 0.58 in the EV-only group vs in the PLTD group, P = 0.19) or the severity of rejection. There was also no significant difference for the number of patients who had treatment of rejection with steroids versus thymoglobulin in addition to steroids. There were 5 (8.1%) EV-only patients who required graft explant for rejection, compared with no PTLD patients (P = 0.122). In the multivariate analysis (Table 4), steroid intensification before EV, age at EV, and median peak EBV level remained statistically significant (P-values 0.001, 0.049, and 0.042, respectively). Patients who had EV and developed PTLD were 143 times more likely to have received steroid intensification compared with patients who had EV and did not develop PTLD. Of the 28 patients that developed PTLD, 11 (39.3%) had WHO grade I, 12 had WHO grade II (42.9%), 3 had WHO grade III (10.7%), and 0 had WHO grade IV disease. The median peak EBV values were 18,708 copies/10 5 lymphocytes (1437 to 174,750) for grade I disease, 6665 copies/ 10 5 lymphocytes (1030 to 459,778) for grade II disease, and 4434 copies/10 5 lymphocytes (1480 to 49,400) for grade III disease (P-values for median peak EBV value for grade I vs. II/III and grade I/II vs. III were and 0.243, respectively). Pathologic grading was unavailable for 2 patients (Table 5). DISCUSSION This study is the largest, single institution, retrospective review of pediatric SOT recipients who developed EV posttransplant. In extensive analysis, younger age at transplant and at first onset of EV, increased immunosuppression before EV, and higher peak EBV level were associated with the development of PTLD in SOT patients with EV. In a multivariate analysis, steroid intensification before EV, age at first-onset EV, and the median peak EBV level remained statistically significant. In addition, we found that patients with PTLD were 143 times more likely to have received steroid intensification before first-onset EV, compared with younger age at EV diagnosis and a high-peak EBV level. All patients with PTLD presented with clinical symptoms that included fever, diarrhea, and/or lymphadenopathy. This variable was not included in the statistical analysis because the data were not available on all EV-only patients. Our study demonstrated that younger age at transplant was associated with the development of PTLD, which is consistent with published literature. 15 Previous studies have also shown that the risk of developing PTLD is higher for an EBV-negative recipient compared with an EBVpositive recipient before transplantation. 16 As younger children are more likely to be EBV negative at the time of transplant, it has been suggested that this is the reason for the increased risk of developing PTLD. Our study, however, did not find a statistical difference in pretransplant EBV status. Younger age at transplant may, therefore, be an independent predictive factor. Transplant from an EBV-seropositive donor into a seronegative recipient has also been associated with the development of PTLD. 17 Unfortunately, these data were not available for all patients in our study. Previous studies have demonstrated that the intensity, duration, and type of immunosuppression related to the type of transplantation are major risk factors for PTLD. 4 e483

4 Weintraub et al J Pediatr Hematol Oncol Volume 36, Number 8, November 2014 TABLE 3. Univariate Analysis PTLD (N = 28) EV-Only (N = 62) P Sex Female Male 9 27 Age at transplant (median months) 11.5 ( ) 21.5 (1-245) Induction medication aside from mycofenolate mofetil, solumedrol (n [%]) Thymoglobulin 1 (3.6) 13 (21) Basiliximab 9 (32.1) 5 (8.1) Daclizumab 18 (64.3) 41 (66.1) Maintenance medication aside from tacrolimus, prednisone (n [%]) Myofenolate Mofetil 10 (35.7) 18 (29.0) Cyclosporine 0 4 (6.5) Sirolimus 2 (7.1) 4 (6.5) Prednisone (> 6 mo) 3 (10.7) 12 (21) Other 0 2 (3.2) Maintenance immunosuppression (n [%]) Monotherapy 15 (53.6) 35 (56.5) Double immunotherapy 9 (32.1) 13 (21) Triple immunotherapy 4 (14.3) 14 (22.6) Steroid Intensification (n [%]) 24 (85.7) 16 (25.8) < Steroid intensification or increased tacrolimus level (n [%]) 28 (100) 23 (37.1) < Tacrolimus level (median) Pre-EV 9.48 ( ) 9.77 ( ) At EV diagnosis 7.90 ( ) 7.60 ( ) Post-EV diagnosis 6.55 ( ) 7.60 ( ) Pretreatment EBV IgG serostatus (n [%]) Positive 6 (21.4) 19 (30.6) Negative 20 (71.4) 38 (61.3) Inconclusive 2 (7.1) 5 (8.1) CMV viremia (n [%]) 6 (21.4) 18 (29) Time from transplant to EV (mo) 9.5 ( ) 6.0 ( ) Age at EV diagnosis (median months) 30.3 ( ) 73 ( ) Median peak EBV value (copies/10 5 lymphocytes) ( ) ( ) Median peak EBV value (square root analysis) ( ) ( ) EBV level (n [%]) 0.039* Primarily < (28.6) 33 (53.2) Intermittently > (46.4) 22 (35.5) Chronically > (25) 7 (11.3) Treatment with antivirals (n [%]) 18 (64.3) 37 (59.7) Change to immunosuppression (n [%]) None 0 6 (9.7) Reduction 26 (92.9) 54 (87.1) Cessation 2 (7.1) 0 Unknown 0 2 (3.2) Post-EBV rejection treated with increased 12 (42.9) 22 (35.5) immunosuppression (n [%]) Treatment of rejection (n [%]) Steroids 12 (42.9) 19 (30.6) Thymoglobulin and steroids 2 (7.1) 3 (4.8) Rejection requiring explant (n [%]) 0 5 (8.1) P-values calculated using Pearson w 2 for categorical data. P-value calculated using Mann-Whitney U for continuous, nonparametric data and median values with interquartile range are reported. *P-value represents w 2 analysis between transient versus intermittent and chronic. Significant P values in bold. EBV indicates Epstein-Barr virus; EV, EBV viremia; PTLD, posttransplant lymphoproliferative disease. Our study showed significant difference between types of induction immunosuppression used; however, these results also corresponded to the type of organ transplanted. For example, of all the patients in this study treated with ATG, only 1 of 14 developed PTLD. This is in contrast to previous studies, which have shown that ATG, which depletes all T cells, is a risk factor for PTLD. 18 However, of the 14 patients treated with ATG, 13 were kidney transplants, a population with a lower incidence of PTLD. In addition, there was a significantly higher percentage of PTLD patients who received basiliximab as induction immunosuppression compared with the EV-only group. Again, only a few renal transplant and the small bowel transplant patients received basiliximab, and it is difficult to draw conclusions from such a small percentage of patients. In our study, such differences seen between different types of immunosuppression could be e484

5 J Pediatr Hematol Oncol Volume 36, Number 8, November 2014 Predictive Factors for PTLD TABLE 4. Multivariate Analysis b SE P Odds Ratio Age at EBV seroconversion (mo) Steroid intensification Peak EBV value (square root) Significant P values in bold. EBV indicates Epstein-Barr virus. due to the type of organ transplanted. A larger patient population is needed to determine if there is a significant difference between types of immunosuppression. Further research is also needed to explain the difference in incidence of PTLD between organs. Factors may include higher immunosuppression used in transplanted organs such as heart and small bowel versus liver, or dose of donor lymphocytes present in the transplanted organ. Dose of donor lymphocytes may increase the risk of PTLD due to the fact that EBV exists inside cells and the increased lymphocytes results in a higher viral load. Lymphocyte dose is expected to be higher in liver and small bowel versus heart and kidney transplants. Therefore, the risk of PTLD in liver and small bowel transplant recipients may be increased due to the dose of donor lymphocytes, independent of the increased immunosuppression required. Regardless of organ transplanted, we found that increased immunosuppression (both steroid intensification and increased level of calcineurin inhibitors) in the face of rejection was significantly associated with the development of PTLD. Prior studies have also found a correlation between the intensity of immunosuppression and PTLD. 4,19 The specific effect of steroids on the development of PTLD is unknown. As with other types of immunosuppression, steroids target T cells. It is interesting, however, that steroids may increase the risk of PTLD in this way, while at the same time are often therapeutic andusedinthemanagementofptld.steroidsareoftenused as first-line therapy for rejection, which may explain the increased association with PTLD. Furthermore, steroid intensification is often accompanied by changes in other immunosuppression. In our study, tacrolimus levels were analyzed; however, did not reach statistical significance and were, therefore, not included in the multivariate analysis. Given that tacrolimus levels were not predictive of PTLD development, tacrolimus together with steroids may be a risk factor for PTLD and should be further studied. Previous studies have shown that an elevated EBV PCR level identifies a patient at risk of PTLD. 4,20,21 Our study demonstrated that a higher peak EBV level had statistically significant predictive value for PTLD. This was also demonstrated in a series of lung transplant patients, in which the majority of patients studied had high viral loads before and at PTLD diagnosis. 22 Prior studies have also shown that persistent high EBV loads are a major risk factor for the development of late-onset PTLD This supports the need for early intervention in patients with high EBV viral loads. Lee et al 26 found that by using frequent EBV viral load monitoring and preemptive immunosuppression modulation they were able to decrease the incidence of PTLD in their pediatric liver transplant population. In a pediatric renal transplant cohort studied by Tanaka and colleagues, high EBV viral load, as defined by having >1000 copies/1 g DNA for over 6 months, was managed by first discontinuing MMF followed by reducing calcineurin inhibition, if no response was seen. No patients in their study developed PTLD. 27 Because of variations in EBV PCR assays at different institutions, it is necessary for a medical center to establish their own cutoff EBV level. At our own institution, we propose that reduction of immunosuppression is warranted if the EBV PCR is >1000 copies/10 5 lymphocytes; however, imaging and/or biopsy for PTLD is not warranted unless there are clinical symptoms such as fever, diarrhea, and/or lymphadenopathy. In addition, patients who have received steroid intensification require close clinical monitoring and parents should be instructed to call immediately for any clinical symptoms. There was no statistically significant difference between pathologic grading of PTLD and peak EBV levels. The ability to detect such a difference was likely limited by the low number of patients with WHO grade III or IV disease. Studies have provided evidence that early diagnosis of PTLD, followed by reduction and/or discontinuation of immunosuppressive agents decreases morbidity and mortality. 28 A high index of suspicion is important for early diagnosis as it has been shown to correlate with more benign lesions and more favorable outcomes. 29 In our experience, not all patients with EV warrant further workup and reduction of immunosuppression is sufficient in an asymptomatic patient population. The data collected in this retrospective review are being used to develop a protocol for management of EV that minimizes the risk of rejection in low-risk patients and minimizes exposure to radiologic studies and invasive procedures, while also still promptly diagnosing PTLD in high-risk patients. By using the TABLE 5. WHO PTLD Classification No. PTLD Patients (n [%]) Median Peak EBV Value (Copies/10 5 Lymphocytes) P WHO grade I 12 (39.3) 18,708 ( ,750) 0.776* WHO grade II 11 (42.9) 6665 ( ,778) 0.243w WHO grade III 3 (10.7) 4434 ( ,400) P-values calculated using w 2 analysis. Data not available on 2 patients. *WHO grade I versus II and III. wwho grade I and II versus III. EBV indicates Epstein-Barr virus; PTLD, posttransplant lymphoproliferative disease; WHO, World Health Organization. e485

6 Weintraub et al J Pediatr Hematol Oncol Volume 36, Number 8, November 2014 predictive factors found to be statistically significant in this study, we can spare patients who do not meet criteria unnecessary radiation, anesthesia, and surgical procedures. Emerging evidence also suggests that genetic factors play a role in the development of PTLD. Polymorphisms in IFN-g have been associated with the development of EBV-positive PTLD in animal models. 30 In addition, in a case-control study done by Babel et al, 31 polymorphisms in the anti-inflammatory cytokines, IL-10, and TFG-bwere associated with the development of PTLD. Such genetic predispositions should be further evaluated in future prospective studies. The limitations of our study included sample size that affects the extent to which an extensive multivariate analysis could be performed. As this was a retrospective analysis, there is also risk of bias if inaccurate data were obtained. In addition, lung transplants are not performed at our institution and may affect the comparison to SOT recipients at other institutions. In summary, we have shown that, in our institution, the children at highest risk for PTLD are those who are transplanted at a younger age, are treated with increased immunosuppression for episodes of rejection before EV, have an elevated EBV viral load, and develop clinical symptoms. Most significantly, we found that steroid intensification before EV has a very high odds ratio for the development of PTLD in the setting of EV. Patients who develop EV and have had rejection requiring intensification of immunosuppression before developing EV are more likely to develop PTLD. Therefore, with regard to steroid intensification in patients with EV, the physician must weigh the risk of allograft loss versus the risk of PTLD, and patients who are treated with steroid intensification should be monitored especially closely. REFERENCES 1. Henle W, Henle G. Seroepidemiology of the virus. In: Epstein MA, Achong BG, eds. The Epstein-Barr Virus. Berlin, Germany: Springer-Verlag; 1979: Crawford DH. Biology and disease associations of Epstein- Barr virus. Philos Trans R Soc Lond B Biol Sci. 2001;356: Rickinson AB, Moss DJ. Human cytotoxic T lymphocyte responses to Epstein-Barr virus infection. Annu Rev Immunol. 1997;15: Faye A, Vilmer E. Post-transplant lymphoproliferative disorder in children: incidence, prognosis, and treatment options. Pediatr Drugs. 2005;7: Smets F, Sokal EM. Epstein-Barr virus-related lymphoproliferation in children after liver transplant: role of immunity, diagnosis, and management. Pediatr Transplant. 2002;6: Finn L, Reyes J, Bueno J, et al. Epstein-Barr virus infection in children after transplantation of the small intestine. Am J Surg Pathol. 1998;22: Webber SA, Naftel DC, Fricker FJ, et al. Pediatric Heart Transplant Study. Lymphoproliferative disorders after pediatric heart transplantation: a multi-institutional study. Lancet. 2006;367: Reyes J, Bueno J, Kocoshis S, et al. Current status of intestinal transplantation in children. J Pediatr Surg. 1998;33: Quintini C, Kato T, Gaynor JJ, et al. Analysis of risk factors for the development of posttransplant lymphoproliferative disorder among 119 children who received primary intestinal transplants at a single center. Transplant Proc. 2006;38: Dror Y, Greenber M, Taylor G, et al. Lymphoproliferative disorders after organ transplantation in children. Transplantation. 1999;67: Lucas KG, Pollok KE, Emanuel DJ. Post-transplant lymphoproliferative disorders. Leuk Lymphoma. 1997;25: McDonald RA, Smith JM, Ho M, et al. CCTPT Study Group. Incidence of PTLD in pediatric renal transplant recipients receiving basiliximab, calcinuerin inhibitor, sirolimus and steroids. Am J Transplant. 2009;8: Leyland V, Sutton L, Doreen R, et al. Lymphoproliferative disorders after organ transplantation: a report of 24 cases observed in a single center. J Oncol. 1995;13: Harris NL, Ferry JA, Swerdlow SH. Posttransplant lymphoproliferative disorders: summary of Society for Hematopathology Workshop. Semin Diagn Pathol. 1997;14: Cox K, Lawrence-Miyasaki L, Garcia-Kennedy R, et al. An increased incidence of Epstein-Barr virus infection and lymphoproliferative disorder in young children on FK506 after liver transplantation. Transplantation. 1995;59: Walker RC, Marshall WF, Strickler JG, et al. Pretransplantation assessment of the risk of lymphoproliferative disorder. Clin Infect Dis. 1995;20: Haque T, Thomas JA, Falk KI, et al. Transfer of donor EBV in transplanted organs causes lymphoproliferative disease in EBV-seronegative recipients. J Gen Virol. 1996;77: Landgren O, Gilbert ES, Rizzo JD, et al. Risk factors for lymphoproliferative disorders after allogeneic cell transplantation. Blood. 2009;113: Juvonen E, Aalto SM, Tarkkanen J, et al. High incidence of PTLD after non-t-cell-depleted allogeneic haematopoietic stem cell transplantation as a consequence of intensive immunosuppressive treatment. Bone Marrow Transplant. 2003;32: Riddler S, Breinig M, McKnight J. Increased levels of circulating Epstein-Barr virus-infected lymphocytes and decreased EBV nuclear antigen antibody responses are associate with the development of post-transplant lymphoproliferative disease in solid-organ transplant recipients. Blood. 1994;84: van Esser JW, van der Holt B, Meijer E, et al. Epstein- Barr virus (EBV) reactivation is a frequent event after allogeneic stem cell transplantation (SCT) and quantitatively predicts EBV-lymphoproliferative disease following T-cell depleted SCT. Blood. 2001;98: Stevens SJ, Verschuuren EA, Pronk I, et al. Frequent monitoring of Epstein-Barr virus DNA load in unfractionated whole blood is essential for early detection of posttransplant lymphoproliferative disease in high-risk patients. Blood. 2001;97: D Antiga L, Del Rizzo M, Mengoli C, et al. Sustained Epstein- Barr virus detection in paediatric liver transplantation. Insights into the occurrence of late PTLD. Liver Transpl. 2007;13: Bingler MA, Feingold B, Miller SA, et al. Chronic high Epstein-Barr viral load state and risk for late-onset posttransplant lymphoproliferative disease/lymphoma in children. Am J Transplant. 2008;8: Green M, Soltys K, Rowe DT, et al. Chronic high Epstein-Barr viral load carriage in pediatric liver transplant recipients. Pediatr Transplant. 2009;13: Lee TC, Savoldo B, Rooney CM, et al. Quantitative EBV viral loads and immunosuppression alterations can decrease PTLD incidence in pediatric liver transplant recipients. Am J Transplant. 2005;5: Tanaka E, Sato T, Ishihara M, et al. Asymptomatic high Epstein-Barr viral load carriage in pediatric renal transplant recipients. Pediatr Transplant. 2011;15: YounesBS,McDiarmidSV,MartinMG,etal.Theeffectofimmunosuppression on post transplant lymphoproliferative disease in pediatric liver transplant patients. Transplantation. 2000;70: Fernández MC, Bes D, De Da vila M, et al. Post-transplant lymphoproliferative disorder after pediatric liver transplantation: characteristics and outcome. Pediatr Transplant. 2009;13: Dierksheide JE, Baiocchi RA, Ferketich AK, et al. IFNgamma gene polymorphisms associate with development of EBV + lymphoproliferative disease in hu PBL-SCID mice. Blood. 2005;105: Babel N, Vergopoulos A, Trappe RU, et al. Evidence for genetic susceptibility towards development of posttransplant lymphoproliferative disorder in solid organ recipients. Transplantation. 2007;84: e486