NEFROLOGÍA. Vol. XXV. Número 2. 2005 Variability of Epstein-Barr virus serological markers in adult kidney transplantation recipients R. Lauzurica*, C. Frías**, B. Bayés*, V. Ausina** and R. Romero* *Nephrology and **Microbiology Units. Germans Trias i Pujol University Hospital. Badalona. Spain. SUMMARY Epstein-Barr virus (EBV) infection is associated with the development of posttransplant lymphoproliferative disorders (PTLD). However, the clinical relevance and criteria for EBV serological reactivation in EBV ­seropositive transplant recipients is unclear. EBV­ specific antibodies: viral capsid immunoglobulm G [IgG (VCA)], nuclear antigen (EBNA) IgG, immunoglobulin M [IgM (VCA)] and early antigen IgG (EA) were prospectively analyzed in 71 adult kidney transplant recipients, before starting immunosuppression, when they were uraemic, and after transplantation. A total of 351 serum samples were tested. Relevance of different EBV reactivation-related variables were analyzed using the chisquare test. In 37 of 71 (52.1%) patients IgM (VCA) or IgG (EA) were detected when they were uraemic. EBV reactivation occurred in 25 of 71 (35.2%) patients, with clinical symptoms (fever, leukopenia, kidney function impairment, and increase in transaminases) in nine cases. One of 71 patients developed a PTLD, without detection of serologically EBV reactivation, but with an increase in EBV viral load. Absence of mycophenolate mofetil, that inhibits lymphocyte proliferation and antibody production, in immunosuppression was statistically significantly associated with EBV reactivation (p = 0.015). Serological diagnosis of EBV reactivation should be based on strict criteria (IgM (VCA) seroconversion, four-fold increase in IgM (VCA) or IgG (EA), or four-fold decrease in IgG (EBNA) titers and on analysis of serial samples. Some EBV-seropositive patients at high risk of developing PTLD could benefit from this diagnostic methodology. Key words: Epstein-Barr virus. Kidney transplant. Serology. Reactivation. Correspondence: Dr. Ricardo Lauzurica Servicio de Nefrología. Hospital Universitari Germans Trias i Pujol Ctra. del Canyet, s/n 08916 Badalona E-mail: rlauzu@ns.hugtip.scs.es Financiado con becas del Fondo de Investigaciones Sanitarias, FIS 98/0331 y Ministerio de Eduación y Ciencia 97-0319. 185 R. LAUZURICA y cols. VARIABILIDAD EN LOS MARCADORES SEROLÓGICOS DEL VIRUS DE EPSTEIN-BARR EN RECEPTORES ADULTOS DE TRASPLANTE RENAL RESUMEN La infección por el virus de Epstein-Barr (VEB) se asocia con el desarrollo de síndromes linfoproliferativos post-trasplante (SLPPT). Sin embargo, la relevancia clínica y el criterio para la reactivación serológica del VEB en pacientes seropositivos para el VEB no está clara. Se analizaron prospectivamente los anticuerpos específicos del VEB: inmunoglubulina IgG frente al antígeno de la cápside viral [IgG (VCA)], IgG frente al antígeno nuclear [IgG (EBNA)], [IgM (VCA)] e IgG frente al antígeno precoz IgG (EA) en un grupo de 71 pacientes adultos trasplantados renales, antes de iniciar el tratamiento inmunosupresor, en uremia y después de efectuado el trasplante. Se analizaron un total de 351 muestras de suero. Asimismo se estudió la relevancia de distintas variables asociadas a la reactivación del VEB mediante el test de la chi-cuadrado. En 37 de 71 (52,1%) pacientes se detectaron anticuerpos IgM (VCA) o IgG (EA) en condiciones de uremia. La reactivación del VEB ocurrió en 25 de 71 (35,2%) pacientes, con manifestaciones clínicas (fiebre, leucopenia, insuficiencia renal, e incremento en las transaminasas) en nueve casos. Uno de los 71 pacientes desarrolló un SLPPT, con incremento en la carga viral pero sin evidencia de reactivacion serológica del VEB. Se encontró una asociación estadísticamente significativa entre la ausencia de micofenolato mofetil, que actúa inhibiendo la proliferación de los linfocitos y la producción de anticuerpos, en el protocolo inmunosupresor y la reactivación del VEB (p = 0,015). El diagnóstico serológico de la reactivación del VEB debería basarse en criterios estrictos: seroconversión de IgM (VCA), incremento de cuatro veces en el título de IgM (VCA) o IgG (EA), o disminución de cuatro veces en el título de IgG (EBNA) así como en el análisis de muestras seriadas de suero. Algunos pacientes seropositivos frente al VEB con un elevado riesgo de desarrollar SLPPT podrían beneficiarse de este método diagnóstico. Palabras clave: Virus de Epstein-Barr. Serología. Trasplante renal. INTRODUCTION Epstein-Barr virus (EBV) is a gamma herpesvirus being able to remain in a latency state in a low number of B lymphocytes after infection. Usually diagnosis of EBV in immunocompetent subjects is made by serology. In general, patients that receive a renal transplantation have a risk for developing post-transplantation lymphoproliferative syndromes (PTLPS) due to primary infection with the virus or its reactivation1-3. Type of transplanted organ, severity and duration of immunosuppression (IS), use of polyclonal or monoclonal anti-lymphocitic antibodies, both for induction or for treatment of acute rejection, and serological disparity towards cytomegalovirus (CMV) between the donor and the recipient (D+/R-) are described among risks factors for developing PTLPS2-4. Mycofenolate mofetil (MMF) is an immunosupressant that works by inhibiting lymphocytes proliferation 186 and antibodies production5. In transplantation, serological diagnosis is made difficult due to patients' inability to create adequate serological responses, which renders the interpretation of serological markers difficult6. VEB infection in seronegative patients is clearly associated with the development of PTLPS3. Interestingly, there are few studies relating to dynamics of EBV serological markers both in uremia7,8 and after renal transplantation4,9-11. There is a lack of consensus in the literature about the criterion for interpreting EBV serological reactivation as well as its clinical relevance in EBV seropositive transplantation recipients4,9-12. Usually, EBV serological diagnosis is based on detection of several viral markers in isolated serum samples and not on markers variability in serial serum samples. With the aim to determine whether serology may be of benefit in the diagnosis of EBV infection in these patients, we performed a prospective study about EBV serological markers in EPSTEIN-BARR VIRUS SEROLOGY IN PATIENTS WITH KIDNEY TRANSPLANTATION a group of 71 adult patients recipients of a renal transplantation (RT) in order to establish the EBV serological reactivation rate and its association with demographic data, immunosuppressive treatment, CMV infection, and clinical manifestations of patients. PATIENTS AND METHODS Patients We prospectively studied EBV serological markers in 71 patients submitted to renal transplantation from cadaver donor at the German Trias i Pujol University Hospital, Barcelona Autonomous University (Badalona), between July of 1997 and June of 2000. The protocol was previously approved by the Ethics Committee and had patients consent. Forty-five men and 26 women were studied; mean age: 49 years (range, 21 to 69 years); mean hemodialysis duration: 29 months (range, 0 to 111 months). Serum samples were collected just before transplantation (uremic samples), and at 15 days, 1, 2, 3, 4, 6, 9, and 12 months after RT, and annually thereafter. Immunosuppressive treatment In all cases Prednisone (corticosteroids) was used, and cyclosporin A (CyA) in 39.4%, tacrolimus (Fk) in 59.1%, mono/polyclonal anti-lymphocitic antibodies in 40%, and MMF in 45% of cases. Serological analysis Serum samples were analyzed for specific EBV antibodies. IgG against viral capsid antigen (VCA) was determined by indirect immunofluorescence (Gull Laboratories, Utah, USA), IgG against nuclear antigen (EBNA) by the anti-complement immunofluorescence technique (Fresenius Diagnostik, Germany), and IgM (VCA) and IgG against early antigen (EA) by enzimoimmunoassay (Biotest, Dreieich, Germany). Criterion for EBV serological reactivation and statistical analysis After reviewing the literature, we decided to use the most stringent criterion of all, one or several of the following: IgM seroconversion (VCA), a four-fold increase in IgM (VCA) or IgG (EA) titers, or a fourfold decrement in IgG (EBNA) titers. IgM (VCA) or IgG (EA) titers were defined by the optical densities (OD) ratios of consecutive serum samples. IgG (EBNA) titer was defined as the inverse of the last dilution of each serum sample in which IgG (EBNA) detection was positive. As well, EBV DNA viral load was determined in polymorphonuclear lymphocytes (PBMN) from the peripheral blood and saliva samples of the 71 patients collected at the same time as serum samples; the results have already been published13. Gancyclovir (GCV) prophylaxis was administered (intravenously during the immediate post-transplantation period and orally thereafter for one month) to all CMV seronegative patients that receive an organ from a seropositive donor. Diagnosis of CMV infection was done through detection of pp65 antigen or CMV IgM seroconversion. Patients with CMV infection were treated with GCV p.o. for two months. In case of CMV disease, GCV was started intravenously (2 months) and continued thereafter orally until completion of 2 months of treatment, usually. Statistical analysis Results were expressed as mean ± standard deviation. Differences between subgroups of RT patients with or without EBV serological reactivation were analyzed by the Chi-squared test with Yates' correction; p values < 0.05 were considered statistically significant. RESULTS Results pertaining to EBV serological markers from the 71 patients are summarized in Table I. Globally, EBV serological reactivation was diagnosed in 25 out of 71 (35.2%) RT patients. In nine out of these 25 patients, at the time of EBV reactivation it was observed: fever [2 out of 9 (22.2%)], leucopoenia [1 out of 9 (11.1%)], fever and leucopoenia [1 out of 9 (11.1%)], isolated renal failure (RF) [2 out of 9 (22.2%)], fever, RF and leucopoenia [1 out of 9 (11.1%)], fever, RF, leucopoenia and transaminases increase [1 out of 9 (11.1%)], and in 1 patient EBV reactivation was concurrent with an episode of cutaneous herpes zoster [1 out of 9 (11.1%)]. In these 9 patients, no changes in EBV DNA levels were detectable at the time of EBV serological reactivation but in one case. In 5 out of 9 patients, EBV serological reactivation occurred along with CMV infection (3 primary infections and 2 reactivations). In these 5 patients, fever and leucopoenia were the most frequent clinical manifestations. Diag187 R. LAUZURICA y cols. Table I. Results of Epstein-Barr Virus serological markers in 71 RT recipients obtained during the study period Serum samples analyzed n = 351 Mean number of serum samples per patient: 4.9 (range, 4-8) EBV serological status Follow-up period: Number of patients Type of dialysis Detection of EBV antibodies D+/R+: 69 D-/R+ > 2 years: 2 1-2 years: 22 1 year: 21 3-11 months: 26 PD: 13 HD: 56 None: 2 Uremic sample1 37/51 (52.1%) Post-transplantation 25/51 (35.2%)2 IgM (VCA): 3 IgM (VCA): 11 · HD: 3 · HD: 9 IgG (EA): 31 · PD: 2 · HD: 26 IgG (EA): 13 · PD: 4 · HD: 11 · None: 1 · PD: 2 IgM (VCA) + lgG (EA): 3 IgM (VCA) + lgG (EA): 1 · HD: 2 · HD: 1 · PD: 1 RT, renal transplantation; D, donor; R, recipient; PD, peritoneal dialysis; HD, hemodialysis; EBV, Epstein-Barr virus; IgM, immunoglubulin M; VCA, viral capsid antigen; IgG, immunoglubulin G; EA, early antigen. 1 The results corresponding to the uremic sample (pre-transplantation) reflect qualitative detection of EBV serological markers and were used as basal values for comparison. 2 Detection of EBV antibodies en post-transplantation samples according to the EBV reactivation criterion. nosis of EBV reactivation was done by IgM (VCA) seroconversion in 3 patients, and in 2 patients by a four-fold increase in IgG (EA) titers. In one patient, EBV serological reactivation preceded CMV infection, in one patient both were concurrent, and in 3 patients it occurred thereafter. There were no variations in EBV viral load in patients with CMV infection at the time of EBV serological reactivation. There were no differences either in both clinical symptoms or in serological profile. In these patients, mean time elapsed to EBV infection diagnosis was 17.8 weeks (range, 8 to 48 weeks), although in 4/5 this event occurred 8 to 12 weeks after transplantation. Most of EBV serological reactivations [20 out of 25 (80%)] were recorded within the first year, and 15 out of 25 (75%) within 3 months after transplantation. In the only patient of the 71 RT patients that developed a PTLPS (1.4%), a brain primary lymphoma, 188 there was no evidence of EBV serological reactivation. However, an increase in EBV viral load did happen at the time of PTLPS diagnosis. PTLPS treatment consisted in IS, chemotherapy and acyclovir withdrawal, but the patient died14. In 24 out of 46 (52.2%) RT patients, IgM (VCA) and/or IgG (EA) antibodies were occasionally detected, although the stringent previously established criterion of EBV serological conversion was not met. A great variability was observed both in EBV antibodies type and titer among RT patients. In 25/71 patients with EBV serological reactivation, mean OD values for IgM (VCA) and IgG (EA) were 0.453 ± 0.2709 (range, 0.27100.631) and 0.8869 ± 0.8329 (range, 0.31791.4559), respectively. Fifteen acute rejection (AR) episodes were detected in 14 out of 71 patients (19.2%). In 2 out of 15 (13.3%) an EBV serological reactivation was detected: one IgM (VCA) seroconversion and a four-fold increase in IgG (EA) titer. After AR, EBV viral load increased in two of these 15 episodes (13.4%). There was no correlation between EBV serological reactivation and frequency and severity (duration and clinical manifestations) of the AR episodes or acyclovir or gancyclovir treatment. In the same way, no differences could be observed either between patients with and without EBV reactivation in relation to demographical data: age, gender, type of dialysis [hemodialysis (HD) vs peritoneal dialysis (PD)], time on dialysis, or frequency or severity of AR episodes. When different IS treatments were compared, a statistically significant association was observed between EBV serological reactivation and MMF absence in the IS protocol (p = 0.015) (Fig. 1). CMV infection occurred in 11 out of 71 RT patients (15.5%), with clinical manifestations in all cases but one. Clinical manifestations included: fever [3 out of 10 patients (30%)]; fever and leucopoenia [4 out of 10 (40%)]; leucopoenia [1 out of 10 (10%)]; fever, transaminases increase, and respiratory disease [1 out of 10 (10%)]; fever, leucopoenia, thrombopenia, transaminases increase, and RF [1 out of 10 (10%)]. Eight patients were diagnosed of primary infection and 3 of reactivation. Mean time elapsed from RT to CMV infection was 7 weeks (range, 2 to 11 weeks). DISCUSSION There exist an association between EBV primary infection in transplanted patients and PTLPS development1. Most of the studies have focused on EBV viral load analysis and on variability of EBV serological markers in the transplantation setting, and par- EPSTEIN-BARR VIRUS SEROLOGY IN PATIENTS WITH KIDNEY TRANSPLANTATION Number of renal transplantation patients 60 50 40 30 20 10 0 EBV reactivation No reactivation MMF 8 30 No MMF 17 16 ALA 10 17 No ALA 15 29 Fk 12 30 Cs A 13 16 HD 21 37 PD 4 9 Fig. 1.--Results of the analysis of EBV serological markers in subcategorized transplanted patients. Differences between groups of patients were analyzed by the Chi-squared test with Yates' correction, considering p values < 0.05 as statistically significant. MMF: mycofenolate mofetil; ALA: anti-lymphocitic antibodies; Fk: tacrolimus; CyA: cyclosporin A; HD: hemodialysis; PD: peritoneal dialysis; MMF vs no MMF (p = 0.099); HD vs PD (p = 0.96). ticularly in EBV seropositive transplanted patients. Generally, several criteria have been used for diagnosis of EBV serological reactivation in EBV seropositive patients: IgM (VCA) qualitative detection10,12, IgG (EA) qualitative detection11,12, increase in the IgG EBNA-1 to IgGEBNA-2 ratio8, four-fold increase in IgG (VCA) antibodies titer10,12, whereas other authors have used the same criteria than we did4,9. We detected EBV serological reactivation in 27 out of 71 (35.2%) patients. Reactivation rates reported in the literature vary from 20% to 54%. Results3,4,9-12,14-15 comparison is difficult likely due to a combination of several factors such as: type of studied transplanted population (pediatric/adult), type of study (prospective/retrospective), absence of a basal uremic sample used for further comparison, and difficulties in the interpretation of serological profiles in transplanted patients. Besides, transplanted patients usually received hemoderived products at the time of transplantation. According to our results, we believe that the highest EBV reactivation rates reported in the literature could be easily overestimated. In 46 RT-recipient patients there was no evidence of EBV reactivation, although in 24 of them IgM (VCA) and/or IgG (EA) antibodies were detected at some point during the follow-up, but in no case our stringent EBV serological reactivation criterion was met. If defined criteria are not strictly observed results from the different studies may be highly variable. On the other hand, serology should always rely on the analysis of several serum samples. We found a high variability in EBV antibodies patterns and titers in the different RT patients, although when analyzed individually, they showed repetitive and constant serological profiles as well as similar OD values throughout the follow up period and when samples were analyzed and reanalyzed, which is in agreement with other authors8,15. In our series, IgG (EA) antibodies were usually detected in RT patients, which is also similar to what researchers observed6. Clinical relevance of the observed findings at the time of what we have termed as "symptomatic" reactivation in 9 out of 25 patients is difficult to interpret since we cannot demonstrate to what extent they were due to reactivation itself. Only in 4 out of 9 patients clinical symptoms were concurrent with CMV clinical infection. In one patient of our series that developed PTLPS serology was useless for PTLPS diagnosis, which is in agreement with Granot et al.15 and is in contrast with EBV viral load that increased at that precise moment13,14. We found an EBV reactivation rate in RT patients with CMV infection (45%) higher than that found in the RT population (35.2%), which is in agreement with others12, although this association lacked statistical significance. The only variable related to transplantation that was significantly associated with EBV reactivation was absence of MMF in IS treatment (p = 0.015). Since MMF inhibits lymphocytes proliferation, suppressing B cells proliferation and antibodies production, patients that did not receive MMF as part of IS treatment may present a higher risk of EBV reactivation. However, the lower frequency of serological reactivation found in these patients might be attributed to inhibition of antibodies synthesis caused by MMF. In our series, EBV reactivation was not associated with AR, acyclovir or gancyclovir, which is in agreement with other groups17,18. Although only one patient in our group developed a PTLPS, she was the only one that died. According to our results, and contrary to EBV viral load13,14, EBV serology is generally useless in diagnosing PTLPS in RT adult patients who are EBV seropositive. However, we do not rule out that EBV serology may be useful in some transplanted patients with a great risk for developing a PTLPS (EBV seronegative, and EBV seropositive patients recipients of certain organ types or submitted to protocols with great risk for IS). There is a need for both a consensus on EBV serological reactivation criterion and further EBV serological prospective studies in transplanted patients in order to determine the usefulness of these markers in PTLPS diagnosis. 189 R. LAUZURICA y cols. ACKNOWLEDGEMENT We want to thank the NIH, NIC, CCR Fellow Editorial Board for reviewing this manuscript. 10. 11. REFERENCES 12. 1. Okano M, Thicle GM, Davis JR, Grierson HL, Purtilo DT: Epstein-Barr virus and human diseases: recent advances in diagnosis. Clin Microbiol Rev 1(3): 300-302, 1998. 2. Cockfield SM, Preitsaikis J, Lewell LD, Parfrey NA: Post-transplant lymphoproliferative disorder in renal allolgraft recipients. Clinical experience and risk factor analysis in a single center. Transplantation 56: 88-96, 1993. 3. Ho M, Jaffke R, Miller G, Breining MK, Dummer JS, Makowka L, Atchinson RW, Karrer F, Nalesnik MA, Starzl TE: The frequency of Epstein-Barr virus infection and associated lymphoproliferative syndrome after transplantation and its manifestations in children. Transplantation 45: 719-727, 1988. 4. Hornef MW, Bein G, Fricke L, Steinhoff J, Wagner HJ, Hinderer W, Sonneborn HH, Kirchner H: Coincidence of Epstein-Barr virus reactivation, cytomegalovirus Infection and rejection episodes in renal transplant recipients. Transplantation 60: 474-480, 1995. 5. Eugui EM, Mircolvich A, Allison AC: Lymphocyte-selective antiproliferative and inmunosuppressive effects of mycophenolic acid in mice. Scand J Immunol 33: 175-183, 1991. 6. Matter L, Gorgievski M, Montandon A, Germann D, Telenti A: Epstein-Barr virus (EBV) antibody patterns and circulating EBV in immunocompromised hosts. Biotest Bull 5: 27-32, 1993. 7. Yamamoto T, Nakajima Y, Yamamoto M, Hironaka T, Hirai K, Nakamura Y: Epstem-Barr virus activity in patients on chronic hemodialysis. Nephron 70: 449-459, 1995. 8. Winkelspecht B, Mueller-Lantzsch N, Köhler H: Serological evidence for reactivation of EBV infection due to uraemic IMmunodeficiency. Nephrol Dyal Transplant 12: 2099-2104, 1997. 9. O'Neill HJ, Shirodaria PV: Virus-specific antibodies to Epstein-Barr virus, varicella-zoster virus and rubella virus in renal 13. 14. 15. 16. 17. 18. transplant patients with cytomegalovirus Infections. J Infect 24: 301-309, 1992. Kaden J, Petersen S, Kaden K, May G: Epstein-Barr virus infections after kidney transplantation. Transplant Int 11 (Supl. 1: S119-S124, 1998. Merlino, C, Giacchino F, Tognarelli G, Cavallo R, Re D, Sinesi F, Segoloni G, Negro Ponzi A: Infezione da virus di Epstein-Barr e disordini linfoproliferativi in portatori di trapianto renale. Minerva Urol Nefrol 48: 139-143, 1996. Aalto SM, Linnavuori K, Peltola H, Vuori E, Weissbrich B, Schubert J, Hedman L, Hedman K: Immunoreactivation of Epstein-Barr virus due to cytomegalovirus infection. J Med Virol 56: 186-191, 1998. Frías C, Lauzurica R, Bayés B, Ausina V: Prospective followup of Epstein-Barr virus load in adult kidney transplant recipients by semiquantitative polymerase chain reaction in blood and saliva samples. Eur J Clin Microbiol Infect Dis 20: 892-895, 2001. Batlle M, Frías C, Lauzurica R, Ribera JM: Aumento de la carga viral del virus de Epstein-Barr en el momento del diagnóstico de un linfoma cerebral primario en una paciente receptora de un trasplante renal. Med Clin 117: 49-50, 2001. Khameneh ZR, Soin J, Durlik M, Lao M, Paczek L, Gaciong Z: Factors afecting reactivation of Epstein-Barr virus infection after kidney allograft transplantation. Ann Transplant 4: 1822, 1999. Granot E, Doron D, Morag R: Monitoring of Epstein-Barr virus serology in children after liver transplant: lack of clinical correlation. Transplant Proc 32: 706-707, 2000. Birkeland SA, Andersen HK, Hamilton-Dutoit SJ: Preventing acute rejection, Epstein-Barr virus Infection, and posttransplant lymphoproliferative disorders after kidney transplantation: use of acyclovir and mycophenolate moletil in a steroid-free immunosuppressive protocol. Transplantation 167: 1209-1214, 1999. Giral M, Nguyen JM, Daguin P, Hourmant M, Cantarivich D, Dantal J, Blancho G, Josien R, Ancelet D, Soulillou JP: Mycophenolate mofetil does not modify the incidence of cytomegalovirus (CMV) disease after kidney transplantation but prevents CMV-induced chronic graft dysfunction. J Am Soc Nephrol 12: 1758-1763, 2001. 190