Evaluation of use of Epstein-Barr viral load in patients after allogeneic stem cell transplantation to diagnose and monitor posttransplant lymphoproliferative disease

Abstract

Epstein-Barr virus (EBV)-induced posttransplant lymphoproliferative disease (PTLD) continues to be a serious complication following transplantation. The aim of the present study was to evaluate the EBV load as a parameter for the prediction and monitoring of PTLD. The EBV load was analyzed by a quantitative competitive PCR with 417 whole-blood samples of 59 patients after allogeneic stem cell transplantation (SCT). The EBV load was positive for all 9 patients with PTLD and for 17 patients without PTLD. The viral loads of patients with manifest PTLD differed from the loads of those without PTLD (median loads, 1.4 x 10(6) versus 4 x 10(4) copies/microg of DNA; P < 0.0001). A threshold value of 10(5) copies/microg of DNA showed the best diagnostic efficacy (sensitivity, 87%; specificity, 91%). However, in patients with less than three major risk factors for PTLD, the positive predictive value of this threshold was rather low. One week prior to the manifestation of PTLD, the EBV load was as low in patients who developed PTLD as in patients without disease (median, 2.2 x 10(4) copies/microg of DNA; P was not significant). EBV DNA tested positive first at 20 to 71 days prior to the clinical manifestation of PTLD and occurred with the same delay after transplantation regardless of disease (median delay, 52 versus 63 days; P was not significant). EBV DNA was detected earlier in patients with primary infections than in those with reactivations (33 versus 79 days; P = 0.01), but the peak levels were similar in the two groups. EBV primary infection or EBV reactivation is frequent in patients after allogeneic SCT but results in PTLD only in a subgroup of patients. Although evaluation of the EBV load has limitations, the EBV load represents a valuable parameter to guide therapy.

FIG. 1.

Comparison of EBV loads in patients with and without PTLD. The EBV loads in EBV DNA-positive samples from patients without PTLD (open squares; median load, 4.0 × 10⁴ copies/μg of DNA; 266 samples were negative) were compared with those in patients with PTLD at the time of diagnosis or during the course of PTLD (closed circles; median load, 1.4 × 10⁶ copies/μg of DNA) and those in the last sample before PTLD was manifested (open circles; median load, 2.2 × 10⁴ copies/μg of DNA; 1 sample was negative).

FIG. 2.

Threshold value calculation for EBV load for diagnosis of PTLD. The corresponding sensitivities (closed symbols) and specificities (open symbols) for different threshold values are shown.

FIG. 3.

EBV loads in patients developing PTLD after SCT. All patients received acyclovir (ACV) prophylaxis until day 30 after SCT. (A) Patient 1 developed high-grade B-cell non-Hodgkin's lymphoma during ganciclovir (GCV) and foscarnet (FC) treatment and received no further PTLD treatment. (B) Patient 3 developed diffuse large-cell B-cell non-Hodgkin's lymphoma during ganciclovir and foscarnet treatment; PTLD regressed after two donor lymphocyte infusions (DLI; solid arrows). (C) Patient 4 developed high-grade B-cell non-Hodgkin's lymphoma during antiviral treatment and received no further PTLD treatment. (D) Patient 6 developed high-grade B-cell non-Hodgkin's lymphoma during foscarnet, ganciclovir, and cidofovir treatment and additionally received donor lymphocyte infusions (solid arrow) and anti-CD20 antibody (Rituximab; dotted arrow). (E) Patient 7 received only acyclovir prophylaxis. (F) Patient 8 developed PTLD after a second transplantation due to primary graft failure. Acyclovir prophylaxis was extended until day 30 after the second transplantation. The patient received anti-CD20 antibody (Rituximab; dotted arrow).

FIG. 4.

Examples of EBV loads after SCT in patients who did not develop PTLD. (A) Adult patient with EBV reactivation without any EBV-related symptoms. (B) Pediatric patient with EBV primary infection and infectious mononucleosis. ACV, acyclovir; FC, foscarnet; GCV, ganciclovir.

FIG. 5.

Comparison of peak EBV load in patients with primary infection and reactivation. Open symbols, EBV primary infection (median load, 6.8 × 10⁴ copies/μg of DNA), closed symbols, EBV reactivation (median load, 6.6 × 10⁴ copies/μg of DNA) (P was not significant).

FIG. 6.

First times of detection of EBV DNA-positive sample after transplantation. (A) First times of EBV DNA detection in patients with (solid line) and without PTLD (dotted line). (B) First times of EBV DNA-positive sample in patients with primary infection (solid line) or reactivation (dotted line).