Abstract
PCR monitoring for EBV reactivation is commonly used to confirm the diagnosis of EBV-LPD in alloSCT recipients. Previous studies using multiple PCR approaches have suggested a correlation between high EBV copy number and impending EBV-LPD, potentially allowing for the administration of preemptive rituximab and/or donor leukocytes prior to disease onset. However, subpopulations of alloSCT recipients with EBV reactivation at higher risk for EBV-LPD remain incompletely defined. We sought to determine the demographic factors among alloSCT recipients that affect the predictive value of our EBV PCR assay. Quantitative real-time PCR was performed on serum samples obtained from alloSCT recipients using primers that amplify a portion of the BNRF1-p143 locus. Clinical patient information was retrieved from institutional research databases. Patients who did not undergo appropriate EBV PCR screening (n=13), had a history of EBV-LPD prior to alloSCT (n=1) or received rituximab for prophylaxis against EBV-LPD (n=2) were excluded. 187 alloSCT recipients underwent EBV PCR monitoring between 7/03–7/05, of whom 58 (31.0%) were <18 years old, 117 (62.6%) were female and 111 (59.4%) received allografts depleted of T cells ex vivo (TCD). 56/187 (29.9%) had EBV reactivation (i.e., PCR >500 copies/mL) at some point after alloSCT. The prevalence of EBV reactivation was similar in recipients of TCD and conventional allografts (28.8% vs. 31.5%). Among patients with EBV reactivation, the risk for EBV-LPD did not differ based on age, sex, underlying disease, donor relation or HLA disparity. However, of the 56 patients with EBV reactivation, 5/32 (15.6%) recipients of TCD allografts and 0/24 (0.0%) recipients of conventional allografts developed EBV-LPD (p<0.03). The highest PCR values within one month prior to presentation of EBV-LPD varied ~1,000-fold between the 5 patients (799, 1007, 2380, 12K and 862K copies/mL). Of the alloSCT recipients without EBV-LPD, 8 had PCR values >500 for >6 months and 3 had ≥1 PCR value >100K. In our study population, the utility of the EBV PCR was maximized in recipients of TCD allografts with PCR values >500 (Table). In conclusion, EBV reactivation is common among recipients of both conventional and T cell depleted allografts, but preemptive therapy guided by EBV PCR should be primarily directed toward the latter. The effects of in vivo TCD with agents such as alemtuzumab on EBV reactivation and EBV-LPD are under investigation.
Patient group . | PCR cutoff (copies/mL) . | Sensitivity . | Specificity . | Positive predictive value . | Negative predictive value . | Number needed to treat . |
---|---|---|---|---|---|---|
The highest PCR value within 1 year prior to LPD presentation was used for patients with LPD. | ||||||
All patients | >500 | 100% | 72% | 9% | 100% | 11.2 |
>1,000 | 80% | 77% | 9% | 99% | 12.5 | |
>5,000 | 40% | 85% | 7% | 98% | 20 | |
>10,000 | 40% | 88% | 9% | 98% | 14.6 | |
>100,000 | 20% | 98% | 25% | 98% | 4.4 | |
T cell depleted | >500 | 100% | 75% | 16% | 100% | 6.4 |
>1,000 | 80% | 77% | 14% | 99% | 7.6 | |
>5,000 | 40% | 85% | 11% | 97% | 12.7 | |
>10,000 | 40% | 89% | 14% | 97% | 8.9 | |
>100,000 | 20% | 98% | 33% | 96% | 3.4 |
Patient group . | PCR cutoff (copies/mL) . | Sensitivity . | Specificity . | Positive predictive value . | Negative predictive value . | Number needed to treat . |
---|---|---|---|---|---|---|
The highest PCR value within 1 year prior to LPD presentation was used for patients with LPD. | ||||||
All patients | >500 | 100% | 72% | 9% | 100% | 11.2 |
>1,000 | 80% | 77% | 9% | 99% | 12.5 | |
>5,000 | 40% | 85% | 7% | 98% | 20 | |
>10,000 | 40% | 88% | 9% | 98% | 14.6 | |
>100,000 | 20% | 98% | 25% | 98% | 4.4 | |
T cell depleted | >500 | 100% | 75% | 16% | 100% | 6.4 |
>1,000 | 80% | 77% | 14% | 99% | 7.6 | |
>5,000 | 40% | 85% | 11% | 97% | 12.7 | |
>10,000 | 40% | 89% | 14% | 97% | 8.9 | |
>100,000 | 20% | 98% | 33% | 96% | 3.4 |
Author notes
Corresponding author