Observational study provides suggestive evidence

Monitoring Epstein-Barr virus (EBV) levels after liver transplantation to avoid over-immunosuppression showed a signal for less post-transplant lymphoproliferative disease (PTLD) over the long term in an observational study.

Standardized incidence of PTLD was consistently numerically lower over time at a hospital that used EBV monitoring for liver transplant patients compared with one that didn’t, reported Bart van Hoek, MD, PhD, of Leiden University Medical Center in the Netherlands, and colleagues in the Annals of Internal Medicineopens in a new tab or window.

Accounting for decreasing PTLD rates over time by looking at the difference between hospitals for the contemporary period versus a historical period at the same hospital, the estimates ranged from 28.7 to 70.6 fewer incident PTLD cases with monitoring per 1,000 patients over 5 to 15 years of follow-up.

However, none of the differences in differences were statistically significant.

The monitoring strategy hospital had more patients require rejection treatment, especially in the first 3 months after transplantation, compared with the control hospital, “which could be associated with a reduction in immunosuppression,” van Hoek and team noted. However, “all of these rejections were easily treatable and did not lead to graft loss.”

By contrast, PTLD is associated with morbidity and mortality, and is thus of “utmost importance” to avoid, they argued. While EBV infection or reactivation is asymptomatic in most liver transplant patients, approximately 70% of PTLD cases that do occur are related to this highly prevalent virus.

“The current data contradict the conclusion from [a prior retrospective] studyopens in a new tab or window that EBV viremia is benign, and on the basis of the current data, we consider detectable EBV VL [viral load] as a sign of over-immunosuppression, which can lead to B-lymphocyte proliferation and PTLD,” van Hoek’s group wrote.

Despite the limitations of the retrospective observational study, “we strongly believe that the reported results merit serious consideration of the EBV VL monitoring policy in an attempt to reduce the incidence of PTLD after LT [liver transplant] in adults,” they noted. “At least such a strategy seems safe.”

van Hoek and colleagues examined health records for adult recipients of a first liver transplant at Leiden University Medical Center after it started routine EBV DNA monitoring on liver transplant patients in September 2003. The program involved weekly monitoring in the first month after transplantation, biweekly monitoring in the second month, and then monthly or when patients came for additional visits for the rest of the first year. Thereafter, the viral load was measured at least yearly.

The 302 patients treated under the EBV viral load monitoring strategy through January 2017 were compared with 116 historical controls with corresponding transplants from September 1992 until the start of monitoring.

Among the 12% of monitored patients who had two or more positive viral load measurements within 2 months in the first year after transplant, 89% had their immunosuppression regimen reduced as required by the protocol. Of the 21% of the monitored patients with a single positive viral load result followed by an undetectable level on the repeat test, 44% had their immunosuppression reduced based on physician judgment. Altogether, 33% had at least one detectable EBV viral load measurement in the first year, and 60% of this group had their immunosuppression reduced.

After the first year, 25% of patients had at least one positive EBV viral load result, with 46% having their immunosuppressive medication reduced.

Another contemporary control group at a second university medical center who didn’t undergo EBV monitoring included 579 liver transplant patients from September 2003 through January 2017. A fourth group, of historical controls at that center, included 284 patients from 1986 through January 2003.

The historical control group at both centers had numerically but not significantly more PTLD events compared with the contemporary era (crude incidence 25.5 vs 13.3 per 1,000 patients at the control center and 40.4 vs 4.2 per 1,000 at the monitoring center), which the researchers suggested was “likely to be related to less immunosuppression in contemporary versus historical patients, similar to renal transplant.”

Their main results utilized an inverse probability of treatment-weighted number of patients for PTLD, because distributions showed “many influential outliers.” Replacing scores above the 95th percentile with the 95th percentile and those below the fifth percentile with the fifth percentile improved but did not achieve complete balance in initial characteristics.

The cohort had an average age of 46.2 to 53.2 across groups, and 53.2% to 71.9% were men. EBV immunoglobulin G positivity was 96.0% to 99.1%.

Other limitations to the study included the use of persistently detectable EBV viral levels as the threshold for treatment decisions, although the level above which action is required has not been well established and even the same plasma assay analyzed in a different laboratory could generate a different detection limit cutoff, as the researchers noted.

“An EBV VL monitoring strategy with immunosuppression reduction may reduce the incidence of PTLD in other adult patients with long-term immunosuppression and may contribute to tumor surveillance and prevention of other infections; however, future studies should confirm this,” they concluded.