Progress and challenges in assessing allograft health in liver transplantation

In this issue of Liver Transplantation, Schaapman and colleagues from 3 leading European transplant centers sought to determine whether multiparametric MRI (mpMRI), in combination with serum tests, could noninvasively identify patients undergoing liver transplant at low risk for rejection and thus avoid liver biopsy. In a real-world, prospective, multinational study, 131 adult and pediatric recipients of liver transplants underwent liver biopsies (protocol 52% and for cause 48%) and an additional research noncontrast MRI. The investigators demonstrated that the composite imaging and serum biomarkers could identify those without either indeterminate or T-cell–mediated allograft rejection and potentially avoid liver biopsy.1 Overall, the study supports the important concept that allograft health may be determined noninvasively without the risk and cost of liver biopsy. The report has significant strengths, most critically that they were able to implement the same tools of assessment effectively across 3 distinct centers. However, because this was a pragmatic real-world study, it lacked many of the strengths of a controlled trial. In addition, important questions are raised for future investigations: (1) When does the mpMRI add value over that of serum liver tests?; (2) Would a marker of immune activation add sensitivity and specificity?; (3) If the threshold for liver tests were lower, would the added value of the mpMRI be attenuated?; (4) How often and under what circumstances were the liver tests and mpMRI not congruent?; and (5) How does mpMRI perform in nonimmunologic injury? In the field of liver transplantation, we have advanced our understanding of the alloimmune response; however, we have not broken the code of immune tolerance to liver allografts. Development of operational tolerance is the exception, even among stable long-term pediatric and adult recipients of liver transplants.2–5 Silent graft damage is prevalent among stable, long-term recipients of liver transplants. Deep inquiry into the biology of silent allograft injury has revealed an alloimmune process suggesting inadequate immunosuppression.6,7 But that is only part of the story. Immunosuppression in excess of that needed to maintain allograft health increases the risk of complications, including cancer, chronic renal injury, and cardiovascular disease.8 Both chronic allograft injury and complications of chronic immunosuppression are similarly insidious as the cumulative impact of inadequate or excessive immunosuppression may take decades, not years, to manifest. Immunosuppression management at most liver transplant centers is directed by empiric, center-based protocols that are stratified according to liver disease etiology and time since transplant. Deviations from protocol are driven by a patient’s medical and/or immunological history and by serial liver tests, even though the latter lacks both sensitivity and specificity to detect graft injury. Given the risk associated with imprecise immunosuppression, it seems we must move away from empiricism. The challenge encourages us to ask, “How can we personalize immunosuppression to avoid allograft damage incurred by too little immunosuppression and complications incurred by too much immunosuppression?” What is the value of assessment of allograft health? Assessment of allograft health seems the cornerstone of personalized immunosuppression. We know from well-constructed clinical trials that liver transplant recipients, both adults and children, with biopsy-proven quiescent allografts and normal liver tests can safely minimize immunosuppression.2–5 The assumption that decreased exposure to immunosuppression will decrease the long-term risk of complications seems sound, but physicians and programs often do not implement surveillance biopsy protocols due to the invasiveness and perceived risks. Liver biopsy remains the gold standard to definitively diagnose T-cell–mediated rejection but still is less than perfect due to often significant interobserver variability across pathologists. Moreover, liver biopsies incur direct and indirect costs, the former related to loss of work or time for the patient and family. This brings us back to the opportunity offered by noninvasive assessment of liver allograft health. What is our current state of knowledge? Sanchez-Fueyo and colleagues recently reported that combined assessment of ALT, class II donor-specific antibodies, and determination of liver stiffness could noninvasively identify stable recipients of liver transplant without significant alloimmunity, a subgroup of patients who might benefit from minimization of immunosuppression.9 This current study, if indeed these findings were extended to larger studies, could be a valuable addition to our toolbox of noninvasive markers of allograft health so that we may progress toward precision immunosuppression and away from empiricism.