The Medication Level Variability Index (MLVI), a Behavioral Biomarker, Predicts Late Allograft Rejection in Pediatric Liver Transplant Recipients – A Prospective Multisite Study.
1Mount Sinai, New York
2Children's Hospital, Cincinnati
3Emmes Corporation, Rockville
4UCLA, Los Angeles
5Lurie Children's Hospital, Chicago
6Children's Hospital, Pittsburgh
7Children's Hospital, Texas.
Meeting: 2016 American Transplant Congress
Abstract number: 358
Keywords: Immunosuppression, Liver
Session Information
Session Name: Concurrent Session: Pediatric Liver Transplant
Session Type: Concurrent Session
Date: Monday, June 13, 2016
Session Time: 4:30pm-6:00pm
Presentation Time: 5:18pm-5:30pm
Location: Room 206
Objective: The MALT [Medication Adherence in children who had a Liver Transplant (LT), NCT-01154075, R01DK080740 (NIDDK)] multisite study prospectively evaluated a biomarker for erratic medication adherence (MLVI), as a predictor of late allograft rejection (LAR). The primary hypothesis was that a higher MLVI will predict a masked central pathology reading of LAR.
Methods: Subjects were 1-17 years old LT recipients, more than one year after LT, who did not have LAR in the year prior to enrollment. MLVI, defined as the standard deviation (SD) of sequential tacrolimus blood levels, was calculated for each subject from at least three out-patient levels. Decisions to do a liver biopsy were clinically determined at sites. Biopsies were read locally to drive immunosuppression strategy and centrally for the study endpoint. Locally diagnosed LAR both histologically and clinically (modification of immunosuppression) were also recorded. Pre-defined secondary analyses evaluated the MLVI-LAR relationship in adolescents.
Results: 5 centers enrolled 400 patients (attrition <2%). A higher MLVI predicted later development of the primary and secondary endpoints: centrally confirmed LAR [mean pre-rejection MLVI for patients with LAR: 2.4 (3.6 SD) vs. without LAR, 1.6 (1.1); p=0.026], site-read LAR [2.6 (3.8) vs. 1.6 (1.1), p=0.007], and immunosuppression modification because of suspected LAR [2.5 (3.5) vs. 1.6 (1.1)], p=0.004]. In adolescents (>12 years old), the Area Under the Receiver-Operator-Characteristic Curve was 0.78 (0.61 for the entire cohort). 45% of adolescents who met the MLVI threshold (>2) in year 1 had LAR in year 2, as compared with LAR rate of 8% for those below the threshold.
Conclusions: This prospective multisite study has shown that the MLVI, a cost-effective innovative extension of existing clinical practice, is a robust predictor of LAR. As one of few known biomarkers of human behavior, the MLVI could inform behavioral interventions to improve post-transplant outcomes in pediatric liver transplantation and in other patient groups.
CITATION INFORMATION: Shemesh E, Bucuvalas J, Anand R, Venick R, Alonso E, Mitchell J, Annunziato R, Mazariegos G, Shneider B. The Medication Level Variability Index (MLVI), a Behavioral Biomarker, Predicts Late Allograft Rejection in Pediatric Liver Transplant Recipients – A Prospective Multisite Study. Am J Transplant. 2016;16 (suppl 3).
To cite this abstract in AMA style:
Shemesh E, Bucuvalas J, Anand R, Venick R, Alonso E, Mitchell J, Annunziato R, Mazariegos G, Shneider B. The Medication Level Variability Index (MLVI), a Behavioral Biomarker, Predicts Late Allograft Rejection in Pediatric Liver Transplant Recipients – A Prospective Multisite Study. [abstract]. Am J Transplant. 2016; 16 (suppl 3). https://atcmeetingabstracts.com/abstract/the-medication-level-variability-index-mlvi-a-behavioral-biomarker-predicts-late-allograft-rejection-in-pediatric-liver-transplant-recipients-a-prospective-multisite-study/. Accessed December 11, 2024.« Back to 2016 American Transplant Congress