Background: Limited studies have compared the intra-patient tacrolimus exposure variability (IP-TEV) between Advagraf and Prograf using different IP-TEV metrics. Moreover, the clinical implications of IP-TEV in kidney transplant recipients on Advagraf are not clear. This study compares IP-TEV of Prograf and Advagraf in the context of a conversion program, and the impact of IP-TEV in Advagraf levels on clinical outcomes one year post-conversion, including acute rejection, and total graft failure (i.e., composite of graft loss or death with graft function (DWGF)).
Methods: The cohort included 471 KTRs, transplanted between 1-Jan-2000 to 31-Dec-2012, who were in a formal Advagraf conversion program between 1-Sep-2012 to 31-Dec-2013, and followed to 31-Dec-2014.  Patients had tacrolimus blood trough measurements for 12-months pre-conversion (Prograf) and 12-months post-conversion (Advagraf) and they received a transplant at least 3-months prior to conversion. IP-TEV was determined using standard deviation (SD), coefficient of variation (CV), and intrapatient variability % (IPV%) calculated with one year of pre- and post-values. SD, CV, and IPV% were compared using Wilcoxon matched-pairs signed-ranked tests. Kaplan-Meier curves as well as univariable and multivariable Cox proportional hazard models were used to analyze graft outcomes.
Results: The difference in median SD, CV, and IPV% pre- and post-conversion were 0.16 (p=0.09), 0.01 (p=0.52), and 1.41 (p=0.32), respectively. Box plots of the distribution of pre- and post-conversion variability measures were symmetric and centred near zero. Multivariable Cox models showed that for every 1 unit increase in Advagraf SD and IPV% there was a 1.19 and 1.02-fold increase in the hazard of graft failure (p=0.01 and 0.03, respectively). Every 0.05 unit increase in Advagraf CV was associated with 1.12 and 1.15-fold increase in the hazard of graft failure (p=0.001 and 0.001. respectively. Upon dividing patients into two groups based on the median, there was a significantly increased risk of developing total graft failure in the SD >0.92 vs. ≤0.92 group and CV >0.16 vs. ≤0.16 group but no association in IPV% >7.7 vs. ≤7.7 group.
Conclusion: Despite no significant difference in IP-TEV upon conversion from Prograf to Advagraf based on SD, CV, and IPV%, all three methods showed similar trends in variability. All three methods showed that increased Advagraf variability post-conversion was associated with an increased risk of total graft failure. Future studies should address the long-term implications of Advagraf intra-patient variability in de novo KTR to corroborate the findings of the current study.