The gold standard for characterizing chronic kidney disease (CKD) is the glomerular filtration rate (GFR), which is commonly estimated in both native and transplanted kidneys for patient monitoring and therapeutic management and ultimately guides decision-making about whether a patient needs renal replacement therapy. In particular, the National Kidney Foundation has defined CKD stages according to estimated GFR (eGFR) values and in several studies, the eGFR slope or change has been found to be strongly associated with end stage renal disease (ESRD).

However, little is known about the heterogeneity of eGFR evolution in time - i.e. eGFR trajectories - and the related progression to ESRD and death. To date, no studies have investigated eGFR trajectories in diversified cohorts and populations worldwide, although this approach could provide a better understanding of CKD evolution and hence improve risk stratification. In addition, determinants of eGFR trajectories remain poorly described.

An unsupervised approach could allow examining eGFR trajectories over time and could lead to the identification of patient groups according to the probability of the progression of their kidney disease.

Therefore, this study aims:

1. To identify the long-term eGFR trajectories after kidney transplantation using latent class mixed models;
2. To identify the clinical, immunological, histological and functional determinants of the eGFR trajectories using multinomial regressions;
3. To investigate the associations of the eGFR trajectories with the progression to ESRD and death.

Based on the results, the investigators will provide an easily accessible tool to calculate personalized probabilities of belonging to eGFR trajectories after kidney transplantation, by using datasets from prospective cohorts and post hoc analysis of randomized control trial datasets.

Background

Chronic kidney disease (CKD) now affects 850 million individuals worldwide, exceeding the global prevalence of diabetes, cancer and HIV/AIDS. In addition, end-stage renal disease affects 7.4 million individuals and mortality rate for individuals burdened by kidney disease is now estimated at 5 to 10 million individuals each year. Therefore, developing better diagnostic and treatment approaches for the kidney disease epidemic is a global priority, as leading professional societies and health agencies have emphasized (the US Food & Drug Administration, the National Kidney Foundation, the European Medicines Agency, the European Society of Organ Transplantation, the American Society for Transplantation and the American Society of Transplant Surgeons).

However, current approaches for investigating the relationship between eGFR course and outcomes such as ESRD and mortality have been limited by registries with an overall lack on granular data, including infrequent eGFR measurements for a single patient and convenience clinical samples. An unsupervised longitudinal approach to determine patient eGFR evolution may bring an original perspective to the traditional clinical interpretation of kidney function based on limited eGFR measurements, short-term follow-up, and standard statistical approach.

Main Outcome(s) and Measure(s)

• eGFR trajectories
• Determinants of eGFR trajectories
• Associations of eGFR trajectories with ESRD and death
• Prediction system that will provide the personalized probabilities of belonging to eGFR trajectories

Inclusion Criteria:

• Kidney recipients transplanted after 2000
• Kidney recipients over 18 years of age
• Kidney recipients with at least two eGFR measurements after transplantation

Exclusion Criteria:

• Combined transplantation

Kidney recipients aged over 18 and of all sexes recruited from 2000 in European and North American centers, who have eGFR repeated measurements, data from protocol and for cause biopsies, as well as immunological, clinical, functional parameters and survival data for allograft and patient survival assessment. Additional kidney recipients from RCTs with longitudinal data including baseline and follow-up clinical, functional, immunological and histological data.