Multiparametric MRI in a Prospective Cohort of Living Kidney Donors, Recipients, and Healthy Controls: Correlations With Markers of Renal Function, Fibrosis and Ageing

Development of renal fibrosis is a characteristic end point of various renal diseases and independently predicts adverse renal outcomes. The most common cause of kidney allograft failure after 1 year is chronic allograft nephropathy which is characterized by interstitial fibrosis and tubular atrophy. At present the only method for quantifying renal fibrosis is a renal biopsy. Although the risk of complications is relatively low, the procedure is invasive, limiting its use for repeated assessments in a clinical setting. Furthermore, as only the renal cortex is usually sampled and because of the miniscule amount of tissue obtained there are issues of sampling bias and information on total renal fibrosis is limited. Estimated glomerular filtration rate (eGFR) is widely used for monitoring renal function but is insensitive for detecting allograft fibrosis.

Epidemiological data suggests that living kidney donors (LKD) are at increased risk of end stage renal disease compared to matched non-donors. Lower predonation renal function and higher age accentuate the risk. In a recent prospective study, renal cortex volume predicted renal outcome after 1 year in LKD.

In recent years there has been a rapid development in renal magnetic resonance imaging (MRI) techniques, allowing assessment of total renal morphology, microstructure, hemodynamics, oxygenation, and diffusion of water. Multiparametric MRI (mpMRI) refers to the combination of several MRI-based measures sensitive to different biophysical tissue properties in a single scan session. The information obtained could possibly provide a means of characterizing several aspects of renal function and morphology.

The MRI based measures that show the greatest promise for clinical application are longitudinal relaxation time (T1), diffusion weighted imaging (DWI), blood oxygen-level dependent imaging (BOLD) and arterial spin labelling (ASL). In a cross-sectional study, patients with chronic kidney disease (n=22) and healthy volunteers (n=22) were assessed with MRI-based measures and renal biopsy. MRI based measures were reproducible and correlated with currently used methods for measuring renal function. Significant differences were found in T1-mapping, apparent diffusion coefficient (ADC) and ASL in persons with low and high degrees of renal fibrosis. In another cross-sectional study (n = 164) the corticomedullary ADC difference was highly correlated to and independently associated with renal fibrosis in both native kidneys and allografts. In a more recent prospective study, the corticomedullary ADC difference was predictive of kidney function decline and dialysis initiation in transplant recipients and patients with CKD. Reduced oxygenation determined by low BOLD has been linked to the progression rate of chronic kidney disease. Interestingly, a study found that diffusion-based MRI measures detected an increase in allograft fibrosis before apparent changes in the estimated glomerular filtration rate.

Before mpMRI of the kidneys can be utilized in the clinical setting there is a need for development of a larger evidence base. To date clinical studies employing MRI-based measures have been mostly cross-sectional. So far, only few studies have been performed in living kidney donors (LKD) and renal transplant recipients. To the best of our knowledge, no studies have been performed with long term follow-up including repeated mpMRI, renal biopsies and accurate measurement of glomerular filtration rate with DTPA clearance. Repeated imaging is necessary to establish how MRI-based measures change over time in association with currently used reference methods for quantifying renal function and fibrosis.

Accelerated renal ageing trough p16INK4a pathway activation, leading to cellular senescence, is involved in the development of renal fibrosis. Senescent cells are characterized by irreversible growth arrest and express a pro-inflammatory and pro-fibrotic senescent associated secretory phenotype (SASP). This biochemical footprint can be detected by immunohistochemistry and may be predictive of renal fibrosis in chronic kidney disease and renal transplantation.

Serum uromodulin (sUmod) and urinary epidermal growth factor (uEGF) originate from kidney tubules and may reflect functional nephron mass and are associated with renal function decline and fibrosis.

Klotho, which was initially known as an anti-ageing gene is expressed on the cell surface membranes of proximal and distal tubules. Decreasing levels of soluble klotho accompany chronic kidney disease and have been linked to the development of renal fibrosis.

The overall aim of this prospective follow-up study is to investigate the utility of repeated mpMRI and biomarkers in monitoring renal function and for detecting the development of fibrosis in renal allografts. MRI-based measures and biomarkers of renal ageing and fibrosis will be compared to current gold standard methods for characterizing and quantifying renal function and fibrosis in a cohort of LKD, recipients and healthy controls.

Several components of this study are exploratory. A novel and non-standardised combination of MRI-based measures performed in a single scan session is used, and specific prior data to inform estimates of accuracy and variance for this combined protocol are limited. The formal power calculation therefore applies specifically to the primary histological validation analysis in transplant recipients, in whom serial protocol biopsies are obtained. Analyses involving living kidney donors and healthy controls, diagnostic performance, circulating biomarkers, and the principal component analysis-derived composite MRI index are secondary or exploratory and are not individually powered. Furthermore, published data on the relationships between MRI-based measures and fibrosis are heterogenous regarding the methods used for quantifying fibrosis histologically and thresholds for fibrosis.

The prevalence of fibrosis in renal allografts, which is of primary interest is expected to increase from baseline to the final visit. Based on previous work it is expected that 40-50 % of renal allografts will have significant levels of fibrosis after 2 years. It is however possible, that present levels of fibrosis in allografts are lower than previous due to optimized immunosuppressive regimens.

The target sample size of 32 evaluable participants per cohort was fixed at the design stage on feasibility grounds, based on the annual volume of living-donor transplantation at the two participating centres and an anticipated 20% attrition. In December 2025 a Monte Carlo simulation (500 replications) was performed to characterise the statistical power of this fixed sample size for the primary longitudinal analysis in transplant recipients. The simulation assumed 32 evaluable recipients with paired mpMRI and biopsy data at four time points (0, 3, 12 and 24 months), an intra-subject correlation of 0.6 for histological fibrosis, a residual standard deviation of 10 percentage points, and an intra-subject correlation of 0.7 for the MRI measure. The model was a linear mixed-effects model of fibrosis on time and the MRI measure entered as a time-varying covariate, with a random intercept for participant. Under these assumptions the study has 84% power to detect an effect of approximately 5 percentage points of fibrosis per 1 SD of the MRI measure at two-sided alpha of 0.05. The simulation did not alter the sample size, the recruitment target, the eligibility criteria or the outcome measures.

During the inclusion period LKD and recipients from Aalborg- and Aarhus University Hospital will be offered to participate in the study. Currently approximately 30-35 living donor pairs are transplanted annually. During the inclusion period of 2 years, the investigators aim to include 40 living donors and recipients. With an expected drop-out of 20% 32 living donors and recipients are expected to complete the study.

Multiparametric MRI mpMRI will be performed according to standard operating procedure in the Department of Diagnostic imaging at the participating hospitals.

The mpMRI scan protocol will be performed on a 3 Tesla MRI scanner (General Electrics (GE) Healthcare) and includes the following MRI measures:

• Structural T2 and/or T1 weighted renal MRI (for volumetric measurement)
• Diffusion weighted imaging (DWI) with b-values 0, 10, 20, 30, 40, 50, 70, 100, 200, 300, 400, 500, 800 s/mm2
• T1- and T2-mapping
• BOLD MRI with T2*
• ASL perfusion
• Non-contrast Magnetic Resonance Angiography
• Quantitative flow measurements in the renal arteries

Scan time is approximately 45 minutes. During this time participants will be required to lie still in the supine position. Participants are instructed to refrain from eating for 3 hours prior to each scan; intake of liquids is permitted. In accordance with PARENCHIMA recommendations, participants attend in a normally hydrated state and no fixed fluid-loading protocol is applied.

Postprocessing of MRI data will be done with the aid of commercially available software and analysis software developed in-house in collaboration with Aalborg- and Aarhus University Hospital and Aarhus university.

99mTc-DTPA clearance is a method for accurately measuring GFR. The DTPA clearance will be performed according to standard operating procedure at the Departments of Diagnostic Imaging at the participating hospitals.

Allograft biopsy Allograft biopsy Biopsies are obtained under ultrasound guidance. Biopsies performed at Aarhus University Hospital, which account for the large majority of study biopsies, use an 18G needle; the smaller number performed at Aalborg University Hospital use a 16G needle. Needle gauge is therefore determined by study site rather than by participant characteristics or clinical circumstance. Specimens are fixed in phosphate-buffered 4% formaldehyde, embedded in paraffin, and stained with hematoxylin and eosin, periodic acid-Schiff, and Masson trichrome.

Quantitative histological evaluation will be used to evaluate biopsies using and Olympus BX50 light microscope (Olympus Denmark, Ballerup, Denmark) equipped with a prior motorized stage, and Olympus DP70 digital camera interfaced to a PC with commercially available newCAST software (Visiopharm, Hørsholm, Denmark).

Extent of fibrosis will be quantified by systematic evaluation of scattered test points (≈100) across the Masson trichrome-stained sections using a x4 lens (NA 0.13) (29).

Additionally, biopsies will be evaluated for markers of renal ageing and fibrosis:

• p16INK4a pathway activation
• klotho
• additional markers of renal ageing and fibrosis

Blood- and urine samples Hemoglobin, hematocrit, leukocytes, platelets, CRP, creatinine, eGFR, urea, sodium, potassium, total phosphate, bicarbonate, ionized calcium, total magnesium, intact PTH, alkaline phosphatase, glucose, soluble klotho and sUmod.

A spot urine sample will be used for the determination of u-albumin/creatinine ratio, glucose, leucocytes, nitrite, blood and uEGF.

Analysis plan. The statistical analysis plan was developed in full during preparation of the study protocol manuscript, after recruitment had commenced, and is described in detail in the published protocol. It comprises linear mixed-effects modelling of the association between MRI measures and histological fibrosis, corticomedullary difference and ratio variables derived from T1, T2, T2* and ADC, a principal component analysis-derived composite MRI index designated as exploratory and internally validated by bootstrap resampling, receiver operating characteristic analysis at fibrosis thresholds of 10% and 25%, and an assessment of intraobserver reproducibility of MRI post-processing. None of these analyses altered the outcome measures, the eligibility criteria or the target enrolment.