Towards clinical translation of urinary vitronectin for non-invasive detection and monitoring of renal fibrosis in kidney transplant patients

Abstract

Background: Interstitial fibrosis and tubular atrophy (IFTA) is a critical factor in the prognosis of kidney health. Currently, IFTA quantitation in kidney biopsy samples is crucial for diagnosis and assessing disease severity, but the available non-invasive biomarkers are not satisfactory. Proteomic studies identified urinary vitronectin (VTN) as a potential biomarker for kidney fibrosis. As mass spectrometry techniques are not practical for use in clinical settings, we tested whether evaluation of urinary VTN levels through enzyme-linked immunosorbent assay (ELISA) can help monitor fibrotic changes in kidney transplant recipients and prove the clinical viability of the assay.

Methods: A total of 58 kidney transplant (KTx) patients who underwent renal biopsy were included in the study. Patients were categorized into two groups referred as no fibrosis (0%) or with fibrosis (≥ 5%) based on their histological findings. In a subsequent/follow-up analysis, the time elapsed from transplantation was also considered. The urinary levels of VTN were measured using ELISA.

Results: VTN (p = 0.0180) and VTN normalized by urinary creatinine levels (p = 0.0037), were significantly increased in patients with fibrotic grafts. When focusing on patients with long-term grafts (> 3 years from transplantation, n = 36), VTN exhibited superior potential in identifying fibrotic grafts compared to albuminuria (VTN p = 0.0040 vs. albuminuria p = 0.0132). Importantly, in this group, while albuminuria correctly identified 71% of fibrotic patients, the combination of VTN plus albuminuria correctly classified 89% of fibrotic grafts detected by renal biopsy.

Conclusions: VTN has emerged as a valid indicator of renal fibrosis. Of interest, urinary levels of VTN in combination with conventional clinical parameters (such as albuminuria) significantly improved the non-invasive detection of renal fibrosis in kidney transplant patients.

Keywords: (3–10): Biomarker; ELISA; Extracellular matrix; Kidney fibrosis; Liquid biopsy; Nephrology; kidney transplant.

Fig. 1

Analysis of all KTx patients. (A) Spearman’s correlation coefficient matrix. Spearman r values are indicated in bold and p-values of the correlation are indicated at the bottom if p-value < 0.2. P-values > 0.2 are indicated as ns. (B) Violin plots showing the descriptive statistics and statistical differences according to the parameter “years from transplantation”, (C) albuminuria levels, (D) VTN levels obtained by ELISA and (E) VTN levels normalized by urinary creatinine, stratified regarding the percentage of fibrosis observed in the kidney biopsies. Statistical differences are indicated for *p < 0.05 and **p < 0.01 by a Mann-Whitney test. (F) ROC curve based on “years from transplantation”, (G) albuminuria (H) VTN, and (I) VTN normalized by urinary creatinine, as a unique biomarker to differentiate the two groups of patients (no fibrosis vs. fibrosis). VTN, Vitronectin; U.Creat, urinary creatinine; mALB, albuminuria

Fig. 2

VTN levels of all the patients stratified by their years from transplantation and fibrosis percentage. Data are shown as mean ± SD in all groups

Fig. 3

Analysis of the “long-term” KTx patients. (A) Spearman’s correlation coefficient matrix. Spearman r values are indicated in bold and p-values of the correlation are indicated at the bottom if p-value < 0.2. P-values > 0.2 are indicated as ns. (B) Violin plots showing the descriptive statistics and statistical differences according to the parameter “years from transplantation”, (C) albuminuria levels, (D) VTN levels obtained by ELISA and (E) VTN levels normalized by urinary creatinine, stratified regarding the percentage of fibrosis observed in the kidney biopsies. The statistical differences are indicated for *p < 0.05 and **p < 0.01 by a Mann-Whitney test. (F) ROC curve based on “years from transplantation”, (G) albuminuria (H) VTN, and (I) VTN normalized by urinary creatinine, as a unique biomarker to differentiate the two groups of patients (no fibrosis vs. fibrosis). VTN, Vitronectin; U.Creat, urinary creatinine; mALB, albuminuria

Fig. 4

Analysis of the “long-term” KTx patients. Venn diagrams showing the number and percentage of patients with fibrosis that would be detected by VTN or VTN/U.Creat levels, albuminuria levels or both, without a biopsy. (A and B) VTN, VTN/U.Creat, and albuminuria threshold levels have been stablished based on the ROC curves of these parameters (VTN considered positive > 12.80 ng/mL; VTN/U.Creat considered positive > 0.13; albuminuria considered positive > 95.65 mg/g). (C and D) The clinical albuminuria threshold (> 300 mg/g) was considered for the analysis. VTN, vitronectin; U.Creat, urinary creatinine

Fig. 5

(A) VTN levels of healthy controls and “long-term” KTx patients stratified by percentage of fibrosis, with descriptive statistics at the bottom. (B) VTN levels normalized by urinary creatinine of healthy controls and “long-term” KTx patients stratified by percentage of fibrosis, with their descriptive statistics at the bottom. Statistical differences are indicated for *p < 0.05, ***p < 0.001 and ****p < 0.0001 by a Kruskal Wallis test with Dunn’s post-hoc analysis. (C) VTN levels obtained by ELISA stratified regarding the percentage of fibrosis observed in the kidney biopsies. Statistical differences are indicated for ****p < 0.0001 by a Mann-Whitney test. (D) ROC curve based VTN as a unique biomarker to differentiate the two groups of patients (healthy controls vs. fibrotic patients). Cntrl, Healthy controls; VTN, Vitronectin; U.Creat, urinary creatinine