Early isolated V-lesion may not truly represent rejection of the kidney allograft.

Intimal arteritis is known to be a negative prognostic factor for kidney allograft survival. Isolated v-lesion (IV) is defined as intimal arteritis with minimal tubulointerstitial inflammation (TI). Although the Banff classification assesses IV as T cell-mediated rejection (TCMR), clinical, and prognostic significance of early IV (early IV, eIV) with negative C4d and donor-specific antibodies (DSA) remains unclear. To help resolve if such eIV truly represents acute rejection, a molecular study was performed. The transcriptome of eIV (n=6), T cell-mediated vascular rejection with rich TI (T cell-mediated vascular rejection, TCMRV, n=4) and non-rejection histologic findings (n=8) was compared using microarrays. A total of 310 genes were identified to be deregulated in TCMRV compared with eIV. Gene enrichment analysis categorized deregulated genes to be associated primarily with T-cells associated biological processes involved in an innate and adaptive immune and inflammatory response. Comparison of deregulated gene lists between the study groups and controls showed only a 1.7% gene overlap. Unsupervised hierarchical cluster analysis revealed clear distinction of eIV from TCMRV and showed similarity with a control group. Up-regulation of immune response genes in TCMRV was validated using RT-qPCR in a different set of eIV (n=12) and TCMRV (n=8) samples. The transcriptome of early IV (< 1 month) with negative C4d and DSA is associated with a weak immune signature compared with TCMRV and shows similarity with normal findings. Such eIV may feature non-rejection origin and reflect an injury distinct from an alloimmune response. The present study supports use of molecular methods when interpreting kidney allograft biopsy findings.

Introduction

Intimal arteritis, also called v-lesion, is known to be a negative prognostic factor for kidney allograft survival [1-3]. According to the Banff classification, intimal arteritis is characterized by infiltration of mononuclear cells beneath the arterial endothelium. It is considered pathognomonic of acute rejection, either T cell- (TCMR) or antibody-mediated rejection (ABMR) [4,5].

Isolated v-lesion (IV), defined as intimal arteritis that occurs with minimal concurrent tubulointerstitial inflammation (TI), poses a challenge of unclear clinical and prognostic significance. According to the Banff classification, arterial involvement is sufficient to classify IV as type II or III acute TCMR regardless of interstitial inflammation (i) and tubulitis (t). IV could also be considered as ABMR in the presence of donor-specific antibodies (DSA) and evidence of antibody interaction with the endothelium (C4d positive staining, moderate microvascular inflammation (MI) or gene expression markers of endothelial injury in the biopsy tissue) [6].

However, the role of the immune response in the pathophysiology of IV remains unclear. Recent studies have reported diverse findings on IV response to antirejection treatment, clinical and prognostic significance [7-12]. While some authors believe in hidden ischemic/reperfusion injury, others are convinced of its rejection origin.

In response to the limitations of conventional histologic assessment, molecular technologies are being explored in the quest for greater precision [13]. In the transplantation field, a few genome-wide transcriptional studies have aimed to elucidate the origin of IV [14-16]. However, inclusion of heterogeneous samples regardless of DSA and time after transplantation, brought no clear findings, particularly on the non-antibody-associated phenotype of IV.

Therefore, this retrospective study was conducted to thoroughly evaluate the transcriptome of early v-lesion within the first month after kidney transplantation with no signs of ABMR (negative DSA, absence of C4d, and MI) and minimal (eIV) or rich (T cell-mediated vascular rejection, TCMRV) tubulointerstitial inflammation.

Materials and methods

Study design and patients

An observational, nested case–control study within a large single-center retrospective cohort study was performed to compare the expression profiles of renal allograft biopsies with histologic findings of eIV and TCMRV. eIV was defined as intimal arteritis (v > 0) with minimal tubulitis (t ≤ 1) and interstitial inflammation (i ≤ 1), no MI [glomerulitis (g) + peritubular capillaritis (ptc) = 0], with C4d and DSA negativity. TCMRV was defined as the presence of intimal arteritis with significant TI (t > 1, i > 1) and negative C4d, MI and DSA and served as a positive control. Only indication kidney allograft biopsies performed within the first month of post-transplant were included. As a negative control group, normal histologic findings from 3-month protocol biopsies without a history of previous rejection were included.

For a training set, 18 patients undergoing kidney transplantation at the Prague-based Institute for Clinical and Experimental Medicine (IKEM) from 2009 to 2014, with eIV (n=6), TCMRV (n=4), or normal histology (n=8) and sufficient biopsy samples for microarray evaluation were selected. A flowchart of the training set is provided in Figure 1, patients’ baseline characteristics are shown in Table 1.

Figure 1

Study flowchart of the training set (cross-sectional screening)

All patients after kidney transplantation at our department in the years 2009–2014 were considered for inclusion in the study. Abbreviations: AVR, acute vascular rejection; DSA, donor specific antibodies; eIV, early isolated v-lesion; IKEM, Institute for Clinical and Experimental Medicine; TCMRV, T cell-mediated vascular rejection; Tx, transplantation.

Table 1

Characteristics of patients in the training set

	TCMRV (n=4)	eIV (n=6)	normal (n=8)	P
Donor characteristics
Age, years	44.5 (40.3; 57)	58.5 (51.3; 68.3)	54 (31; 58)	0.186
Gender (female)	2 (50)	5 (83.3)2	1 (14.3)2	0.045
Type of donor (deceased)	3 (75)	6 (100)	8 (100)	0.157
ECD	1 (15)	5 (83.3)	4 (57)	0.175
Hypertension	1 (15)	5 (83.3)	3 (43)	0.125
Recipient characteristics
Age, years	55.2 (44; 66)	59.5 (53.3; 66.3)	62.5 (56.5; 66.2)	0.792
Gender (female)	1 (25)	1 (16.7)	2 (25)	0.923
Dialysis vintage, months	39.7 (24.7; 45)	30 (11; 43)	21.6 (15.6; 30)	0.678
PRA max, %	3 (0.5; 32.5)	22 (6.5; 34.75)	1 (0; 3.5)	0.062
HLA mismatch	4.5 (4; 5)	3 (2.75; 4.25)	3 (2; 4.75)	0.22
HLA at biopsy (neg./pos./not known)	2/0/2	6/0/02	0/0/82	0.001
DSA at biopsy (neg./pos./not known)	2/0/2	6/0/02	0/0/82	0.001
Transplantation characteristics
Retransplantation	0	0	0	N/A
Cold ischemia, hours	15.25 (6.5; 19.8)	15 (14; 16)	18.1 (7.4; 19.5)	0.696
Delayed graft function	1 (15)	2 (33)	3 (43)	0.908
Biopsy characteristics
Postoperative day	6.5 (6; 8)3	7.5 (7; 10)2	97 (93.5; 101)2,3	0.001
Clinical characteristics
Induction therapy				0.005
None	3 (75)1	01	8 (100)
Basiliximab	1 (25)	5 (83.3)2	02
Thymoglobulin	0	1 (16.7)	0
Maintenance immunosuppression				0.245
TAC/MMF/steroids	3 (75)	6	5 (62.5)
CsA/MMF/steroids	1 (25)	0	3 (37.5)
TAC level at the biopsy (µg/l)	14.6 (10.3; 17.2)	10.95 (8; 16.6)	8.7 (8.3; 10.4)	0.646
CsA level at the biopsy (µg/l)	231	NA	191 (176; 278)	1
Anti-rejection treatment
Steroids	2 (50)	5 (83)	N/A	0.3745
Thymoglobulin	1 (25)	1 (17)	N/A
Steroids + Thymoglobulin	1 (25)	0	N/A
Serum creatinine (µmol/l)
At biopsy	378 (307; 448)3	441 (226; 514)2	115 (97; 154)2,3	0.002
12 months	119 (100; 121)	161 (151; 172)	115 (93; 148)	0.171
24 months	114 (100; 121)	186 (158; 187)2	106 (99; 144)2	0.017
Proteinuria (g/24 h)
At biopsy	0.21 (0.105; 1.37)1	1.52 (1.25; 2.27)1,2	0.36 (0.27; 0.55)2	0.038
12 months	0 (0; 0.11)	0.26 (0.23; 0.26)	0.26 (0.18; 0.40)	0.144
24 months	0.12 (0.06; 0.54)	0.29 (0.09; 0.3)	0.19 (0.10; 0.33)	0.913
Banff scores in diagnostic biopsy (grade)
mm	0 (0; 0)	0 (0; 0)	0 (0; 0)	0.535
g	0 (0; 0)	0 (0; 0)	0 (0; 0)	1
cg	0 (0; 0)	0 (0; 0)	0 (0; 0)	1
i	2 (2; 2)1,3	0 (0; 0)1	0 (0; 0)3	0.0001
t	2.5 (2; 3)1,3	0 (0; 0)1	0 (0; 0)2	0.001
v	1(1; 1.75)3	1 (1; 1.25)2	0 (0; 0)2,3	0.0001
ptc	0 (0; 0)	0 (0; 0)	0 (0; 0)	1
ti	2 (2; 2)3	0 (0; 1)2	0 (0; 0)2,3	0.001
ci	0 (0; 0.75)	1 (0.75; 1)2	0 (0; 0)2	0.027
ct	0.5 (0; 1)	0 (0; 1)	1 (0; 1)	0.871
ah	1 (1; 1)	1.5 (1; 1.25)2	0 (0; 0.75)2	0.004
cv	1 (0.25; 1)	2 (1; 3)2	0 (0; 0)2	0.002
C4d positivity, n	0	0	0	N/A
Pathologic diagnosis
pure TCMR	33	42	02,3	0.0012
TCMR + ATN	1	2	0
Normal	03	02	82,3
Number of glomeruli	8.5 (7.25; 9)	12 (7; 15)	8 (7; 12.25)	0.522

Data are presented as medians (interquartile (IQ) range) or n (%). Differences were calculated by the Kruskal–Wallis test or χ2 Fisher exact test and significant results of post hoc comparisons were adjusted by the Bonferroni correction for multiple tests (1TCMRV compared with eIV, 2eIV compared with control, 3TCMRV compared with control). Abbreviations: ah, arteriolar hyaline thickening; ATN, acute tubular necrosis; cg, transplant glomerulopathy; ci, interstitial fibrosis; ct, tubular atrophy; cv, vascular intimal fibrosis; CsA, cyclosporine A; ECD, expanded criteria donor; g, glomerulitis; HLA, human leukocyte antigen; i, interstitial inflammation; mm, mesangial matrix expansion; MMF, mycophenolate mofetil; ptc, peritubular capillaritis; PRA, panel reactive antibody; t, tubulitis; TAC, tacrolimus; ti, total interstitial inflammation; v, intimal arteritis.

For a validation set, 12 eIV and 8 TCMRV biopsy samples from patients transplanted between 2010 and 2016 were retrospectively identified and validated by RT-qPCR (Table 2). The Institutional Review Board (IRB) of IKEM approved the study protocol (G09-12-20), and all patients provided informed consent to participate in the study.

Table 2

Characteristics of patients in the validation set

	TCMRV (n=8)	eIV (n=12)	P
Donor characteristics
Age, years	46 (41; 62)	57 (50; 66)	0.082
Gender (female)	4 (50)	7 (51.3)	0.535
Type of donor (deceased)	5 (62.5)	10 (83.2)	0.296
ECD	3 (37.5)	6 (50)	0.465
Recipient characteristics
Age, years	52 (43; 62)	58 (54; 60)	0.3841
Gender (female)	1 (12.5)	2 (16.7)	0.656
Dialysis vintage, months	41 (17.5; 51)	20.4 (7.1; 25)	0.698
PRA max, %	3 (0; 10)	2 (0.5; 32.3)	0.973
HLA mismatch	5 (4; 5)	3 (3; 4.75)	0.39
HLA at biopsy (neg./pos./not known)	1/1/6	4/1/7	0.571
DSA at biopsy (neg./pos./not known)	2/0/6	5/0/7	0.392
CKD diagnosis, n (%)			0.339
Diabetes	1 (12.5)	5 (41.7)
Glomerulonephritis	2 (25)	2 (16.7)
Polycystosis	1 (12.5)	2 (16.7)
TIN	2 (25)	0
Hypertension	1 (12.5)	2 (16.7)
Ischemic nephropathy	1 (12.5)	0
Other	0	1 (8.3)
Transplantation characteristics
Retransplantation	1 (12.5)	1 (8.3)	0.653
Cold ischemia, hours	14 (1.25; 14)	15 (11.7; 18.2)	1
Delayed graft function	4 (50)	6 (50)	0.675
Biopsy characteristics
Postoperative day	6.5 (6; 18)	7 (6; 13)	1
Clinical characteristics
Induction therapy			0.966
None	1 (12.5)	2 (16.7)
Basiliximab	5 (62.5)	7 (58.3)
Thymoglobulin	2 (25)	3 (25)
Maintenance immunosuppression			1
TAC/MMF/steroids	8 (100)	11 (100)
TAC level at the biopsy (µg/l)	11.6 (8.3; 15.1)	10.3 (7.55;14.7)	0.629
Rejection treatment			0.4177
Steroids	4(50)	8 (67)
Thymoglobulin	3 (37.5)	4 (33)
Steroids + Thymoglobulin	1 (12.5)	0
Serum creatinine (µmol/l)
At biopsy	378 (166; 470)	464 (231; 609)	0.305
12 months	154 (112; 167)	154 (135; 188)	0.571
24 months	142 (118; 172)	144 (113; 184)	1.0
Banff scores in diagnostic biopsy (grade)
mm	0 (0; 0)	0 (0; 0)	1
g	0 (0; 0)	0 (0; 0)	1
cg	0 (0; 0)	0 (0; 0)	1
i	2 (2; 2)	0 (0; 0.75)	0.000016
t	2 (1.25; 3)	0 (0; 1)	0.0002
v	1 (1; 1.75)	1 (1; 2)	0.734
ptc	0 (0; 0)	0 (0; 0)	1
ti	2 (2; 2)	0 (0; 1)	0.000016
ci	0 (0; 0.75)	0.5 (0; 1)	0.384
ct	0.5 (0; 1)	1 (0.25; 1)	0.384
ah	1 (1; 1)	1 (1; 2)	0.181
cv	1 (1; 1)	2 (1; 2)	0.02
C4d positivity, n	0	0	1
Pathologic diagnosis			0.2421
Pure TCMR	8	9
TCMR + ATN	0	3
Number of glomeruli	9 (7–19)	12 (7.5–14)	0.851

Data are presented as medians (interquartile (IQ) range) or n (%). Differences were calculated by the Mann–Whitney test or χ2 Fisher exact test. Abbreviations: ah, arteriolar hyaline thickening; ATN, acute tubular necrosis; cg, transplant glomerulopathy; ci, interstitial fibrosis; CKD, chronic kidney disease; ct, tubular atrophy; cv, vascular intimal fibrosis; ECD, expanded criteria donor; g, glomerulitis; HLA, human leukocyte antigen; i, interstitial inflammation; mm, mesangial matrix expansion; MMF, mycophenolate mofetil; ptc, peritubular capillaritis; PRA, panel reactive antibody; t, tubulitis; TAC, tacrolimus; TCMR, T cell- mediated rejection; ti, total interstitial inflammation; TIN, tubulointerstitial nephritis; v, intimal arteritis.

Induction immunosuppression was based on the immunological risk of recipients according to panel reactive antibodies, DSA, dialysis vintage, and transplantation history. High-risk patients received depletive induction treatment with Thymoglobulin, medium-risk patients received basiliximab, and low-risk patients received no induction treatment. Maintenance immunosuppression was based on standard triple immunosuppression regimen with tacrolimus in the majority of patients. T cell-mediated rejection was treated by steroid pulses. Thymoglobulin was administered in the case of type 2B TCMR or in steroid-resistant rejections.

Biopsy of kidney allografts

All the kidney allograft biopsies were obtained under ultrasound guidance using a 16-gauge Tru-Cut needle (Somatex®; Medical Technologies GmbH, Germany). A small portion (2 mm) of the cortical zone of the biopsy specimen was immediately placed in the RNA later Stabilization Reagent (Qiagen) and stored at −20 or −80°C until RNA extraction. The remaining part of each particular biopsy was sent for histological examination and found representative.

Microarray analysis

RNA preparation and hybridization

Total RNA was isolated from 16-gauge renal biopsies using the RNeasy Micro Kit (Qiagen, Hilden, Germany) according to manufacturer’s instructions. Samples with RNA integrity number < 6 (measured using the Agilent RNA 6000 Nano Kit on the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, U.S.A.) were excluded from analysis. A total of 200 ng of purified RNA served as a template for the amplification and biotinylation of cRNA using the Illumina® TotalPrep™ RNA Amplification Kit (Thermo Fisher Scientific, Waltham, MA, U.S.A.) according to manufacturer’s instructions. Yields of labeled cRNA were determined using the Agilent RNA 6000 Nano Kit on the Agilent 2100 Bioanalyzer (Agilent Technologies). Labeled cRNA (750 ng) was hybridized at 58°C for 16 h to Illumina HumanHT-12 v4.0 Expression BeadChips (Illumina, Inc., San Diego, CA, U.S.A.), and then washed and Cy3-stained according to manufacturer’s instructions.

Scanning and microarray data analysis

After hybridization, the chip was scanned using the iScan System (Illumina), and raw data were extracted with the BeadStudio Data Analysis Software (Illumina). The lumi package in R software was used to process the raw data. The quantile method was used for normalization [17]. The given workflow represents one of the standard preprocessing pipelines for Illumina microarrays with verified applicability under similar conditions [18].

Only probes with detectable signal intensity in at least five samples were included in the analysis. After normalization, probes with log of average expression intensity less than 3.5 were removed from further analysis. Differentially expressed genes were chosen as those with a fold change > 2 and adjusted P<0.05 (Benjamini–Hochberg method). The affected genes were functionally annotated, and the deregulated pathways were identified using the David database (http://david.abcc.ncifcrf.gov). For comparing the lists of deregulated genes, an interactive online tool for Venn diagrams http://bioinformatics.psb.ugent.be/webtools/Venn/ was used. Complete raw and normalized data were deposited in the NCBI Gene Expression Omnibus (GEO) database [19] and are accessible through the GEO Series accession number GSE114712 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE114712).

Validation of microarray data

In order to validate the microarray results, internal validation and validation by RT-qPCR analysis was performed.

Internal validation

To identify the most relevant genes that distinguish between the TCMRV and eIV groups, the well-known support vector machine (SVM) classification algorithm [20], specifically e1071 R library implementation was applied [21]. The negative effect of overfitting caused by the small number of samples and the large number of evaluated genes was minimized by the recursive feature elimination (RFE) algorithm [22]. We first trained the classifiers on microarray data. Performance of SVM–RFE was evaluated by leave-one-out cross validation (LOOCV) [23]. Since the group sizes were not perfectly balanced in their size, the usual measure of classification accuracy (ACC) was complemented by another measure robust to skewed groups, the area under the ROC curve (AUC) [24]. The LOOCV demonstrated that TCMRV and eIV groups can be perfectly differentiated by a large scale of gene sets of various sizes, i.e., the groups can be separated by a small number of genes as well as by much larger gene sets. The gene set size ranges from 5 to 500 genes and showed absolute 100% ACC and maximum AUC (Supplementary Figure S1). However, the single small set of genes that separate the study groups was not easy to determine due to the large number of differentially expressed microarray genes. SVM–RFE perfectly split the groups with a large scale of potentially different gene sets of various sizes. For this reason, the final selection of genes for external validation was not induced directly from SVM-RFE as described below.

Validation by RT-qPCR analysis

Using a custom-made Taqman low-density array (Applied Biosystems, Foster City, CA, U.S.A.) on a validation set of patients (n=20), RT-qPCR was employed to analyze 38 genes identified by microarray to be deregulated between TCMRV and eIV and found to be of biological interest in the rejection process (Supplementary Table S1). Quantitative RT-qPCR based on TLDA technology was carried out as described elsewhere [25]. Real-time RT-qPCR data were quantified using the SDS 2.4 software package (Applied Biosystems) and relative gene expression values were determined using the comparative 2−ΔΔ method of the relative quantification (RQ) Manager Software v 1.2.1 (Applied Biosystems) with normalization to an endogenous control (HPRT1). The endogenous control was chosen from three candidate housekeeping genes (GAPDH-Hs99999905_m1, PGK1-Hs99999906_m1, HPRT1- Hs01003267_m1) using NormFinder (www.mdL.dk) as the gene with the most stable expression (HPRT1 with a stability value of 0.003). As a calibrator, one of the samples with a good expression profile on all of the target genes was used. All investigated mRNAs were measured in triplicate for each sample.

Risk of overfitting

In our study, we deal with the well-known n<