Background
The diagnosis of rejection in renal transplantation is currently reliant on core needle biopsies given the lack of specificity of renal function measurements in blood or urine. Core biopsies are generally accepted to be safe but are expensive, labor-intensive, and can be accompanied by serious complications that may require hospitalization, such as gross hematuria requiring blood transfusions as well as even surgical interventions or angiographic embolizations (1–2.5% of kidney biopsies) [
1,
2]. Frequent monitoring of transplanted kidney status using repeated core needle biopsies is therefore not routinely feasible.
A less-invasive method of monitoring allograft status would facilitate understanding and management of allograft dysfunction. Such a tool would be especially relevant where protocol biopsies are recommended to detect subclinical T cell or antibody-mediated rejection, prior to overt functional allograft deterioration, or where response to therapy may have been suboptimal in routine practice [
3].
Molecular analyses are emerging as important techniques to complement or in certain cases to even replace tissue histology in diagnosis of transplant dysfunction. Robust transcript sets have been identified which predict future tissue alterations and/or reflect early acute or chronic allograft rejection in renal allografts [
4,
5]. A set of 13 transcripts were found to be predictive of fibrosis at 1 year and subsequent loss of allograft function [
6]. Endothelial cell-derived transcript expression was shown to reflect active antibody-mediated microvascular injury and poor transplant outcome despite absence of histologic C4d staining [
7]. A gene set has been described which differentiated polyomavirus-associated nephropathy from acute renal allograft rejection [
8]. In each of these circumstances, the molecular diagnosis provided relevant clinical information not revealed by histology alone. Gene profiling has also been extended to other organs and disease processes owing to the highly stereotyped and organ-unspecific molecular patterns associated with inflammation and immune responses [
9].
Given the inherent risk of repeated core needle biopsies, but the high potential clinical value of obtaining transcriptome information, we have investigated whether FNA, being less invasive, yields equivalent molecular information to regular core needle biopsies. In this proof-of-principle study we compared RNA sequencing results from two well defined gene sets (cell cycle and Wnt panels) in aspirates and biopsies from nephrectomy specimens.
Methods
Patients and tissue samples
Kidney tissue samples were obtained from patients (
n = 11) with suspected renal cell carcinoma at the time of radical nephrectomy. All patients had normal kidney function (eGFR > 60 ml/min/1.73m
2) and had undergone total nephrectomy. Key characteristics of the patients are outlined in Table
1. One 16 gauge (G16) core biopsy and four fine needle aspirate samples (FNA) using 19-, 21-, 23- and 25-gauge needles were obtained with one pass (attempt) per individual biopsy or FNA from normal appearing kidney tissue as far as possible away from the tumor. Kidneys with warm ischemia time over 2 hours were excluded. Ethics approval was granted by the regional ethics committee of the Western Norway Regional Health Authority (REK vest); approval number: 78–05. Written informed consent was obtained from all patients.
Table 1
Patient/sample overview at the time of nephrectomy (n = 11)
39 N | X | X | X | X |
42 N | X | X | | |
44 N | X | X | X | X |
47 N | X | X | | |
49 N | X | X | | |
50 N | X | X | X | X |
57 N | X | X | X | X |
64 N | X | X | X | X |
65 N | X | X | X | X |
66 N | X | X | | |
69 N | X | X | | |
Concurrently harvested kidney biopsies (G16) from the tumor and adjacent normal tissue from a nested subset (
n = 6) of the total group of subjects (
n = 11), stored as formalin-fixed, paraffin-embedded (FFPE) tissues, had previously been analyzed by whole transcriptome RNA sequencing (TruSeq RNA Access kit®, Illumina, USA) (Table
1) [
10,
11]. The complete transcriptome RNA sequencing data from these patients (
n = 6) were included to investigate potential differences between whole transcriptome and targeted sequencing.
RNA extraction and analysis
Core biopsy samples (G16) were immediately placed in RNA later® (Qiagen, Netherlands). FNA samples (G19–25) were placed directly into 700 μl of Qiazol lysis buffer (Qiagen, Netherlands). Samples were stored at − 80 °C.
Core needle kidney biopsy samples were weighed and cut into pieces of approximately 5 mg. Homogenization of biopsy cores and FNA samples was performed in 700 μl Qiazol lysis buffer with ceramic beads. All samples were homogenized for 3 × 10 seconds at 6800 rpm on a Precellys homogenizer (Bertin Technologies, USA). Total RNA was extracted using the miRNeasy Micro kit (Qiagen, Netherlands). RNA was stored at -80C until further use for RNA sequencing.
RNA concentration was measured using a NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific, USA). RNA integrity was assessed using the Agilent RNA 6000 Nano kit on a 2100 Bioanalyzer (Agilent Technologies, USA). The DV200 value (the percentage of RNA fragments longer than 200 nucleotides) was used to reflect RNA quality for subsequent sequencing from FFPE tissues [
10,
11]. A minimum DV200 value of 30% is recommended and required for RNA sequencing [
12]. RNA integrity number (RIN) values were also determined.
RNA sequencing
RNA sequencing and analyses were conducted using targeted RNA expression, heatmaps and PCA. RNA sequencing was performed using two different predesigned TruSeq Targeted RNA panels from Illumina©. Panel sequencing was selected because, if successful, sequencing of defined gene panels (e.g. related to rejection processes) is much more likely to be integrated into routine clinical practice than full transcriptome sequencing.
Sequencing libraries were generated according to manufacturer’s instructions (Illumina, Inc., San Diego, USA) with RNA normalized to inputs of 200 ng total RNA from each sample. Equal amounts of libraries were pooled (normalized) to a final concentration of 18 pM and subjected to cluster and single end read sequencing.
Sequencing was performed for 50 cycles using MiSeq Reagent Kit v3 on a MiSeq® instrument (Illumina, Inc., San Diego, USA). Base calling was done on the instrument and Fastq files were generated using bcl2fastq v.2.18 with default settings for the RNA-seq data. Transcript expression values were determined using the Salmon software [
15] with the GRCh37 transcriptome downloaded from the Ensembl database (
http://www.ensembl.org/).
The datasets created during and/or analysed during the current study are available from the corresponding author upon reasonable request.
Discussion
This study demonstrates that RNA sequencing of FNA samples provides equivalent molecular information to that of core needle biopsy samples. These proof of principle findings may justify a less invasive approach to immune monitoring in organ transplantation, which thus far has heavily relied on full core biopsies.
Currently percutaneous FNA is often performed for cancer diagnosis in various solid organs, such as breast, liver, and thyroid gland as well as in accessible lymph nodes [
18]. FNAs are much easier to perform than core needle biopsies and are associated with significantly fewer complications. A recent metaanalysis of more than 4500 patients undergoing core-needle biopsy and/or FNA for thyroid cancer diagnosis reported an essentially 0 % complication rate [
18]. In addition, cessation of oral anticoagulation or anti-platelet agents might not be necessary before fine needle aspiration [
19,
20]. Importantly also, patients with potentially malignant renal masses, who have many comorbidities (e.g. cardiac events necessitating stents and platelet inhibitors) and a high operative risk, could benefit from the less invasive initial fine needle approach that includes transcriptome analyses. Indeed, a recent study demonstrated very good diagnostic accuracy - and a synergistic diagnostic advantage - of both FNA and core needle biopsy for renal masses in native kidneys [
21]. Similar evidence in renal transplantation is lacking.
The present study demonstrates that sufficient quantities of RNA suitable for RNA sequencing can be obtained with a wide variety of fine needle sizes. Moreover, the comparative analyses indicate excellent correlation of mRNA expression patterns abundances obtained by FNA and full core biopsies. The internal comparator group with complete RNA sequencing further highlighted the similarity between fine and core needle biopsies with targeted sequencing.
This study has several important strengths. Current FNA and core biopsy samples were obtained under direct vision after nephrectomy and permitting stringent comparability of RNA expression from the various procedures taken from the same tissue area. The quality and quantity of RNA obtained by FNA was robust and therefore strongly suggest that FNA may be a viable diagnostic tool in selected patients. Despite these encouraging findings however, our study has several limitations that must be acknowledged. Samples were obtained from explanted kidneys which were no longer perfused, therefore potential confounding of FNA reproducibility by circulating blood cells as compared with core biopsy samples cannot be addressed. Comparative studies in perfused organs are required to assess the influence of peripheral blood and biopsy/aspiration site on the results. Clinical complications could not be assessed in this study, however a proof-of-principal study such as ours is a necessary ethical pre-requisite prior to consideration of clinical studies. Further limitations to our study include the facts that we provide panel sequencing and not complete RNA sequencing of the whole genome and that we did not determine the variability of RNA expression between multiple FNA aspirates and the same biopsy. Moreover, FNA is not able to deliver the transcriptome of individual cell types and a detailed pathology report cannot be assigned to an FNA. In addition, as with all transcriptome studies, it remains to be shown if mRNA results of our selected panel genes can be extrapolated to the respective gene product on the protein level and to other genes and pathways.
Transcriptome analysis has not yet reached the clinical mainstream and therefore the true clinical impact of facilitated transcriptome/RNA monitoring also requires prospective studies. It could be argued that the current hesitation in performing multiple biopsies in the same patient may have hindered progress in understanding the clinical value of the transcriptome. The use of FNA may therefore accelerate understanding of its clinical relevance. The close correlation of gene expression from RNA obtained with FNA as compared with core biopsies, as well as the relative safety, justifies further investigation and confirmation of the value of FNA in prospective clinical studies.
It is important to acknowledge however that FNA is not an adequate source for histological investigations. Therefore, we do not suggest that FNA can replace all histologic analysis, which remains important for diagnosis of disease recurrence, borderline rejection changes, viral infections etc. We continue to support the need for histology especially for the first biopsy to ensure that unexpected and transplant-unrelated diagnoses are not missed. It is conceivable however that FNA could replace protocol biopsies, especially where allograft function appears normal. More frequent monitoring with serial protocol FNA in immunologically high-risk kidney allografts might detect rejection processes earlier than histological changes and/or an overt decline in kidney function. Candidate gene panels for transplant rejection have already been defined [
22,
23]. FNA would also be an attractive procedure for follow up monitoring of tissue response to therapy after a core biopsy baseline histologic and transcriptome diagnosis is made. Availability of diagnosis using FNA could optimize timing and reduce frequency of indication of core needle biopsies not only in kidney but also in other transplanted organs.
Acknowledgements
We are thankful to the other local urologists for participation in biopsy harvesting. The library preparation, sequencing and partly the bioinformatics analysis were provided by the Genomics Core Facility (GCF), Norwegian University of Science and Technology (NTNU). GCF is funded by the Faculty of Medicine and Health Science at NTNU and Central Norway Regional Health Authority.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (
http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.