Discovery and validation of a molecular signature for the noninvasive diagnosis of human renal allograft fibrosis

Dany Anglicheau, Thangamani Muthukumar, Aurélie Hummel, Ruchuang Ding, Vijay K. Sharma, Darshana Dadhania, Surya V. Seshan, Joseph E. Schwartz, Manikkam Suthanthiran

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

Background: Tubulointerstitial fibrosis (fibrosis), a histologic feature associated with a failing kidney allograft, is diagnosed using the invasive allograft biopsy. A noninvasive diagnostic test for fibrosis may help improve allograft outcome. Methods: We obtained 114 urine specimens from 114 renal allograft recipients: 48 from 48 recipients with fibrosis in their biopsy results and 66 from 66 recipients with normal biopsy results. Levels of messenger RNAs (mRNAs) in urinary cells were measured using kinetic, quantitative polymerase chain reaction assays, and the levels were related to allograft diagnosis. A discovery set of 76 recipients (32 with allograft fibrosis and 44 with normal biopsy results) was used to develop a diagnostic signature, and an independent validation set of 38 recipients (16 with allograft fibrosis and 22 with normal biopsy results) was used to validate the signature. Results: In the discovery set, urinary cell levels of the following mRNAs were significantly associated with the presence of allograft fibrosis: vimentin (P<0.0001, logistic regression model), hepatocyte growth factor (P<0.0001), α-smooth muscle actin (P<0.0001), fibronectin 1 (P<0.0001), perforin (P=0.0002), plasminogen activator inhibitor 1 (P=0.0002), transforming growth factor β1 (P=0.0004), tissue inhibitor of metalloproteinase 1 (P=0.0009), granzyme B (P=0.0009), fibroblast-specific protein 1 (P=0.006), CD103 (P=0.02), and collagen 1A1 (P=0.04). A four-gene model composed of the levels of mRNA for vimentin, NKCC2, and E-cadherin and of 18S ribosomal RNA provided the most accurate, parsimonious diagnostic model of allograft fibrosis with a sensitivity of 93.8% and a specificity of 84.1% (P<0.0001). In the independent validation set, this same model predicted the presence of allograft fibrosis with a sensitivity of 77.3% and a specificity of 87.5% (P<0.0001). Conclusions: Measurement of mRNAs in urinary cells may offer a noninvasive means of diagnosing fibrosis in human renal allografts.

Original languageEnglish
Pages (from-to)1136-1146
Number of pages11
JournalTransplantation
Volume93
Issue number11
DOIs
Publication statusPublished - 15 Jun 2012
Externally publishedYes

Fingerprint

Allografts
Fibrosis
Kidney
Biopsy
Messenger RNA
Vimentin
Logistic Models
18S Ribosomal RNA
Granzymes
Perforin
Tissue Inhibitor of Metalloproteinase-1
Hepatocyte Growth Factor
Plasminogen Activator Inhibitor 1
Transforming Growth Factors
Cadherins
Fibronectins
Routine Diagnostic Tests
Smooth Muscle
Actins
Real-Time Polymerase Chain Reaction

Keywords

  • Fibrosis
  • Gene expression
  • Kidney transplantation

ASJC Scopus subject areas

  • Transplantation

Cite this

Discovery and validation of a molecular signature for the noninvasive diagnosis of human renal allograft fibrosis. / Anglicheau, Dany; Muthukumar, Thangamani; Hummel, Aurélie; Ding, Ruchuang; Sharma, Vijay K.; Dadhania, Darshana; Seshan, Surya V.; Schwartz, Joseph E.; Suthanthiran, Manikkam.

In: Transplantation, Vol. 93, No. 11, 15.06.2012, p. 1136-1146.

Research output: Contribution to journalArticle

Anglicheau, D, Muthukumar, T, Hummel, A, Ding, R, Sharma, VK, Dadhania, D, Seshan, SV, Schwartz, JE & Suthanthiran, M 2012, 'Discovery and validation of a molecular signature for the noninvasive diagnosis of human renal allograft fibrosis', Transplantation, vol. 93, no. 11, pp. 1136-1146. https://doi.org/10.1097/TP.0b013e31824ef181
Anglicheau, Dany ; Muthukumar, Thangamani ; Hummel, Aurélie ; Ding, Ruchuang ; Sharma, Vijay K. ; Dadhania, Darshana ; Seshan, Surya V. ; Schwartz, Joseph E. ; Suthanthiran, Manikkam. / Discovery and validation of a molecular signature for the noninvasive diagnosis of human renal allograft fibrosis. In: Transplantation. 2012 ; Vol. 93, No. 11. pp. 1136-1146.
@article{4abe7444f7ca4bf181c7c075e86a185c,
title = "Discovery and validation of a molecular signature for the noninvasive diagnosis of human renal allograft fibrosis",
abstract = "Background: Tubulointerstitial fibrosis (fibrosis), a histologic feature associated with a failing kidney allograft, is diagnosed using the invasive allograft biopsy. A noninvasive diagnostic test for fibrosis may help improve allograft outcome. Methods: We obtained 114 urine specimens from 114 renal allograft recipients: 48 from 48 recipients with fibrosis in their biopsy results and 66 from 66 recipients with normal biopsy results. Levels of messenger RNAs (mRNAs) in urinary cells were measured using kinetic, quantitative polymerase chain reaction assays, and the levels were related to allograft diagnosis. A discovery set of 76 recipients (32 with allograft fibrosis and 44 with normal biopsy results) was used to develop a diagnostic signature, and an independent validation set of 38 recipients (16 with allograft fibrosis and 22 with normal biopsy results) was used to validate the signature. Results: In the discovery set, urinary cell levels of the following mRNAs were significantly associated with the presence of allograft fibrosis: vimentin (P<0.0001, logistic regression model), hepatocyte growth factor (P<0.0001), α-smooth muscle actin (P<0.0001), fibronectin 1 (P<0.0001), perforin (P=0.0002), plasminogen activator inhibitor 1 (P=0.0002), transforming growth factor β1 (P=0.0004), tissue inhibitor of metalloproteinase 1 (P=0.0009), granzyme B (P=0.0009), fibroblast-specific protein 1 (P=0.006), CD103 (P=0.02), and collagen 1A1 (P=0.04). A four-gene model composed of the levels of mRNA for vimentin, NKCC2, and E-cadherin and of 18S ribosomal RNA provided the most accurate, parsimonious diagnostic model of allograft fibrosis with a sensitivity of 93.8{\%} and a specificity of 84.1{\%} (P<0.0001). In the independent validation set, this same model predicted the presence of allograft fibrosis with a sensitivity of 77.3{\%} and a specificity of 87.5{\%} (P<0.0001). Conclusions: Measurement of mRNAs in urinary cells may offer a noninvasive means of diagnosing fibrosis in human renal allografts.",
keywords = "Fibrosis, Gene expression, Kidney transplantation",
author = "Dany Anglicheau and Thangamani Muthukumar and Aur{\'e}lie Hummel and Ruchuang Ding and Sharma, {Vijay K.} and Darshana Dadhania and Seshan, {Surya V.} and Schwartz, {Joseph E.} and Manikkam Suthanthiran",
year = "2012",
month = "6",
day = "15",
doi = "10.1097/TP.0b013e31824ef181",
language = "English",
volume = "93",
pages = "1136--1146",
journal = "Transplantation",
issn = "0041-1337",
publisher = "Lippincott Williams and Wilkins",
number = "11",

}

TY - JOUR

T1 - Discovery and validation of a molecular signature for the noninvasive diagnosis of human renal allograft fibrosis

AU - Anglicheau, Dany

AU - Muthukumar, Thangamani

AU - Hummel, Aurélie

AU - Ding, Ruchuang

AU - Sharma, Vijay K.

AU - Dadhania, Darshana

AU - Seshan, Surya V.

AU - Schwartz, Joseph E.

AU - Suthanthiran, Manikkam

PY - 2012/6/15

Y1 - 2012/6/15

N2 - Background: Tubulointerstitial fibrosis (fibrosis), a histologic feature associated with a failing kidney allograft, is diagnosed using the invasive allograft biopsy. A noninvasive diagnostic test for fibrosis may help improve allograft outcome. Methods: We obtained 114 urine specimens from 114 renal allograft recipients: 48 from 48 recipients with fibrosis in their biopsy results and 66 from 66 recipients with normal biopsy results. Levels of messenger RNAs (mRNAs) in urinary cells were measured using kinetic, quantitative polymerase chain reaction assays, and the levels were related to allograft diagnosis. A discovery set of 76 recipients (32 with allograft fibrosis and 44 with normal biopsy results) was used to develop a diagnostic signature, and an independent validation set of 38 recipients (16 with allograft fibrosis and 22 with normal biopsy results) was used to validate the signature. Results: In the discovery set, urinary cell levels of the following mRNAs were significantly associated with the presence of allograft fibrosis: vimentin (P<0.0001, logistic regression model), hepatocyte growth factor (P<0.0001), α-smooth muscle actin (P<0.0001), fibronectin 1 (P<0.0001), perforin (P=0.0002), plasminogen activator inhibitor 1 (P=0.0002), transforming growth factor β1 (P=0.0004), tissue inhibitor of metalloproteinase 1 (P=0.0009), granzyme B (P=0.0009), fibroblast-specific protein 1 (P=0.006), CD103 (P=0.02), and collagen 1A1 (P=0.04). A four-gene model composed of the levels of mRNA for vimentin, NKCC2, and E-cadherin and of 18S ribosomal RNA provided the most accurate, parsimonious diagnostic model of allograft fibrosis with a sensitivity of 93.8% and a specificity of 84.1% (P<0.0001). In the independent validation set, this same model predicted the presence of allograft fibrosis with a sensitivity of 77.3% and a specificity of 87.5% (P<0.0001). Conclusions: Measurement of mRNAs in urinary cells may offer a noninvasive means of diagnosing fibrosis in human renal allografts.

AB - Background: Tubulointerstitial fibrosis (fibrosis), a histologic feature associated with a failing kidney allograft, is diagnosed using the invasive allograft biopsy. A noninvasive diagnostic test for fibrosis may help improve allograft outcome. Methods: We obtained 114 urine specimens from 114 renal allograft recipients: 48 from 48 recipients with fibrosis in their biopsy results and 66 from 66 recipients with normal biopsy results. Levels of messenger RNAs (mRNAs) in urinary cells were measured using kinetic, quantitative polymerase chain reaction assays, and the levels were related to allograft diagnosis. A discovery set of 76 recipients (32 with allograft fibrosis and 44 with normal biopsy results) was used to develop a diagnostic signature, and an independent validation set of 38 recipients (16 with allograft fibrosis and 22 with normal biopsy results) was used to validate the signature. Results: In the discovery set, urinary cell levels of the following mRNAs were significantly associated with the presence of allograft fibrosis: vimentin (P<0.0001, logistic regression model), hepatocyte growth factor (P<0.0001), α-smooth muscle actin (P<0.0001), fibronectin 1 (P<0.0001), perforin (P=0.0002), plasminogen activator inhibitor 1 (P=0.0002), transforming growth factor β1 (P=0.0004), tissue inhibitor of metalloproteinase 1 (P=0.0009), granzyme B (P=0.0009), fibroblast-specific protein 1 (P=0.006), CD103 (P=0.02), and collagen 1A1 (P=0.04). A four-gene model composed of the levels of mRNA for vimentin, NKCC2, and E-cadherin and of 18S ribosomal RNA provided the most accurate, parsimonious diagnostic model of allograft fibrosis with a sensitivity of 93.8% and a specificity of 84.1% (P<0.0001). In the independent validation set, this same model predicted the presence of allograft fibrosis with a sensitivity of 77.3% and a specificity of 87.5% (P<0.0001). Conclusions: Measurement of mRNAs in urinary cells may offer a noninvasive means of diagnosing fibrosis in human renal allografts.

KW - Fibrosis

KW - Gene expression

KW - Kidney transplantation

UR - http://www.scopus.com/inward/record.url?scp=84862298374&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862298374&partnerID=8YFLogxK

U2 - 10.1097/TP.0b013e31824ef181

DO - 10.1097/TP.0b013e31824ef181

M3 - Article

C2 - 22592886

AN - SCOPUS:84862298374

VL - 93

SP - 1136

EP - 1146

JO - Transplantation

JF - Transplantation

SN - 0041-1337

IS - 11

ER -