1 What’s New in Transplant Innovative Approaches to Detect Graft Loss and Novel Treatments to Support Graft Acceptance.

1

What’s New in Transplant Innovative Approaches to Detect Graft Loss and Novel Treatments to Support Graft Acceptance

2

Overview

• Current methods of detecting graft loss in transplantation have advanced the field, but have limitations1

• Technologies that might address those limitations and create a better diagnostic gold standard are being investigated and developed2

• The pathway to get these new technologies into clinical practice has yet to be defined.2 This presentation is a preview of future technologies

• New ways to preserve and treat grafts are also being investigated3

1. Bromberg JS, et al. Am J Transplant. 2012;(12):2573-2574.2. Bromberg JS, et al. Am. J Transplant. 2009;9:11-13.3. Bon D, et al. Nat Rev Nephrol. 2012; 8(6):339-347.

3

Getting Laboratory Technologies Into Clinical Practice Is Neither a Linear Nor Smooth Path

1. Bromberg JS, et al. Am J Transplant. 2009;9:11-13.2. Huskey J, et al. Clin J Am Soc Nephrol. 2011;6:423-429.

• Developing methods to detect graft loss is challenging when the mechanisms of graft loss are not fully known1

• New methods have the double duty of1

– Offering more precise detection

– Furthering the understanding of biological mechanisms of graft loss

• To better gauge new approaches, the strengths and weaknesses of the laboratory science used to create the approach must be identified2

Example: ImmuKnowTM Assay2

• Measure of intracellular adenosine triphosphate (ATP) released from CD4+ T cells

• Quantify cell-mediated immunity in transplant recipients

• Identify patients at risk for acute rejection (AR) or opportunistic infections

4

New Techniques Are Exciting, But Discovery Must Be Counterbalanced With Trial Limitations

• If these factors are taken into account, landmark advances may be realized– Adequate sample size

– Prospective vs retrospective analysis

– Detection of biomarkers indicating concurrent diagnosis vs prediction of an outcome

– Validation of each new biomarker

– Correlating genes to biopsies may have limitations, when biopsies themselves are imperfect gold standards

Bromberg JS, et al. Am J Transplant. 2012;(12):2573-2574.

5

Why Go Molecular?

6

Why Use Genomics?

1. Bromberg JS, et al. Am J Transplant. 2012;(12):2573-2574.2. Li L, et al. Am J Transplant. 2012;12(10):2710-2718.3. Anglicheau D, et al. Transplantation. 2012;93:1136-1146.4. Flechner SM, et al. Am J Transplant. 2004;4(9):1475-1489.

• Current markers for rejection—serum creatinine and biopsies—are imperfect as the gold standards1

– Biopsies are also subject to reader bias and sampling error

– They are invasive procedures

• New assays that look at biomarkers – Are less invasive than biopsies2,3

– May yield more precise measurements then currently used methods by virtue of being objective2,4

7

A Brief Look at How Genomic Biomarkers Are Discovered: Polymerase Chain Reaction

1. Muthukumar T, et al. N Engl J Med. 2005;353:2342-2351.2. Anglicheau D, et al. Transplantation. 2012;93:1136-1146.3. Ben-Dov IZ, et al. Transplantation. 2012;94:1086-1094.

• Polymerase chain reaction (PCR) – Method of RNA/DNA amplification used to detect or quantify

whether gene expression in a sample. It can be used as a validation strategy for other detection methods1-3

• eg, if gene expression is discovered using microarray technology, it can then be independently validated using PCR3

8

A Brief Look at How Genomic Biomarkers Are Discovered: MicroRNA

Anglicheau D, et al. Transplantation. 2010 ;90(2):105-112.

• MicroRNA (miRNA)– Thought to regulate gene expression at the posttranscriptional

level

– Plays a central role in cell proliferation, differentiation, apoptosis, fat metabolism, and oncogenesis

– Several cancers are associated with dysregulation of miRNA expression

– Messenger (mRNA) can be targeted by multiple miRNAs

9

DNA Microarray Technology: How the Genes That Might Be Rejection Culprits Are Identified

Principle of cDNA microarray

assay for gene expression

• Make cDNA (complementary DNA) reverse transcript• Label cDNAs with fluorescent dyes

Control Experimental

Hybridizationto microarray

Laser excitationat dye-specific Hz

Laser emission Computer calculatesratio of intensity

Red = up-regulationGreen = down-regulationBlack = constitutive expression

+ =

Reproduced with permission: © 2012 by Steven M. Carr, after Gibson & Muse 2002.

10

Gene Expression

Histology

Strong correlations yield effective gene profiling

From PBLswith analysisusing DNAmicroarrays

Frompathologicbiopsies

Flechner SM, et al. Am J Transplant. 2004;4(9):1475-1489.

Mapping DNA to Tissues to Yield Gene Profiles

PBL, peripheral blood lymphocyte.

11

Current Molecular Diagnostic PursuitsOverview

12

Molecular Diagnostics in Current Research

• The following examples look at methods for detecting biomarkers in urine and peripheral blood (PB) in AR

• The next set of examples includes noninvasive detection for fibrosis and interstitial fibrosis tubular atrophy (IFTA), including 1study that yielded unexpected results

• Finally, the last section focuses on whether a molecular scorecard can be created to predict graft loss

• If these new detection methods pass the test of utility and accuracy in the clinical mainstream, they could offer noninvasive and precise detection methods

13

Current Molecular Diagnostic PursuitsAcute Rejection

14

mRNA in Urinary Cells Is a Biomarker for AR

Lowest third, Middle third, Highest third,0

10

20

30

40

50

60

Log FOXP3 mRNA

Gra

ft L

oss

Wit

hin

6 M

on

ths

Aft

erA

cute

-Rej

ecti

on

Ep

iso

de,

%

P=0.02RR=6.0

RR=3.0

RR=1.0

<3.17 3.17-4.84 >4.84

Muthukumar T, et al. N Engl J Med. 2005;353:2342-2351.

• FOXP3 mRNA is a marker for regulatory T (Treg) cells and AR– High levels of FOXP3 are associated with a lower risk for graft

failure after acute rejection

RR, relative risk.

15

A 5-Gene Signature Suggested to Identify AR in Pediatric Renal Allograft

Li L, et al. Am J Transplant. 2012;12(10):2710-2718.

• Aim: To provide a PB diagnostic test for AR in pediatric patients

• Methods: 367 unique PB samples were paired with a graft biopsy for blinded phenotype classification. A logistic regression model was designed for analysis

• Findings: A 5-gene set classified– AR from stable (absence of AR), with 91% sensitivity and 94%

specificity

– AR from all other non-AR phenotypes, with 91% sensitivity and 90% specificity

• 5-gene set was shown to diagnose AR– DUSP1, PBEF1, PSEN1, MAPK9, and NKTR

16

A Measured Response to PBL Diagnostics and Biomarker Studies: A Bit of Fair Balance

• PBL signature for AR is important in making microarray a clinical reality, but it has important limitations– Majority of biopsies from white children/adolescents

– Correlating genes to biopsies—the latter of which are a subpar gold-standard measurement—might not be predictive

– The study designs and statistical methods suitable to make a prediction are much more complex than those to make a diagnosis of concurrent rejection

• The study by Li et al is a landmark study because of its further ability to diagnose AR using PB; however, challenges remain in validating the signature

Bromberg JS, et al. Am J Transplant. 2012;(12):2573-2574.

17

Current Molecular Diagnostic PursuitsFibrosis and Interstitial Fibrosis and Tubular Atrophy (IFTA)

18

mRNA Offers a First Step Toward a Noninvasive Test for Fibrosis

Urine samples from 114 patients with and withoutfibrosis assigned to discovery and validation sets

Discovery set of 76 urine samples

Fibrosis biopsy group32 urine samples

Preamplificationenhanced real-timequantitative PCRof 22 mRNAs and

18 rRNA

Evaluation of bivariate association of individualmRNA measures to likelihood of fibrosis

Construction of diagnosticequation by logistic regression

4-gene model

Normal biopsy group44 urine samples

Discovery Set

Stage 1:Discovery Phase

Validation set of 38 urine samples

Fibrosis biopsy group16 urine samples

Preamplificationenhanced real-timequantitative PCR

Normal biopsy group22 urine samples

Validation Set

Stage 2:IndependentValidation Phase

Diagnostic accuracy of the4-gene model

Reproduced with permission from Anglicheau D, et al. Transplantation. 2012;93:1136-1146.

Anglicheau D, et al. Transplantation. 2012;93:1136-1146.

19

Using Gene Expression to Better Understand IFTA

• Building on previous research that identified mRNA expression signatures in fibrosis and IFTA, investigators analyzed miRNA (which regulates mRNA) to– Understand the mechanics of pathophysiology of fibrogenesis

– Determine whether IFTA is associated with altered expression of miRNA and if miRNA can be used diagnostically

• The miRNA transcriptome of allografts with or without fibrosis were sequenced, catalogued, and validated

• IFTA miRNA patterns were distinct from that of healthy biopsy specimens

– Independent of clinical patient characteristics

– Predicted allograft outcome

Ben-Dov IZ, et al. Transplantation. 2012;94:1086-1094.

20

Discovery of IFTA Biomarkers Also Yielded Unpredicted Results

• A gene expression‒based signature in PBL discovered – Predictive accuracy of 80% for mild IFTA and 92% for moderate

to severe IFTA*

• Unpredicted for outcome

Banff 1 (mild IFTA)

Banff 2,3 (moderate to severe IFTA)

• Associated with number of genes representing different pathways connected to immune/ inflammatory and tissue injury mechanisms

• Significantly fewer immune/inflammatory genes expressed than in mild IFTA

• More genes linked to metabolic and other pathways

* Chronic allograft nephropathy (CAN) is now called IFTA and has been changed here to IFTA to match the current lexicon

Kurian SM, et al. PLoS One. 2009;4(7):e6212.

21

Current Molecular Diagnostic PursuitsMolecular Modeling to Predict Graft Loss

22

Creating a Molecular Risk Score to Predict Late Graft Loss Using Microarray

• Aim: Identify molecules that predict organ failure after signs of renal dysfunction or proteinuria

• Findings: Genes associated with graft failure were related to tissue injury, epithelial dedifferentiation, matrix remodeling, and TGF-β effects

• Grafts that failed had– Higher incidence of proteinuria and rapid deterioration in function

before biopsy and a lower glomerular filtration rate at time of biopsy

• Grafts that failed had no differences in– Primary disease, time after transplantation, maintenance

immunosuppression, or incidence of anti-HLA antibodies

Einecke G, et al. J Clin Invest. 2010;120(6):1862-1872.

HLA, human leukocyte antigen.

23

• Molecular risk scores in individual biopsies; each triangle is a biopsy

Creating a Molecular Risk Score to Predict Late Graft Loss (cont)

• Kaplan-Meier plots for the 2 risk groups

Einecke G, et al. J Clin Invest. 2010;120(6):1862-1872.

0 20 40 60 80 100 120-3

-2

-1

0

1

2

3

4

High Risk

Biopsies Orders by Risk Score

CensoredGraft failure

Low Risk

0 500 100 1500 20000.0

0.2

0.4

0.6

0.8

1.0

Su

rviv

al P

rob

abil

ity

Time After Biopsy, days

Ris

k S

core

Risk groups Low (n=53, Failed = 5) High (n=52, Failed = 25)

P = 3 x 10-7

Reproduced with permission from Einecke G, et al. J Clin Invest. 2010;120(6):1862-1872.

24

Current Molecular Diagnostic PursuitsLooking at the Antibody Targets: Endothelial Cells

25

• Given that endothelial cells (ECs) are antibody targets, altered expression of EC genes—identified by their transcripts—may identify ABMR

• Expression of these transcripts, if antibodies were also present, predicted graft loss with higher sensitivity (77% vs 31%) and lower specificity (71% vs 94%) than did the presence of C4d

• Clinical affect – Reveals that C4d-negative stain does not rule out

ABMR

Sis B, et al. Curr Opin Organ Transplant. 2010;15(1):42-48.

Analyzing Altered Gene Expression in Endothelial Cells Could Identify Antibody-Mediated Rejection (ABMR)

26

2 Types of ABMR: C4d Positive and C4d Negative With a Common Target

• Complement activation• Ifn-g effects• Endothelial activation• Leukocyte recruitment• Fc receptors• Platelets

• Ifn-g effects• Endothelial activation• Leukocyte recruitment• Fc receptors• Platelets

C4d-Positive ABMR C4d-Negative ABMR

Complement activationC4d deposition

No detectable C4d depositionby immunofluorescence

Platelets

CTL

NK

MØ

NK

MØ

CTL

MØ

NKCytokines

Cytokines

C1q

Platelets

Microcirculation is the main target of ABMR

Endothelial activation

Increased gene expression

Sis B, et al. Curr Opin Organ Transplant. 2010;15(1):42-48.

IFN, interferon. CTL, cytotoxic T lymphocytes.NK, natural killer.

Reproduced with permission from Sis B, et al. Curr Opin Organ Transplant. 2010;15(1):42-48.

27

Innovative Treatment Protocols

28

0

10

20

30

40

50

60

70

80

90

100

0 12 24 36 48 60 72 84 96

HLA Desensitization Provides a Survival Advantage to Patients

• Compared rates of death between one group undergoing desensitization treatment and 2 control groups. Control groups included

– Waiting list + dialysis (dialysis only)– Patients who underwent either dialysis or HLA-compatible transplantation (dialysis or

transplantation group)

No. at Risk

Desensitization 210 170 143 11 75 58 42 28 14treatment

Dual therapy 1027 854 688 497 321 230 157 96 41

Dialysis only 1012 822 626 419 250 159 93 54 17

Su

rviv

al,

%

Months

Desensitization treatment

Dialysis only

Dialysis ortransplantation

Montgomery RA, et al. N Engl J Med. 2011;365(4):318-326.

Reproduced with permission from Montgomery RA, et al. N Engl J Med. 2011;365(4):318-326.

29

Innovative Treatments for Graft Preservation or Recipients May Improve Kidney Recovery

• Ischemic preconditioning• Regional normothermic circulation• Resuscitation

Donor

Kidney evaluation(metabolomic NMR)

Graftpreservation

Recipient

• Preservation solution content(polyethylene glycols)

• Perfusion machine(oxygenation, temperature)

• Pharmacologic supplementation(inflammation, coagulation,energy metabolism, oxygen)

• Gene expression modulation (siRNA)

• Cell therapy(adipocytes, amniocytes, stem cells)

• Postconditioning (erythropoietin, adenosine)

Diagnostic Treatments

Bon D, et al. Nat. Rev Nephrol. 2012;8(6):339-347.

Reproduced with permission from Bon D, et al. Nat. Rev Nephrol. 2012;8(6):339-347.

siRNA, small interfering RNA.NMR, Nuclear magnetic resonance

30

Banff Guidelines

Molecular Understanding Is Creating Change

31

Molecular Data Have Compelled Banff to Refine Current Guidelines

• Recognizing the limitations of the previously empirically based histologic definitions and scoring thresholds, Banff is working to refine its guidelines

Isolated v-lesionworking group

Fibrosis scoringworking group

Glomerular lesionworking group

Molecular pathologyworking group

Polyomavirus nephropathy

working group

Quality assuranceworking group

data-driven and validated

refinement ofBanff criteria

Sis B, et al. Am J Transplant. 2010;10(3):464-471.

Permission Pending from Sis B, et al. Am J Transplant. 2010;10(3):464-471.

32

In Addition to Using Working Groups, Banff Looks to the Future

• Compelling molecular research data led to discussion of updating Banff to utilize – “omics” technologies

– New tissue markers

• Combine histopathology and molecular parameters within the Banff working classification

Sis B, et al. Am J Transplant. 2010;10(3):464-471.

33

Summary

34

Summary

• Short-term transplant outcomes have seen remarkable improvements in the recent decade, and this seems to be a result of advances in the field, such as more targeted therapies, diagnostics, and clinical procedures

• However, long-term outcomes remain a concern and, as a result, different methods and techniques are being addressed to further advance the field

• The power of genomics, coupled with new ways of looking at treatment, may advance the field so long-term outcomes are as improved as short-term outcomes were during the past decade

35

© 2012 sanofi-aventis U.S. LLC US.THY.12.12.003