The Molecular Phenotype of Heart Allograft Biopsies

Mario C Deng

Associate Professor of Medicine

Director of Cardiac Transplantation Research

Center for Advanced Cardiac Care

Department of Medicine

Columbia University

USA

modern medicine & immortality

New York Times Magazine Jan 30, 2000

heart transplant milestones

Electrical VAD (Portner)

Human HLtx (Reitz)

Fk506MMFAzathioprin

SirolimusNeoral

ATGAMSteroids

OKT3

1965 1975 1985 1995 2005

Animal Htx (Shumway) First human Htx (Barnard)

Endomyocardial Biopsy (Caves) Copeland re-Htx (Copeland)

Baby Htx (Bailey)

CsA

H. Genome (Lander)

Daclizumab

(1959)

(1967)

(1973)

(1974)

(1981)

(1984)

(2000)

Chemogenomics

Allomap (Deng)(2005)

2009

heart transplant survival by era

0

20

40

60

80

100

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Years

1982-1991 (N=18,854)

1992-2001 (N=35,146)

2002-6/2006 (N=12,369)

All comparisons significant at p < 0.0001

HALF-LIFE 1982-1991: 8.8 years; 1992-2001: 10.5 years; 2002-6/2006: NA

Su

rviv

al (

%)

ISHLT Taylor D et al. J Heart Lung Transplant 2008;27: 937-983

heart transplant cause of death

CAUSE OF DEATH

0-30 Days

(N = 3,006)31 Days –

1 Year

(N = 2,722)

>1 Year –

3 Years

(N = 2,135)

>3 Years –

5 Years

(N = 1,857)

>5 Years –

10 Years

(N = 4,054)

>10 Years

(N = 2,107)

CARDIAC ALLOGRAFT VASCULOPATHY 52 (1.7%) 127 (4.7%) 298 (14.0%) 299 (16.1%) 581 (14.3%) 309 (14.7%)

ACUTE REJECTION 193 (6.4%) 338 (12.4%) 220 (10.3%) 82 (4.4%) 69 (1.7%) 26 (1.2%)

LYMPHOMA 2 (0.1%) 54 (2.0%) 85 (4.0%) 96 (5.2%) 195 (4.8%) 73 (3.5%)

MALIGNANCY, OTHER 1 (0.0%) 57 (2.1%) 218 (10.2%) 340 (18.3%) 749 (18.5%) 392 (18.6%)

CMV 4 (0.1%) 34 (1.2%) 16 (0.7%) 3 (0.2%) 5 (0.1%) 1 (0.0%)

INFECTION, NON-CMV 393 (13.1%) 896 (32.9%) 276 (12.9%) 180 (9.7%) 442 (10.9%) 213 (10.1%)

GRAFT FAILURE 1,257 (41.8%) 500 (18.4%) 499 (23.4%) 379 (20.4%) 765 (18.9%) 353 (16.8%)

TECHNICAL 233 (7.8%) 28 (1.0%) 17 (0.8%) 17 (0.9%) 36 (0.9%) 20 (0.9%)

OTHER 162 (5.4%) 175 (6.4%) 187 (8.8%) 147 (7.9%) 339 (8.4%) 175 (8.3%)

MULTIPLE ORGAN FAILURE 356 (11.8%) 268 (9.8%) 117 (5.5%) 102 (5.5%) 309 (7.6%) 190 (9.0%)

RENAL FAILURE 20 (0.7%) 25 (0.9%) 36 (1.7%) 65 (3.5%) 225 (5.6%) 173 (8.2%)

PULMONARY 133 (4.4%) 108 (4.0%) 96 (4.5%) 85 (4.6%) 172 (4.2%) 99 (4.7%)

CEREBROVASCULAR 200 (6.7%) 112 (4.1%) 70 (3.3%) 62 (3.3%) 167 (4.1%) 83 (3.9%)

ISHLT Taylor D et al. J Heart Lung Transplant 2008;27: 937-983

CD8+

Allo Agallo MHC

allo APC

Allo Agself MHC

self APC

TCR

rejection

tolerance

T-cell activating signals#1 T-cell receptor#2 CD28 #2a CD40L#3 IL2, IL15 etc

Bone marrow

Recipientimmune response

Cardiac allograft

B-cellspleenlymphnode

thymus

CD8+ Cytotox

CD8+ supprCD8+28- suppr

monocyte CD3+ T-cell

CD4+

CD4+Th1

CD4+Th2

CD4+/25+

CD4+/45RO+

allograft rejection

systems biology strategy

RejectionRejection QuiescenceQuiescence

genome

transcriptome

proteome

metabolome

phenome

CD8+

Allo Agallo MHC

allo APC

Allo Agself MHC

self APC

TCR

rejection

tolerance

T-cell activating signals#1 T-cell receptor#2 CD28 #2a CD40L#3 IL2, IL15 etc

Recipientimmune response

Cardiac allograft

B-cell

CD8+ Cytotox

CD8+ supprCD8+28- suppr

monocyte CD3+ T-cell

CD4+

CD4+Th1

CD4+Th

2

CD4+/25+

CD4+/45RO+

allograft rejection

RejectionRejection QuiescenceQuiescence

genome

transcriptome

proteome

metabolome

phenome

organ systemic

phenome echo clinical

proteome histo, Cd4 biomarkers

transcriptome RT-PCR, ISH, array Allomap

genome DNA-sequencing DNA-sequencing

endomyocardial biopsy

mild

severe

invasive & complication-prone late-stage cellular rejection diagnosis insensitive for humoral rejection significant variability no insight into molecular mechanisms resource-intense

status quo monitoring

rejection grading

0 1

2 3

4

no rejection A, focal infiltrate without necrosis B, diffuse sparse infiltrate w/o necrosis one focus aggressive infiltration/focal myocyte damage A, multifocal aggr infiltr or myoc damage B, diffuse inflamm process with necrosis diffuse, necrosis, edema, hemorrhage

Billingham ME et al. J Heart Lung Transplant 1990;9:587Stewart S et al. J Heart Lung Transplant 2005;24:1710

0 R 1 R

2 R

3 R

AMR

Quilty

current standard of literature

Nielsen H., F.B. Sorensen and B. Nielsen et al., Reproducibility of the acute rejection diagnosis in human cardiac allografts. The Stanford Classification and the International Grading System, J Heart Lung Transplant 12 (1993), pp. 239–243

Fishbein M.C., G. Bell and M.A. Lones et al., Grade 2 cellular heart rejection does it exist?, J Heart Lung Transplant 13 (1994), pp. 1051–1057

Winters G.L., E. Loh and F.J. Schoen, Natural history of focal moderate cardiac allograft rejection. Is treatment warranted?, Circulation 91 (1995), pp. 1975–1980

Milano A., A.L. Caforio and U. Livi et al., Evolution of focal moderate rejection of the cardiac allograft, J Heart Lung Transplant 15 (1996), pp. 456–460

Brunner-La Rocca H.P., G. Sutsch, J. Schneider, F. Follath and W. Kiowski, Natural course of moderate cardiac allograft rejection early and late after transplantation, Circulation 94 (1996), pp. 1334–1338

Mills R.N., D.C. Naftel and J.K. Kirklin et al., Heart transplant rejection with hemodynamic compromise a multiinstitutional study of the role of endomyocardial cellular infiltrate. Cardiac Transplant Research Database, J Heart Lung Transplant 16 (1997), pp. 813–821

Winters G.L., B.M. McManus and Rapamycin Cardiac Rejection Treatment Trial Pathologists, Consistencies and controversies in the application of the ISHLT working formulation for cardiac transplant biopsy specimens, J Heart Lung Transplant 17 (1998), p. 754

Rodriguez E.R. and International Society for Heart and Lung Transplantation, The pathology of heart transplant biopsy specimens revisiting the 1990 ISHLT working formulation, J Heart Lung Transplant 22 (2003), pp. 3–15

Marboe CC, Billingham M, ... Berry G. JHLT 2005;24:S219

CD8+

Allo Agallo MHC

allo APC

Allo Agself MHC

self APC

TCR

rejection

tolerance

T-cell activating signals#1 T-cell receptor#2 CD28 #2a CD40L#3 IL2, IL15 etc

Recipientimmune response

Cardiac allograft

B-cell

CD8+ Cytotox

CD8+ supprCD8+28- suppr

monocyte CD3+ T-cell

CD4+

CD4+Th1

CD4+Th

2

CD4+/25+

CD4+/45RO+

allograft rejection

RejectionRejection QuiescenceQuiescence

genome

transcriptome

proteome

metabolome

phenome

organ systemic

phenome echo clinical

proteome histo, Cd4 biomarkers

transcriptome RT-PCR, ISH, array Allomap

genome DNA-sequencing DNA-sequencing

intragraft cytokine expression

Alvarez CM et al. Clin Transplant 2001;15:228

intragraft cytokine expression

Alvarez CM et al. Clin Transplant 2001;15:228

Author Year No Time Method Result

Zhao 1994 21 early RT-PCR IL6, TGFß+

Van Hoffen 1996 40 first mo‘s ISH/IHC IL6,8,9,10+

Baan 1996 16 < 12 mo RT-PCR IL4,IL6+

Van Besouw 1997 85 > 1 y GIL-ELISA IL6+ TxV

Kimball 1996 62 < 30d ELISA IL6,8+

correlation of IL6 with rejection

Author Year No Time Method Result

Salom 1998 22 early IHC IL6-, IL1, TNF+

Fyfe 1993 40 var ELISA IL4, IL6,TNF-

George 1997 484 < 8 wk ELISA IL6,IL8,TNF-

Ruan 1992 113 ? IHC IL6-,IL2+,IFN+

Van Besouw 1995 49 < 90d culture IL6-,IL4-,IL2,IFN+

Lagoo 1996 328 < 8 wk RT-PCR IL6-

Grant 1996 187 <2y RT-PC/ELISA IL6-,IL2+

Grant 1996 259 <2y ELISA IL6-

no correlation of IL6 with rejection

Deng 1998 115 <3mo ELISA IL6-

systems biology strategy

00

Phenotype 1Phenotype 1

proteome

transcriptome

metabolome

phenome/physiome

genome

Phenotype 2Phenotype 2

bot

tom

-to-

top

top-to-b

ottom

...We used mouse transplants to annotate pathogenesis-based transcript sets (PBTs) that reflect major biologic events in allograft rejection—cytotoxic T-cell infiltration, interferon-γ effects and parenchymal deterioration. We examined the relationship between PBT expression, histopathologic lesions and clinical diagnoses in 143 consecutive human kidney transplant biopsies for cause. PBTs correlated strongly with one another, indicating that transcriptome disturbances in renal transplants have a stereotyped internal structure. This disturbance was continuous, not dichotomous, across rejection and nonrejection. PBTs correlated with histopathologic lesions and were the highest in biopsies with clinically apparent rejection episodes. Surprisingly, antibody-mediated rejection had changes similar to T-cell mediated rejection. Biopsies lacking PBT disturbances did not have rejection. PBTs suggested that some current Banff histopathology criteria are unreliable, particularly at the cut-off between borderline and rejection...many transcriptome changes previously described in rejection are features of a large-scale disturbance characteristic of rejection but occurring at lower levels in many forms of injury. PBTs represent a quantitative measure of the inflammatory disturbances in organ transplants, and a new window on the mechanisms of these changes.

Mueller TF et al. Am J Transplant 2007;7:1

pathogenesis-based transcript sets

endomycardial array & rejection

Karason K et al. BMC Cardiov Dis 2006;6:29

endomycardial array & rejection

Karason K et al. BMC Cardiov Dis 2006;6:29

endomycardial array & rejection

Karason K et al. BMC Cardiov Dis 2006;6:29

endomycardial array & rejection

Karason K et al. BMC Cardiov Dis 2006;6:29

endomycardial array & rejection

Karason K et al. BMC Cardiov Dis 2006;6:29

…Methods: Endomyocardial tissue samples and serum were obtained in connection with clinical biopsies ... Endomyocardial RNA,..were analysed with DNA microarray. Genes showing up-regulation during rejection followed by normalization after the rejection episode were evaluated further with real-time RT-PCR…ELISA was performed to investigate whether change in gene-regulation during graft rejection was reflected in altered concentrations of the encoded protein in serum…Results…CCL9 was significantly upregulated during rejection (p < 0.05)…There were no changes in CXCL9 and CXCL10 serum concentrations during cardiac rejection…Conclusion: We conclude, that despite a distinct up-regulation of CXCL9 mRNA in human hearts during cardiac allograft rejection, this was not reflected in the serum levels of the encoded protein. Thus, in contrast to previous suggestions, serum CXCL9 does not appear to be a promising serum biomarker for cardiac allograft rejection. The lack of success in the identification of cardiac rejection biomarkers in the current study indicates that expression profiling of immunological active cells of the heart recipient may be a better way to identify cardiac rejection biomarkers.

intragraft & PBL expression

Flechner SM et al. Am J Transplant 2004;4:1475

…Our results demonstrate that PBL gene expression profiles in acute rejection are distinctly different from those of normal controls and from patients with well-functioning transplants. Therefore, acute rejection does influence the gene expression profile of the circulating lymphocyte pool. Moreover, despite the fact that surprisingly we found very little common gene expression between PBLs and kidney biopsies, we did identify a large number of lymphocyte-specific genes in the kidney tissue. One interpretation is that there are compartment-specific differences between the PBLs in the circulation and the subset of lymphocytes that are activated and recruited to the transplant kidney during acute rejection. ..these results…may explain the failure of more than a decade of work testing PBLs for an array of activation antigens based on findings in rejecting allografts and other immune models…It is possible that the gene expression profile of the PBLs represents the adequacy of immunosuppression such that the rejecting patients reflect the profile of inadequate immunosuppression as compared with the PBLs sampled from patients with well-functioning transplants... there is a distinct gene expression profile in the PBL pool that correlates with acute rejection and immunosuppression. If these results can be confirmed in a large, prospective trial it would support the use of such profiles as a minimally invasive monitoring strategy for the immunological status of the graft and support the potential of using them to monitor the adequacy of immunosuppression…

current standard of care

…the variability in the grading of heart transplant biopsies suggests the biopsy itself may not be a true gold standard against which all subsequent tests should be compared; this has clear implications for the evaluation of any new molecular diagnostic test if the only end-point is comparison with biopsy grade. Multifactorial end-points combining clinical, hemodynamic and biopsy data would provide a better standard. Indeed, the correlation of these multiple factors and peripheral blood gene expression with biopsy histology may provide a basis for further refining of the biopsy grading system by providing insight into the histologic features that best correlate with immunologic status and clinical outcomes.

Marboe CC, Billingham M, ... Berry G. JHLT 2005;24:S219

networked alloimmunity

Orosz CG. J Heart Lung Transplant 1996;15:1063Baan et al. Transplant Int 1998;11:160Bumgardner et al. Sem Liver Dis 1999;19:189

detection of cytokine transcripts does not imply protein detection of cytokine protein does not imply function function may vary in different contexts composite effects of multiple cytokines are rarely tested unknown cytokines may be involved in rejection cytokine polymorphisms may explain variations in-vitro effects may not reflect in-vivo effects animal data may not translate into in clinical data > nonreductionist research approach necessary

CD8+

Allo Agallo MHC

allo APC

Allo Agself MHC

self APC

TCR

rejection

tolerance

T-cell activating signals#1 T-cell receptor#2 CD28 #2a CD40L#3 IL2, IL15 etc

Recipientimmune response

Cardiac allograft

B-cell

CD8+ Cytotox

CD8+ supprCD8+28- suppr

monocyte CD3+ T-cell

CD4+

CD4+Th1

CD4+Th

2

CD4+/25+

CD4+/45RO+

allograft rejection

RejectionRejection QuiescenceQuiescence

genome

transcriptome

proteome

metabolome

phenome

organ systemic

phenome echo clinical

proteome histo, Cd4 biomarkers

transcriptome RT-PCR, ISH, array Allomap

genome DNA-sequencing DNA-sequencing

Grant AJ et al. Lancet 2002;359:150

differential lymphocyte homing

endomyocardial biopsy

mild

severe

invasive & complication-prone late-stage cellular rejection diagnosis insensitive for humoral rejection significant variability no insight into molecular mechanisms resource-intense

highly sensitive for rejection strong negative predictive value positive test >need for further workup non-invasive easily repeatable on outpatient basis low complication rate decreased costs

status quo monitoring future monitoring

systemic IL6 & HTx

1 2 3 4 5 6 7 8Bx time

050

100150200

pg/ml

MOF

stable

15 pts, < 3 mo posttx, at EMB time IL6 ELISA, RHC, echo IL2 & rej 2+ IL6 & prognosis

0 5 10 15 20 25 30RAP

050

100150200250300

IL6

Deng et al. Transplantation 1995;60:1118

malaise

graft dysfunction

under

cellul rej humor rej infectionSIRS

overimmunosuppression

bolus steroids cyclophosph reduced IS antibiotics

HTx management

Deng et al. Transplantation 1998;65:1255

CARGO clinical study summary Overview

Cardiac Allograft Rejection Gene expression Observational study = “CARGO” 8 center, 4-year observational study initiated in 2001 (22% of US HTx). 629 patients, 4917 post-transplant encounters

Hypothesis Gene expression profiling of peripheral blood mononuclear cells can discriminate ISHLT grade 0 rejection (quiescence) from moderate/severe (ISHLT grade ≥ 3A) rejection

Design & ResultProspective, blinded validation study of 20 gene algorithm demonstrated ability to distinguish Grade 3A rejection from quiescence

Deng/Eisen/Mehra et al. Am J Transplant 2006;6:150

Algorithm development Real-time PCR

20-gene algorithm to distinguish rejection from quiescence (AlloMap molecular testing)

Candidate gene selection 285 Leukocyte microarray

Database / literature mining

252 candidate genes

Validation Prospective, blinded, statistically-powered (n = 270)

Additional samples tested to further define performance (n > 1000)

DevelopmentDevelopment~1 year~1 year(PCR)(PCR)

ClinicalClinical ValidationValidation

~1 year~1 year(Molecular Test)(Molecular Test)

DiscoveryDiscovery~2 years~2 years

(microarray)(microarray)

II

IIII

IIIIII

Study Design Prospective Multi-center Non-blinded Randomized Non-inferiority

Patients 2-5 years post-Tx ≥ 18 years old Stable outpatients

Invasive Monitoring Attenuation through Gene Expression IMAGEQuestionHow do theGEP-based study restrictions affect clinical implementation?

Pham/Deng/Kfoury et al. J Heart Lung Transplant 2007;26:808ClinicalTrials.gov identifier NCT00351559

• HypothesisTo determine whether the monitoring of acute

rejectionusing GEP is not inferior compared to the use of the EMB with respect to the event-free survival

Decrease in LV function, defined as LVEF change ≥ 25% compared with the baseline, or enrollment value, as measured by echocardiography

Development of clinically overt rejection (heart failure, hemodynamic compromise)

Death from any cause

Invasive Monitoring Attenuation through Gene Expression IMAGE

EnrollmentVisit &

Randomization

Gene Profiling Arm

Year 2 - 3 Year 4 - 5

Clinic x x x x x

Echo x x x x x

GEP/EMB x x x x x

Study End

Biopsy Arm

~2 year follow-up

CARGOII Study

2000 2005 2010

CARGO study start

CARGO study completion/CLIA approval

Allomap Medicare reimbursement

FDA approval IVDMIA

IMAGE Study

(2001)

(2005)

(2006)

(2008)

(2009)

(2010)

Allomap implementation milestones

regulatory transitions – CLIA>FDA

CLIA approval 2005 FDA approval 2008 – safety & efficacy Center for Devices & Radiological Health In-vitro diagnostic multivariate index assay IVDMIA FDA-director Daniel Schultz comment:

“…Allomap is an example of how advancements in science and technology are leading to new medical care diagnostics…“

Phase

Figure 2: Strategies in gene expression based biomarker test development.

Tasks

Phase 1

Phase 2

Phase 3

Phase 4

Phase 5

Phase 6

Phase 7

• Imperfect Clinical/Phenotype Standards

• Dichotomous vs. Continuous Phenotype Choices

Challenges

• Multicenter Study

• General Gene Discovery Strategy

• Focused vs. Whole Genome Microarray

• Real time-PCR Validations

• Discriminatory vs. Classifier Genes

• Mathematical Modeling

• Biological Plausibility of the Diagnostic Test Gene List

• Clinical Test Replicability

• Independence of Primary Clinical Validation Cohort

• Prevalence Estimation of Clinical Phenotype of Interest

• Regulatory Approval

• Payer Reimbursement

• Clinical Acceptance

CLINICAL PHENOTYPE CONSENSUSDEFINITION

GENE DISCOVERY

INTERNAL DIFFERENTIAL GENE LIST VALIDATION

DIAGNOSTIC CLASSIFIER DEVELOPMENT

EXTERNAL CLINICAL VALIDATION

CLINICAL IMPLEMENTATION

POST-CLINICAL IMPLEMENTATION STUDIES

Phase

Figure 2: Strategies in gene expression based biomarker test development.

Tasks

Phase 1

Phase 2

Phase 3

Phase 4

Phase 5

Phase 6

Phase 7

• Imperfect Clinical/Phenotype Standards

• Dichotomous vs. Continuous Phenotype Choices

Challenges

• Multicenter Study

• General Gene Discovery Strategy

• Focused vs. Whole Genome Microarray

• Real time-PCR Validations

• Discriminatory vs. Classifier Genes

• Mathematical Modeling

• Biological Plausibility of the Diagnostic Test Gene List

• Clinical Test Replicability

• Independence of Primary Clinical Validation Cohort

• Prevalence Estimation of Clinical Phenotype of Interest

• Regulatory Approval

• Payer Reimbursement

• Clinical Acceptance

CLINICAL PHENOTYPE CONSENSUSDEFINITION

GENE DISCOVERY

INTERNAL DIFFERENTIAL GENE LIST VALIDATION

DIAGNOSTIC CLASSIFIER DEVELOPMENT

EXTERNAL CLINICAL VALIDATION

CLINICAL IMPLEMENTATION

POST-CLINICAL IMPLEMENTATION STUDIES

strategies in GEP test development

Shahzad ,Sinha , Latif, Deng. Standardized operational procedures in clinical gene expression biomarker panel development. In: Scherer A (Ed).John Wiley & Sons 2009

Clarke R et al. Nat Rev Cancer 2008;8:37

dimensionality problem overviewThe application of several high-throughput genomic and proteomic technologies generate high-dimensional data sets•The multimodality of high-dimensional expression data can confound both simple mechanistic interpretations of biology and the generation of complete or accurate gene signal transduction pathways or networks.•The mathematical and statistical properties of high-dimensional data spaces are often poorly understood or inadequately considered, particularly if the number of data points obtained for each specimen greatly exceed the number of specimens.•Data are rarely randomly distributed in high-dimensions and are highly correlated, often with spurious correlations.•The distances between a data point and its nearest and farthest neighbours can become equidistant in high dimensions, potentially compromising the accuracy of some distance-based analysis tools.•Owing to the ‘curse of dimensionality’ phenomenon and its negative impact on generalization performance, the large estimation error from complex statistical models can easily compromise the prediction advantage provided by their greater representation power. •Some machine learning methods address the ‘curse of dimensionality’ in high-dimensional data analysis through feature selection and dimensionality reduction, leading to better data visualization and improved classification.•It is important to ensure that the generalization capability of classifiers derived by supervised learning methods from high-dimensional data is independently validated

invasive vs noninvasive algorithm

Allomap-score?

team patientassessment

clinical status?

graft function?

no biopsy

clinical biopsy

treatment & re-biopsy

3-6mo <207-12 mo < 30> 12 mo < 34

Allomap-score?

team patientassessment

clinical status?

graft function?

no biopsy

clinical biopsy

treatment & re-biopsy

3-6mo <207-12 mo < 30> 12 mo < 34

team patientassessment

clinical status?

graft function?

clinical biopsy

treatment & re-biopsy

protocol biopsy grade?

team patientassessment

clinical status?

graft function?

clinical biopsy

treatment & re-biopsy

protocol biopsy grade?