Importance of Formal Classification of Rejection for Collaborative

Importance of Formal Classification of Rejection for Collaborative Studies in Transplantation L. G. Hunsicker, M. D. U. of Iowa College of Medicine Stephen M. Rose, Ph. D. NIAID

Outline • Major NIH initiatives in transplantation. • The importance of histology for understanding mechanism, and relating this to practice. • The future endpoints for efficacy in clinical trials: structure and function. • Early changes and the importance of structure. • What the transplant pathology community needs to address.

NIH Program in Transplantation • NIH provides > $260 M to support basic, preclinical and clinical studies and trials in transplantation designed to: – reduce graft rejection, and – increase long-term graft survival, – eventually leading to immunosuppression free indefinite graft survival (i. e. transplantation tolerance).

NIH/NIAID Support Relevant to Transplantation • Adult and Pediatric Kidney Transplantation trials • Immune Tolerance Network • Islet transplantation • Kidney transplantation • Liver transplantation (too be implemented) • Autoimmunity • Asthma and allergic diseases

NIH/NIAID Support Relevant to Transplantation • Autoimmunity Centers of Excellence • Stem Cell Transplantation for Autoimmune Diseases • Asthma and Allergic Diseases Cooperative Centers • Innovative Research in Tolerance for Immune-mediated diseases (PPGs)

Other NIH Support Relevant to Transplantation • Genetics of Renal Diseases • Type I diabetes • DPT-1 and Trial. Net • • Lung transplantation trials Heart transplantation trials Occular transplantation trials Bone Marrow/Stem Cell transplantation

Standardization is essential! All NIH efforts require standardized, validated, and community accepted histopathology as the platform for development of new, less invasive techniques that can detect clinically relevant changes BEFORE irreversible damage is done!

Outline • Major NIAID initiatives in transplantation. • The importance of histology for understanding mechanism, and relating this to practice. • The future endpoints for efficacy in clinical trials: structure and function. • Early changes and the importance of structure. • What the transplant pathology community needs to address.

Current Status of Renal Transplantation • One year graft survival has improved over the past 30 years from about 40% to almost 90%. • Recent analyses (Hariharan S, N Engl J Med 2000; 342: 605) have documented improved long-term renal graft survival over the past decade, possibly due to: – Better immunosuppressives – Better/earlier diagnosis of rejection – Better treatments for rejection BUT

Current Status of Renal Transplantation • Data still shows that even just one serious rejection episode significantly compromises long-term survival. Therefore NIH is focusing its resources on studies to develop validated surrogates and biomarkers of rejection AND induction, maintenance and loss of tolerance.

Surrogate- and biomarkers of rejection and graft survival • All NIAID supported transplant trials have integrated mechanistic studies (ex. gene expression analyses, ELISPOT, tetramer analyses for Tc. R expression) BUT • The utility of these new assays need to be validated against established clinical data. Today’s gold standard is still classic histopathology!

Immune Tolerance Network Tissue Analysis Core Facility • Scope of Work Includes: – Interpreting biopsies from transplanted kidneys, livers, islet transplants, and other tissues, – Performing assays for tolerance in clinical studies (in situ hybridization, TUNEL assays, etc. ) – Performing new protocol specific assays for same. – Developing a tissue analysis database. – Advancing tissue analysis assays such as biopsy grading, laser-mediated capture dissection, etc. • Interested parties should visit ITN web site: http: //www. immunetolerance. org

Outline • Major NIAID initiatives in transplantation. • The importance of histology for understanding mechanism, and relating this to practice. • The future endpoints for efficacy in clinical trials: structure and function. • Early changes and the importance of structure. • What the transplant pathology community needs to address.

Major Clinical Goals in Renal Transplantation • Reduce early (e. g. , 1 year) graft and patient losses (failure from acute rejection). • Reduce long term (e. g. , 10 year) graft and patient losses (loss from “chronic rejection”). • Achieve donor specific unresponsiveness so as to make long-term immunosupression, with its toxicities and expense, unnecessary. • Reduce other morbidity, improve quality of life, reduce expense.

Proposed “Hard” End Points for Clinical Trials in Renal Transplantation • Reduce graft loss to acute rejection: Patient and graft survival at one year • Reduce graft loss to “chronic rejection”: Pt. and graft survival over many years. • Induce tolerance: ? Withdrawal of immunosuppression? • Reduce cost and improve quality of life: Eliminate acute rejection episodes

Difficulty of Studying Strategies to Improve 1 Year Graft Survival • In 1997 -1998, U. S. 1 year renal allograft survival (all comers) was: – 89. 4% for cadaveric donor organs, and – 94. 5% for living donor organs. • To have 80% power to detect a 30% improvement in 1 year cadaveric graft survival (p=0. 05) requires 2, 627 patients. • The same study with living donor patients requires 5, 371 patients. Studies with these endpoints are no longer feasible.

Proposed “Hard” End Points for Clinical Trials in Renal Transplantation • Reduce graft loss to acute rejection: Patient and graft survival at one year • Reduce graft loss to “chronic rejection”: Pt. and graft survival over many years. • Induce tolerance: ? Withdrawal of immunosuppression? • Reduce cost and improve quality of life: Eliminate acute rejection episodes

Difficulty of Studying Strategies to Improve 5 Year Graft Survival • Currently, 5 year graft survival for cadaveric renal transplants is 65% – to detect a 30% reduction in graft loss (75% survival) requires 800 patients recruited over 2 years and followed for an additional 3 years. • However, about 50% of graft losses are due to death.

Difficulty of Studying Strategies to Improve 5 Year Graft Survival • To detect a 30% reduction in graft loss due to “chronic rejection” would take 1600 pts. recruited over 4 years and followed for 3 more. These endpoints are not feasible, either.

Proposed “Hard” End Points for Clinical Trials in Renal Transplantation • Reduce graft loss to acute rejection: Patient and graft survival at one year • Reduce graft loss to “chronic rejection”: Pt. and graft survival over many years. • Induce tolerance: ? Withdrawal of immunosuppression? • Reduce cost and improve quality of life: Eliminate acute rejection episodes

Drugs Approved by the FDA Based on Reduction of Acute Rejection • Chemical immunosuppressants – mycophenolate mofetil – sirolimus (rapamycin) • Antibodies – daclizumab – basiliximab – thymoglobulin

Present Incidence of Acute Rejection in 1 st 6 Months • Recent sirolimus trials: – acute rejection rate within 6 months was 11% - 19%. • Thymoglobulin induction trial: – acute rejection rate was 6%. (Brennan et al. ) • Daclizumab 2 dose regimen in high risk patients trial: – acute rejection was 6%. (Golconda et al. ) Trials to reduce acute rejection are no longer feasible.

Proposed Alternative End Points for Clinical Trials in Renal Transplantation • Reduce graft loss to acute rejection: Patient and graft survival at one year • Reduce graft loss to “chronic rejection”: Surrogate endpoints for graft survival • Induce tolerance: ? Withdrawal of immunosuppression? • Reduce cost and improve quality of life: Eliminate acute rejection episodes

“Surrogate Endpoints” for Late Renal Allograft Failure • Even when there is currently a strong relationship between a surrogate endpoint and a “clinically meaningful” endpoint, that relationship may not persist following an intervention. • For this reason, the FDA is very reluctant to accept surrogate outcomes in clinical trials. There may be an exception to this generality.

Alternative Explanations for Correlation of Outcomes

“On the Way” Surrogate Endpoints: The Possible Exception

Proposal for Use of “On the Way” Surrogate Endpoints for CAN • Randomize patients at 6 months with high risk for CAN and early graft failure. • Provisionally establish the impact of the intervention using change of glomerular filtration rate (6 month baseline to 2. 5 years) for FDA approval.

Proposal for Use of “On the Way” Surrogate Endpoints for CAN • Estimate changes in histology from 6 months to protocol biopsy at 2. 5 years. • Confirm provisional findings with long term graft outcomes, possibly using UNOS registry.

Early Access to Drugs • “Subpart E in Section 312 of the Code of Federal Regulations establishes procedures to expedite the development, evaluation, and marketing of new therapies intended to treat people with lifethreatening and severely-debilitating illnesses, especially where no satisfactory alternatives exist. ” (Federal Register, October 21, 1988). • Designed primarily in response to the demands of HIV/AIDS advocates.

Accelerated Development/Review • “Can be used under two special circumstances: – when approval is based on evidence of the product's effect on a ‘surrogate endpoint, ’ and – when the FDA determines that safe use of a product depends on restricting its distribution or use. ”

Accelerated Development/Review • “The fundamental element of this process is that the manufacturers must continue testing after approval to demonstrate that the drug indeed provides therapeutic benefit to the patient. ” • May be an avenue for provisional approval of a drug that requires very long trials for definitive endpoints -- e. g. for treatment of chronic renal allograft rejection.

Outline • Major NIAID initiatives in transplantation. • The importance of histology for understanding mechanism, and relating this to practice. • The future endpoints for efficacy in clinical trials: structure and function. • Early changes and the importance of structure. • What the transplant pathology community needs to address.

Functions of the Glomerulus • Filtration of water and small solutes • Retention of larger elements – plasma proteins – formed elements (RBCs, platelets, etc. ) • Maintenance of near constancy of GFR – short term autoregulation – longer term compensatory hyperfiltration in response to reductions in glomerular number or filtration surface area.

GLOMERULAR HEMODYNAMICS Afferent Arteriole Efferent Arteriole Normal Pressure: 105 mm. Hg 50 mm. Hg 15 mm. Hg EFFECTS OF CHANGES IN ARTERIOLE TONE Afferent arteriolar dilatation leads to: Flow PGC Efferent arteriolar dilatation leads to: Flow PGC

HYPOTHETICAL EFFECT OF ACE INHIBITION ON GLOMERULAR FILTRATION AREA (GA) AND GFR IN DIABETIC NEPHROPATHY 100 GA % nl 50 GFR % nl Placebo Ace. INH 0 0 Intervention 1 10 Years 20 Intervention 2

CSG Trial of Captopril in Type I Diabetic Nephropathy: By Entry Scr Lewis EJ et al N Engl J Med 1993; 329: 1456 -62

Outline • Major NIAID initiatives in transplantation. • The importance of histology for understanding mechanism, and relating this to practice. • The future endpoints for efficacy in clinical trials: structure and function. • Early changes and the importance of structure. • What the transplant pathology community needs to address.

Requirements for Evaluation of Histological Outcomes • Consensus on methods for evaluation and grading of biopsies -- the Banff process. • Demonstration of ability to achieve good intraand interobserver consistency in grading. • Determination of an appropriate baseline for studies of progressive CAN. • Validation of correlation of histological changes with longer term functional changes and “hard” outcomes.

Definition of a Standard Does Not Imply Reproducibility • Five experienced renal pathologists graded twice each 25 biopsies of patients with lupus nephropathy for Austin “chronicity index (CI). ” • Grades: – low-risk: 0 or 1 – intermediate-risk: 2 or 3 – high risk: >3 • Analyses examined range of mean values, and intra- and interobserver variability using the kappa statistic. Wernick RM et al, Ann Int Med 1993; 119: 805

Wernick Study: Results • Range of average CI over readers : – 2. 3 - 4. 8 • Intraobserver consistency of reading low or intermediate vs. high risk: – Mean kappa (range): 0. 62 (0. 44 - 0. 78) • Interobserver consistency: – Range of agreement: 44% - 72% – Mean kappa, first reading: 0. 53 – Mean kappa, second reading : 0. 48 Wernick RM et al, Ann Int Med 1993; 119: 805

Wernick Study: Conclusions • “…the incorporation of index scoring into patient management may lead to erroneous decision making. Small differences in chronicity index scoring may profoundly alter risk group assignment…” Wernick RM et al, Ann Int Med 1993; 119: 805

Wernick Study: Conclusions • “Experienced reviewers from referral institutions [may] benefit from ‘mutual education, ’ [leading] to a more uniform standard of grading. ” • “…biopsy specimens should be scored by pathologists working in a quality-controlled reference center. ” Wernick RM et al, Ann Int Med 1993; 119: 805

Reproducibility of the Banff Classification of Renal Allograft Pathology Marcussen N et al. Transplantation 1995; 60: 1083 • Five experienced transplant renal pathologists scored 77 renal transplant biopsies for the components of the 1995 Banff classification of acute allograft rejection, including estimation of volume fraction of inflammation. • Intraobserver kappa score for dx of ACR: 0. 38 • Interobserver kappa score for dx of ACR: 0. 40

Some Possible Conclusions • The Banff Consortium must go beyond defining a standard method for the analysis of transplant biopsies to establish the ability of the definition to be used in a reproducible way. • The Banff Consortium should seek opportunities for pathologists to experience joint reading sessions with experts to improve consistency.

Some Possible Conclusions • Consideration should be given to development of centralized Pathology Cores. • Consideration should be given to the use of morphometry for things such as fibrosis.

Requirements for Evaluation of Histological Outcomes • Consensus on methods for evaluation and grading of biopsies -- the Banff process. • Demonstration of ability to achieve good intraand interobserver consistency in grading. • Determination of an appropriate baseline for studies of progressive CAN. • Validation of correlation of histological changes with longer term functional changes and “hard” outcomes.

Possible Etiologies of Chronic Allograft Nephropathy • Initial, but not ongoing, renal damage from ATN, advanced donor age, etc. • Ongoing immunological damage: – Impact of HLA mismatching on long term outcomes – Possibly antibody mediated - C 4 d staining • Ongoing non-immunological damage: – calcinerurin inhibitor toxicity – accelerated graft loss in hypertensive African. Americans

ALTERNATIVE MECHANISMS FOR EARLY KIDNEY GRAFT FAILURE Good Function GFR Accelerated Slope Reduced Intercept GFR at return to dialysis Time

Implantation renal biopsy showing ATN, but minimal interstitial fibrosis. Courtesy of Lorraine Racusen

Renal biopsy in the same patient 5 weeks after transplant, showing extensive interstitial fibrosis. Courtesy of Lorraine Racusen

Nicholson ML et al, Transplantation 1999; 68: 236

What is the Basis for Fibrosis after (Transplant) Kidney Damage? • Progressive fibrosis might be due to: – overactivity of profibrotic mechanisms, “squeezing out” recovery of normal tubules. – failure of recovery of normal tubular cells, leading to tubular atrophy and “filling in” with fibrous tissue. • Can the ultimate extent of “baseline” fibrosis be predicted by analyses of: – tubular cell apoptosis or mitosis rates? – early overactivity of profibrotic cytokines?

Requirements for Evaluation of Histological Outcomes • Consensus on methods for evaluation and grading of biopsies -- the Banff process. • Demonstration of ability to achieve good intraand interobserver consistency in grading. • Determination of an appropriate baseline for studies of progressive CAN. • Validation of correlation of histological changes with longer term functional changes and “hard” outcomes.

Conclusions - 1 • NIH and the transplant community continue to need consistent and reproducible consensus criteria for the diagnosis of: – acute rejection -- primarily for the testing of new proposed surrogates of rejection and tolerance. – chronic allograft nephropathy -- for the performance of definitive therapeutic trials.

Conclusions - 2 • The NIH is focusing major resources on development of minimally and noninvasive diagnostic tools. • The only way to be ready for this is to have standardized histopathology AND be ready for the landscape to change to more molecular and bioimaging technologies.

Conclusions - 3 • The Banff process permits consensus to develop. Now the Banff process has to be expanded to maximize consistency of reading and answer new questions raised about clinical interpretation.