IgA nephropathy databank: Development of a system for management of renal biopsy acquired data

Share Embed


Descrição do Produto

me OwlclalJournalof the National

Kidney

Foundation

American Journal ofKidnej Diseases SPECIAL

VOL 29, NO 6, JUNE 1997

REPORT

IgA Nephropathy Databank: Development of a System for Management of Renal Biopsy Acquired Data Robert J. Wyatt, MD, Steven N. Emancipator, MD, Valentina Kon, MD, F. Bryson Waldo, MD, James Donadio, MD, Joseph P. Grande, MD, PhD, Sharon P. Andreoli, MD, and Richard J. Glassock, MD INDEX

WORDS:

IgA nephropathy;

renal

biopsy;

prognosis.

T

HE DEVELOPMENT of comprehensive systems for the identification and follow-up of patients with IgA nephropathy (IgAN) will better establish the importance of IgAN as a cause of end-stage renal disease (ESRD); provide a readily available resource for testing hypotheses about the pathogenesis, natural history, and possible treatment interventions for IgAN; and gain epidemiologic data about IgAN in the United States. On April 12 and 13, 1996, a group of renal pathologists and nephrologists met at the Medical Education and Research Institute in Memphis, TN, to discuss the development of a regional databank for IgAN and to reach a consensus on how renal biopsy data will be managed in that databank. The term “databank” is used instead of “registry” to emphasize the concept that the information stored in the database is meant to be accessible to all serious investigators interested in IgAN. For over 15 years, Drs Robert J. Wyatt and Bruce A. Julian, with the collaboration of many renal pathologists and nephrologists in the fourstate region of Kentucky, Tennessee, Alabama, and Mississippi, have identified many of the known cases of IgAN in the region. The IgAN databank has been established in Memphis for entry of the clinical data for these patients. At the time of this conference, 987 patients from this region had been identified, and clinical data American

Journal

of Kidney

Diseases,

Vol 29,

No 6 (June),

for almost 300 (mostly from the Lexington, KY, and Memphis, TN, areas) have been entered into the IgAN databank. The database was designed using the Macintosh-based relational database, 4th Dimension (ACIUS, Cupertino, CA) and may be accessed via modem. The latest version of 4th Dimension is platform independent, allowing use by both Windows and Macintosh operating systems. The purpose of the presentations on April 12, 1996, was to develop the framework for a discussion on the development of a system for the management of renal biopsy data in the IgAN dataFrom the Department of Pediatrics, University of Tennessee, Memphis; Department of Pediatrics, Vanderbilt University, Nashville, TN; Department of Pathology, Case Western Reserve University, Cleveland, OH; Department of Pediatrics, University of Alabama, Birmingham, AL; Departments of Nephrology and Pathology and Laboratory Medicine, Mayo Clinic, Rochester, MN; Department of Pediatrics, Indiana University, Indianapolis, IN; and Department of Internal Medicine, University of Kentucky, Lexington, KY. Received September 25, 1996; accepted in revised form January 7, 1997. Support for the conference was provided by a grant from the Methodist Hospitals Foundation and a gift from the Conwood Company, L.P. Address reprint requests to Robert J. Wyatt, MD, Crippled Children’s Foundation Research Center, Room 301 WPT, 50 N Dunlap, Memphis, TN 38103. 0 1997 by the National Kidney Foundation, Inc. 0272.6386/97/2906-0003$3.00/O 1997:

pp 817-828

817

818

bank. Dr Robert Wyatt began the conference with a description of clinical and epidemiologic studies derived from the patients with IgAN from Kentucky and Tennessee. An ideal use of the IgAN databank is as a source of patients for prognostic marker studies. Two of the investigators within the study region, Drs Wyatt and Valentina Kon, have studied prognostic markers in IgAN. Dr Wyatt gave an overview of work on immunogenetic markers of prognosis and Dr Kon presented work from her group on the angiotensinconverting enzyme (ACE) DD genotype and poor prognosis in IgAN. Dr Bryson Waldo reviewed pathologic scoring systems that have previously been applied to IgAN, and Drs James Donadio and Joseph Grande from the Mayo Nephrology Collaborative Group and Dr Sharon Andreoli showed how their respective classification systems have been used for clinicopathologic correlations. Dr Steven Emancipator discussed issues related to the development of a pathologic database for IgAN. After these presentations, the conference attendees undertook a lengthy discussion on establishing criteria for entry of pathologic data into the IgAN databank. A system of criteria for scoring and recording renal biopsy findings was then developed. The morning session on April 13, 1996, was an open discussion related mostly to administrative issues surrounding clinical and pathologic data in the IgAN databank. Dr Richard Glassock closed the conference with his ideas about the utility of the IgAN databank and recommendations for future initiatives. PROGRESS TO DATE ON THE REGIONAL IGA NEPHROPATHY DATABANK

Robert J. Wyatt, MD

Our initial report from 1984 that used patient data now incorporated in the IgAN databank included 82 patients, both pediatric and adult, monitored in Lexington, KY.’ This study contained the first kidney survival analysis for patients from the United States which indicated that at 10 years from apparent onset of disease, 22% of the 56 adult patients with IgAN were predicted to progress to ESRD. This study showed that the outcome for IgAN in a center from the United States was similar to that described in European centers. Patients in that study were not categorized according to a renal histopathologic grade.

WYA-iT

ET AL

Long-term outcome was recently reported for the pediatric patients identified in Memphis, TN, and Lexington, KY.’ All of these patients are included in the regional IgAN databank. Renal histology was classified according to the grade 1,2, or 3 system devised by Dr Fred Silva and the Southwest Pediatric Nephrology Study Group.3 Progression to ESRD occurred in 14 of the 103 patients who underwent biopsy before the age of 18 years. Actuarial predicted kidney survival from time of biopsy was 85% at 10 years and 73% at 20 years. For the patients with type 3 lesions, predicted kidney survival was 75% at 10 years and 49% at 20 years. The actuarial predicted kidney survival was significantly worse for patients with type 3 lesions than for those with type 1 and 2 lesions (P = 0.0038, log-rank test). Preliminary data from the southeastern IgAN databank indicate that over the past decade in the Lexington, KY, region, a substantial number of patients were older than 60 years at the time of diagnosis and an increased number of patients presenting with chronic renal insufficiency have undergone biopsy. The incidence of IgAN for the region of central and eastern Kentucky for the period 1985 through 1994 is 10 to 12 cases per one million population per year. In comparison, European centers report the incidence of IgAN to be in the range of 20 to 40 cases per one million population per year.4 In Brittany, France, for the period from 1981 to 1990, the incidence of ESRD in patients with IgAN was 8 cases per million per year. In comparison, in central and eastern Kentucky for the period from 1991 to 1995, the incidence of ESRD was 6 cases per million per year. These data tend to confirm the notion that IgAN is underdiagnosed in the United States compared with Europe, but that the disease is just as frequent in the United States as a cause of ESRD. IMMUNOGENETIC MARKERS OF PROGNOSIS IN IGA NEPHROPATHY

Robert J. Wyatt, MD

For almost two decades, major investigative efforts have searched for association between particular immunogenetic markers and IgAN, or a subgroup of patients with IgAN having a poor outcome. The initial putative marker for poor outcome, HLA B35, was suggested by data from

MANAGEMENT

OF

BIOPSY

DATA

FOR

IGA

Berthoux’s group in the Rhone-Alps region of France.’ Other markers of poor prognosis located in the major histocompatibility complex include HLA DR4 in Japan6 HLA DRl in transplanted patients followed by the South Eastern Organ Procurement Foundation7 and C4A deficiency in the United States8 The major problem with these data is that virtually all markers for poor prognosis found in one study have not been confirmed in other regions.’ Recently, C4A deficiency was shown to associate with poor prognosis in Swedish patients with 1gAN.l’ However, C4A deficiency only occurs in 2% to 3% of control subjects and in approximately the same frequency in patients with IgAN.8%‘o Thus, its use as a marker of poor prognosis is very limited. One of the major problems for interpretation of data from prognostic marker studies is a lack of uniformity in collection and interpretation of the clinical data. An early use of the IgAN databank would be the identification of patients for these studies who are well characterized with respect to clinical manifestations, follow-up, and pathologic findings. Cohorts of patients large enough for meaningful statistical analysis of marker data could be generated for use in prognostic marker studies. One extremely interesting new area of investigation, applicable to many patients with IgAN, is the role of the ACE DD genotype as a prognostic marker for IgAN. ACE GENOTYPE

AND

PROGNOSIS

819

NEPHROPATHY

IN IGAN

Valentina Kon, MD Polymorphisms of three genes in the renin angiotensin system have been linked to variability of outcome in several cardiovascular diseases and more recently with disease progression in IgAN.‘l These genetic variants include (1) ACE deletion polymorphism in intron 16 (ACE/ID), (2) a point mutation in the angiotensinogen gene resulting in methionine to threonine substitution at residue 235, and (3) an angiotensin receptor type I A to C transition at base pair 1,166. The study population included 64 white patients with IgAN who were aged 6 to 83 years at time of renal biopsy. Mean follow-up was 6.8 years. For patients presenting with and maintaining serum creatinine less than 1.5 mg/dL, ACE genotype frequencies were 35% for II, 61% for ID, and 4% for DD. In comparison, for patients with pro-

gressive disease, ACE genotype frequencies were 22% for II, 44% for ID, and 33% for DD (nonprogressors v progressors: P = 0.057, Fisher’s exact test). When patients with hypertension and/or heavy proteinuria were excluded, the association with the DD phenotype was even more striking. The genotype frequencies for those with serum creatinine less than 1.5 mg/dL were 53% for II, 47% for ID: and 0% for DD; for those with progressive disease they were 0% for II, 60% for ID, and 40% for DD (nonprogressors v progressors: P = 0.009, Fisher’s exact test). Sequence analysis of the I gene polymorphism showed a potential 13 base pair silencer motif. Neither the angiotensinogen 235T nor the ATR A 1166C gene variant was associated with progressive deterioration of renal function. The results of this study indicate that although polymorphism in each of the three genes in the renin angiotensin system have been associated with cardiovascular diseases, only the ACE/ID polymorphism is associated with progressive deterioration of renal function in patients with IgAN. Since previous observations link ACE polymorphisms with ACE activity, these findings imply a role for ACE in modulating destructive processes in different organs. PREVIOUS

PATHOLOGIC

SCORING

SYSTEMS

Bryson Waldo, MD A number of pathologic systems have been used to classify patients with IgAN. Different investigators have approached the classification of renal pathology in IgAN with different goals in mind. These goals have included improvement in diagnostic accuracy, correlation of various pathologic changes, correlation with prognosis, and assessment of treatment effect. These systems can very simply be divided into two groups: lumped and split. Examples of lumped systems include those of the Southwest Pediatric Nephrology Study Group,3 Lee et al,” and Mina and Murphy.13 Examples of split systems include those of Kobayashi et all4 and Alamartine et all5 Andreoli and Bergstein16 and Waldo et all7 devised split systems similar to some used in systemic lupus erythematosus to evaluate effect of treatment on activity and chronicity of disease. A clear advantage of the lumped systems is their simplicity and ease of application in large

WYATT

multicenter studies that lack central pathology core support. Although very simplistic, the Southwest Pediatric Nephrology Study Group system of scoring as mild, moderate, or severe has proved to be extremely valuable in the analysis of a large clinical database. The obvious weakness of a lumped system is the lack of flexibility in interpretation and the chance that important isolated pathologic changes will be missed. The split systems allow detailed analysis of each individual finding included in the system. This obviously adds significant complexity to the analysis and in a multicenter study could result in significant intercenter variation in classification. In all the split systems described, a global or aggregate score could be obtained that proved more informative than any individual finding and that was very comparable to the findings obtained with lumped systems. Based on this past experience, the working group from this conference should first decide the goals of any scoring system. These goals should include prognostication to permit selection of patients at risk for progression and evaluation of activity and chronicity in patients both at diagnosis and potentially at follow-up after therapeutic interventions. A split system that includes evaluation of glomeruli, tubules, interstitium, and vessels should be used. The system should rely primarily on light microscopy, since this can be most easily reviewed centrally. Such a system would provide maximal future flexibility and minimal intercenter variability. CLINICOPATHOLOGIC CORRELATIONS FROM THE IGAN DATABASE OF THE MAYO NEPHROLOGY COLLABORATIVE GROUP

James Donadio, MD, and Joseph Grande, MD, PhD The Mayo Nephrology Collaborative Group has accumulated one of the largest cohorts of closely followed-up patients with IgAN in the United States. The study from this group in 198118 represents the first report from the United States that comprehensively examined clinical and pathologic indicators of progressive disease. Between 1973 and 1995, 206 patients were diagnosed with primary IgAN by renal biopsy performed at the Mayo Clinic. Renal tissue blocks were obtained and at least 3 years of clinical

ET AL

follow-up were completed on 148 patients. Histopathologic scoring was recorded using a semiquantitative (0 to 3+) system adapted from Donadio et all9 for assessment of six glomerular, eight interstitial, and six vascular histopathologic components. Glomerular and interstitial proliferative activity was evaluated by immunostaining archival biopsy specimens with Mib-1, an antibody directed against the Ki-67 antigen. At the time of diagnostic renal biopsy, 47% of these patients were hypertensive, 59% had elevated serum creatinine levels, and 76% had proteinuria exceeding 1 g/24 hr. Thirty-nine patients (26%) developed ESRD, the primary end point of this analysis. Survival free of renal failure was 79% 5 3.8% (mean 2 SE) at 5 years and 67% t 5.0% at 10 years. Kaplan-Meier survival analysis was performed, with renal failure defined as onset of dialysis or transplantation. A number of clinical and pathologic features was univariately associated with adverse outcome, including elevated serum creatinine levels, the presence of hypertension, proteinuria, and positive glomerular or interstitial Mib- 1 scores. Total histopathologic scores higher than 10 were associated with adverse outcome, as were glomerular scores higher than 5, interstitial scores higher than 8, and vascular scores higher than 5. Spearman rank correlation analysis indicated that each component score contributed independent information to the total score, ie, that there was little redundant information provided by each subscore. All histopathologic parameters were significantly associated with renal failure, except mesangial hypercellularity, tubular epithelial cell vacuolation, presence of red blood cell casts, and interlobular arterial sclerosis. The total glomerular score demonstrated the strongest univariate association with renal survival, followed by the overall scores, the interstitial scores, and the vascular scores. By multivariate analysis, the total glomerular score was the strongest predictor of renal failure. When histopathologic data were added to the clinical and demographic data, multivariate analysis indicated that increased serum creatinine, high total glomerular histopathologic score, and younger age at diagnosis were significant independent predictors of renal failure. Interstitial or glomerular Mib-1 scores were not independent predictors of renal failure by multivariate analysis.

MANAGEMENT

OF

BIOPSY

DATA

FOR

IGA NEPHROPATHY

The clinical and histopathologic variables associated with progressive renal failure in our study cohort agree with other clinicopathologic studies with one exception. Among the independent risks (ie, impaired renal function, younger age at diagnosis, and total glomerular score), younger age is at variance with previous reports indicating that older age was associated with a poor outcome, including a previous report from our center. l2 In our current study group, among the 26% of patients who presented with a serum creatinine level greater than 2 mg/dL, there was a strong correlation with poor prognosis in those younger than 20 years. Stated another way, there was an unusually high number of young patients who had advancing renal disease in this referralbased cohort. In fact, the overall lo-year renal survival rate of 67% in our study group was below the survival rates in recent large cohort studies from Europe, Australia, and Japan, which reported survival rates between 74% and 94%. The three most consistent risks for progression in these cohort studies were hypertension, elevated serum creatinine level, and increased protein excretion. The frequency of all three risks was much greater in our study group than in the other studies. Many of our patients carried all three risks, reflecting the fact that, in general, more advanced cases are seen in referral centers such as ours and that patients are coming to renal biopsy late in their disease course. In the United States, it is common practice to avoid biopsy of patients presenting with isolated hematuria or mild proteinuria, with renal biopsy held in reserve for individuals who develop recognized risk factors portending a worse prognosis. This is likely the major reason for the differences noted in renal survival between our study group and that of other populations, rather than the geographic and ethnic differences usually offered to account for variable disease progression. Our data provide a rationale for the implementation of a composite histopathologic scoring system in future clinical outcome studies of patients with IgAN. The clinical utility of this clinicopathologic scoring system is currently being prospectively evaluated in a variable dose trial of omega-3 fatty acids in IgAN. Reproducible histopathologic scoring systems, combined with the important well-recognized clinical markers of

821

disease progression, will be essential for the proper stratification of at-risk patients with IgAN into promising therapeutic trials. HISTOLOGIC GRADATION OF BIOPSY SPECIMENS FROM PATIENTS WITH IGAN

Sharon Andreoli, MD We have used a simplified system, adapted from a scoring system used to define activity and chronicity scores in patients with lupus nephritis,20 to grade biopsy specimens from patients with IgAN.‘6,2’ This grading system allows for comparison of biopsy specimens among patient groups and for comparison of histology before and after therapy. The activity score is determined by grading mesangial and glomerular proliferation and interstitial inflammation from 0 to 3. Crescent formation is scored as 0 to 3 according to the percentage of glomeruli involved with cellular crescent formation. The sum of the degree of mesangial and glomerular proliferation (0 to 3), interstitial inflammation (0 to 3), and the percent crescents (0 to 3) comprised the activity score, with a maximum score of 9. The chronicity score is determined by grading the degree of tubular atrophy and interstitial fibrosis as 0 to 3, while the number of glomeruli demonstrating fibrous crescents or segmental or global sclerosis were counted and each scored as 0 to 3 according to the percentage of glomeruli involved. The sums of these numbers (tubular atrophy and fibrosis 0 to 3, percent global sclerosis 0 to 3, percent segmental sclerosis 0 to 3, and percent fibrous or fibrocellular crescents 0 to 3) composed the chronicity score, with a maximum of 12. In addition, the percentage of glomeruli demonstrating segmental or global sclerosis was added together to reflect, on a scale of 0% to lOO%, the severity of chronic irreversible glomerular changes. Using this grading system, we have demonstrated that glomerular basement membrane deposition of IgA is strongly correlated with crescent formation, chronic changes of segmental or global sclerosis, and persistent proteinuria.2’ In other studies we have also used the grading system to compare the activity of lesions and chronic changes before and following therapy in children with severe IgAN.r6 When comparing the activity score pretherapy and posttherapy, the mean activity score decreased significantly following

WYATT

822

therapy, while the chronicity score remained stable in most patients. This scoring system can be used in patients with IgAN to compare histologic features in a semiquantitative manner. ISSUES

RELATED TO THE DEVELOPMENT A PATHOLOGIC DATABASE

Steven N. Emancipator,

OF

MD

The goals of medical tests are diagnosis, prognosis, assessment of a patient’s course, and gauging response to therapy. Ultimately, fulfillment of these goals resides with the reproducible classification of patients, ideally by definable criteria. The parameters to be measured, and their critical values, are proposed from predictions based on understanding the pathogenesis and pathophysiology of the relevant conditions, and become ratified for clinical use when predictive value is established by correlation of the test results with objective outcomes. As knowledge about a disease evolves, diagnostic modalities become revised. Therefore, the selection of parameters depends on sound theory, whereas the establishment of criteria resides with trial and error. Powerful diagnostic tests can also serve to illuminate pathophysiologic processes and guide therapeutic initiatives: consider the ancient observation of sweet urine in diabetic patients that underlies the discovery of insulin. Renal biopsy interpretation evolved as morphologic patterns became associated with particular clinicopathologic entities.22-24 Traditionally, glomerulonephritis is classified based on the extent and severity of mesangial proliferation, mesangial matrix expansion, endocapillary proliferation, glomerulosclerosis, leukocytic infiltration, thrombosis, necrosis, synechiae, and crescents. The distribution, size, extent, and composition of the deposits are also central elements of diagnosis, and variability of the changes within and between glomeruli is sometimes emphasized. Inflammation (tubulitis), atrophy, degeneration, and resorptive changes within the cortical tubules, and interstitial inflammation, edema, and fibrosis are now also evaluated carefully, along with medial hypertrophy and medial fibroelastosis, hyalinosis, and necrosis or frank vasculitis in arteries and arterioles. Among these parameters, only glomerulosclerosis, interstitial fibrosis, and in some cases, vascular sclerosis reach significance as an independent variable

ET AL

correlated with progressive loss of renal function in patients with IgAN.25 With the addition of tubular inflammation in some instances, these same parameters are the ones associated with progression in all other renal diseases, whether they are primarily glomerular or not.26327Yet, none of these, with the possible exception of tubulitis, is of prognostic portent in that the presence of scarring is a consequence rather than a premonitor of progression. Moreover, although the stigmata of progression and the rate of functional nephron loss are closely correlated at the populational level, they are not useful in predicting the course of an individual patient at a particular point in time. Thus, the limitations of current renal biopsy interpretation are patent: single variables offer no predictive power, very limited correlative power at the level of the individual patient, and no insight to pathophysiologic processes or therapeutic objectives.26,28,29 The limitations do not necessarily refute the notion that progressive renal disease is heralded by discretely identifiable morphologic features and that function follows form. Instead, these limitations may be emblems of failure to consider the appropriate criteria, irreproducibility in the grading of the selected features, or a fundamental interaction of different parameters such that sets, rather than individual elements, are the appropriate variables to consider. Alternately, studies to date may have missed choosing the proper break point for a particular parameter that is discontinuous with respect to progression of renal injury; small changes in critical values often have profound ramifications in outcomes analysis. Effective stratifications need to be learned, but an approach of trial and error is cumbersome. If morphologic features have any predictive power, computer analysis of a comprehensive database from a large population of IgAN patients offers potential resolution of the aforementioned limitations. Beyond serving as a passive repository for clinical and pathologic information and retrieval, a database of morphologic parameters can be evaluated continuously and repetitively against the clinical course followed by the entire cohort of patients. Dynamic algorithms can be applied, weighting parameters according to their partial correlation coefficients, to develop a progression index that takes the covariance of all the other parameters into account. For example,

MANAGEMENT

OF

BIOPSY

DATA

FOR

IGA NEPHROPATHY

the computer might associate the percentage of glomeruli bearing endocapillary lesions as a premonitor for progression only if there is associated capillary wall immune deposition; positive capillary fluorescence could be applied as a discontinuous modifier to the coefficient for percent endocapillary proliferation. In essence, the computer can learn what combinations and stratifications of variables offer the most effective prognostic power, and refine the criteria according to objective indicators. A number of logistical issues must be confronted. Since textual reports are not easily interpreted by machine, a simple quantification of appropriate morphologic features must be extracted objectively, based on readily applied criteria, from the histologic material. Simplicity will promote compliance by pathologists, and objective criteria will promote reproducibility between pathologists and between temporally separate evaluations of the same material by the same pathologist. For example, the Banff schema for evaluation of allografts codifies, with clearly delineated and easily applied criteria, all the traditional elements of light microscopic evaluation of a renal biopsy, except some of the finer intraglomerular elements.30 Reproducibility of this schema is excellent because of the objectivity of the criteria. A similar schema, with finer intraglomerular, immunofluorescence, and electron microscopic parameters added, could be applied to &AN. An additional benefit of the development of a relational computer database derives from the capacity to integrate novel techniques and parameters into existing classification algorithms. This feature can serve to reduce the risk that the appropriate criteria are omitted from analysis. Groups of investigators could choose different nontraditional parameters to include with the basic database; subsequently, the predictive power of separate nontraditional values could be compared with each other and with the traditional data set. Again, the computer can serve a dynamic function, relating the substudies within the larger population and guiding selection of criteria. Among the nontraditional parameters, some are based on histochemical (immunohistochemical, lectin histochemical, or in situ hybridization) approaches using traditional biopsy processing. For example, cellular proliferation, and to some

823

degree the lineage of the proliferating cellular elements, can be identified by use of Ki67/PCNA antibodies31; matrix composition can be assessed and/or semiquantified with antibodies or nucleic acid probes specific for collagens (aI-IV) or proteoglycans, among others.28*32Apoptosis can be assessed33234and phenotype modulation of selected cell populations might be indicated by expression of vasoactives (eg, angiotensin II), cytokines and growth factors and/or their receptors, phospholipase isozymes, a-actin, or chemokines.35”7 Altered levels of tyrosine phosphates, integrins, cell infiltrates, or products secreted from lymphocytes also can be detected by such methods.38-40 More biochemically based methods (such as reverse transcriptase-polymerase chain reaction or Northern hybridization) or analysis of functional states of specific macromolecules (such as quantitative measurement of the tissue content and percent tyrosine phosphorylation of specific immunoprecipitated molecules like ZO1, focal adhesion kinase, or integrins) might also be warranted, although they would render the tissue unsuitable for morphologic assessment.41 In any case, a data-dense, machine-interpretable database analysis of a large, multicenter cohort of IgAN patients using prescribed, codified, reproducible data elements is likely to offer powerful support of clinical research and promote conceptual advances surrounding the pathogenesis and pathophysiology of disease. By using dynamic relational software and some plasticity of nontraditional elements within the framework of traditional data elements, the computer will support direct comparisons among substudies and can actually guide ongoing investigation and hone our approach to the criteria. The issues that remain center on logistics: the database elements, the commitment of money, memory, and process time, access to the data, and control of the data emerge as issues. To begin, we must devise a comprehensive but prudent and practical morphometric database. DISCUSSION

OF ADMINISTRATIVE

ISSUES

The morning session on April 13, 1996, was an open discussion related mostly to administrative issues surrounding both pathologic and clinical data management. The participants of the conference agreed that the databank should be available for entry of patients residing outside of the four-

824

WYA-iT

state southeastern region. However, the goal of the pilot study to identify and enter all patients within the region will be maintained to demonstrate the feasibility in terms of cost and effort in obtaining data for all patients within a geographically defined region. Issues related to confidentiality and informed consent were discussed at length. It was proposed that a subject would give signed consent, allowing contact to be made soliciting participation in studies involving treatment or blood drawing. A committee should be formed to evaluate proposals from investigators for use of subjects or biopsy material from subjects who are entered into the IgAN databank. This would ensure that a patient with IgAN can only be contacted for studies approved by the committee. There was unanimous agreement to include individuals having a biopsy diagnosis of IgAN and the clinical features of SchBnlein-Henoch purpura in the IgAN databank. Overall, a high level of enthusiasm was expressed by all participants for the development of the IgAN databank. CONCLUDING REMARKS RECOMMENDATIONS

AND

Richard J. Glassock, MD Fundamentally, a disease registry or databank is a centralized repository of information secured from the process of disease diagnosis at a population level. As such, the goal of most registries or databanks is to accumulate information on a large population of patients, greater than that obtainable at a single institution, in order that inferences regarding the disease as it presents in the population can be drawn. Uncommon events, which would seldom be observed in single institutions, can be accumulated for study more readily. These events can have potential explanatory power. Demographic trends also can be monitored more closely, with caveats regarding ascertainment bias. In an ideal databank, the patients enrolled would represent the results of a random selection process. Thus, the patients enrolled in the databank would be representative of the universe of patients with the disease in the population. For IgAN, this is almost certainly not the case, and the fact that a databank for IgAN will not be truly representative of the population or universe of patients must be assumed. Patients with later

ET AL

stages of disease are and will continue to be overrepresented among the databank enrollees. Efforts to secure a more representative population will be frustrating, since presently there is no clear evidence that earlier diagnosis of IgAN will lead to better management.42343 Nevertheless, one of the central issues that a databank could address is the identification of the earliest marker(s) of future progressive disease. All patients with IgAN presumably begin with a normal-appearing kidney structure. With the gradual accumulation of IgA polymers in the mesangium, a threshold is ultimately reached whereby pathologic and clinical abnormalities are expressed. Identification of the processes that drive certain patients into a progressive pathway while sparing others is one of the central questions for our understanding of IgAN. However, it is doubtful that a databank, as presently constructed, could directly address such a question. Retrospective study of late-stage disease is not likely to yield critical new insights. However, a comparison of histologic and clinical features between progressors and nonprogressors, defined in functional terms, would be interesting. Confounded by random applications of therapy, this process may be clouding natural history studies. A search for parameters that could identify patients in each of these functional categories at the earliest possible stage would be a useful and eminently practical objective of the databank. The question of the impact of ACE genotype and urinary p-2 microglobulin excretion come to mind.44,45 One might ask how much precision is clinically required in the process of separating progressor from nonprogressor. Is the separation into good versus bad prognosis sufficient, or is more detailed information necessary? It is important to contrast the largely intellectual exercise of population prognostication with individual prognostication. Clinicians need the latter, while large-scale studies involving therapeutic interventions require the former. One might ask whether prediction of prognosis at an earlier stage, before gross markers of progression appear, is essential to the ultimate achievement of the maximum therapeutic utility of intervention. I remain quite skeptical that a purely morphologic approach will have the power to move the threshold of prognostication backwards in time.

MANAGEMENT

OF

BIOPSY

DATA

FOR

I also believe that the application of nonmorphologic methodologies to prognosis will be difficult and time consuming. Each incremental shift of the threshold backward in time will come at great expense. The thesis that information gleaned from a single renal biopsy can be used as a prognostic tool is in fact based on the premise that in the absence of any external intervention, what happened in the past, retained to a precise moment in time, will bear significance for the future: “What’s past is prologue” (The Tempest, Shakespeare). Carrying this thesis one step further, one could posit that the predictive ability of a static record of past events will require a certain threshold of cumulative events to have occurred. If true, then the challenge is to move the threshold backward in time without any loss of prognostic

Table Light microscopy No. of glomeruli Mesangial cellular Mesangial matrix Endocapillary

1. Memphis

System

proliferation increase (subjective)

occlusion

Glomerulosclerosis-global Glomerulosclerosis-segmental Cellular crescent, nonfibrous adhesion, tuft necrosis Fibrous or fibrocellular crescent, fibrous adhesion Interstitial edema and/or leukocyte infiltrate Interstitial Tubular

fibrosis atrophy

Tubular injury: loss of microvilli, cytoplasmic nuclear hyperchromasia, coarse vacuolation, desquamation (2 3/6 criteria)

amphiphilia, tubulitis,

Arteriosclerosis Arteriolosclerosis lmmunofluorescence IgG, IgA, IgM,

kappa,

microscopy lambda,

C3,

C4, Clq,

Electron microscopy Mesangial deposits Subendothelial deposits Subepithelial deposits Basement membrane changes

* If present, variability t Predominantly.

is noted

825

IGA NEPHROPATHY

P, fibrin

for

precision. One might question whether a very dense database of clinical and pathologic information stored digitally and retrieved electronically will disclose features which permit a more accurate and individualized prognosis that would add significantly to prognosis determined solely on clinical grounds. I am optimistic that the answer to this question will be resoundingly “Yes!” Major progress was made at this conference to establish a rational system for the evaluation of renal biopsy information in IgAN. This system now needs to be tested against the actual database, both retrospectively and prospectively. Nevertheless, it is an excellent start toward what might ultimately be the Memphis classification of the renal pathology in IgAN. The scoring system

Scoring

Renal

Biopsy

Findings

Percentage of glomeruli involved i- variability* None, mild, moderate, severe 2 variability; grade based on typical glomerulus None, mild, moderate, severe ? variability; grade based on typical glomerulus No. of glomeruli, percentage of glomeruli No. of glomeruli, percentage of glomeruli No. of glomeruli, percentage of glomeruli No. of glomeruli, percentage of glomeruli Percentage (best estimate of involvement of cortical tissue) Percentage (best estimate of involvement of cortical tissue) Percentage (best estimate of involvement of cortical tissue) None, mild, moderate, severe i variability; grade based on visual average of sample (ie, if half of sample is severe and other half normal, score would be moderate with variability flag set) None, mild, moderate, severe i variability None, mild, moderate, severe t variability

Separate mesangial and capillary scores, negative, trace, 1+, 2+, 3+, not done Absent or present; lucent or Absent or present; lucent or Absent or present; lucent or None, thin (adult 450

WYATT

developed at this meeting through consensus development will very likely provide a useful new tool to be added to the existing database of clinical information. A series of next steps should follow this conference. 1. The Memphis scoring system for evaluating renal pathology should be piloted using reports already available to determine the extent of missing data in the retrospective information. Simultaneously, the Memphis scoring system should be applied prospectively to all new biopsy information acquired (Table 1). 2. An attempt should be made to develop an agreement on a standardized approach to obtaining clinical information. 3. A merged clinical and pathologic database could then be created. 4. Data on the maximum number of patients in the four-state region (Kentucky. Tennessee, Alabama, and Mississippi) could be secured and entered. 5. Once the database is complete for these patients, one could begin to test hypotheses by querying the databank. Some interesting hypotheses discussed during the meeting include: 1. Will renal pathologic data, as obtained using the Memphis scoring system, improve the prognostic accuracy, in terms of development of renal failure, over and above the use of clinical information alone (eg, serum creatinine, urinary protein excretion, elevated blood pressure at the time of biopsy)? 2. Which parameter or grouping of parameters in the Memphis scoring system has the greatest independent prognostic predictive power? 3. Among familial cases of IgAN (an index case with at least one affected first- or seconddegree relative), will prognosis, in terms of renal failure, cosegregate with susceptibility? 4. Does the combined entity of IgAN and thin basement membrane nephropathy have a more benign prognosis than IgAN without thin basement membrane disease? Overall, the conference succeeded in its initial purpose that a clear agreement be reached regarding strategies for the further evolution of a regional IgAN databank. The development of a rational system (Memphis scoring system) for evaluating renal pathologic material obtained from patients with IgAN is a singular achieve-

ET AL

ment. It is hoped that this system will eventually evolve into one that can be usefully merged with clinical information obtained both retrospectively and prospectively. It is hoped that this will be the first of many such conferences related to the IgAN databank, and the work accomplished at the conference will assist in the development of extramural funding for the registry process itself. CONFERENCE

PARTICIPANTS

Nephrologists: Sharon Andreoli, MD, Bettina Ault, MD, Radhakrishna Baliga, MD, Richard Baehler, MD, James Donadio, MD, John Galla, MD, Richard J. Glassock, MD, Juacinto Hemandez, MD, Ronald J. Hogg, MD, Jeffrey Hymes, MD, Bruce A. Julian, MD, Valentina Kon, MD, Sangiv Restogi, MD, Shane Roy III, MD, Bryson Waldo, MD, Lucius Wright, MD, Barry Wall, MD, and Robert J. Wyatt, MD. Pathologists: Steven Bigler, MD, Steven N. Emancipator, MD, Agnes Fogo, MD, Lillian W. Gaber, MD, Joseph Grande, MD, PhD, Michael Kashgarian, MD, Bonnie Mitchell, MD, and Max Solano, MD. Epidemiologist: Stephen Kritchevsky, PhD. Research nurses: Paula M. Miller, MSN, and Susan Y. Woodford, BSN. Database consultant: Derrick Pulliam, BS.

ACKNOWLEDGMENT The authors are indebted to the Medical Education and Research Institute, 50 N Cleveland Ave, Memphis, TN, 38 104 for the organization and management of the conference and the use of their conference facility.

REFERENCES 1. Wyatt RJ, Julian BA, Bhathena DB, Mitchell BL, Holland NH, Malluche HH: IgA nephropathy: Presentation, clinical course and prognosis in children and adults. Am J Kidney Dis 4:192-200, 1984 2. Wyatt RJ, Kritchevsky SB, Woodford SY, Miller PM, Roy S III, Holland NH, Bishof NA: IgA nephropathy: Longterm prognosis for pediatric patients. J Pediatr 127:913-919, 1995 3. Southwest Pediatric Nephrology Study Group: A multicenter study of IgA nephropathy in children: A report of the Southwest Pediatric Nephrology Study Group. Kidney Int 22:643-652, 1982 4. Simon P, Ram&e M-P, Autuly V, Laruelle E, Charasse C, Cam G, Ang KS: Epidemiology of primary glomerular diseases in a French region. Variations according to period and age. Kidney Int 46:1192-1198, 1994 5. Berthoux FC, Gagne C, Gagne A, Le Petit JC, Sabatier JC: HLA-Bw35 antigen and mesangial IgA glomerulonephritis: A poor prognosis marker. Proc Eur Dial Transplant Assoc 16551-555, 1979 6. Kashiwabara H, Shishido H, Tomura S, Tuchida H, Miyajima T: Strong association between IgA nephropathy and HLA-DR4 antigen. Kidney Int 22:377-382, 1982 7. Freeman BI, Spray BJ, Heise ER: HLA associations in

MANAGEMENT

OF

BIOPSY

DATA

FOR

IGA

NEPHROPATHY

IgA nephropathy and focal and segmental glomerulosclerosis. Am J Kidney Dis 23:352-357, 1994 8. Wyatt RJ, Julian BA, Woodford SY, Wang C, Roberts .I, Thompson JS, Christienson MJ, McLean RH: C4A denciency and poor prognosis in patients with IgA nephropathy. Clin Nephrol 36:1-5, 1991 9. Wyatt RJ, Julian BA, Woodford SY, McLean RH, Thompson JS: Immunogenetic markers as prognostic features in patients from Kentucky with IgA nephropathy. Semin Nephrol 7:389-392, 1987 10. Wopenka U, Thysell H, Sjoholm AG, Trnedsson L: C4 phenotypes in IgA nephropathy: Disease progression associated with C4A deficiency but not with C4 isotype concentrations. Clin Nephrol 45:141-145, 1996 11. Hunley TE, Julian BA, Phillips JA III, Summar ML, Yoshida H, Horn RG, Brown NJ, Fogo A, Ichikawa I, Kon V: Angiotensin converting enzyme gene polymorphism: Potential silencer motif and impact on progression in IgA nephropathy. Kidney Int 49:571-577, 1996 12. Lee K, Rao WV, Franklin W, Schiffer MS, Aronson AJ, Spargo BH, Katz AI: IgA nephropathy: Morphologic predictors of progressive disease. Hum Path01 12:314-322, 1982 13. Mina SN, Murphy WM: IgA nephropathy: A comparative study of the clinicopathologic features in children and adults. Am J Clin Path01 83:669-675, 1985 14. Kobayashi Y, Tateno S, Hiki Y, Shigematsu H: IgA nephropathy: Prognostic significance of proteinmia and histologic alterations. Nephron 34:146-153, 1983 15. Alamartine E, Sabatier JC, Bertoux FC: Comparison of pathological lesions on repeated renal biopsies in 73 patients with primary IgA glomerulonephritis: Value of quantitative scoring and approach to final prognosis. Clin Nephrol 34:45-51, 1990 16. Andreoli SP, Bergstein JM: Treatment of severe IgA nephropathy in children. Pediatr Nephrol 3:248-253,1989 17. Waldo FB, Wyatt RJ, Kelly DR, Herrera GA, Benefield MR, Kohaut EC: Treatment of IgA nephropathy in children: Efficacy of alternate-day oral prednisone. Pediatr Nephrol 7:529-532, 1993 18. Hood SA, Velosa JA, Holley KE, Donadio JV Jr: IgAIgG nephropathy: Predictive indices of progressive disease. Clin Nephrol 16:55-62, 1981 19. Donadio JV Jr, Bergstralh EJ, Offord KP, Holley KE, Spencer DC, for the Mayo Nephrology Collaborative Group: Clinical and histopathological associations with impaired renal function in IgA nephropathy. Clin Nephrol 41:65-71, 1994 20. Austin HA, Muenz LR, Joyce KM, Antonovgeh TA, Kullick ME, Klippel JH, Decker JL, Balow JE: Prognostic factors in lupus nephritis: Contribution of renal histologic data. Am J Med 75:382-391, 1983 21. Andreoli SP, Yum NM, Bergstein JM: IgA nephropathy in children: Significance of glomerular basement membrane deposition of IgA. Am J Nephrol 6:28-33, 1986 22. Pirani CL: Renal biopsy: An historical perspective, in Silva FG, D’Agati VD, Nadasdy T (eds): Renal Biopsy Interpretation. New York, NY, Churchill Livingstone, 1996, pp 1-19 23. Heptinstall RH: The development of renal pathology. Am J Kidney Dis 16:568-573, 1990

827 24. Churg J, Sobin LH: Renal Disease: Classification and Atlas of Glomemlar Diseases. Tokyo, Japan, Igaku-Shoin, 1982 25. Emancipator SN: Primary and secondary forms of IgA nephritis, Schonlein-Henoch syndrome, in Heptinstall RH (ed): Pathology of the Kidney. Boston, MA, Little, Brown, 1991, pp 389-476 26. Nath KA: Tubulointerstitial changes as a major determinant in the progression of renal damage. Am J Kidney Dis 20:1-17, 1992 27. Ong ACM, Fine LG: Loss of glomerular function and tubulointerstitial fibrosis: Cause or effect? Kidney Int 45:345351, 1994 28. He J, Yang CW, Peten EP, Liu ZH, Pate1 A, Striker LJ, Striker GE: Collagen and collagenase mRNAs in normal and sclerotic glomertdi: Predictors of progression and response to therapy. Kidney Int 49:S39-43, 1995 (suppl) 29. Cameron JS: Proteinuria and progression in human glomerular diseases. Am J Nephrol 10:81-87, 1990 (suppl 1) 30. Solez K, Axelsen RA, Benediktssin H, Burdick JF, Cohen AH, Colvin RB, Croker BP, Droz D, Dunnill MS, Halloran PF, HLyry P, Jennette JC, Keown PA, Marcussen N, Mihatsch MJ: Morozumi K, Myers BD, Nast CC, Olsen S, Racusen LC, Ramos EL, Rosen S, Sachs DH, Salomon DR, Sanfilippo F, Verani R, von Willebrand E, Yamaguchi Y: International standardization of criteria for the histologic diagnosis of renal allograft rejection: The Banff working classification of kidney transplant pathology. Kidney Int 44:41 l422, 1993 31. Takemura T, Okada M, Akano N, Murakami K, Hino S, Yagi K, Takekoshi Y, Yoshioka K: Proto-oncogene expression in human glomerular diseases. J Path01 178:343351, 1996 32. Floege J, Johnson RJ, Gordon K. Iida H, Pritzl P, Yoshimura A; Campbell C, Alpers CE, Couser WG: Increased synthesis of extracellular matrix in mesangial proliferative nephritis. Kidney Int 40:477-488, 1991 33. Takemura T, Murakami K, Miyazato H, Yagi K, Yoshioka K: Expression of fas antigen and bcl-2 in human glomerulonephritis. Kidney Int 48:1886-1892, 1995 34. Makino H, Kashihara N, Sugiyama H, Sekikawa T, Ota Z: Role of apoptosis in the progression of glomertdosclerosis. Contrib Nephrol 118:41-47, 1996 35. Alpers CE, Hudkins KL, Gown AM, Johnson RJ: Enhanced expression of “muscle-specific” actin in glomerulonephritis. Kidney Int 41:1134-1142, 1992 36. Abboud HE: Growth factors in glomerulonephritis. Kidney Int 43:252-267, 1993 37. Emancipator SN, Sedor JR: Cytokines in renal diseases, in Remick DG, Kunkel SL (eds): Cytokines in Health and Disease. Boston, MA, Marcel Dekker, 1992, pp 476-488 38. Hooke DH, Gee DC, Atkins RC: Leukocyte analysis using monoclonal antibodies in human glomerulonephritis. Kidney Int 31:964-972, 1987 39. Shikata K, Makino H, Morioka S, Kashitani T; Hirata K, Ota Z, Wada J, Kanwar YS: Distribution of extracellular matrix receptors in various forms of glomerulonephritis. Am J Kidney Dis 25:680-688, 1995

40. Rabb HA: Cell adhesion molecules and the kidney. Am J Kidney Dis 23:155-166, 1994 41. Kurihara H, Anderson JM, Farquhar MG: Increased Tyr phosphorylation of ZO-1 during modification of tight junctions between glomerular foot processes. Am J Physiol 268:F514-F524, 1995 42. Glassock R, Adler S, Cohen A: Primary glomemlar disease, in Brenner EB (ed): The Kidney (ed 5). Philadelphia, PA, Saunders, 1996, pp 1419-1421 43. Lee GSL, Glassock R: Immunoglobulin A nephropathy, in Ponticelli C, Glassock R (eds): Treatment of Primary

Glomerulonephritis. New York, NY, Oxford University Press, 1997, pp 186-217 44. Toshida Y, Mitarai T, Kawamura T, Kitajima T, Kanai T, Miyazaki Y, Kawaguchi Y, Kubo H, Nagasawa R, Ichikawa I, Sakai 0: Role of the deletion polymorphism of the angiotensin converting enzyme gene in the progression and therapeutic responsiveness of IgA nephropathy. J Clin Invest 86:2126-2169, 1995 45. Woo KT, Lau YK, Lee GSL, Wong KS, Wei SS, Chiang GSC, Lim C: Pattern of proteinuria in IgA nephropathy. Nephrology 3:31-34, 1997

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.