An international evidence-based classification of IgA nephropathy: the Oxford Classification

An international evidence-based classification of IgA nephropathy:

the Oxford Classification

Ian Roberts

Oxford, UK

IgA nephropathy

• Biopsy is usually performed at or near to the time of renal presentation

• Roles: Diagnosis

Renal prognosis

Therapeutic decisions• Light microscopical changes are highly variable

IgA nephropathy is heterogenous

Prognostic information within a biopsy

• Active lesions (necrosis, inflammation, proliferation) – potentially reversible

• Chronic lesions (global glomerulosclerosis; tubular atrophy, interstitial fibrosis) – largely irreversible

• Which histological lesions are relevant/useful in clinical practice?

• How should the histological changes be expressed?

Clinicopathological correlation in IgA nephropathy

Reference mesangial endocapillary crescents capillary wall focal seg glomerulo- interstitial fibrosis/

severity proliferation IgA lesions sclerosis tubular atrophy

Nozawa et al, 2005X

Ballardie et al, 2002 X

To et al, 2000 X

Mera et al, 2000 X

Daniel et al, 2000 X

Vleming et al, 1998 X

Freese et al, 1998 X X X

Hogg et al, 1994 X X

Katafuchi et al, 1994 X X

Ibels et al, 1994 X X

Okada et al, 1992 X X

Bogenschutz et al, 1990 X

Rekola et al, 1989 X

D'Amico et al, 1986 X X X

Boyce et al, 1986 X

Which classification schema do nephropathologists use?

• Renal Pathology Society survey 2006

In reporting a diagnosis of IgA nephropathy, do you use a specific histologic classification system? 63% No37% Yes – 5 different schemas

most popular Haas (14% of total respondents)

Do you think that a more universal histologic classification system for IgA nephropathy would be worthwhile?94% Yes (if it can be demonstrated to be of clinical value)

Histological classification of IgA nephropathy

• Is important for clinical trials, comparison of different studies, communication between pathologists and nephrologists, management of individual patients…

• A classification schema must be evidence-based, clinically relevant, simple, precise in its definitions and reproducible.

Traditional approach to development of new histological classifications in renal pathology:

1. One individual or a group of experts publish a classification

2. Studies attempt to validate the classification and test reproducibility

3. Subsequent revision or rejection of the classification

Define the histological lesions, test the reproducibility, collect evidence from a clinicopathological study…..

Then publish.

Towards a new clinicopathological classification of IgA nephropathy: the

Oxford classification

NephrologistsJonathan Barratt [UK]Francois Berthoux [France]Roberta Camilla [Italy]Daniel Cattran [Canada]Rosanna Coppo [Italy]Beppe D’Amico [Italy]John Feehally [UK]Ron Hogg [USA]Stephen Hsu [USA] Bruce Julian [USA]Tetsuya Kawamura [Japan]Philip Li [Hong Kong]Paolo Schena [Italy]Stephan Troyanov [Canada]

PathologistsVivette d’Agati [USA]Charles Alpers [USA]Stephen Bonsib [USA]Jan Bruijn [Netherlands]Terry Cook [UK]Steven Emancipator [USA]Franco Ferrario [Italy]Sandrine Florquin [Netherlands]Agnes Fogo [USA] Hermann-Josef Groene [Germany]Mark Haas [USA]Andrew Hertzenberg [Canada]Prue Hill [Australia]Charles Jennette [USA]Kensuke Joh [Japan]Fernand Lai [Hong Kong]Ian Roberts [UK]Patrick Walker [USA]Jan Weening [Netherlands]

A Working Group of the International IgA Nephropathy Network & Renal Pathology Society Nov 2004 First meeting of International working group at ASN meeting. April 2005 Workshop, Magdalen College, OxfordApril 2006 Pathology review meeting, USCAP meetingApril 2008 Final workshop, Magdalen College

Overall strategy for development of evidence-based consensus classification of IgAN

Agree pathological definitions

Agree pathological scoring process

Agree clinical dataset

Select patient cohorts

Agree data collection process and data verification

Data collection

Data analysis to identify elements with

prognostic predictive power

Refine and agree aclinico-pathological classification

Publish

Achieve international usage Test on further patient cohorts

Health warning!

The clinicopathological study:

Retrospective study Selected patient groupNot controlled for immunosuppression or other therapy

Requires validation in other patient groups and in prospective studies

The Study Group

Inclusion criteria:Biopsy proven IgANStatus within 3 months of biopsy

Proteinuria > 0.5g/24hr; children [age < 18 yrs] ≥ 0.5g/24hr /1.73m2Estimated GFR [eGFR] > 30ml/min/1.73m2 at presentation [eGFR calculated using 4 variable MDRD equation in adults or Schwartz formula in children]

Sufficient clinical data within 3 months of biopsy Sequential follow-up data at least annually for >3 years from biopsy

Retrospective recruitment, aiming for:

300 cases comprising 250 adults and 50 children.

15 centres in 11 countries in 4 continents collaborated

Patients who had received a range of different antihypertensive agents and different immunosuppressive treatment schedules were included.

Primary clinical end point: Rate of loss of renal function.

Secondary end points: ESRD/50% loss of renal function by end of follow-up

Demographics/Clinical Laboratory data

Date of birth Serum creatinine

Ethnicity Estimated GFR

Gender Serum albumin

Date of first clinical presentation Serum cholesterol

Date of renal biopsy Serum triglycerides

Presenting clinical featuresMacroscopic haematuriaAsymptomatic microhaematuria Asymptomatic microhaematuria and proteinuriaNephrotic syndromeAcute renal failureCKD Stage 3[eGFR< 60mL/min]

Urine protein excretion

Height Microscopic haematuria0-3+ on dipstick analysis

Weight Medications – before biopsy

Body Mass Index Number of blood pressure medications prescribed

Smoker ACE inhibitorsAngiotensin receptor blockersNon dihydropyridine calcium channel blockersDihydropyridine calcium channel blockersStatinsDiureticsFish oilCorticosteroidsCyclophosphamideMycophenolateAzathioprinePlatelet aggregation inhibitors

Systolic blood pressure Tonsillectomy

Diastolic blood pressure

Clinical dataset (data collected within 3 months of biopsy)

Follow up dataset (data collected on at least annually over >3 years)

Demographics/Clinical Medications – before biopsy

Height Number of blood pressure medications prescribed

Weight ACE inhibitorsAngiotensin receptor blockersNon dihydropyridine calcium channel blockersDihydropyridine calcium channel blockersStatinsDiureticsFish oilCorticosteroidsCyclophosphamideMycophenolateAzathioprinePlatelet aggregation inhibitors

Body Mass Index

Smoker

Systolic blood pressure

Diastolic blood pressure

Laboratory data

Serum creatinine

Estimated GFR

Serum albumin

Serum cholesterol

Serum triglycerides

Urine protein excretion

Microscopic haematuria0-3+ on dipstick analysis

Tonsillectomy

The Study Group

Exclusion criteria:

Henoch-Schȍnlein purpura, diabetes mellitusProteinuria , < 0.5g/24hr; children [age < 18 yrs] < 0.5g/24hr /1.73m2

eGFR < 30ml/min/1.73m2 at presentationInsufficient clinical data within 3 months of biopsy or on follow upEnd-stage renal disease < 12 months from time of biopsyBiopsy containing <8 glomeruli

Started with 320 patients 265 patients with adequate biopsies and full histological and clinical dataset.

Adults Children[age < 18yrs at biopsy]

Total 265 206 59

Asia 48 14

China Beijing 12 2

Hong Kong 9 1

Nanjing 7 1

Japan Tokyo 19 1

Wakayama 1 9

Europe 73 21

France St Etienne 23 1

Italy Bari 23 1

Milano 16 3

Roma - 9

Torino 3 7

United Kingdom Glasgow 8 -

Americas 82 24

Canada Toronto 32 0

United States Birmingham 12 1

Mayo Clinic 14 4

South West Study Group 24 19

South America 3 0

Chile Santiago 3 0

Pathology dataset

Histological lesions were defined by consensus at initial Oxford meeting in 2005.

Provisional analysis of the first 40 cases, to identify areas of high inter-observer variation were identified.

In order to improve reproducibility, definitions were refined at a meeting of pathologists in Atlanta in 2006

Philosophy:

Keep an open mind on which histological lesions are important.

Collect as much information as possible – simplify only after data analysis

Pathology dataset

IgA nephropathy:

IgA nephropathy in the native kidney is defined as dominant or co-dominant staining with IgA in glomeruli by immunofluorescence or immunoperoxidase. Not all glomeruli need show this positivity. SLE-related nephritis should be excluded. The intensity of IgA staining should be more than trace. The distribution of IgA staining should include presence in the mesangium, with or without capillary loop staining, excluding a pure membranous, diffuse, global granular GBM staining pattern or linear GBM staining pattern. IgG and IgM may be present, but not in greater intensity than IgA, except that IgM may be prominent in sclerotic areas. Complement C3 may be present. The presence of C1q staining in more than trace intensity should bring up consideration of lupus nephritis.

Pathology datasetGlomerular definitions:Diffuse: a lesion involving most (≥ 50%) glomeruli.Focal: a lesion involving <50% of glomeruli.Global: a lesion involving more than half of the glomerular tuft (NB see below for definitions of segmental and global sclerosis).Segmental: a lesion involving less than half of the glomerular tuft (i.e. at least half of the glomerular tuft is spared). N.B. see below for definitions of segmental and global sclerosisEndocapillary hypercellularity: hypercellularity due to increased number of cells within glomerular capillary lumina causing narrowing of the lumina.Karyorrhexis: presence of apoptotic, pyknotic and fragmented nuclei.Necrosis: is defined by (i) disruption of the glomerular basement membrane with (ii) fibrin exudation and (iii) karyorrhexis. At least two of these three lesions need to be present to meet the criteria for necrosis. (*2008 amendment: Necrosis should not be scored on the PAS-stained section alone; fibrin is more easily identified on H&E or MSB-stained sections and breaks in the glomerular basement membrane are more easily identified on silver-stained sections. A minimum requirement for the definition of a necrotising lesion is extracapillary fibrin exudation.)GBM duplication: a double contour of the GBM with or without endocapillary hypercellularity. Increased mesangial matrix: an increase in the extracellular material in the mesangium such that the width of the interspace exceeds two mesangial cell nuclei in at least two glomerular lobules.Sclerosis: obliteration of the capillary lumen by increased extracellular matrix, with or without hyalinosis or foam cells. An adhesion: an area of continuity between the glomerular tuft and Bowman's capsule separate from an extracapillary lesion or area of segmental sclerosis.Segmental sclerosis: any amount of the tuft involved with sclerosis, but not involving the whole tuft. Global sclerosis: the entire glomerular tuft involved with sclerosis. Collapsed/ischaemic glomerulus: A glomerulus showing collapse of the capillary tuft with or without thickening of Bowman’s capsule and fibrosis in Bowman’s space

Pathology datasetExtracapillary lesions subclassified as follows: Extracapillary proliferation or cellular crescent: extracapillary cell proliferation of more than two cell layers with >50% of the lesion occupied by cells. It is further classified by the percentage of glomerular circumference involved <10%, 10-25%, 26-50%, >50%.Extracapillary fibrocellular proliferation or fibrocellular crescent: an extracapillary lesion comprising cells and extracellular matrix, with <50% cells and <90% matrix. This is further classified by the percentage of the glomerular circumference involved <10%, 10-25%, 26-50%, >50%.Extracapillary fibrosis or fibrous crescent: >10% of the circumference of Bowman's capsule covered by a lesion composed of >90% matrix. It is further classified by the percentage of the glomerular circumference involved 10-25%, 26-50%, >50%. Ischaemic, obsolescent glomeruli should be excluded.A crescent is one of these extracapillary lesions which involves >10% of the circumference of Bowman’s capsule.

Mesangial hypercellularity is subclassified as follows: If <4 mesangial cells/mesangial area = normal,4-5 mesangial cells/mesangial area = mild mesangial hypercellularity,6-7 mesangial cells/mesangial area = moderate mesangial hypercellularity,8 or more mesangial cells/mesangial area = severe mesangial hypercellularity. Note: This is scored for each glomerulus by assessing the most cellular mesangial area. Mesangial areas immediately adjacent to the vascular stalk should not be scored. Individual mesangial areas showing hypercellularity are separated by areas of narrowing to the width of less than 2 mesangial cell nuclei (ie count clusters, not files of mesangial cell nuclei).

Pathology datasetTubulointerstitial definitions:Tubular atrophy: is defined by thick irregular tubular basement membranes with decreased diameter of tubules. It is scored according to the percent of cortical area involvement with 1-5% rounded to 5% and other values rounded to the closest 10%.Interstitial fibrosis: is defined as increased extracellular matrix separating tubules in the cortical area. It is scored as percentage involvement with 1-5% rounded to 5% and other values rounded to the closest 10% Interstitial inflammation: is defined as inflammatory cells within the cortical interstitium in excess. It is scored as percentage involvement with 1-5% rounded to 5% and other values rounded to the closest 10%. It should be noted whether the inflammation is confined to areas of interstitial fibrosis or not.Additional tubular lesions are noted as follows: The presence of numerous red blood cells, defined as tubules completely filled with red blood cells with or without casts, is noted as a lesion when it involves >20% of tubules. Acute tubular injury of the proximal tubular epithelium is defined by simplification of the epithelium without tubular basement membrane thickening. Vascular definitions:Arterial lesions are scored based on the most severe lesions. Interlobular and larger arteries are scored separately. An interlobular artery is one surrounded by cortex; an arcuate artery is one at the corticomedullary junction. Intimal thickening is scored by comparing the thickness of the intima to that of the media in the same segment of vessel. Score the intima variously as normal, and thickened to more or less than the thickness of the media.Arteriolar hyaline is noted as the proportion of arterioles affected (0, 1-25%, 26-50%, >50%).

Pathology dataset

Pathology dataset

The slide circulation:Histological score sheet with instructions agreed at Atlanta 2006.Batches of 5 cases circulated between 5 pathologists in a rolling fashion – no two batches scored by the same pathologists.Glomerular lesions scored on a circled PAS-stained section.

Aims:Median scores taken for clinicopathological correlations, to achieve robust quantitative/semiquantitative measure of histological changes, independent of a single pathologist’s idiosyncrasies.Measure interobserver variation – exclude histological lesions with poor reproducibility.

Circulation 2 Last date by which these slides should be sent on to the next pathologist on the list

Pathologist Pathologist No. 03-Jan-07 29-Jan-07 26-Feb-07 19-Mar-07 9-Apr-07 30-Apr-07I Roberts 1 Cases sent out 96-100 191-195 186-190 181-185 Return 176-180 to OxfordF Lai 2 Cases sent out 106-110 96-100 191-195 186-190 Return 181-185 to OxfordA Herzenberg 3 Cases sent out 111-115 106-110 96-100 191-195 Return 186-190 to OxfordA Fogo 4 Cases sent out 116-120 111-115 106-110 96-100 Return 191-195 to OxfordS Bonsib 5 Cases sent out 121-125 116-120 111-115 106-110 Return 96-100 to OxfordJ Weening 6 Cases sent out 126-130 121-125 116-120 111-115 Return 106-110 to OxfordS Florquin 7 Cases sent out 131-135 126-130 121-125 116-120 Return 111-115 to OxfordH-J Grone 8 Cases sent out 136-140 131-135 126-130 121-125 Return 116-120 to Oxford C Alpers 9 Cases sent out 141-145 136-140 131-135 126-130 Return 121-125 to OxfordP Walker 10 Cases sent out 146-150 141-145 136-140 131-135 Return 126-130 to OxfordJ Bruijn 11 Cases sent out 151-155 146-150 141-145 136-140 Return 131-135 to OxfordF Ferrario 12 Cases sent out 156-160 151-155 146-150 141-145 Return 136-140 to OxfordS Emancipator 13 Cases sent out 161-165 156-160 151-155 146-150 Return 141-145 to OxfordM Haas 14 Cases sent out 166-170 161-165 156-160 151-155 Return 146-150 to OxfordC Jennette 15 Cases sent out 171-175 166-170 161-165 156-160 Return 151-155 to OxfordV D'Agati 16 Cases sent out 176-180 171-175 166-170 161-165 Return 156-160 to OxfordK Joh 17 Cases sent out 181-185 176-180 171-175 166-170 Return 161-165 to OxfordP Hill 18 Cases sent out 186-190 181-185 176-180 171-175 Return 166-170 to OxfordT Cook 19 Cases sent out 191-195 186-190 181-185 176-180 Return 171-175 to Oxford

Global slide circulation

Column A I SN/ RPS I gA nephropathy Working Group score sheetMesangial cell hypercellularity Total Mesangial scoreno hypercellularity (0) Case Numbermild (1) (4-5 cells) Scorermoderate (2) (6-7 cells) Date severe (3) (≥8 cells)Total number of scorable glomeruli A B Mean mesangial score (B divided by A)Indeterminate Indeterminate mesangial cellularity due to: TotalTotal number of glomeruli Global sclerosis/advanced segmental sclerosis

Column B Global endocapillary hypercellularityNormal glomerulus Retracted glomerular tuft (ischaemic/collapse)Segmental sclerosis Incomplete mesangial areaAdhesion Crescent only (type in col. B)Ischaemia/collapse Column CEndocapillary hypercellularity Mesangial matrix expansion out of proportion to cellularitySegmental Tubular atrophy (%) score 0%, 1-5% as 5%, >5% to nearest 10%Global Interstitial fibrosis (%) score as for tubular atrophyGBM duplication Interstitial inflammation (%) score as for tubular atrophyNecrosis check in scarred areas onlyExtracapillary lesions - cellular in scarred and non-scarredTiny focus (<10%) Arteriosclerosis interlobular arteries None presentCrescent (10-25%) No intimal thickeningCrescent (26-50%) intima thickened and < thickness of mediaCrescent (>50%) intima thickened and > thickness of mediaExtracapillary lesions - fibrocellular arcuate arteries & larger None presentTiny focus (<10%) No intimal thickeningCrescent (10-25%) intima thickened and < thickness of mediaCrescent (26-50%) intima thickened and > thickness of mediaCrescent (>50%) Arteriolar hyalinosis absentExtracapillary lesions - fibrous 1-25% of arterioles presentCrescent (10-25%) 26-50% of arterioles presentCrescent (26-50%) >50% of arterioles presentCrescent (>50%) Other

Total Glom

Mesangial 1 Median mesangial score

Mesangial 2 % scorable glomeruli showing severe mesangial hypercellularity (median of group)

Global GS % of total glomeruli showing global sclerosis or retracted glomerular tuft (median of group)

Normal glomeruli % of total glomeruli noted as normal (median of group)

Segmental GS % of total glomeruli showing segmental sclerosis (median of group)

Adhesion % of total glomeruli showing adhesions (median of group)

Endocapillary 1 % of total glomeruli showing segmental endocapillary hypercellularity (median of group)

Endocapillary 2 % of total glomeruli showing segmental + global endocapillary hypercellularity (median of group)

GBM duplication % of total glomeruli showing GBM duplication (median of group)

Necrosis % of total glomeruli showing necrosis (median of group)

Extracapillary 1 % total glomeruli showing cellular crescents (median)

Extracapillary 2 % total glomeruli showing cellular + fibrocellular crescents (median)

Extracapillary 3 mean cellular + fibrocellular crescent score (median of group)

Extracapillary 4 % total gloms showing fibrous crescents (median)

Extracapillary 5 mean fibrous crescent score (median of group)

Tubular atrophy % of cortex showing tubular atrophy (median of group)

Interstitial fibrosis % of cortex showing interstitial fibrosis (median of group)

Interstitial inflammation 1 % of cortex showing interstitial inflammation (median of group)

Interstitial inflammation 2% of cortex showing interstitial inflammation if majority (3 or more) checked scarred and non-scarred. Score as

0 if majority checked scarred areas only. Scarred only 0; Scarred and non-scarred 1

Arterial 1 Median arcuate artery score. Leave blank if none present.

Arterial 2 Median interlobular artery score. Leave blank if none present.

Arterial 3 Median artery score - worst of arcuate and interlobular. Leave blank if none present.

Arteriole 1 Absent=0; Present=1. Take majority verdict

Arteriole 2 Median arteriolar hyalinosis score

Histol No. Mesang1 Mesang2 Global GS Norm glomSeg GS Adhesion Endocap1 Endocap2 GBM dup Necrosis Extracap1 Extracap2 Extracap35 1.3 10 17 0 27 7 0 0 0 0 7 29 0.75

Total glom29 1.3 10 17 0 27 7 0 0 0 0 7 29 0.7530 1.3 22 17 0 23 10 10 10 10 0 7 13 0.3329 1.6 19 24 0 34 28 0 0 0 0 10 34 0.930 1.3 0 10 33 27 3 0 0 0 0 10 33 0.8728 0.65 0 11 11 39 4 0 0 0 0 7 14 0.3628 1 10 21 0 14 7 11 0 0 0 7 29 0.75

Extracap3 Extracap4 Extracap5 Tub atrophyInt fibrosis Int inflam1 Int inflam2 Arterial1 Arterial2 Arterial3 Arteriole1 Arteriole20.75 7 0.11 20 20 10 0 0 0 0 0

0.75 7 0.11 20 20 10 0 0 00.33 3 0.03 20 20 0 0 0 1 10.9 28 0.62 20 15 15 1 0 0 0 0

0.87 3 0.1 20 20 20 0 0 0 0 0 00.36 21 0.5 40 50 10 0 0 0 0 00.75 7 0.11 30 30 0 0 0 0 0

Study No. Histol No. Mesang1 Mesang2 Global GS Norm glomSeg GS Adhesion Endocap1 Endocap2 GBM dup Necrosis Extracap1 Extracap2501 1 0.89 0 0 14 11 11 0 0 0 0 0 0502 2 0.69 0 0 15 0 0 0 0 0 0 0 0503 3 0.2 0 40 40 17 0 0 0 0 0 0 0504 4 0.14 0 21 36 0 0 0 0 0 0 0 0505 5 1.3 10 17 0 27 7 0 0 0 0 7 29506 6 0.6 0 19 62 7 0 0 0 0 0 0 5507 7 1 8 7 50 0 0 0 0 0 0 0 0508 8 0.84 0 2 79 0 6 0 0 0 0 0 0509 9510 10 1.5 8 40 0 17 13 0 0 0 0 0 0511 11 0.8 0 30 20 18 22 0 0 0 0 0 0512 12 0.6 0 30 25 20 11 0 0 0 0 0 0513 13 0.09 0 0 77 0 0 0 0 0 0 0 0514 14 1.13 13 0 0 0 0 9 9 0 0 0 9515 15 1.5 8 0 21 0 0 0 0 0 0 0 0516 16 1.32 0 0 13.5 10 10 0 0 0 0 9 15517 17 0.55 0 21 50.5 0 4.5 0 0 0 0 0 0518 18 0.91 0 30 19.5 0 2 5.5 5.5 0 0 0 0519 19 0.6 0 11 42 1 4.5 1 1 0 0 0 0520 20 1.13 5.5 28 10.5 5 12.5 0 0 0 0 0 0521 21 0.67 0 5 42.5 5 23 0 0 0 0 0 0522 22 2.2 50 0 0 13 16 0 0 0 0 0 0

23 0.73 0 25.5 13.5 0 5 0 0 0 0 0 0524 24 0.8 3 1 33 0 0 0 0 0 0 0 0

25 1.5 8 16 0 0 11 31 39 0 0 8 29526 26 0.6 0 0 8 0 0 0 0 0 0 0 0527 27 1.12 6 3 3 21.5 7 0 0 0 0 0 11528 28 0.9 8 4 0 0 4 0 0 0 0 0 0529 29 1.7 0 67 0 6 0 0 0 0 0 0 0530 30 0.8 0 10 0 13 10 0 0 0 0 0 10531 31 2.24 47 0 0 6 11 18 24 11 0 0 0

32533 33 1.4 14 0 0 11 11 0 0 0 0 0 0534 34 1.5 40 43 0 29 25 0 0 14 0 0 0

35 0 0 0 0 0 0 0 0 0 0 0 0

• 83% of the cases were scored by four or more pathologists, 17% by three pathologists.

• Median 18 glomeruli per biopsy. • Some lesions were seen infrequently: necrosis was seen in only 6

cases and GBM duplication in 30 cases.

How to simplify in to a smaller number of lesions/patterns to

include in a classification?

Selection of which pathological variables to include in the final classification based on:

• Reproducibility

• Independence from other lesions

• Clinical correlation

Definition ICC

Mesangial 1 Mesangial score 0.63

Mesangial 2 % scorable glomeruli showing severe mesangial hypercellularity 0.54

Global GS % of total glomeruli showing global sclerosis or retracted glomerular tuft 0.89

Normal glomeruli % of total glomeruli noted as normal 0.28

Segmental GS % of total glomeruli showing segmental sclerosis 0.50

Adhesion % of total glomeruli showing adhesions 0.21

Seg GS & adhesion Summation of %segmental GS and adhesion 0.49

Endocapillary 1 % of total glomeruli showing segmental endocapillary hypercellularity 0.35

Endocapillary 2 % of total glomeruli showing segmental + global endocapillary hypercellularity 0.57

GBM duplication % of total glomeruli showing GBM duplication 0.11

Necrosis % of total glomeruli showing necrosis 0.31

Extracapillary 1 % total glomeruli showing cellular crescents 0.62

Extracapillary 2 % total glomeruli showing cellular + fibrocellular crescents 0.63

Extracapillary 3 mean cellular + fibrocellular crescent score 0.65

Extracapillary 4 % total gloms showing fibrous crescents 0.30

Extracapillary 5 mean fibrous crescent score 0.33

Tubular atrophy % of cortex showing tubular atrophy 0.76

Interstitial fibrosis % of cortex showing interstitial fibrosis 0.73

Interstitial inflammation 1 % of cortex showing interstitial inflammation 0.54

Interstitial inflammation 2 Inflammation if involving scarred and non-scarred cortex 0.09

Arterial 1 Arcuate artery score 0.79

Arterial 2 Interlobular artery score. 0.69

Arterial 3 Artery score - worst of arcuate and interlobular* 0.71

Arteriole 1 Absent=0; Present=1. Take majority verdict 0.36

Correlations between pathology variables.

“R” values (correlation coefficients) Statistically significant correlations were determined using the Holm-Bonferroni method to minimize the probability of making a type I statistical error.

mesang GSC endocap Extracap Interstitium vessels

mes1 Mes2 GlobGS SeGS Adh End1 End2 GBMdup Necr Extr1 Extr2 Extr3 Extr4 Extr5 TubAt IntFib Intinfl1 Intinfl2 Art1 Art2 Art3 Artiol1 Artiol2

Mesang1 - 0.7 0.2 0.3 0.3 0.3 0.2 0.2 0.2 0.3

Mesang2 - 0.3 0.3 0.3

Global GS - 0.4 0.2 0.7 0.8 0.6 0.3 0.3 0.4 0.3

Seg GS - 0.5 0.4 0.4 0.3 0.3 0.5 0.5 0.4

Adhesion - 0.3 0.3 0.2 0.2 0.3 0.3 0.3

Endocap1 - 0.99 0.3 0.4 0.5 0.5

Endocap2 - 0.3 0.4 0.5 0.5

GBM dup -

Necrosis -

Extracap1 - 0.7 0.7

Extracap2 - 0.99 0.4 0.4

Extracap3 - 0.4 0.4 0.2

Extracap4 - 0.99

Extracap5 -

Tubatrophy - 0.98 0.9 0.8 0.3 0.3 0.4 0.4

Int fibrosis - 0.9 0.8 0.3 0.3 0.4 0.4

Int inflam1 - 0.3

Int inflam2 -

Arterial1 - 0.8 0.9 0.6

Arterial2 - 0.9 0.5 0.5

Arterial3 - 0.5 0.5

Arteriole1 - 0.95

Arteriole2 -

Pathology variables selected for inclusion in the clinicopathological analysis

Mesangial cellularity score

Segmental glomerulosclerosis/adhesion

Endocapillary hypercellularity (segmental + global)

Cellular/fibrocellular crescents

Tubular atrophy/interstitial fibrosis

Arterial score

Can these histological lesions add value to clinical variables (at the time of biopsy and follow-up) in predicting outcome?

Can a change in a biopsy predict what will happen to renal function years later?

Clinical characteristics at time of biopsy and follow-up (n=265) At time of biopsy Follow=up

Age (years) 32 ± 15 Duration of follow-up (mo) 69 (12*-268) Female 28% MAP (mmHg) 95 ± 11 Ethnicity No. of antihypertensive drugs 0.9 (0-4.7) Caucasian/African/Asian/Other 66, 3, 27, 4% Use of RAS blockade % (ACEi, ARB) 75 (68, 22) BMI 25 ± 5 Proteinuria (g/day)* 1.4 (0.1-9.3) MAP (mmHg) 98 ± 18 Serum albumin 40 ± 6 % taking bp meds (% RAS blockade) 31 (23) Serum cholesterol (mmol/L) 5.3 ± 1.2 GFR (ml/min/1.73m2) 82 ± 36 Serum triglycerides (mmol/L) 1.9 ± 1.0 % Stage I, II, III, IV 35, 49, 22, 4 Immunotherapy (any form) 26% Proteinuria (g/day)* 1.7 (0.5-18.5) Prednisone 26% Previous macroscopic haematuria 33% Second line (cyclophosphamide, others) 8% (6%, 4%) Serum albumin (g/L) 38 ± 6 Use of fish oil 16% Serum cholesterol (mmol/L) 5.5 ± 1.5 Use of statins 13% Serum triglycerides (mmol/L) 1.9 ± 1.6 Tonsillectomy during follow-up 0 Previous immunosuppression 11% Rate of renal function decline (ml/min/1.73m2/y) - 3.5 ± 8.4 Previous use of fish oil (%) 6% 50% decline in renal function 22% Known previous tonsillectomy 6% Renal failure (<15 ml/min/1.73m2) 13%

*proteinuria in children expressed in g/day/1.73m2.

Survival Functions

Time to combine event

120967248240

Cu

m S

urv

ival

1.2

1.0

.8

.6

.4

.2

mes4_0.5_1_1.5

1.5+

1.5+- censored

1- 1.5

1- 1.5- censored

0.5- 1

0.5- 1- censored

<0.5

<0.5- censored

Survival Functions

Time to combine event

120967248240

Cu

m S

urv

ival

1.1

1.0

.9

.8

.7

.6

.5

.4

.3

SEGADH2

>20

>20- censored

<20%

<20%- censored

absent

absent- censored

Survival Functions

Time to combine event

120967248240

Cu

m S

urv

ival

1.2

1.0

.8

.6

.4

.2

0.0

Interstitial fibrosi

severe (>50)

severe (>50)

- censored

moderate (25- 50)

moderate (25- 50)

- censored

mild (<25)

mild (<25)- censored

none

none- censored

What are the clinically relevant cut-offs for continuous quantitative lesions?

MAP p GFR p Proteinuria p mmHg mL/min/1.73m2 g/day

Mesangial score ≤0.5 100 ± 18 >0.1 84 ± 28 >0.1 1.4 (0.6-9.2) 0.001 Mesangial score >0.5 98 ± 17 82 ± 38 2.0 (0.5-18.5)

No segmental GS or Adhesion 94 ± 16 0.04 94 ± 40 0.004 1.5 (0.5-7.2) 0.01 Any segmental GS or adhesion 99 ± 18 79 ± 34 1.9 (0.6-18.5)

No endocapillary hypercellularity 101 ± 18 0.003 76 ± 31 0.002 1.5 (0.5-11.3) 0.008 Any endocapillary hypercellularity 94 ± 16 92 ± 40 2.0 (0.5-18.5)

No crescent 98 ± 17 >0.1 84 ± 37 >0.1 1.5 (0.5-18.5) 0.002 Any crescents 98 ± 18 80 ± 35 2.2 (0.5-12.0)

Tubular atrophy None (0%) 91 ± 17 0.04 109 ± 35 <0.001 1.5 (0.5-7.2) 0.06 Mild (1-25%) 99 ± 18 85 ± 35 1.7 (0.5-18.5) Moderate (26-50%) 99 ± 11 59 ± 17 1.7 (0.6-7.5) Severe (51%+) 105 ± 24 46 ± 27 3.0 (1.1-9.0)

Artery score None 95 ± 17 0.01 91 ± 40 <0.001 1.8 (0.5-18.5) >0.1 Mild 103 ± 15 67 ± 19 1.5 (0.6-4.6) Moderate 101 ± 20 70 ± 25 1.6 (0.8-.7.3) Severe 102 ± 7 72 ± 33 1.7 (1.1-2.2)

Mean ± SD, Median (range),

Correlations between pathology and initial clinical presentation

Therapy received during follow-up*

% with received RAS blockade

p % given Immunosuppression

p

Mesangial score

≤0.5 72 19 >0.5 75 >0.1 30 >0.1

Segmental GS or Adhesion absent 54 <0.001 27 >0.1 present 81 29

Endocapillary hypercellularity absent 76 17 present 73 >0.1 44 <0.001

Crescent absent 72 20 <0.001 present 78 >0.1 39

Tubular atrophy 0-25% 72 >0.1 28 >0.1 26-50% 83 24 >50% 85 50

Artery score absent 69 32 >0.1 Mild 83 0.006* 24 Moderate 86 19 Severe 75 50 * Intend to treat.

Rate of renal function decline Survival from renal failure or a 50% drop in GFR

(Linear regression) (Cox regression)

Multivariate Multivariate

Univariate slope Model A Model B

Univariate Hazard ratio Model A Model B

(ml/min/1.73m2/y) ß ß (95% CI) Mesangial hypercellularity score ≤0.5 -0.5 ± 3.3 0.06 (0.01-0.45) 0.07 (0.01-0.53) 0.11 (0.01-0.80) >0.5 -4.2 ± 9.0

-2.2 -0.8 1 1 1

p<0.001 0.10 p>0.1 p=0.006 p=0.01 p=0.03

Segmental glomerulosclerosis absent -0.5 ± 7.5 1 1 1 present -4.4 ± 8.4 -3.6 -2.5 3.1 (1.4-7.3) 1.8 (0.6-5.3) 2.5 (0.9-7.3) p=0.001 p=0.005 p=0.03 p=0.009 p>0.1 p=0.09

Tubular atrophy/interstitial fibrosis 0-25% -2.5 ± 7.6 1 1 1 26-50% -5.7 ± 8.8 3.5 (1.9-6.5) 6.0 (2.7-13.9) 5.0 (2.3-11.1) >50% -11.1 ± 12.6

-5.2 -3.7

15.5 (7.5-31.9) 17.3 (5.9-50.9) 8.8 (2.9-26.4) p<0.001 p<0.001 p<0.001 p<0.001 p<0.001 p<0.001

Pathology and outcomes

Model A: multivariate - 3 pathological features + initial GFR, MAP, proteinuria. Model B: multivariate - 3 pathological features + initial GFR + follow-up MAP, proteinuria

Interaction of pathological features with immunosuppressive therapy

• Relationship between pathology variables and the rate of renal function decline was not influenced by immunosuppression except for endocapillary lesions.

• Patients with endocapillary proliferation:

Rate of renal function decline + immunosuppression -2.6+/-5.1 ml/min/1.73m2 /yr no immunosuppression -5.4+/-11.1 ml/min/1.73m2 /yr

p=0.006

Interaction of pathological features with age and ethnicity

• Younger patients:Presented with significantly less segmental glomerulosclerosis, tubular atrophy/interstitial fibrosis and vascular lesions, but had significantly more endocapillary lesions. The predictive value of each pathology variable on the rate of renal function decline was not influenced by the age at the time of biopsy (p>0.1 for interaction term).

• Ethnicity (Caucasian vs Asian patients):For each pathology variable, the interaction term with ethnicity was not statistically significant except for endocapillary lesions (p=0.02); the rate of renal function decline in Asian subjects was significantly better compared to Caucasians. But - Asian patients were significantly more likely to receive immunosuppressive therapy during follow-up (42% compared to 22% in Caucasians, p=0.002).

Recommendations for the pathology report

Minimum prognostic data:

Glomerular “pattern”: Mesangial hypercellularity in > or <50% of glomeruli (M 0/1)Endocapillary hypercellularity – present/absent (E 0/1)Segmental sclerosis/adhesions – present/absent (S 0/1)Tubular atrophy/interstitial fibrosis – 0-25%, 26-50%, >50% (T 0/1/2)

In addition: Total number of glomeruliEndocapillary proliferation - %Cellular/fibrocellular crescents - %Necrosis - %Global glomerulosclerosis - %

Example summary line: There is an IgA nephropathy showing diffuse mesangial proliferation with focal segmental sclerosis and moderate chronic tubulointerstitial damage (M1,E0,S1,T1)

Slope:

ml/min/1.73m2/yr

Minimal mesangial without segmental sclerosis M0,E0,S0 12 0.7 ± 2.5

with segmental sclerosis M0,E0,S1 22 -1.5 ± 2.7

Mesangial hypercellularity without segmental sclerosis M1,E0,S0 31 -2.2 ± 4.3

with segmental sclerosis M1,E0,S1 88 -4.7 ± 7.6

Endocapillary proliferation without segmental sclerosis M0/1,E1,S0 21 1.2 ± 1.2

with segmental sclerosis M0/1,E1,S1 90 -4.9 ± 10.0

CriteriaNo. of

patients

Prognostic impact of combinations of histological lesions

Combining glomerular patterns:

Slope:

ml/min/1.73m2/yr

Minimal mesangial ≤25% M0,E0,T0 30 -0.6 ± 3.0

> 26% M0,E0,T1-2 5 -1.0 ± 1.2

Mesangial hypercellularity ≤25% M1,E0,T0 89 -2.7 ± 5.5

> 26% M1,E0,T1-2 30 -7.9 ± 9.1

Endocapillary proliferation ≤25% M0/1,E1,T0 88 -3.0 ± 1.9

> 26% M0/1,E1,T1-2 23 -6.9 ± 1.2

Glomerular lesions TA/IF CriteriaNo. of

patients

Prognostic impact of combinations of histological lesions

Combining glomerular and tubulointerstitial lesions:

Why not a classification? (eg. class I, class II, etc)

How should the histological data be combined with clinical indices?

Because the data does not support this approach – the MEST lesions are independent predictors of outcome and the relative risks can not be simply summed.

?

This work was supported by: International Society of NephrologyKidney Research UK

Vivor Pharma Aspreva plc