Revision of the ISN/RPS classification for lupus nephritis: of … · 2019-02-16 · For Peer Review Only 4 1 On May 9-11th 2016, a Working Group for Lupus Nephritis Classification

For Peer Review Only

Revision of the ISN/RPS classification for lupus nephritis:

modified NIH activity and chronicity indices and clarification

of definitions

Journal: Kidney International

Manuscript ID KI-08-17-1259.R1

Article Type: Meeting Report

Date Submitted by the Author: 13-Nov-2017

Complete List of Authors: Bajema, Ingeborg; Leiden University Medical Center, Department of Pathology Wilhelmus, Suzanne; Leiden University Medical Center, Pathology Alpers, Charles Bruijn, Jan Colvin, Robert; Mass Gen Hosp, Pathology Cook, Terence; Imperial College London, Medicine D'Agati, Vivette; Columbia University, Pathology Ferrario, Franco; University of Milano-Bicocca, Surgery and Translational Medicine Haas, Mark; Cedars-Sinai Medical Center, Dept of Pathology and Laboratory Medicine; Cedars-Sinai, Jennette, Charles Joh, Kensuke Nast, Cynthia; Cedars-Sinai Medical Center, Pathology Noël, Laure-Hélène Rijnink, Emilie; Leiden University Medical Center, Pathology Roberts, Ian; Oxford University Hospital, Cellular Pathology Seshan, MD, Surya V; Cornell University,New York, pathology Sethi, Sanjeev; Mayo Clinic, Laboratory Medicine and Pathology Fogo, Agnes; Vanderbilt University, Pathology

Keywords: systemic lupus erythematosus, renal pathology

Subject Area: Renal Pathology

The International Society of Nephrology (http://www.isn-online.org/site/cms)

Kidney International


1

Revision of the ISN/RPS classification for lupus nephritis: 1

clarification of definitions, and modified NIH activity and chronicity 2

indices 3

4

Ingeborg M. Bajema1, Suzanne Wilhelmus

1, Charles E. Alpers

2, Jan A. Bruijn

1, Robert B. 5

Colvin3, H. Terence Cook

4, Vivette D. D’Agati

5, Franco Ferrario

6, Mark Haas

7, J. Charles 6

Jennette8, Kensuke Joh

9, Cynthia C. Nast

7, Laure-Hélène Noël

10, Emilie C. Rijnink

1, Ian S.D. 7

Roberts11

, Surya V. Seshan12

, Sanjeev Sethi13

, Agnes B. Fogo14

8

Affiliations: see page 2 9

10

11

Corresponding author 12

Ingeborg M. Bajema, MD PhD 13

Department of Pathology, Leiden University Medical Center 14

PO Box 9600 15

2300 RC Leiden 16

The Netherlands 17

Telephone number: +31 71 526 6621 18

Fax number: + 31 71 526 6952 19

[email protected] 20

21

22

Running title: revision of lupus nephritis classification 23

Key words: lupus nephritis, renal biopsy, systemic lupus erythematosus 24

Word count abstract: 197 25

Word count text (excluding references, figures and tables): 3616 26

Page 1 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


2

Affiliations: 1

2 1Department of Pathology, Leiden University Medical Center, Leiden, the Netherlands 3

4 2Department of Pathology, University of Washington, Seattle, WA, USA 5

6 3Department of Pathology, Massachusetts General Hospital, Harvard Medical School, Boston 7

MA 02114, USA 8

9 4Department of Medicine, Imperial College, London, UK 10

11 5Department of Pathology and Cell Biology, Columbia University Medical Center, New York, 12

New York, USA 13

14 6Nephropathology Center, San Gerardo Hospital-Monza, Milan Bicocca University, Milan, 15

Italy 16

17 7Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, Los 18

Angeles, California USA 19

20 8Department of Pathology and Laboratory Medicine, University of North Carolina at Chapel 21

Hill, Chapel Hill, NC, USA 22

23 9Department of Pathology, Tohoku University Graduate School of Medicine, Sendai, Japan 24

25 10

Department of Pathology, Necker Hospital, Paris, France 26

27 11

Department of Cellular Pathology, Oxford University Hospitals, Oxford, UK 28

29 12

Department of Pathology and laboratory Medicine, Weill Cornell University Medical 30

Center, New York, New York, USA 31

32 13

Department of Laboratory Medicine and Pathology, Mayo Clinic Rochester, MN, USA 33

34 14

Dept. of Pathology, Vanderbilt University School of Medicine, MCN, USA 35

Page 2 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


3

ABSTRACT 1

We present a consensus report pertaining to the improved clarity of definitions and 2

classification of glomerular lesions in lupus nephritis that derived from a meeting of 18 3

members from an international nephropathology working group in Leiden, 2016. Here we 4

report detailed recommendations on issues for which we can propose adjustments based on 5

existing evidence and current consensus opinion (phase 1). New definitions are provided for 6

mesangial hypercellularity and for cellular, fibrocellular and fibrous crescents. The term 7

endocapillary proliferation is eliminated and the definition of endocapillary hypercellularity 8

considered in some detail. We also eliminate the Class IV-S and IV-G subdivisions of Class IV 9

lupus nephritis. The active/chronic (A/C) designations for Class III/IV lesions are replaced by 10

a proposal for activity and chronicity indices which should be applied to all classes. In the 11

activity index, we include fibrinoid necrosis as a specific descriptor. We also make 12

recommendations on issues for which there are limited data at present, and that can best be 13

addressed in future studies (phase 2). We propose to proceed to these investigations, with 14

clinicopathologic studies and tests of inter-observer reproducibility to evaluate the 15

applications of the proposed definitions and to classify lupus nephritis lesions. 16

Page 3 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


4

On May 9-11th 2016, a Working Group for Lupus Nephritis Classification (the Group) met at 1

Leiden University Medical Center to reach a consensus on recently raised issues concerning 2

problems with definitions of lupus nephritis lesions.1 Prior to the meeting, those attending 3

received a questionnaire asking for anonymous suggestions for improving the definitions. 4

The responses served as a starting point for making adjustments to the definitions of lupus 5

nephritis lesions, a process that was further accomplished by group discussions and a multi-6

head microscopy session. The Group decided that consensus had to be reached for any 7

proposed changes and that recommendations should be divided in two types: Phase 1 8

recommendations are clarifying modifications for which we could propose adjustments 9

based on existing published evidence and mutual agreement. Phase 2 recommendations will 10

address issues that can best be validated or modified through an evidence-based process. 11

These include more problematic lesion definitions and adjustments to the lupus nephritis 12

classification. We now report on phase 1 recommendations, and provide a framework for 13

phase 2 issues. 14

Our immediate aim is to improve problematic definitions that form the basis of the 15

lupus nephritis classification and thereby increase the interobserver agreement between 16

nephropathologists worldwide who apply these definitions to classify lupus nephritis. Our 17

eventual goal is to improve the lupus nephritis classification using an evidence-based 18

approach, but refining the definitions for lesions is necessary because they form the 19

essential elements for the classification. Here we describe a plan to proceed in the near 20

future to gather data and, as indicated, modify the lupus nephritis classification. 21

Many renal lesions encountered in lupus nephritis also are present in other renal 22

diseases, providing a rationale for harmonizing definitions for lesions irrespective of the 23

Page 4 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


5

disease context. Depending on the setting, i.e. evaluation of renal biopsies in a clinical 1

setting or for research purposes, different guidelines may apply regarding biopsy 2

requirements. In a clinical setting, it is necessary to obtain as much information as possible 3

from the biopsy by evaluating all stains in all levels and sections and to apply a basic format 4

of the kidney biopsy report. As a general rule, 10 glomeruli for evaluation seem to be 5

appropriate. By studying definitions of frequently occurring lesions as currently formulated 6

(e.g., as in classification systems for IgA nephropathy2-4

and ANCA-associated 7

glomerulonephritis,5 as well as definitions created by the Neptune,

6 CureGN

7 and Banff

8 8

workgroups), we strove for uniform definitions but recognized that certain thresholds may 9

be different among diseases. For example, it remains to be determined (an evidence-based 10

phase 2 issue) which thresholds – for instance for mesangial hypercellularity – have clinical 11

and prognostic value in lupus nephritis, and whether these should be different from those 12

for another disease such as IgA nephropathy. Below, we discuss our modifications of 13

definitions by class. Glomerular, tubulointerstitial and vascular lesions are discussed 14

separately. At this stage, we mainly focus on lesions evaluable by light microscopy, although 15

we do take into account findings by immunofluorescence (IF) and electron microscopy (EM) 16

if they are helpful in the decision-making process. An overview of the alterations to the 17

ISN/RPS lesion definitions and classification9 that we propose is found in Table 1. 18

19

Page 5 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


6

Table 1. Phase 1 recommendations for lupus nephritis classification

1

2

Category Recommendation Comments on ISN/RPS guidelines

Class II Definition for mesangial hypercellularity adjusted: Cuttoff for mesangial hypercellularity unclear

Four or more nuclei fully surrounded by matrix in

the mesangial area not including the hilar region (A)

Class III and IV The term endocapillary proliferation is replaced by Definition for endocapillary proliferation unclear;

endocapillary hypercellularity (B) the term proliferation was considered imprecise

The term crescent is used for a lesion consisting of Extracapillary proliferation involving < 25% of the circumference

extracapillary hypercellularity, composed of a variable mixture of Bowman's capsule was original cutoff. There were no

of cells. Fibrin and fibrous matrix may be present; 10% or more definitions for fibrous or fibrocellular crescents

of the circumference of Bowman's capsule should be involved.

Cellular crescent: more than 75% cells and fibrin

and less than 25% fibrous matrix (C)

Fibrous crescent: more than 75% fibrous matrix

and less than 25% cells and fibrin (D)

Fibrocellular crescent: 25-75% cells and fibrin

and the remainder fibrous matrix (E)

Adhesion: an area of isolated continuity of extracellular matrix There was no definition for an adhesion

material between the tuft and capsule even when the underlying

segment does not have overt sclerosis (F)

Fibrinoid necrosis: fibrin associated with glomerular basement membrane There was no definition for fibrinoid necrosis

disruption and/or lysis of the mesangial matrix; this lesion does not

require the presence of karyorrhexis

Elimination of segmental and global subdivions of class IV Definitions for segmental and global were unclear;

interobserver variability was large; clinical significance uncertain

Modification of the NIH lupus nephritis activity and chronicity scoring Designation of activity/chronicity through A, C and A/C

system (Table 2) to be used instead of the currently used A, C considered too broad and nonspecific; preference for a

and A/C parameters semiquantitative approach to describe active and chronic lesions

Tubulointerstitial lesions Indicate whether interstitial inflammation occurs in presence or Lack of cut-off values for reporting the severity of

absence of interstitial fibrosis tubulointerstitial lesions

Page 6 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


7

Glomerular lesions, Class I – VI: 1

Class I and II 2

An important question was the threshold between class I and class II. We questioned 3

whether the current cutoff for mesangial hypercellularity implies that hypercellularity in 4

merely one mesangial area in one glomerulus in the entire biopsy would suffice. We agreed 5

that mesangial hypercellularity in one area in one glomerulus seems rather low. The 6

appropriate threshold will be investigated in phase 2. In the meantime, we propose 7

increasing the threshold of mesangial hypercellularity from 3 cells/mesangial area to 4 cells, 8

not including the hilar region, in line with the Oxford classification of IgA nephropathy2-4

and 9

specify that mesangial cell nuclei be fully surrounded by matrix. An evidence-based 10

approach to define an appropriate threshold was judged to be necessary. Whether and to 11

what extent mesangial matrix expansion should be incorporated in the definition together 12

with this cell number cut-off level, also needs to be investigated in an evidence-based 13

approach. Of note, only peripheral mesangial areas should be assessed for cellularity, with 14

central and perihilar areas excluded, as described for IgA nephropathy. Importantly, we 15

discussed whether or not hypercellularity within the mesangial zone caused by 16

monocytes/macrophages, lymphocytes, or neutrophils should be considered as mesangial 17

hypercellularity or as endocapillary hypercellularity. This topic will be discussed in more 18

detail below. 19

Class III and IV 20

A substantial amount of discussion centered on Class III and IV lesions. We agreed, as 21

previously noted, that the definitions of endocapillary ‘’proliferation’’ are unclear and 22

inconsistent,1 with issues raised about the types and numbers of cells involved in 23

Page 7 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


8

endocapillary lesions, the definition of lumen reduction, and the specific contribution of 1

endothelial cells. The Group decided that the term ‘’endocapillary proliferation’’ is a 2

misnomer, which should be abandoned and replaced by the term “endocapillary 3

hypercellularity” because most of the hypercellularity in glomerular capillaries in lupus 4

nephritis is caused by influx of inflammatory cells rather than by actual cell proliferation. 5

Phase 2 will address if there should be, for instance, an overall glomerular inflammation 6

score. 7

Endocapillary Hypercellularity 8

Hypercellularity in lupus nephritis may be due to increase in cells in mesangial, endocapillary 9

and/or extracapillary locations. With regard to mesangial hypercellularity, it could be argued 10

that this should be named mesangial hyperplasia in lesions purely consisting of an 11

abundance of mesangial cells (Figure 1), representing lupus nephritis class II lesions. It is 12

unknown if the presence of inflammatory cells in the mesangium indicates a more active 13

lesion; the cut-off values for mesangial hypercellularity and significance of mesangial 14

inflammation remain to be determined in phase 2. Likewise, the cut-off levels for number of 15

inflammatory cells, extent of capillary luminal narrowing and role of endothelial cell swelling 16

need to be defined in a phase 2 exercise. Figure 1 shows ultrastructural features of a single 17

glomerular capillary affected by lupus glomerulonephritis. Inflammatory cells can be in the 18

capillary lumen, beneath endothelial cells in capillary walls, and in the mesangial 19

extracellular compartment. It has to be decided in phase 2 whether endocapillary 20

hypercellularity should encompass all glomerular hypercellularity internal to the capillary 21

wall glomerular basement membrane (GBM) and paramesangial GBM (excluding pure 22

mesangial hyperplasia), or whether it should be restricted to an increase of cells within 23

Page 8 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


9

capillary lumens. Endothelial cell swelling alone was considered insufficient for a lesion to be 1

regarded as representing endocapillary hypercellularity. If endothelial cell swelling is 2

encountered in the absence of inflammatory cells, TMA should be considered, taking into 3

account the number and extent of swollen endothelial cells and whether or not this is 4

accompanied by subendothelial expansion or thrombosis. Relevant thresholds for this 5

situation should be defined more precisely in phase 2. Hypercellularity external to the GBM 6

is extracapillary hypercellularity. 7

Membranoproliferative glomerulonephritis (MPGN) pattern 8

We discussed the value of using the term ‘MPGN pattern’ in relation to the modifications of 9

definitions for lesions within class III and IV lupus nephritis. The group concluded that an 10

MPGN pattern of injury is subsumed in class III/IV lupus nephritis as a form of endocapillary 11

injury. We agreed with the ISN/RPS approach of considering subendothelial deposits that 12

can be seen by light microscopy (i.e. wire loops) and hyaline masses within capillary lumens 13

caused by immune complexes (i.e. hyaline thrombi) as lesions indicative of class III/IV. 14

Determination of the clinical significance of the MPGN pattern and whether or not acute and 15

chronic variants should be distinguished, is a phase 2 exercise. 16

Crescents 17

The term crescent is used for a lesion consisting of extracapillary hypercellularity, composed 18

of a variable mixture of cells. Because some crescents may have predominantly epithelial 19

proliferation, whereas others consist predominantly of monocytes/macrophages, the Group 20

chose the description of a ‘variable mixture of cells’. Fibrin and fibrous matrix may also be 21

present. The ISN/RPS criterion of a crescent involving 25% or more of the glomerular 22

capsular circumference was discussed. It was decided that this threshold should be 10% or 23

Page 9 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


10

more in accord with evidence from the Oxford Classification of IgA nephropathy and 1

standard approaches used in clinical reporting of renal biopsy lesions. Crescents should be 2

composed of more than 2 cell layers in order to distinguish them from apposition of the 3

single layers of hypertrophied visceral and parietal cells. We propose definitions for the 4

distinction of cellular, fibrous and fibrocellular crescents, which were lacking in the ISN/RPS 5

lupus nephritis classification (Table 1). Some glomeruli may have more than one type of 6

crescent. In line with conventional nephropathology practice, extracapillary hypercellularity 7

attributable to concurrent collapsing glomerulopathy lesions should not be designated as 8

crescents. We propose to preserve the term ‘adhesion’ for a lesion characterized by an area 9

of isolated continuity of extracellular matrix material between the tuft and capsule even 10

when the underlying segment does not have overt sclerosis. This is a lesion that is distinct 11

from both crescents and segmental sclerosis, although this differs from the Oxford 12

classification of IgA nephropathy in which such lesions are considered manifestations of 13

segmental sclerosis. How to deal with a lesion characterized by cellular continuity between 14

the tuft and Bowman’s capsule, which is not extensive enough to be called a crescent, needs 15

to be further discussed based on evidence acquired in phase 2. 16

Global/segmental 17

We previously discussed problematic issues concerning the distinction between segmental 18

and global lesions in class III and IV.10

As recently demonstrated in a meta-analysis by Haring 19

et al,11

the clinical importance of distinguishing between segmental and global lesions in 20

class IV as defined by the ISN/RPS Classification System has been questioned. A caveat is that 21

there is poor interobserver agreement among nephropathologists worldwide in determining 22

whether a glomerular lesion in lupus nephritis is segmental or global.10

The latter, in 23

Page 10 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


11

combination with the uncertainty of how to evaluate the combination of intra- and 1

extracapillary lesions into a segmental/global division, may have added greatly to the 2

variability in outcomes of studies addressing the clinical impact of segmental and global 3

lesions. The ISN/RPS approach to subclassifying class IV did not use segmental necrosis as a 4

defining feature of a segmental variant of lupus nephritis, which may have reduced 5

reproducibility and precluded recognition of a pathophysiologically distinct variant of lupus 6

nephritis. Therefore, for phase 1, based on the lack of reproducibility and weak evidence of 7

clinical significance, we propose to eliminate the S and G subdivisions of Class IV. We believe 8

that the distinction between segmental and global lesions should be retained in the 9

microscopic description, as such distinctions still may prove to have clinical value with 10

further study using the modified classification of lupus nephritis. As noted below, segmental 11

fibrinoid necrosis as well as segmental endocapillary hypercellularity will be evaluated. 12

Fibrinoid Necrosis 13

It was emphasized during the meeting that another potentially important feature not taken 14

into consideration in the ISN/RPS identification of segmental lesions is the occurrence in 15

lupus nephritis of fibrinoid necrosis. This is usually segmental and resembles lesions caused 16

by anti-neutrophil cytoplasmic antibodies (ANCA)-associated vasculitis. Conceptually, 17

segmental endocapillary lesions may be merely a distribution variant of global endocapillary 18

hypercellularity caused by the same pathogenic mechanisms, whereas segmental fibrinoid 19

necrosis may be a pathogenetically distinct lesion. Therefore, for classes III and IV, we 20

recommend noting the presence of fibrinoid necrosis. In phase 1, we propose to adjust the 21

definition for fibrinoid necrosis in lupus nephritis as follows: fibrin associated with GBM 22

disruption and/or lysis of the mesangial matrix; this lesion does not require the presence of 23

Page 11 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


12

karyorrhexis. Furthermore, karyorrhexis is defined as the presence of apoptotic, pyknotic 1

and fragmented nuclei – a definition used uniformly in other classification systems. In phase 2

2, we propose that the clinical significance of fibrinoid necrosis should be reevaluated in 3

addition to assessing the coexistence of necrotizing lesions with crescents. Whether these 4

lesions have “pauci-immune” features by IF and/or co-existent ANCA serology12

requires 5

further study. The Group concluded that phase 2 studies should determine and compare the 6

clinical and pathogenic significance of karyorrhexis alone, which is common in class III and IV 7

lupus nephritis, versus fibrinoid necrosis, which is less common. 8

Activity and Chronicity Assessment 9

For a critical re-evaluation of activity and chronicity, we turned to the paper by Austin et al13

10

on which the currently widely used NIH activity and chronicity index in lupus nephritis was 11

based. This system can be used to report the extent of overall activity and chronicity in a 12

semiquantitative way. It should be emphasized that this system has a number of limitations: 13

the cut-offs for the scores 0-3 are arbitrary and the scoring system was not developed using 14

an evidence- based approach. Therefore, in phase 2, we will use an evidence- based 15

approach without prior assumptions to modify these indices. We propose in phase 2 that the 16

evaluation for active and chronic lesions in lupus nephritis should be refined to improve 17

interobserver reproducibility and to validate prognostic value. In the meantime, we advocate 18

the usage of these indices, but modified as indicated below, for all classes in a modified 19

classification for lupus nephritis, thereby not restricting them to class III and IV. The modified 20

NIH activity and chronicity scoring system provides more information than the shorthand A, 21

C and A/C parameters currently used. We decided it is important to retain total scores of 24 22

in the activity index and 12 in the chronicity index, in particular for comparing the scores 23

Page 12 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


13

recorded for earlier renal biopsy samples from the same patient using the original NIH 1

scoring system. We therefore continue to accord double weight to the presence of fibrinoid 2

necrosis and cellular/fibrocellular crescents; results from our phase 2 study will be used to 3

find out whether this approach is valid. 4

In the modified NIH activity index, we have made the following modifications: 5

because in the original description of the combined karyorrhexis/fibrinoid necrosis category, 6

the emphasis was on the presence of fibrinoid necrosis,13

we now have modified this into a 7

stand-alone category of fibrinoid necrosis whereas we link the presence of karyorrhexis to 8

neutrophil infiltration. Because most karyorrhexis represents apoptotic cell death of 9

neutrophils, and because the original description for ‘leukocyte infiltration’ refers to the 10

presence of neutrophils only, we have changed the name and description of this category to 11

indicate the presence of neutrophils and/or karyorrhexis. In the original description of the 12

category ‘cellular crescents’, it was unclear to what extent fibrocellular crescents should be 13

included. Whereas fibrous crescents are part of the chronicity Index, we have now included 14

fibrocellular crescents in the activity index (see our proposal for definitions in Table 1). We 15

refer to our considerations about the definition for endocapillary hypercellularity above: in 16

the original description of Austin et al,13

only monocytes were taken into account in this 17

category with respect to inflammatory cells leading to hypercellularity. We have to 18

investigate in phase 2 how to approach the presence of inflammatory cells in endocapillary 19

hypercellularity in more detail. For the moment, we find it important to specifically score 20

neutrophils as a separate entity. Also the composition of the interstitial infiltrate as well as 21

its presence in areas affected by interstitial fibrosis and tubular atrophy (IFTA) should be 22

studied in further detail in phase 2. 23

Page 13 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


14

Table 2. Modified NIH lupus nephritis activity and chronicity scoring system proposal 1

Modified NIH Activity Index Definition Score

Endocapillary Hypercellularity Endocapillary hypercellularity in <25% (1+), 25-50% (2+), or

>50 (3+) of glomeruli.

0-3

Neutrophils/Karyorrhexis

Neutrophils and/or karyorrhexis in <25% (1+), 25-50% (2+), or

>50 (3+) of glomeruli.

0-3

Fibrinoid Necrosis FFibrinoid necrosis in <25% (1+), 25-50% (2+), or >50 (3+) of

glomeruli.

(0-3) x 2

Hyaline Deposits Wire loop lesions and/or hyaline thrombi in <25% (1+), 25-50%

(2+), or >50 (3+) of glomeruli.

0-3

Cellular/Fibrocellular Crescents Cellular and/or fibrocellular crescents in <25% (1+), 25-50%

(2+), or >50 (3+) of glomeruli.

(0-3) x 2

Interstitial Inflammation Interstitial leukocytes in <25% (1+), 25-50% (2+), or >50 (3+) in

the cortex.

0-3

Total 0-24

Modified NIH Chronicity Index Definition Score

Total Glomerulosclerosis Score Global and/or segmental sclerosis in <25% (1+), 25-50% (2+),

or >50 (3+) of glomeruli.

0-3

Fibrous Crescents Fibrous crescents in <25% (1+), 25-50% (2+), or >50 (3+) of

glomeruli.

0-3

Tubular Atrophy Tubular atrophy in <25% (1+), 25-50% (2+), or >50 (3+) of the

cortical tubules.

0-3

Interstitial Fibrosis Interstitial fibrosis in <25% (1+), 25-50% (2+), or >50 (3+) in the

cortex.

0-3

Total 0-12

2

Page 14 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


15

Global and segmental glomerulosclerosis 1

We previously mentioned difficulties in the attribution of global glomerulosclerosis to either 2

lupus nephritis or another cause,1 an issue related to the classification of lesions into class 3

III/IV, which requires deciding if global sclerosis is the sequel of an active class III/IV lesion. It 4

was discussed that some globally sclerotic glomeruli show features that indicate lupus 5

nephritis was the cause of the global sclerosis, for example fragmented tuft with 6

surrounding fibrosis, and extensive disruption of Bowman’s capsule. Moreover, residual 7

deposits other than IgM and C3 found by IF in globally sclerotic glomeruli should also be 8

considered evidence that the lesion is the result of lupus nephritis. Globally sclerotic 9

glomeruli should not be considered in the evaluation of lupus chronicity if they have a typical 10

pattern of arterionephrosclerosis, e.g. subcapsular clusters of glomeruli with ischemic tuft 11

collapse surrounded by collagen in Bowman’s space. Incorporating globally sclerotic 12

glomeruli that resulted from non-lupus injury in a chronicity index would overestimate the 13

lupus nephritis chronicity and severity, but still may correlate with outcome. This issue 14

should be addressed by phase 2 studies. It should be emphasized that in lupus nephritis the 15

term global sclerosis is used for those glomeruli that are completely sclerotic, and that any 16

other form of glomerulosclerosis should be regarded as segmental sclerosis. As mentioned 17

previously, it is difficult to reliably differentiate segmental sclerosis as a consequence of class 18

III/IV lupus nephritis from segmental sclerosis due to other causes (e.g., post-adaptive 19

segmental sclerosis). Still, only the former should be designated as lesions of class III/IV, 20

employing the same features noted above to differentiate segmental sclerosis due to prior 21

active lupus nephritis from that due to other causes. Within the chronicity index, segmental 22

and global glomerulosclerosis are considered together. 23

Page 15 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


16

Class V 1

Two issues were addressed in Class V. First, we considered whether or not to distinguish 2

Class V with and without mesangial hypercellularity and agreed this decision should be 3

evidence based (phase 2). Second, phase 2 studies should determine the allowable extent of 4

non-wire-loop subendothelial deposits within class V, above which one would have to 5

classify as III + V. 6

Class VI 7

We recognize that the RPS/ISN category VI is rarely seen in renal biopsy specimens, and 8

propose that this class be re-evaluated in phase 2, especially in view of our discussion above 9

on the recognizability of globally sclerotic glomeruli resulting from preceding lupus nephritis 10

active lesions versus nonspecific global sclerosis associated with other factors (e.g., aging, 11

hypertension or healed TMA lesions). Class VI will either need to be eliminated, or some 12

fraction of globally sclerotic glomeruli designated as the cut-off between chronic class IV and 13

class VI. 14

Tubulointerstitial lesions 15

We previously indicated the lack of cut-off values in the ISN/RPS classification for reporting 16

of severity of tubulointerstitial lesions.1 This deficiency will be addressed by the 17

development of a valid activity and chronicity index that scores the severity of 18

tubulointerstitial injury. We propose that in phase 2 we gather data on IFTA, and interstitial 19

infiltrates in a semi-quantitative fashion, rounding fibrosis to nearest 10%, with minimal 20

fibrosis stated as 5%. These values then can be translated into reproducible scoring 21

categories (e.g. on a scale of 0-3+) based on cut-off values to be evaluated for prognostic 22

Page 16 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


17

significance as currently done for the Banff and Oxford classifications. We advocate at this 1

time, as a phase 1 recommendation, to indicate in biopsy reports whether interstitial 2

inflammation occurs in the presence or absence of interstitial fibrosis. For the present, 3

interstitial inflammation remains part of the activity index as proposed above, while 4

interstitial fibrosis and tubular atrophy remain as separate entities within the chronicity 5

index. It has to be determined in phase 2 whether interstitial fibrosis and tubular atrophy 6

should be considered separately or combined into one parameter (as in the Oxford 7

classification for IgA nephropathy) and whether making a distinction between interstitial 8

inflammation in areas with or without interstitial fibrosis has clinical significance. 9

Vascular lesions 10

Currently, the ISN/RPS lupus classification does not evaluate vascular lesions. We believe it is 11

important to have a standardized approach and terminology to distinguish ordinary 12

arterial/arteriolar sclerosis from lupus-related lesions such as vasculopathy associated with 13

immune complex deposition, vasculitis, and TMA. In phase 2, a grading system for vascular 14

lesions could be used following the Banff classification, and the definitions of the Cure GN 15

group for evaluating hyalinosis. Definitions for TMA and vasculitis in lupus nephritis still have 16

to be created, as they can occur in an isolated manner with or without associated specific 17

serologic findings (ANCA, APA, etc.) or coexist with immune complex-mediated lupus 18

glomerular lesions. We propose that lupus vasculopathy be defined as luminal narrowing of 19

arterioles or terminal interlobular arteries by intramural immune deposits, typically admixed 20

with fibrinoid changes, without inflammation of the vessel wall. The immunoglobulin and 21

complement composition of the deposits is confirmed by IF.14-16

Future studies will 22

determine any impact of such lesions on prognosis and response to treatment. 23

Page 17 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


18

Summary and Future Directions 1

In our efforts to work towards a more effective, more reproducible and more valuable 2

classification for lupus nephritis, we have proposed improvements in the definitions of lupus 3

nephritis lesions, which could impact treatment and prognosis. We have also proposed two 4

important alterations in the classification system, namely to abandon the segmental/global 5

designations in class IV, and to replace the A/C/AC designations of class III and IV by use of 6

modified NIH lupus nephritis activity and chronicity scoring indices. Our proposed 7

modifications are based on published evidence, expert experience and mutual agreement; 8

we regard this as a phase 1 venture. The new definitions for lesions in lupus nephritis and 9

the modifications to the classification system address some of the disagreement that 10

currently exists among nephropathologists world-wide in classifying lupus nephritis 11

according to the ISN/RPS lupus nephritis classification. 12

We would like to emphasize that many of the proposed changes involve descriptions of 13

lesions that in other settings, such as the Oxford classification of IgA nephropathy, have 14

proved challenging even for experienced pathologists to reliably reproduce. We will proceed 15

to a phase 2 evidence-based approach using clinicopathologic studies and tests of inter-16

observer reproducibility to evaluate and validate the value of the newly proposed definitions 17

and adjustments to the classification system and include tubulointerstitial and vascular 18

lesions. This will entail a process similar to that used to develop the Oxford classification for 19

IgA nephropathy, i.e. scoring of individual lesions by multiple pathologists in lupus nephritis 20

biopsies obtained from patients for whom clinical outcomes are available. Once these data 21

become available, it will be possible to modify the currently used cut-off points to optimize 22

Page 18 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


19

the classes of lupus nephritis and to develop and validate reproducible activity and 1

chronicity indices that are of clinical utility. 2

3

Page 19 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


20

References 1

1. Wilhelmus S, Alpers CE, Cook HT, Ferrario F, Fogo AB, Haas M, Joh K, Noël LH, Seshan SV, Bruijn JA, 2

Bajema IM: The Revisited Classification of GN in SLE at 10 Years: Time to Re-Evaluate Histopathologic 3

Lesions. J Am Soc Nephrol 26: 2938-2946, 2015 4

2. Working Group of the International IgA Nephropathy Network and the Renal Pathology Society, 5

Cattran DC, Coppo R, Cook HT, Feehally J, Roberts IS, Troyanov S, Alpers CE, Amore A, Barratt J, 6

Berthoux F, Bonsib S, Bruijn JA, D'Agati V, D'Amico G, Emancipator S, Emma F, Ferrario F, Fervenza 7

FC, Florquin S, Fogo A, Geddes CC, Groene HJ, Haas M, Herzenberg AM, Hill PA, Hogg RJ, Hsu SI, 8

Jennette JC, Joh K, Julian BA, Kawamura T, Lai FM, Leung CB, Li LS, Li PK, Liu ZH, Mackinnon B, 9

Mezzano S, Schena FP, Tomino Y, Walker PD, Wang H, Weening JJ, Yoshikawa N, Zhang H: The Oxford 10

classification of IgA nephropathy: rationale, clinicopathological correlations, and classification. Kidney 11

Int 76: 534-545, 2009 12

3. Working Group of the International IgA Nephropathy Network and the Renal Pathology Society, 13

Roberts IS, Cook HT, Troyanov S, Alpers CE, Amore A, Barratt J, Berthoux F, Bonsib S, Bruijn JA, 14

Cattran DC, Coppo R, D'Agati V, D'Amico G, Emancipator S, Emma F, Feehally J, Ferrario F, Fervenza 15

FC, Florquin S, Fogo A, Geddes CC, Groene HJ, Haas M, Herzenberg AM, Hill PA, Hogg RJ, Hsu SI, 16

Jennette JC, Joh K, Julian BA, Kawamura T, Lai FM, Li LS, Li PK, Liu ZH, Mackinnon B, Mezzano S, 17

Schena FP, Tomino Y, Walker PD, Wang H, Weening JJ, Yoshikawa N, Zhang H: The Oxford 18

classification of IgA nephropathy: pathology definitions, correlations, and reproducibility. Kidney Int 19

76: 546-556, 2009 20

4. Trimarchi H, Barratt J, Cattran DC, Cook HT, Coppo R, Haas M, Liu ZH, Roberts IS, Yuzawa Y, Zhang 21

H, Feehally J: Oxford Classification of IgA nephropathy 2016: an update from the IgA Nephropathy 22

Classification Working Group. Kidney Int 91:1014-1021, 2017 23

5. Berden AE, Ferrario F, Hagen EC, Jayne DR, Jennette JC, Joh K, Neumann I, Noël LH, Pusey CD, 24

Waldherr R, Bruijn JA, Bajema IM: Histopathologic classification of ANCA-associated 25

glomerulonephritis. J Am Soc Nephrol 21: 1628-1636, 2010 26

6. Barisoni L, Troost J, Nast C, Bagnasco S, Avila-Casado C, Hodgin J, Palmer M, Rosenberg A, Gasim A, 27

Liensziewski C, Merlino L, Chang A, Meehan S, Gaut J, Song P, Holzman L, Gibson D, Kretzler M, 28

Gillespie B, Hewitt S: Reproducibility of the NEPTUNE descriptor-based scoring system on whole slide 29

images and histologic and ultrastructural digital images. Modern Pathol 29: 671-684, 2016 30

7. NIH/NIDDK Cure Glomerulonephropathy Network (CureGN) Pathology Scoring System Definitions, 31

provided by JC Jennette, CureGN Core Scoring Committeee Chair. 32

8. Loupy A, Haas M, Solez K, Racusen L, Glotz D, Seron D, Nankivell BJ, Colvin RB, Afrouzian M, Akalin 33

E, Alachkar N, Bagnasco S, Becker JU, Cornell L, Drachenberg C, Dragun D, de Kort H, Gibson IW, 34

Kraus ES, Lefaucheur C, Legendre C, Liapis H, Muthukumar T, Nickeleit V, Orandi B, Park W, Rabant 35

M, Randhawa P, Reed EF, Roufosse C, Seshan SV, Sis B, Singh HK, Schinstock C, Tambur A, Zeevi A, 36

Mengel M: The Banff 2015 Kidney Meeting Report: Current Challenges in Rejection Classification and 37

Prospects for Adopting Molecular Pathology. Am J Transplant 17: 28-41, 2017 38

Page 20 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


21

9. Weening JJ, D'Agati VD, Schwartz MM, Seshan SV, Alpers CE, Appel GB, Balow JE, Bruijn JA, Cook T, 1

Ferrario F, Fogo AB, Ginzler EM, Hebert L, Hill G, Hill P, Jennette JC, Kong NC, Lesavre P, Lockshin M, 2

Looi LM, Makino H, Moura LA, Nagata M: The classification of glomerulonephritis in systemic lupus 3

erythematosus revisited. J Am SocNephrol 15: 241-250, 2004 4

10. Wilhelmus S, Cook HT, Noël LH, Ferrario F, Wolterbeek R, Bruijn JA, Bajema IM: Interobserver 5

agreement on histopathological lesions in class III or IV lupus nephritis. Clin J Am Soc Nephrol 10: 47-6

53, 2015 7

11. Haring CM, Rietveld A, vanden Brand JA, Berden JH: Segmental and global subclasses of class IV 8

lupus nephritis have similar renal outcomes. J Am Soc Nephrol 15: 241-250, 2004 9

12. Nasr SH, D'Agati VD, Park HR, Sterman PL, Goyzueta JD, Dressler RM, Hazlett SM, Pursell RN, 10

Caputo C, Markowitz GS: Necrotizing and crescentic lupus nephritis with antineutrophil cytoplasmic 11

antibody seropositivity. Clin J Am Soc Nephrol 3: 682-690, 2008 12

13. Austin HA 3rd, Muenz LR, Joyce KM, Antonovych TT, Balow JE: Diffuse proliferative lupus 13

nephritis: Identification of specific pathologic features affecting renal outcome. Kidney Int 25: 689–14

695, 1984 15

14. Bhathena DB, Sobel BJ, Migdal SD: Noninflammatory renal microangiopathy of systemic lupus 16

erythematosus ('lupus vasculitis'). Am J Nephrol 1: 144-159, 1981 17

15. Appel GB, Pirani CL, D'Agati V: Renal vascular complications of systemic lupus erythematosus. J 18

Am Soc Nephrol 4: 1499-1515, 1994 19

16. Wu LH, Yu F, Tan Y, Qu Z, Chen MH, Wang SX, Liu G, Zhao MH: Inclusion of renal vascular lesions 20

in the 2003 ISN/RPS system for classifying lupus nephritis improves renal outcome predictions. 21

Kidney Int 83: 715-723, 2013 22

23

Page 21 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


22

ACKNOWLEDGMENTS 1

Boehringer Ingelheim gave financial support (Contract No.: 43074068) for the organization 2

of the meeting of this Workgroup in Leiden, the Netherlands, May 2016. 3

Page 22 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


Figure 1 as TIFF file Legend: Drawings depicting the ultrastructural features of a single glomerular capillary affected by lupus

glomerulonephritis: Class I with mesangial immune deposits (black) but no mesangial cell (red)

hypercellularity or influx of leukocytes; Class II with mesangial immune deposits and mesangial cell hypercellularity but no influx of leukocytes; and Class III/IV (upper right) with mesangial and capillary influx of leukocytes; Class III/IV (lower right) with subendothelial capillary wall immune deposits that can be seen

by LM and mesangial but no capillary influx of leukocytes (dark green neutrophils and light green monocytes/macrophages); Class III/IV + V with influx of leukocytes and numerous subepithelial immune deposits in addition to subendothelial deposits; and Class V with numerous subepithelial immune deposits but no influx of leukocytes (podocyte = outer green cell, endothelial cell = yellow cell, mesangial cell = red

cell, neutrophil = green cell with segmented nucleus, monocyte/macrophage = light green cell)

216x149mm (120 x 120 DPI)

Page 23 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


Category Recommendation Comments on ISN/RPS guidelines

Class II Definition for mesangial hypercellularity adjusted: Cuttoff for mesangial hypercellularity unclear

Four or more nuclei fully surrounded by matrix in

the mesangial area not including the hilar region (A)

Class III and IV The term endocapillary proliferation is replaced by Definition for endocapillary proliferation unclear;

endocapillary hypercellularity (B) the term proliferation was considered imprecise

The term crescent is used for a lesion consisting of Extracapillary proliferation involving < 25% of the circumference

extracapillary hypercellularity, composed of a variable mixture of Bowman's capsule was original cutoff. There were no

of cells. Fibrin and fibrous matrix may be present; 10% or more definitions for fibrous or fibrocellular crescents

of the circumference of Bowman's capsule should be involved.

Cellular crescent: more than 75% cells and fibrin

and less than 25% fibrous matrix (C)

Fibrous crescent: more than 75% fibrous matrix

and less than 25% cells and fibrin (D)

Fibrocellular crescent: 25-75% cells and fibrin

and the remainder fibrous matrix (E)

Adhesion: an area of isolated continuity of extracellular matrix There was no definition for an adhesion

material between the tuft and capsule even when the underlying

segment does not have overt sclerosis (F)

Fibrinoid necrosis: fibrin associated with glomerular basement membrane There was no definition for fibrinoid necrosis

disruption and/or lysis of the mesangial matrix; this lesion does not

require the presence of karyorrhexis

Elimination of segmental and global subdivions of class IV Definitions for segmental and global were unclear;

interobserver variability was large; clinical significance uncertain

Modification of the NIH lupus nephritis activity and chronicity scoring Designation of activity/chronicity through A, C and A/C

system (Table 2) to be used instead of the currently used A, C considered too broad and nonspecific; preference for a

and A/C parameters semiquantitative approach to describe active and chronic lesions

Tubulointerstitial lesions Indicate whether interstitial inflammation occurs in presence or Lack of cut-off values for reporting the severity of

absence of interstitial fibrosis tubulointerstitial lesions

Page 24 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


A B C

D E F

Page 25 of 25



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Revision of the ISN/RPS classification for lupus nephritis: of … · 2019-02-16 · For Peer Review Only 4 1 On May 9-11th 2016, a Working Group for Lupus Nephritis Classification

Documents