Emerging Concepts and Controversies in Renal Pathology C4d-Negative and Arterial Lesions as Manifestations of Antibody-Mediated Transplant Rejection Mark Haas, MD, PhD Q2 Q3 ABSTRACT T he consensus classification of antibody- mediated rejection (AMR) of renal allografts developed at the Sixth Banff Conference on Allograft Pathology Q5 , in 2001, identified three find- ings necessary for the diagnosis of active AMR: histologic evidence, antibodies against the graft, and immunohistologic evidence. Morphologic and molecular studies have noted evidence of microvascular injury, which, in the presence of donor-specific antibodies (DSAs) but the absence of C4d deposition, is associated with development of transplant glomerulopathy and graft loss. Recent studies suggest that intimal arteritis may in some cases be a manifestation of DSA- induced graft injury. These newly recognized le- sions of AMR have now been incorporated into a revised Banff diagnostic schema. INTRODUCTION Q6 Q7 During the past 2 decades there has developed an increasing awareness of antibody-mediated rejec- tion (AMR) as an important cause of both short- and long-term injury to renal allografts. 1–3 Part of the difficulty in recognizing the importance of AMR was related to the lack of specific morpho- logic findings that could be used to identify changes of AMR on allograft biopsies. Acute rejec- tion in the presence of donor-specific anti-HLA antibodies (DSAs) shows histologic changes very different from those of acute T cell–mediated rejection (CMR), characterized by microvascular injury and inflammation, including margination of neutrophils and mononuclear leukocytes (later documented as monocytes/macrophages 4 ) Q8 in peritubular and glomerular capillaries, thrombotic microangiopathy (TMA), and, in severe cases, fib- rinoid necrosis of arterioles and small arteries. 5 Antibody usually cannot be detected in the micro- vasculature by immunofluorescence microscopy, however. 5 Furthermore, these morphologic changes are not specific for AMR and can be seen with other causes of endothelial injury, including acute calcineurin inhibitor nephrotoxicity and recurrent TMA. 6–9 The use of staining for C4d by immunofluores- cence or immunohistochemistry (IHC) has been an important development in allowing pathologists to more accurately diagnose acute AMR in allo- graft biopsies and in recognizing the contribution of humoral immunity to lesions of chronic renal allograft rejection. 10–18 Although C4d, formed on cleavage of complement factor C4, a component of the classical complement pathway, is itself bio- logically inactive, it binds covalently at the site of C4 cleavage, rendering it a long-lived marker for Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Los Angeles, CA 91403, USA Q4 E-mail address: [email protected]KEYWORDS Renal transplant Transplant rejection Antibody-mediated rejection Transplant glomerulopathy Transplant glomerulitis Transplant arteriopathy Intimal arteritis Complement C4d PATH262_proof ■ 27 May 2014 ■ 4:08 pm Surgical Pathology - (2014) -–- http://dx.doi.org/10.1016/j.path.2014.04.003 1875-9181/14/$ – see front matter Ó 2014 Published by Elsevier Inc. surgpath.theclinics.com 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
Transcript
Q2Q3
Q5
Q6Q7
Q4
12
3456
789
101112
1314
1516
17181920
212223
2425
2627
282930
313233
343536
373839
4041
424344
454647
4849
50
Emerging Concepts andControversies in Renal
PathologyC4d-Negative and Arterial Lesions asManifestations of Antibody-MediatedTransplant Rejection
T he consensus classification of antibody-mediated rejection (AMR) of renal allograftsdeveloped at the Sixth Banff Conference on
Allograft Pathology, in 2001, identified three find-ings necessary for the diagnosis of active AMR:histologic evidence, antibodies against the graft,and immunohistologic evidence. Morphologicand molecular studies have noted evidence ofmicrovascular injury, which, in the presence ofdonor-specific antibodies (DSAs) but the absenceof C4d deposition, is associated with developmentof transplant glomerulopathy and graft loss.Recent studies suggest that intimal arteritis mayin some cases be a manifestation of DSA-induced graft injury. These newly recognized le-sions of AMR have now been incorporated into arevised Banff diagnostic schema.
INTRODUCTION
During the past 2 decades there has developed anincreasing awareness of antibody-mediated rejec-tion (AMR) as an important cause of both short-and long-term injury to renal allografts.1–3 Part ofthe difficulty in recognizing the importance ofAMR was related to the lack of specific morpho-logic findings that could be used to identify
Department of Pathology and Laboratory Medicine, CedAngeles, CA 91403, USAE-mail address: [email protected]
PATH262_proof ■ 2
Surgical Pathology - (2014) -–-http://dx.doi.org/10.1016/j.path.2014.04.0031875-9181/14/$ – see front matter � 2014 Published by E
changes of AMR on allograft biopsies. Acute rejec-tion in the presence of donor-specific anti-HLAantibodies (DSAs) shows histologic changes verydifferent from those of acute T cell–mediatedrejection (CMR), characterized by microvascularinjury and inflammation, including margination ofneutrophils and mononuclear leukocytes (laterdocumented as monocytes/macrophages4) inperitubular and glomerular capillaries, thromboticmicroangiopathy (TMA), and, in severe cases, fib-rinoid necrosis of arterioles and small arteries.5
Antibody usually cannot be detected in the micro-vasculature by immunofluorescence microscopy,however.5 Furthermore, these morphologicchanges are not specific for AMR and can beseen with other causes of endothelial injury,including acute calcineurin inhibitor nephrotoxicityand recurrent TMA.6–9
The use of staining for C4d by immunofluores-cence or immunohistochemistry (IHC) has beenan important development in allowing pathologiststo more accurately diagnose acute AMR in allo-graft biopsies and in recognizing the contributionof humoral immunity to lesions of chronic renalallograft rejection.10–18 Although C4d, formed oncleavage of complement factor C4, a componentof the classical complement pathway, is itself bio-logically inactive, it binds covalently at the site ofC4 cleavage, rendering it a long-lived marker for
ars-Sinai Medical Center, 8700 Beverly Boulevard, Los
humoral immunity.19 At the Sixth Banff Conferenceon Allograft Pathology, in 2001,20 consensus diag-nostic criteria for acute AMR in renal allograftswere adopted that require immunopathologic evi-dence in the form of C4d deposition in peritubularcapillaries (ptc) as well as morphologic evidence(microvascular lesions as discussed previously)and the presence of DSAs.During the past several years, however, limita-
tions of these diagnostic criteria have been recog-nized. It is now well documented that DSAs maycause graft injury in the absence of C4d deposition.Furthermore, there is now evidence that intimal,non-necrotizing arteritis, previously thought torepresent acute cell-mediated rejection, may inthe presence of DSAs be (at least in part) humorallymediated. This review focuses on evidence sup-porting these paradigm shifts within the mor-phologic spectrum of AMR and reviews newconsensus diagnostic criteria for AMR developedat the 12th Banff Conference on Allograft Pathol-ogy in 2013.21
14
159160
161162163164
165166167
168169170
171172173174
175176177
178179180
181182183
184185186187
188189190
191192
C4D-NEGATIVE ANTIBODY-MEDIATED
REJECTION
Sis and coworkers22 used gene expressionmicroarrays done on tissue from 173 indicationrenal allograft biopsies (performed for acute orpersistent graft dysfunction or for proteinuria)to examine expression of a set of knownendothelial-associated genes (ENDATs). Theyfound that the combination of DSAs and highENDAT expression (but not DSAs alone) was asso-ciated with a significantly increased rate of graftloss, even in the absence of C4d, although diffuseptc C4d staining further increased the rate of graftloss.22 They found that overexpression of 12 indi-vidual ENDATs was correlated with an increasedrisk of graft loss; the gene with the highest associ-ated risk was that for von Willebrand factor.22 Asecond key study supporting the existence ofC4d-negative AMR and its association with devel-opment of graft scarring came from Loupy andcolleagues.23 These investigators examined clin-ical and pathologic findings at 1 year posttrans-plantation in 45 recipients of deceased donorrenal allografts with known DSAs, based on find-ings observed on protocol biopsies done 3monthsposttransplantation. Based on these 3-monthbiopsies, the patients of Loupy and colleagues23
fell into 3 groups: those with no evidence ofAMR, those with subclinical AMR characterizedby both ptc C4d staining and histologic featuresof AMR (glomerulitis, peritubular capillaritis, orboth), and those with histologic features of AMR
PATH262_proof ■ 27 May
but no ptc C4d staining. At 1 year posttransplanta-tion, patients whose 3-month biopsy had no evi-dence of AMR had good graft function andinterstitial fibrosis/tubular atrophy (IF/TA) in onlyone-third of cases. These 1-year biopsies alsoshowed no evidence of transplant glomerulopathy(TG), a lesion that is most often a manifestation ofchronic AMR. TG is characterized histologically byduplication of glomerular basement membranes(GBMs), best demonstrated on sections stainedwith periodic acid–Schiff (PAS) or Jones methena-mine silver stains to highlight GBMs. Patientswhose 3-month biopsy showed subclinical AMRhad reduced graft function, IF/TA in all cases,and TG in approximately half. Patients whose3-month biopsy showed glomerulitis and/or peri-tubular capillaritis but no C4d had a mean creati-nine clearance and frequencies of IF/TA and TGintermediate between the no AMR and C4d-positive subclinical AMR cohorts.23 These findingsare highly consistent with those of Sis andcolleagues,22 supporting the existence of a C4d-negative form of antibody-mediated graft injurythat is less severe than C4d-positive AMR but isnonetheless associated with the development ofchronic changes within the graft, including TG.Fig. 1 illustrates the key microvascular lesions ofAMR: glomerulitis, peritubular capillaritis, and TG.Although TG is most often a manifestation of
chronic AMR, approximately 25% of cases of TGoccur in DSA-negative patients and are likely dueto other causes (see Key Points).24–27 Diagnosisof TG is important because this lesion is associ-ated with poor graft outcome,28 regardless ofcause, although lesions associated with hepatitisC (rather than those associated with DSAs) werefound to have the fastest rate of progression tograft loss.26 In diagnosing TG, however, it is impor-tant to differentiate this from other lesions that alsoshow GBM duplication, notably membranoproli-ferative glomerulitis (MPGN) type I and forms ofC3 glomerulopathy (C3G), including dense depositdisease (formerly termed MPGN type II), bothof which frequently recur in renal allografts.29
MPGN type I is often associated with hepatitisC30; in addition to GBM double contours, theglomeruli appear hypercellular and hyperlobulardue to mesangial cell proliferation and matrixexpansion as well as endocapillary hypercellularitythat is in part due to mononuclear leukocyteswithin the glomerular tuft. Although MPGN type Iis easily distinguished from “pure” QTG by lightmicroscopy due to a lack of glomerular hypercellu-larity in the latter, the distinction by light micro-scopy alone can be more difficult when TG isaccompanied by moderate or severe glomeruli-tis.31 MPGN type I, however, shows immune
2014 ■ 4:08 pm
Q15
Fig. 1. Histopathologic lesions of AMR. (A) Glomerulitis and peritubular capillaritis. There are many mononuclearleukocytes as well as a small number of neutrophils in glomerular and ptc, with endothelial swelling and partialocclusion of some capillary lumens (PAS, original magnification �200). (B) Glomerulitis. Marginated mononuclearleukocytes are present in multiple glomerular capillaries, in some associated with endothelial cell swelling andpartial or nearly complete occlusion of the capillary lumen (PAS �400). (C) Peritubular capillaritis. The ptc ofthe renal cortex (note that most tubules present are proximal tubules) contain prominent numbers of leukocytes,many of which have the typical appearance of monocytes (hematoxylin-eosin stain �400). (D) TG. Several GBMdouble contours are evident on the PAS stain, notably at the lower portion of the glomerulus. Segmental glomer-ulitis is also present, consistent with chronic, active AMR in this patient who had circulating DSAs (see Box 2) (PAS�400).
print&web4C/FPO
Q1 Controversies in Renal Pathology 3
193
194195196197
198199200
201202203
204205206207
208209210
211212213
214215216
217218219220
221222223
224225226
227228229230
231232233
234235236
237238239
240241242243
244245246
247248249
250
251252253254
255256257
258259260
261262263264
265266267
268269270
271272273
274275276277
278279280
281282283
284285286287
288289290
291292293
294295296
297298299300
301302303
304305306
complex deposits (most often containing IgG andIgM, sometimes with IgM and C1q) by immunoflu-orescence and electron microscopy (EM), and us-ing these modalities it is readily distinguished fromTG, which lacks such deposits. C3G is not associ-ated with hepatitis C but results from abnormalactivation of the alternative pathway of comple-ment, due to mutations in complement regulatoryproteins, including factor H and factor H–relatedproteins, autoantibodies against factor H, and/orC3 nephritic factor, an IgG autoantibody that bindsto and prevents the inactivation of C3 convertase,resulting in the persistent cleavage of C3.32–34 Thehistology of C3G is variable; GBM double contoursmay be present and in some cases glomerular
PATH262_proof ■ 2
hypercellularity may be mild, resembling that inFig. 1D. Glomeruli in C3G, however, show strongC3 staining by immunofluorescence (with theintensity of C3 staining exceeding that of anyimmunoglobulin by at least 21 on a 0–31 scale35)and deposits (although not true immune com-plexes per se) by EM that may include highlyelectron-dense, elongated intramembranous de-posits (typical of dense deposit disease), lesselectron-dense subendothelial or intramembra-nous/transmembranous deposits, and even sube-pithelial “humps” resembling those of acutepostinfectious glomerulonephritis (see Pickeringand colleagues34 for a description of morphologicfindings in C3G).
7 May 2014 ■ 4:08 pm
Key PointsMAJOR CAUSES OF TRANSPLANT GLOMERULOPATHY
� Chronic (or chronic, active) AMR
� Accounts for approximately 75% of cases of TG24
� In chronic, active AMR, there is frequently concurrent glomerulitis (see Fig. 1D).
� Chronic or persistent TMA
� May be recurrent (eg, recurrent hemolytic-uremic syndrome) or de novo (eg, due to calcineurininhibitor nephrotoxicity)
� Changes related to hepatitis C virus infection
� May be related to TMA associated with anticardiolipin antibodies25
� Associated with worse prognosis for graft survival than TG secondary to AMR or TMA in hepatitisC–negative patients26
� Must be distinguished from MPGN type I, which may also be related to hepatitis C. In MPGN type I,the glomeruli show immune complex deposits by immunofluorescence and EM (C3G), although thisis not the case with TG, even when glomerulitis is also present.
� Changes related to T-cell activation
� A recent study showed that in cases of TG without C4d deposition or DSAs, there was increasedexpression of cytotoxic T cell–associated transcripts.27
Haas4
307
308309310311
312313314
315316317
318319320321
322323324
325326327
328329330
331332333334
335336337
338339340
341342343344
345346347
348349350
351352353
354355356357
358359360
361362363
364
365366367368
369370371
372373374
375376377378
379380381
382383384
385386387
388389390391
392393394
395396397
398399400401
402403404
405406407
408
Although TG, as evident by light microscopy onPAS and silver-stained sections, is rarely observedduring the first 6 months posttransplantation,36
Wavamunno and colleagues37 showed that mor-phologic changes associated with the subsequentdevelopment of overt TG may be seen duringthe first 1 to 3 months posttransplantation byEM. These ultrastructural changes include swe-lling and vacuolization of glomerular endothelialcells, subendothelial electron-lucent widening (orwidening of the lamina rara interna), and earlyduplication/multilayering of GBMs (Fig. 2). Re-cently, it was shown that one or more of these ul-trastructural findings are seen in most biopsieswith microvascular injury in DSA-positive patientsduring the first 3 months posttransplantation,whether or not C4d is present.38
409410
411412413414
415416417
418419420
ROLE OF ANTIBODY IN ARTERIAL LESIONS
Certain arterial lesions in renal allografts are wellaccepted as having a pathogenesis that often in-volves humoral immunity. As discussed previously,transmural necrosis in one or more arteries isincluded among those lesions satisfying the histo-logic component for diagnosis of acute AMR intheBanff classification.20 Some recent studies pro-vide evidence, however, for expanding the range of
PATH262_proof ■ 27 May
arterial lesions associated with antibody-mediatedgraft injury.Intimal arteritis is currently classified as a lesion
of CMR and defines type 2 acute rejection accord-ing to the Banff classification of CMR.39 There isemerging evidence, however, that intimal arteritismay in some cases be, at least in part, humorallymediated. In some cases, such as that shownin Fig. 3, lesions of intimal arteritis contain a pre-dominance of CD68-positive macrophages, whichare also highly prominent in glomerulitis and peri-tubular capillaritis,4,40 as opposed to CD3-positive T lymphocytes that typically predominatein the tubulointerstitial lesions of acute CMR.41–43
Most recently, Lafaucheur and colleagues44 re-ported on 64 cases of intimal arteritis in DSA-positive patients. They found that compared withcases of AMR without intimal arteritis, those withintimal arteritis had a 3-fold higher rate of graftloss. Although a majority of cases of intimal arter-itis with DSAs also showed interstitial inflammationand tubulitis, indicating a combined lesion of AMRand CMR, it was also noteworthy that intimal arter-itis with DSAs was associated with a significantly(and approximately 6-fold) higher rate of graftloss than intimal arteritis without DSAs, the latterrepresenting pure CMR.44 These findings suggestan association of lesions of intimal arteritis withAMR when DSAs are present and may account
2014 ■ 4:08 pm
Fig. 2. Early ultrastructural lesions of glomerular capillaries in patients with acute/active AMR. (A) Glomerularendothelial swelling and vacuolization in a biopsy showing C4d-positive AMR, done on posttransplantationday (PTD) 9. Uranyl acetate and lead citrate stain (original magnification �7200). (B) Severe glomerular endothe-lial swelling and segmental subendothelial electron-lucent widening in a biopsy showing C4d-negative AMR, PTD29 (uranyl acetate and lead citrate �7200). (C) Subendothelial electron-lucent widening in a biopsy showing C4d-positive AMR, PTD 10 (uanyl acetate and lead citrate �10,000). (D) Subendothelial electron-lucent widening andearly GBM duplication in a biopsy showing C4d-negative AMR, PTD 52 (uranyl acetate and lead citrate �7500).
Controversies in Renal Pathology 5
421
422423424425
426427428
429430431
432433434435
436437438
439440441
442443444
445446447448
449450451
452453454
455456457458
459460461
462463464
465466467
468469470471
472473474
475476477
478
479480481482
483484485
486487488
489490491492
493494495
496497498
499500501
502503504505
506507508
509510511
512513514515
516517518
519520521
522523524
525526527528
529530531
532533534
in part for the frequent lack of a complete responseof lesions reported as Banff type 2 acute CMR totherapy directed at T cells.45,46
The data of Lefaucheur and colleagues44 arguethat testing for DSAs is indicated in renal allograftrecipients with a biopsy showing intimal arteritis,even if C4d staining is negative. Likewise, treat-ment in these cases should include measures to
PATH262_proof ■ 2
remove DSAs if the latter are present, becausepersistence of DSAs, microvascular inflammation,and intimal arteritis is associated with poor graftoutcomes in patients with AMR.47 Still, althoughthese findings suggest that the presence of intimalarteritis may identify a more severe form of AMRthan those cases without intimal arteritis, the evi-dence directly linking intimal arteritis to humoral
7 May 2014 ■ 4:08 pm
Fig. 3. Macrophages in acase of acute rejection withintimal arteritis, microvas-cular inflammation, andDSAs. The biopsy was per-formed 2 months posttrans-plantation for an acute risein serum creatinine level.The PAS stain (A) shows glo-merulitis and an artery withintimal arteritis. There isonly very mild interstititalinflammation and minimaltubulitis. An IHC stainfor CD68 (B) shows manyCD68-positive macrophageswithin glomerular and ptcas well as within the lesionof intimal arteritis (lowerleft of the photomicro-graph). Only small numbersof CD3-positive T cells wereseen in the artery withintimal arteritis (not shown).Immunofluorescence stain-ing of tissue from this biopsyfor C4d was negative,although the patient hadanti-HLA class II DSAs atthe time of biopsy. (Originalmagnification of bothphotomicrographs �200;bar 5 50 mm Q23.)
print&web4C/FPO
Haas
PATH262_proof ■ 27 May 2014 ■ 4:08 pm
6
535
536537538539
540541542
543544545
546547548549
550551552
553554555
556557558
559560561562
563564565
566567568
569570571572
573574575
576577578
579580581
582583584585
586587588
589590591
592
593594595596
597598599
600601602
603604605606
607608609
610611612
613614615
616617618619
620
621622
623624625626
627628629
630631632
633634635
636637638639
640641642
643644645
646647
Controversies in Renal Pathology 7
648
649650651652
653654655
656657658
659660661662
663664665
666667668
669670671
672673674675
676677678
679680681
682683684685
686687688
689690691
692693694
695696697698
699700701
702703704
705
706707708709
710711712
713714715
716717718719
720721722
723724725
726727728
729730731732
733
immunity is indirect and still preliminary. Of the 64DSA-positive cases studied by Lefaucheur andcolleagues,44 only 8 showed intimal arteritis asan isolated finding, in the absence of glomerulitisand/or peritubular capillaritis, and isolated intimalarteritis may also be an infrequent manifestationof acute CMR.21 Also, as discussed previously,72% and 63% of these 64 cases had interstitialinflammation and tubulitis, respectively, consistentwith earlier findings of a strong association be-tween glomerulitis and intimal arteritis48 and sug-gesting the presence of CMR as well as AMR.Finally, the link between predominance of CD68-positive cells in lesions of intimal arteritis andAMR is hypothetical, and one study found no dif-ference in clinical outcomes in patients having bi-opsies with intimal arteritis with a predominanceof CD68-positive cells compared with intimalarteritis with primarily CD3-positive T cells.49
Thus, it is important that additional studies be un-dertaken to determine if lesions of intimal arteritiscontaining mainly CD68-positive cells (see Fig. 3)are significantly associated with the presence ofDSAs and other histologic lesions of active AMRand if these lesions respond to treatment ofAMR, particularly in those cases where initial treat-ment with agents used for the treatment of acuteCMR fail to produce a return of graft function tobaseline level.
Fig. 4. Transplant arterio-pathy. This arcuate arteryshows marked intimalthickeningandnarrowingof the lumen. There areleukocytes as well asactivated-appearing fib-roblastswithin the intima,especially portions closerto the lumen, althoughthere are no leukocytesdirectly beneath theendothelium to indicateacute rejection (intimalarteritis) (hematoxylin-eosin stain, originalmagnification �200).
PATH262_proof ■ 2
The association of arterial lesions with humoralimmunity also seems to extend to intimal fibrosis.Lesions of arterial intimal fibrosis with multipleintimal leukocytes (sometimes termed, transplantarteriopathy), such as those shown in Fig. 4,have been shown associated with DSAs andC4d.17,18 Furthermore, in a study of protocol bi-opsies done at 3 and 12 months posttransplanta-tion, Hill and colleagues50 found that in renaltransplant recipients with DSAs, but not thosewithout, the mean Banff intimal fibrosis (cv) scoreincreased significantly from month 3 to month 12.At later times, the mean cv score continued to in-crease at a faster rate in DSA-positive patients.By contrast, hypertension was not significantlyassociated with the increase in cv score. Althoughintimal fibrosis in a majority of DSA-positive pa-tients did show a hypercellular zone close to theendothelium that was not observed (or was muchless pronounced) in those arteries from DSA-negative patients with intimal fibrosis, there wereDSA-positive patients with a prior history of acuteAMR whose biopsies showed bland intimalfibrosis, indistinguishable from banal arterioscle-rosis.50 It is possible that those lesions of blandintimal fibrosis in DSA-positive patients representa late, quiescent stage of the more cellular lesions,because biopsies from DSA-positive patientsshowing bland arterial intimal fibrosis showed no
print&web4C/FPO
7 May 2014 ■ 4:08 pm
734735
736737738
739740741742
743744745
746747748
749750751
752753754755
756757758
759760761
Q16
Q17Q18
Q21
Haas8
762
763764765766
767768769
770771772
773774775776
777778779
780781782
783784785
786787788789
790791792
793794795
796797798799
800801802
803804805
806807808
809810811812
813814815
816817818
819
820821822823
824825826
827828829
830831832833
834835836
837
glomerulitis or peritubular capillaritis to suggest anactive component of AMR.50
REVISED (2013) BANFF CRITERIA FOR
ANTIBODY-MEDIATED REJECTION
A consensus reached during and after the mostrecent Banff Conference, held in August 2013,has resulted in modification of the Banff criteriafor diagnosis of both acute/active (Box 1) andchronic, active AMR (Box 2) in renal allografts.21
These revised criteria include C4d-negative le-sions. Lesions of intimal arteritis are also included,although, as discussed previously, intimal arteritisas the sole histologic manifestation of AMR is un-common and the current evidence directly linkingintimal arteritis to humoral immunity is indirect andsomewhat preliminary. Still, Banff criteria are (andhave always been) a working classification open
Box 1Revised (Banff 2013) classification of acute/active antfeatures must be present for diagnosisa,b)
� Histologic evidence of acute tissue injury, including
� Microvascular inflammation (g score >0c and/or p
� Intimal or transmural arteritis (v score >0)d
� Acute TMA, in the absence of any other cause
� Acute tubular injury, in the absence of any other
� Evidence of current/recent antibody interaction withe following:
� Linear C4d staining in ptc (involving at least 10%or any ptc by IHC on paraffin sections)
� At least moderate microvascular inflammation (
� Increased expression of gene transcripts in thethoroughly validatedf
� Serologic evidence of DSAs (HLA or other antigens
a For all AMR diagnoses, it should be specified in the repoptc by immunofluorescence on frozen sections or any ptc bysition (<10% of ptc by immunofluorescence on frozen sectio
b These lesions may be clinically acute, smoldering, or subcwith DSAs and C4d without histologic abnormalities potentiseen mainly in ABO-incompatible renal allografts) may be d
c Recurrent/de novo glomerulonephritis should be excludd These arterial lesions may be indicative of AMR, CMR,
arteries having a continuous media with 2 or more smoothe In the presence of acute CMR, borderline infiltrates, or ev
define moderate microvascular inflammation, and glomerulf At present, the only validated molecular marker meetin
been validated in a single center (University of Alberta). Theof gene expression within the biopsy as evidence of AMR muENDAT expression by Sis and colleagues.22
PATH262_proof ■ 27 May
to future revision if warranted by new data.Revised portions of the classification are indicatedin boldface type in Boxes 1 and 2. Biopsiesshowing AMR should be designated as C4dpositive or without evident C4d deposition (seeBoxes 1 and 2) and, in the absence of C4d,more stringent evidence of current or recent anti-body interaction with the vascular endotheliummust be present; this helps avoid overdiagnosisof AMR. Such evidence may be morphologic, inthe form of at least moderate microvascularinflammation, or molecular, provided the latter isbased on a thoroughly validated test. The latterwas mainly included to allow the classification toadapt to emerging data, because at present theonly validated molecular marker of antibody-mediated endothelial injury is ENDAT expres-sion,22 and even this has been validated only ata single center.
ibody-mediated rejection in renal allografts (all 3
one or more of the following:
tc score >0)
apparent cause
th vascular endothelium, including at least one of
of ptc by immunofluorescence on frozen sections
sum of g and ptc scores ‡2)e
biopsy tissue indicative of endothelial injury, if
)
rt whether the lesion is C4d positive (involving �10% ofIHC on paraffin sections) or without evident C4d depo-ns; completely negative by IHC on paraffin sections).
linical. Biopsies showing 2 of the 3 features, except thoseally related to AMR or cell-mediated rejection (the latteresignated as “suspicious” for acute/active AMR.ed.or mixed AMR/CMR. Arterial lesions are only scored inmuscle layers.idence of infection, ptc score�2 alone is not sufficient toitis must be present (g score �1).g this criterion is ENDAT expression,22 and this has onlyuse of ENDATexpression at other centers or other test(s)st first undergo independent validation as was done for
2014 ■ 4:08 pm
838839
840841842
843844845846
847848849
850851852
853854855856
857858859
860861862
863864865
866867868869
870871872
873874875
Q19
Box 2Revised (Banff 2013) classification of chronic, active antibody-mediated rejection in renal allografts (all3 features must be present for diagnosisa,b)
� Morphologic evidence of chronic tissue injury, including one or more of the following:
� TG (cg score >0)c, if no evidence of chronic TMA
� Severe peritubular capillary basement membrane multilayering (requires EM)d
� Arterial intimal fibrosis of new onset, excluding other causese
� Evidence of current/recent antibody interaction with vascular endothelium, including at least one ofthe following:
� Linear C4d staining in ptc (involving at least 10% of ptc by immunofluorescence on frozensections, or any ptc by IHC on paraffin sections)
� At least moderate microvascular inflammation (sum of g and ptc scores ‡2)f
� Increased expression of gene transcripts in the biopsy tissue indicative of endothelial injury, ifthoroughly validatedg
� Serologic evidence of DSAs (HLA or other antigens)
a For all AMR diagnoses, it should be specified in the report whether the lesion is C4d positive (involving �10% ofptc by immunofluorescence on frozen sections or any ptc by IHC on paraffin sections) or without evident C4d depo-sition (<10% of ptc by immunofluorescence on frozen sections; completely negative by IHC on paraffin sections).
b Lesions of chronic, active AMR can range from primarily active lesions with early TG evident only by EM to thosewith advanced TG and other chronic changes in addition to active microvascular inflammation. In the absence ofevidence of current/recent antibody interaction with the endothelium (those features in section 2)Q22 , the term, active,should be omitted; in such cases, DSAs may be present at the time of biopsy or at any previous timeposttransplantation.
c Includes GBM duplication by EM only or GBM double contours by light microscopy.d �7 Layers in 1 cortical peritubular capillary and�5 in 2 additional capillaries, avoiding portions cut tangentially.51e Although leukocytes within the fibrotic intima favor chronic rejection, these are seen with chronic CMR as well as
chronic AMR and are, therefore, helpful only if there is no history of CMR. An elastic stain may be helpful as absenceof elastic lamellae is more typical of chronic rejection and multiple elastic lamellae are most typical of arteriosclerosis,although these findings are not definitive.
f In the presence of acute CMR, borderline infiltrates, or evidence of infection, ptc score�2 alone is not sufficient todefine moderate microvascular inflammation, and glomerulitis must be present (g score �1).
g At present, the only validated molecular marker meeting this criterion is ENDAT expression,22 and this has onlybeen validated in a single center (University of Alberta). The use of ENDATexpression at other centers or other test(s)of gene expression within the biopsy as evidence of AMR must first undergo independent validation as was done forENDAT expression by Sis and colleagues.22
Controversies in Renal Pathology 9
876
877878879880
881882883
884885886
887888889890
891892893
894895896
897898
899900
901902903
904905906
907908909910
911912913
914915916917
918919920
921922923
924925926
927928929930
931
932933934935
936937938
939940941
942943944945
946947948
949950951
952953954
955956957958
959960961
962963964
965966967968
969970971
972973974
975976977
ACKNOWLEDGMENTS
The author has no commercial or other relation-ships that have influenced, could be perceived ashaving influenced, or give the appearance ofpotentially influencing any of the material pre-sented in this article. I thank Bill Pollard for assis-tance in preparation of figures.
978979980981
982983984
985986987
REFERENCES
1. Terasaki PI, Cai J. Human leukocyte antigen anti-
