renal monoclonal immunoglobulin deposition disease: the...
Post on 06-Feb-2018
217 Views
Preview:
TRANSCRIPT
Renal Monoclonal Immunoglobulin Deposition Disease:The Disease Spectrum
JULIE LIN,* GLEN S. MARKOWITZ,‡ ANTHONY M. VALERI,*NEERAJA KAMBHAM,‡ WILLIAM H. SHERMAN, † GERALD B. APPEL,*and VIVETTE D. D’AGATI‡
Divisions of *Nephrology and†Oncology, Department of Medicine, and‡Department of Pathology, ColumbiaUniversity, New York-Presbyterian Hospital, New York, New York.
Abstract.This study reports the clinicopathologic findings andoutcome in 34 patients with renal monoclonal immunoglobulindeposition disease (MIDD), which included 23 light-chain DD(LCDD), 5 light- and heavy-chain DD (LHCDD), and 6 heavy-chain DD (HCDD). A total of 23 patients had pure MIDD,whereas 11 patients had LCDD with coexistent myeloma castnephropathy (LCDD & MCN). Renal biopsy diagnosis pre-ceded clinical evidence of dysproteinemia in 68% of all cases.By immunofluorescence, the composition of deposits included11k/1l (LCDD), 3IgGk/2IgGl (LHCDD), 5g/1a (HCDD),and 10k/1l (LCDD & MCN). Patients with pure MIDD pre-sented with mean serum creatinine of 4.2 mg/dl, nephroticproteinuria, and hypertension. Cases of HCDD were associatedwith a CH1 deletion and frequently had hypocomplementemiaand a positive hepatitis C virus antibody but negative hepatitisC virus PCR. LCDD & MCN is a morphologically and clini-
cally distinct entity from pure MIDD, presenting with highercreatinine (mean, 7.8 mg/dl;P 5 0.01), greater dialysis depen-dence (64versus26%; P 5 0.053), subnephrotic proteinuria,and less nodular glomerulopathy (18versus 100%; P ,0.0001). Multiple myeloma was more frequently diagnosed inLCDD & MCN than in pure MIDD (91versus31%; P 50.025). Renal and patient survivals were significantly worse inpatients with LCDD & MCN (mean, 4 and 22 mo, respective-ly), compared with patients with pure MIDD (mean, 22 and 54mo). Chemotherapy stabilized or improved renal function in 10of 15 patients (67%) with pure MIDD who presented withcreatinine of,5.0 mg/dl, emphasizing the importance of earlydetection. On multivariate analysis, initial creatinine was theonly predictor of renal and patient survival in pure MIDD,underscoring the prognostic significance of the renalinvolvement.
Nonamyloidotic monoclonal Ig deposition disease (MIDD) ischaracterized by nodular sclerosing glomerulopathy, protein-uria, renal insufficiency, and an association with dysproteine-mias (1). Histologic evaluation reveals monoclonal light-and/or heavy-chain deposits within basement membranes ofglomeruli, tubules, and vessels. Three subtypes of MIDD havebeen reported, including light-chain DD (LCDD) (2–4), light-and heavy-chain DD (LHCDD) (5–6), and heavy-chain DD(HCDD) (7–10). Among these conditions, LCDD is the mostprevalent and, in one series, constituted 19% of 118 renalbiopsies from patients with multiple myeloma (11). Reports ofLHCDD and HCDD are rare, with fewer than 2 dozen docu-mented cases in the literature. HCDD cases have included thefull spectrum ofg1, g2, g3, g4, anda-heavy-chain subtypes(7–10,12). A deletion of the CH1 constant domain of theg-heavy chain underlies the secretion of heavy chains by alymphocyte or plasma cell clone (13,14).
At the time of renal biopsy, up to 30% of patients with renalMIDD have no detectable monoclonal protein in serum orurine (1). However, the number of patients who later developa clinical dysproteinemia is unclear. A retrospective analysis ofa small series of 19 patients with LCDD who received chemo-therapy (typically melphalan and prednisone) suggested thatchemotherapy may stabilize or improve renal function (14).Cases of LHCDD and HCDD are too few to draw conclusionsabout therapy, although there is a report of a patient whoseLHCDD responded to pulse steroids (15).
We reviewed retrospectively 34 renal biopsies processed atColumbia Presbyterian Medical Center from patients withMIDD. The clinical, pathologic, and outcome data from thislarge series expand the available literature on these importantand not infrequent disease entities.
Materials and MethodsStudy Design and Patient Selection
The renal pathology files at New York-Presbyterian Hospital, Co-lumbia-Presbyterian Campus, from 1982 to February 2000, werereviewed retrospectively, and 34 cases of LCDD, LHCDD, andHCDD were identified among the 7241 cases (0.47%) processedduring this period. Entry into the study was based on the presence ofrenal biopsy findings diagnostic of MIDD (LCDD, LHCDD, orHCDD). The 34 cases were analyzed with respect to pathologic andclinical findings, as well as outcome data. This information was
Received September 20, 2000. Accepted December 4, 2000.Correspondence to Dr. Vivette D. D’Agati, Department of Pathology, Colum-bia Presbyterian Medical Center, 630 West 168th Street, VC 14-224, NewYork, NY 10032. Phone: 212-305-7460; Fax: 212-342-5380; E-mail:vdd1@columbia.edu
1046-6673/1207-1482Journal of the American Society of NephrologyCopyright © 2001 by the American Society of Nephrology
J Am Soc Nephrol 12: 1482–1492, 2001
obtained mainly by chart review and direct contact with referringphysicians.
Pathology StudiesAll renal biopsies were processed for light microscopy, immuno-
fluorescence, and electron microscopy according to standard tech-niques. For each case, 11 glass slides stained with hematoxylin andeosin, periodic acid-Schiff (PAS), trichrome, and Jones methenaminesilver were reviewed.
Immunofluorescence was performed on 3-mm cryostat sections byuse of a panel of FITC-conjugated rabbit anti-human antibodies toIgG, IgM, IgA, C3, C1, fibrinogen, albumin, andk andl light chains(Dako Corporation, Carpenteria, CA). Immunofluorescence stainingintensity was graded on a scale of 0 to 31. For HCDD, g-chainsubtypes were determined by direct immunofluorescence using mono-clonal FITC-conjugated antibodies to IgG1 (clone 8c/6 to 39), IgG2(clone HP6014), IgG3 (clone HP6050), and IgG4 (clone HP6023; TheBinding Site, Birmingham, United Kingdom). Frozen sections alsowere stained with monoclonal antibodies specific for the constantdomains (CH1, CH2, and CH3) of IgG heavy chain (clones HP6044,HP6018, HP6016; Centers for Disease Control and Prevention, At-lanta, GA), followed by FITC-conjugated sheep anti-mouse secondaryantibody, as described previously (7).
Inclusion criteria were based entirely on pathologic findings.MIDD was defined by immunofluorescence as paraprotein deposits of(1) clear monoclonal composition (i.e., staining exclusively fork orfor l in the case of LCDD, staining for a single class of Ig (g, m, ora) with light-chain restriction in the case of LHCDD, or staining fora single class of Ig (g, m, or a) with no corresponding light chain inthe case of HCDD; (2) 21 or greater intensity of staining; and (3) alinear distribution within glomerular and/or tubular basement mem-branes. The presence of deposits was confirmed subsequently byelectron microscopy in all cases of pure MIDD.
Clinical Studies and Laboratory EvaluationDemographic, clinical, and laboratory information was obtained on
each patient at the time of renal biopsy. Follow-up information wasobtained on all but two patients. Hypertension was defined as systolicBP .140 or diastolic BP.90 or the use of antihypertensive medi-cations at the time of biopsy. Renal insufficiency was defined asserum creatinine.1.2 mg/dl. Acute renal failure was defined by a6-mo interval change in serum creatinine of.0.5mg/dl. Nephroticsyndrome was defined as 24-h urinary protein$3 g/d, edema, andhypoalbuminemia (serum albumin#3.6 g/dl).
Positive urine protein electrophoresis (UPEP) and serum proteinelectrophoresis (SPEP) were defined by the presence of a monoclonal(M protein) spike. Hypercalcemia was defined as Ca21 .10.5 mg/dl(corrected for albumin,4.0 g/dl). Positive bone marrow biopsy wasdefined as$15% plasma cells. Multiple myeloma (MM) was definedby renal MIDD plus at least one of the following: (1) positive bonemarrow biopsy, (2) presence of osteolytic lesions, (3) hypercalcemiawith positive UPEP or SPEP, or (4) $10% bone marrow plasmacy-tosis with low quantitative serum immunoglobulins. Monoclonalgammopathy of unknown significance (MGUS) was defined as pres-ence of MIDD and a positive SPEP or UPEP without other clinicalfeatures of myeloma.
At the conclusion of the study, worsening of renal function wasdefined as a.50% increase in serum creatinine from baseline. End-stage renal disease (ESRD) was defined as serum creatinine$5.0mg/dl or dependence on dialysis. Progression to renal failure wasdefined as ESRD or a doubling of serum creatinine.
Statistical AnalysesResults that involve continuous variables are expressed as mean6
SEM. For analysis of the clinical and laboratory characteristics be-tween the different groups, nonparametric statistical methods thatused exact inference were applied, including the Fisher’s exact test,Mann-WhitneyU test, and Kruskal-Wallis test as appropriate. Foranalysis of renal and patient survival, Kaplan-Meier estimates wereperformed. Multivariate analysis was performed by use of logisticregression models and discriminant analysis. Cox regression was usedin multivariate analysis of predictors of patient and renal outcome.SPSS for Windows (Version 10.0; SPSS, Inc., Chicago, IL) was usedto perform all analyses. Statistical significance was assumed atP ,0.05.
ResultsPathology Studies
Renal biopsies from 34 patients met morphologic entrycriteria for the diagnosis of MIDD. Twenty-three patients had“pure MIDD” (Figure 1). In 11 cases of LCDD, immunofluo-rescence also revealed tubular casts staining with high intensityfor a single light chain (k or l) accompanied by light micro-scopic findings of atypical, fractured, polychromatic tubularcasts, diagnostic of myeloma cast nephropathy (MCN). These11 cases are classified as combined “LCDD & MCN” (Figure1).
In all 23 biopsies from patients with pure MIDD, lightmicroscopy revealed a nodular sclerosing glomerulopathy.Glomeruli were enlarged with a diffuse and nodular expansionof the mesangial matrix, often accompanied by mild mesangialhypercellularity (Figure 2A). The mesangial nodules stainedPAS positive, trichrome red-blue, and nonargyrophilic. Typi-cally, there was little or no thickening of glomerular basementmembranes (GBM). There were occasional membranoprolif-erative features in the form of circumferential mesangial inter-position surrounding some nodules. Global glomerulosclerosiswas common, involving a mean of 20% of glomeruli (range, 0to 81%). Tubular basement membrane (TBM) thickening was
Figure 1.Composition of deposits in 34 cases of monoclonal immuno-globulin deposition disease (MIDD). LCDD, light-chain depositiondisease; LHCDD, light- and heavy-chain deposition disease;HCDD, heavy-chain deposition disease; LCDD & MCN, LCDDplus myeloma cast nephropathy.
J Am Soc Nephrol 12: 1482–1492, 2001 Monoclonal Immunoglobulin Deposition Disease 1483
variable. The degree of tubular atrophy and interstitial fibrosisranged from mild in nine cases to severe in eight cases. In somecases, vascular basement membranes surrounding medial myo-cytes were thickened by PAS-positive material. Congo redstaining for amyloid was performed in all cases; in 3 of the 23cases (allk), there was focal positivity consistent with super-
imposed amyloidosis, which was confirmed by the demonstra-tion of 8- to 12-nm fibrillar deposits on electron microscopy.Nonetheless, the dominant finding was MIDD, and, therefore,these three cases were included in the group of pure MIDD forpurposes of analysis. Two cases also displayed overlappingfeatures of diabetic nephropathy.
Figure 2.A representative case of HCDD is illustrated. (A) A glomerulus displaying a nodular sclerosing glomerulopathy with global mesangialnodules, some surrounded by mesangial interposition, without obvious thickening of the glomerular basement membranes (GBM). Staining forthe constant domains of theg-heavy chain shows negativity for CH1 (B) and strong linear positivity for CH2 (C) and CH3 (D) along GBMand tubular basement membranes (TBM). Staining for complement component C1 is present along TBM, with a granular to linear texture.Ultrastructural evaluation reveals finely granular electron dense deposits that involve the inner aspect of the GBM. Magnifications:3400 inA (periodic acid-Schiff), B, C, and D;3800 in E;35000 in F.
1484 Journal of the American Society of Nephrology J Am Soc Nephrol 12: 1482–1492, 2001
The following patterns of paraprotein deposition were iden-tified by immunofluorescence: among the 12 cases of pureLCDD, the dominant light chain wask in 11 cases andl in 1case. The five cases of LHCDD included three with IgGk andtwo with IgGl. The six cases of HCDD includedg-heavychain in five cases anda-heavy chain in one case. Tissue wasavailable for further study in four cases ofg HCDD, andamong these cases, the heavy-chain component was identifiedas g1 in two cases,g3 in one case, andg4 in one case.Furthermore, in all four cases, a deletion of the CH1 constantdomain of theg-chain was identified by immunostaining (Fig-ure 2, B through D). In two of the five cases of LHCDD andin four of the five cases ofg HCDD, the monoclonal depositsfixed complement components C3 and C1 in a granular orlinear pattern (Figure 2E). Complement fixation was identifiedin the three cases withg1 org3 heavy chains but not in the casewith g4 heavy chain.
In all cases of MIDD, immunofluorescence revealed lineardeposits within TBM. In addition, GBM deposits were identi-fied in 20 of 23 cases (87%), mesangial deposits in 19 of 23(83%), interstitial deposits in 7 of 23 (30%), and vascular
deposits in 15 of 23 (65%). None of the 23 cases of pure MIDDdisplayed significant monoclonal staining of tubular casts orother histologic evidence of MCN.
Ultrastructural evaluation of the 23 cases of pure MIDDdemonstrated deposits in GBM (100% of cases), mesangium(96%), and TBM (96%) (Figure 2F). The deposits typicallyappeared to be granular-powdery and were identified in thelamina rara interna and/or permeating the lamina densa of theGBM. Interstitial deposits also were identified in a minority ofcases (18%). In the 18 cases in which vessels were sampled forelectron microscopy, deposits commonly were seen surround-ing the basement membranes of individual myocytes (78%).The degree of foot process fusion was variable but typicallyextensive (mean, 55%; range, 5 to 100%).
The appearance of the 11 cases of combined LCDD & MCNdiffered significantly from those with pure MIDD. The majorlight microscopic findings were those of atypical, fractured,polychromatic casts, typical of MCN, associated with intersti-tial edema, inflammation, fibrosis, and diffuse tubular degen-erative changes (Figure 3A). Although a nodular sclerosingglomerulopathy was seen in all cases of pure MIDD, this
Figure 3. A representative case of LCDD & MCN. (A) A low-power view that shows focal hard polychromatic casts, diffuse tubulardegenerative changes, and interstitial fibrosis, typical of MCN. A glomerulus appears normal by light microscopy. (B) Although glomerulidisplay no abnormalities by light microscopy, immunofluorescence staining fork reveals intense linear staining of GBM and TBM, typical ofLCDD. In addition to linear TBM staining, there is strong staining of the atypical casts fork (C) and complete negativity forl (D).Magnifications:3200 in A (trichrome);3400 in B; 3800 in C and D.
J Am Soc Nephrol 12: 1482–1492, 2001 Monoclonal Immunoglobulin Deposition Disease 1485
pattern of disease was seen in only 2 of the 11 cases of LCDD& MCN (18%; P , 0.0001). Two additional cases displayedmild mesangial expansion, whereas in seven cases, glomeruliappeared histologically unremarkable. By immunofluores-cence, the monoclonal light chain wask in 10 cases andl in1 case. Paraprotein deposits were identified in all cases withinGBM and TBM and in the majority of cases (82%) withinvessel wall basement membranes (Figure 3B). In all cases,immunofluorescence also revealed the diagnostic finding ofMCN: large, fractured casts that stain intensely with a singlelight-chain component (Figure 3, C and D).
The cases of combined LCDD & MCN were subdividedfurther into two subgroups, depending on whether electron-dense deposits were identified. Four cases had monoclonallight-chain staining by immunofluorescence but no corre-sponding electron-dense deposits by electron microscopy. Ac-cordingly, the findings in these four cases are referred to as“LCDD & MCN (by IF only).” Among the seven cases ofLCDD & MCN, electron-dense deposits were identified inGBM in all seven cases and in the mesangium and TBM in sixcases. Typically, the extent of GBM and mesangial depositswas less than that in pure MIDD, and only mild foot processfusion was identified (mean, 23%; range, 10 to 40%).
Clinical Characteristics at Biopsy: Demographics andRenal Presentation
The 23 patients with pure MIDD consisted of 12 men and 11women with a mean age of 57.4 yr (Table 1). A total of 74%were Caucasian, and 5 of 23 (22%) were African American;notably, among the 5 African American patients, 4 had heavy-chain deposition (either HCDD or LHCDD). There was a trendtoward greater age in the group with LCDD & MCN, com-pared with those with pure MIDD (67.1versus57.4 yr; P 50.066); however, racial composition and gender were not sig-nificantly different. Hypertension was present in the majorityof cases but was seen less frequently in patients with LHCDD(40%) than in those with LCDD (83%) or HCDD (100%) (P 50.03). Four patients (12%) had a clinical history of type 2diabetes mellitus, but only two had biopsy findings suggestiveof diabetic nephropathy.
With the exception of one case ofde novoLCDD in a renalallograft, all other cases of MIDD were diagnosed in the nativekidney. Patients with pure MIDD typically presented withrenal insufficiency at the time of biopsy, as evidenced by 96%with serum creatinine.1.2 mg/dl (Table 1). Patients withLCDD & MCN had a significantly higher serum creatinine (7.8versus4.5 mg/dl;P 5 0.01) and a lower creatinine clearance(13.8versus37.3 cc/min;P 5 0.02) when compared with thegroup with pure MIDD.
Nephrotic-range proteinuria was seen in almost half (48%)of patients with pure MIDD, and the mean 24-h urine proteinwas 4.2 g/d. In contrast, the mean 24-h proteinuria was signif-icantly less in patients with LCDD & MCN (2.2 g/d;P 50.01), and nephrotic-range proteinuria was seen in only 2 of 11patients (18%). Furthermore, the degree of hypoalbuminemiaand hypercholesterolemia and the incidence of peripheraledema were greater in patients with pure MIDD, such that full
nephrotic syndrome was present in 6 of 23 cases but in none ofthe 11 patients with LCDD & MCN. Hypocomplementemiawas present in one of five patients with LHCDD and in threeof six patients with HCDD.
Among patients with pure MIDD, the incidences of acuterenal failure and dialysis dependence (at the time of biopsy)were 30 and 26%, respectively. These incidence rates weresignificantly lower than in patients with LCDD & MCN, ofwhom 82% had acute renal failure and 64% required dialysis(P 5 0.02 andP 5 0.053, respectively).
Of note, five of six HCDD patients who were tested forhepatitis C virus (HCV) antibody by second generation enzymeimmunoassay were positive, whereas none of the eight patientswith LCDD or four patients with LCDD & MCN studied wereHCV antibody positive (P , 0.001). All five patients’ bilirubinand serum transaminase were within the normal range, andfour were HCV PCR negative. HCV data were not availablefor any patients with LHCDD.
Oncologic CharacteristicsRenal biopsy diagnosis of MIDD preceded any other clinical
evidence of dysproteinemia in 16 patients (70%) with pureMIDD and in 7 patients (64%) with LCDD & MCN (Table 2).In only 11 of 34 patients was an M spike tested and identifiedon SPEP and/or UPEP before biopsy. After renal biopsy diag-nosis of pure MIDD, an M spike was identified on SPEP in48% and on UPEP in 52%; in 3 of 23 patients with pure MIDD(13%), both SPEP and UPEP (and immunofixation) were neg-ative. A positive SPEP was present in 80% of patients withLHCDD, compared with 25% of patients with LCDD (P 50.04) and 67% of patients with HCDD.
Oncologic workup of patients with pure MIDD revealed apositive bone marrow biopsy in 35%, hypogammaglobuline-mia in 30%, osteolytic lesions in 13%, and hypercalcemia in asingle patient (4.3%). Thirty-nine percent of patients with pureMIDD met criteria for MM, and 39% were diagnosed withMGUS. At presentation, the only oncologic parameter thatdiffered significantly between the pure MIDD and LCDD &MCN groups was the higher incidence of multiple myeloma inthe latter (39versus91%; P 5 0.025).
OutcomeAmong the seven patients with LCDD and MM, four were
treated with melphalan and prednisone (MP) and one each withregimens of vincristine-adriamycin-dexamethasone; steroids;and vincristine, cyclophosphamide, and MP. At follow-up(mean, 31.5 mo), five had stable renal function, one developedan increase in serum creatinine from 1.5 to 2.3 mg/dl over 47mo, and one progressed to ESRD requiring hemodialysis. Fourof the seven patients with LCDD and MM died during thefollow-up period; exact causes of death are not known. Incontrast, the two patients with LCDD and MGUS were treatedwith MP and plasmapheresis, respectively, and both remainalive at 24 and 27 mo. One patient progressed from a serumcreatinine of 2.9 mg/dl to dialysis dependence over 24 mo,whereas the other had an increase in serum creatinine from 6.2to 6.7 mg/dl over 27 mo but has not yet required dialysis. The
1486 Journal of the American Society of Nephrology J Am Soc Nephrol 12: 1482–1492, 2001
Ta
ble
1.D
emog
raph
ics
and
rena
lpre
sent
atio
na
Cha
ract
eris
ticLC
DD
(n5
12)
LHC
DD
(n5
5)H
CD
D(n
56)
Pur
eM
IDD
(n5
23)
LCD
D&
MC
N(n
511
)P
Val
ue
Gen
der
(M/F
)7/
52/
33/
312
/11
5/6
NS
Age
(yr)
56.6
62.
8763
.86
5.88
53.8
62.
6357
.46
2.12
67.1
63.
920.
066b
Hyp
erte
nsio
n10
(83%
)2
(40%
)6
(100
%)
18(7
8%)
7(6
4%)
0.03
c
Cre
atin
ine
(mg/
dl)
4.06
0.9
5.36
2.0
4.86
1.5
4.56
0.7
7.86
1.2
0.01
b
Ren
alin
suffi
cien
cy(s
Cr$
1.2
mg/
dl)
11(9
2%)
5(1
00%
)6
(100
%)
22(9
6%)
11(1
00%
)N
SP
rote
inur
ia(g
/24
h)4.
260.
82.
961.
05.
362.
24.
260.
72.
260.
70.
01b
Nep
hrot
icpr
otei
nuria
($3
g/24
h)6
(50%
)1
(20%
)4
(67%
)11
(48%
)2
(18%
)N
SA
lbum
in(g
/dl)
3.66
0.2
2.86
0.2
2.96
0.3
3.26
0.2
3.66
0.2
NS
Cho
lest
erol
(mg/
dl)
275.
4617
.221
5612
.121
2.56
26.7
246.
3614
.021
8623
.10.
05d
Ede
ma
5(4
2%)
5(1
00%
)5
(83%
)15
(65%
)2
(18%
)0.
01b
Nep
hrot
icsy
ndro
me
2(1
7%)
1(2
0%)
3(5
0%)
6(2
6%)
0(0
%)
NS
Mic
rohe
mat
uria
(.5
RB
C/h
pf)
5(4
2%)
3(6
0%)
4(6
7%)
12(5
2%)
4(3
6%)
NS
Hyp
ocom
plem
ente
mia
0(0
%)
1(2
0%)
3(5
0%)
4(1
7%)
0(0
%)
NS
Acu
tere
nalf
ailu
re4
(33%
)1
(20%
)2
(33%
)7
(30%
)9
(82%
)0.
02b
Dia
lysi
sat
time
ofbi
opsy
2(1
6%)
2(4
0%)
2(3
3%)
6(2
6%)
7(6
4%)
0.05
3b
aR
esul
tsar
egi
ven
asno
.(%
)or
mea
n6
SE
M.
NS
,no
tsi
gnifi
cant
;H
TN
,hy
pert
ensi
on;
LCD
D,
light
-cha
inde
posi
tion
dise
ase;
LHC
DD
,lig
ht-
and
heav
y-ch
ain
depo
sitio
ndi
seas
e;H
CD
D,
heav
y-ch
ain
depo
sitio
ndi
seas
e;M
IDD
,m
onoc
lona
lim
mun
oglo
bulin
depo
sitio
ndi
seas
e;R
BC
,re
dbl
ood
cells
;hp
f,hi
gh-p
ower
field
.b
Pur
eM
IDD
vers
usL
CD
D&
MC
N.
cLH
CD
Dve
rsu
sLC
DD
&H
CD
D.
dLC
DD
vers
usL
HC
DD
&H
CD
D.
J Am Soc Nephrol 12: , 1492 Monoclonal Immunoglobulin Deposition Disease 1487
Ta
ble
2.O
ncol
ogic
findi
ngs
and
clin
ical
outc
omesa
Fin
ding
san
dO
utco
mes
LCD
D(n
512
)LH
CD
D(n
55)
HC
DD
(n5
6)P
ure
MID
D(n
523
)LC
DD
&M
CN
(n5
11)
PV
alue
Kno
wn
Msp
ike
prio
rto
bx2
(17%
)4
(80%
)1
(17%
)7
(30%
)4
(36%
)N
S(1
)S
PE
Pe3
(25%
)4
(80%
)4
(67%
)11
(48%
)3
(27%
)0.
04c
(1)
UP
EPe
5(4
2%)
4(8
0%)
3(5
0%)
12(5
2%)
7(6
4%)
NS
(2)
UP
EP
AN
D(2
)S
PE
Pe2
(17%
)0
(0%
)1
(17%
)3
(13%
)1
(9%
)N
SS
erum
imm
unof
ixat
ionf
1k,
4lgG
k,
5nl
1de
crg
and
1in
cra
2Ig
Gk
,1
IgG
l2
unkn
own
1Ig
Gk
,3
low
g1
IgA
kan
d1
IgM
l
1IgG
k,
1k,
5nl
,2
gsp
ike,
1de
crg
1un
know
nH
ypog
amm
aglo
bulin
emia
5(4
2%)
0(0
%)
2(3
4%)
7(3
0%)
5(4
5%)
NS
(1)
BM
biop
sy$
15%
plas
ma
cells
6(5
0%)
2(4
0%)
1(1
7%)
8(3
5%)
8(7
3%)
NS
(1)
Ske
leta
lsur
vey
2(1
7%)
0(0
%)
1(1
7%)
3(1
3%)
1(9
%)
NS
Cor
rect
edC
a21
.10
.5m
g/dl
0(0
%)
1(2
0%)
0(0
%)
1(4
.3%
)3
(27%
)N
SC
linic
aldi
agno
sisb
mul
tiple
mye
lom
a7
(58%
)1
(20%
)1
(17%
)9
(39%
)10
(91%
)0.
025
d
MG
USb
2(1
7%)
3(6
0%)
4(6
6%)
9(3
9%)
0(0
%)
isol
ated
HC
DD
0(0
%)
0(0
%)
1(1
7%)
1(4
%)
0(0
%)
insu
ffici
ent
data
3(2
5%)
1(2
0%)
0(0
%)
4(1
7%)
1(9
%)
Mea
nfo
llow
-up
time
(mo)
33.26
9.5
9.16
5.1
14.8
66.
622
.66
5.6
14.7
63.
4ra
nge
(mo)
(7–9
9)(1
–27)
(2–4
2)(1
–99)
(0.2
–36)
Ren
alou
tcom
e0.
057d
ES
RD
/dia
lysi
sat
pres
enta
tion
3(2
5%)
2(4
0%)
3(5
0%)
8(3
5%)
9(8
2%)
0.05
3d
stab
le/im
prov
edb5
(42%
)2
(40%
)3
(50%
)10
(43%
)1
(9%
)w
orse
ning
rena
lfun
ctio
nb1
(8%
)1
(20%
)0
(0%
)2
(9%
)0
(0%
)pr
ogre
ssio
nto
ES
RD
/dia
lysi
sb
3(2
5%)
0(0
%)
0(0
%)
3(1
3%)
1(9
%)
ES
RD
/dia
lysi
sat
stud
yen
d6
(50%
)2
(40%
)3
(50%
)11
(48%
)10
(91%
)0.
063
d
Mea
ntim
eto
ES
RD
(mo)
278
322
40.
0196d
Mea
ntim
eto
deat
h(m
o)69
1342
5422
0.05
d/0
.01c
Pat
ient
deat
hs6
(50%
)3
(60%
)1
(17%
)10
(43%
)6
(55%
)N
S
aR
esul
tsar
egi
ven
asno
.(%
)or
mea
n6
SE
M.
BM
,bo
nem
arro
w;
UP
EP
,ur
ine
prot
ein
elec
trop
hore
sis;
SP
EP
,se
rum
prot
ein
elec
trop
hore
sis;
MC
N,
mye
lom
aca
stne
phro
path
y;M
GU
S,
mon
oclo
nalg
amm
opat
hyof
unkn
own
sign
ifica
nce;
ES
RD
,en
d-st
age
rena
ldis
ease
;nl
,no
rmal
.b
Def
ined
inM
ater
ials
and
Met
hods
sect
ion.
cLC
DD
vers
usL
HC
DD
.d
Pur
eM
IDD
vers
usL
CD
D&
MC
N.
eA
fter
rena
lbio
psy
diag
nosi
s.fA
tan
ytim
edu
ring
clin
ical
cour
se.
1488 Journal of the American Society of Nephrology J Am Soc Nephrol 12: , 1492
three patients with LCDD and insufficient data for oncologicdiagnosis all required dialysis within 1 mo of presentation.
Follow-up data were available on all five patients withLHCDD, of which two received MP and one received cyclo-phosphamide and prednisone. Information on treatment wasnot available for two patients. Renal function remained stablein two patients (one of whom met criteria for MM), one had aworsening of renal function, and two already required dialysisat the time of presentation. Follow-up time is limited to 9.1 moin the LHCDD group, in part because of the early death ofthree of the five patients.
Among the six cases of HCDD, the only patient who metcriteria for MM was the single patient witha-HCDD. Thispatient was not treated, and renal function remained stable at2-mo follow-up. The remaining five patients received eitherMP (one patient), pulse decadron (one patient), prednisone pluschlorambucil (one patient), or no treatment (two patients).Follow-up data revealed two patients with stable serum creat-inine (over 5 mo each) and three who presented with eitherESRD or immediate requirement for dialysis. Of interest, thesingle patient who received pulse decadron had an initialincrease in serum creatinine from 1.6 to 2.2 mg/dl over 1 mo,followed by a decrease to 1.0 mg/dl at the end of 5 mo. At lastfollow-up (mean, 14.8 mo), all six patients with HCDD werealive. One patient with HCDD received a living related renaltransplant from her sister and is doing well 8 mo posttrans-plantation without recurrence of proteinuria.
Ten of the 11 patients with LCDD & MCN met clinicalcriteria for the diagnosis of MM. Four of the patients receivedMP, five patients received vincristine-adriamycin-decadron
(four also received MP, prednisone, cyclophosphamide, orthalidomide), and one was treated solely with steroids. Eight ofthe 10 patients presented with ESRD without subsequent re-covery, and 1 patient had an increase in serum creatinine from2.6 to 6.8 mg/dl over 2 mo. Of interest, a single patient withMM who was treated with MP had a decline in serum creati-nine from 3.5 to 2.0 mg/dl over 36 mo. The single patient withinsufficient data to establish an oncologic diagnosis requireddialysis at the time of presentation. At last follow-up (mean,14.7 mo), 6 of the 11 patients with LCDD & MM had died.Causes of death included sepsis (two patients), hypercalcemia(one patient), and progressive myeloma (three patients).
Renal (P 5 0.0196) and patient (P 5 0.0453) survivals weresignificantly better in patients with pure MIDDversusLCDD& MCN (Figures 4 and 5). Improvement in renal function wasseen in 10 of 23 patients with pure MIDD (43.5%) but only ina single patient with LCDD & MCN (9.1%;P 5 0.0487).When patients were reanalyzed for the presence of clinicalcriteria for MM, 10 of 19 patients with MM (52.6%) died at theend of the study, as opposed to 2 of 10 (20%;P 5 0.0956) whohad been diagnosed with MGUS. Median patient survival timewas 36 mo in the MM group and 42 mo in the MGUS group(P 5 0.74). Mean patient survival in LCDD was 69versus13mo in LHCDD versus42 mo in HCDD, with a statisticallysignificant difference between the LCDD and LHCDD groups(P 5 0.01). The end points of progression of renal insuffi-ciency or ESRD were not significantly different among thethree groups (27 mo in LCDDversus8 mo in LHCDDversus3 mo in HCDD;,P 5 0.34) when adjusted for the relatively
Figure 4.Life-table analysis of renal survival in pure MIDDversusLCDD & MCN. Solid line, pure MIDD; dotted line, LCDD & MCN.
J Am Soc Nephrol 12: 1482–1492, 2001 Monoclonal Immunoglobulin Deposition Disease 1489
short follow-up time in the LHCDD (mean, 9.1 mo) andHCDD (mean, 15.7 mo) groups.
Limited follow-up data are available regarding proteinuria.Among the five patients with LCDD and stable/improved renalfunction, proteinuria declined in two, increased in one, and wasunavailable in the remaining two patients. In the four patientswith either LHCDD or HCDD and stable renal function, adecline in proteinuria was seen.
On multivariate analysis of the pure MIDD group, there wasno correlation between renal or patient survival and any his-tologic parameter, including severity of glomerulosclerosis,interstitial inflammation, interstitial fibrosis, tubular atrophy,or vascular disease. Oncologic diagnosis and treatment also didnot correlate with either renal or patient survival. The onlypredictor of renal and patient survival was the initial serumcreatinine at the time of biopsy (P 5 0.003 and 0.042,respectively).
DiscussionWe report the clinical, pathologic, and outcome data in a
large series of patients with renal MIDD. MIDD is a parapro-tein deposition disease characterized by monoclonal depositswithin renal basement membranes, indicative of underlyingplasma cell dyscrasia. However, not all cases have a demon-strable monoclonal protein by SPEP or UPEP.
One third of our 34 cases of MIDD had evidence of coex-istent MCN. The pathologic findings in these 11 cases ofLCDD & MCN were dominated by the MCN component andmanifested severe tubular damage by cast nephropathy, with
less extensive glomerular pathology. Although linear depositswere noted in basement membranes by immunofluorescence inall 11 cases, they typically were less extensive than in pureMIDD. Furthermore, in 4 of the 11 cases, corresponding elec-tron-dense deposits were not identified by electron microscopy(LCDD & MCN [by IF only]). Potential explanations for thisfinding include nonspecific “trapping” of the circulating mono-clonal light chain in renal basement membranes reflecting highserum levels without true deposit formation or insufficientaggregation of the deposits to be visualized at the ultrastruc-tural level. LCDD & MCN tended to occur in older patientswho presented with more severe renal insufficiency and lesssevere proteinuria and were more likely to fulfill criteria forMM. The clinical, pathologic, and outcome data of this groupare more typical of MCN than MIDD. Unfortunately, paststudies of LCDD have not always distinguished between pureMIDD and cases with overlapping MCN. For example,Ganevalet al. (4) and Buxbaumet al. (5) each reported seriesof LCDD in which almost 50% of patients had concurrent castnephropathy, but they did not perform subgroup analysis.
Examination of the histologic findings in patients withMIDD reveals multiple interesting trends.k is the predominantlight chain deposited within renal basement membranes inLCDD, as identified in 91% of our cases, and is similar to thereported incidences of 73 to 87.5% in other series (2–5,15).This is in clear contrast to the increasedl-to-k ratio seen inamyloidosis and correlates with the reported predominance ofVk4 and Vl6 as precursor proteins in LCDD and amyloidosis,respectively (16,17). Among cases of HCDD,g is the predom-
Figure 5.Life-table analysis of patient survival in pure MIDDversusLCDD & MCN. Solid line, pure MIDD; dotted line, LCDD & MCN.
1490 Journal of the American Society of Nephrology J Am Soc Nephrol 12: 1482–1492, 2001
inant class of heavy chain. Our single case ofa-HCDD is thesecond such case reported (10). A CH1 deletion was noted inall four of our g-HCDD cases studied, correlating with theprevious finding that this deletion is critical for prematureheavy-chain secretion by plasma cell clones (13,18). Althougha nodular sclerosing glomerulopathy is the most common find-ing on light microscopy in MIDD, its incidence ranges from 31to 74% (2–5,15). We identified this pattern in only 2 of 11cases of LCDD & MCN, possibly because of their earlypresentation with MCN-induced acute renal failure, with in-sufficient time for the development of nodular sclerosing glo-merulopathy. Alternatively, the pathogenic light-chain proteinsin this entity may be less sclerogenic, as suggested by theirvariable ability to upregulate mesangial synthesis of TGFb(19). If all previously reported LCDD cases with overlappingMCN are excluded, the true incidence of nodular sclerosingglomerulopathy in pure MIDD likely would approach the100% found in our series. Similarly, in our series, depositswere readily identified by immunofluorescence or electronmicroscopy in GBM (87 and 100%, respectively), mesangium(83 and 96%), TBM (100 and 96%), and vessel wall basementmembranes (65 and 78%). Our rate of detection of depositswas higher than that in previous studies of MIDD (2–5,15),likely because of our exclusion of cases in which MCN dom-inated. Despite the broad pathologic spectrum of MIDD, nomorphologic parameter was found to be predictive of renal orpatient survival in our series, as noted in a previous analysis(2).
MIDD typically presents in the sixth decade, although, inour cohort, an earlier age of presentation was seen in patientswith pure MIDD than in those with LCDD & MCN. Thetypical renal presentation in patients with MIDD includesproteinuria, hypertension, and renal insufficiency. We identi-fied renal insufficiency in 96% of patients with pure MIDD,compared with 92% of cases compiled from five previousseries (2–5,15). Nephrotic range proteinuria ($3 g/d) waspresent in 48% of patients in our cohort, compared with 57%in previous series (2–5,15). Of interest, hypocomplementemiawas present in three of our six patients with HCDD, all ofwhom had heavy-chain deposits composed ofg1 or g3. Theseresults are consistent with the known complement-fixing abil-ity of theseg subclasses, a property that is dependent on anintact CH2 domain.
Consistent with previous literature, the incidence of overtMM was greater in patients with LCDD & MCN than in thosewith pure MIDD (91versus39%;P 5 0.025). Previous seriesrevealed a 45% incidence of MM at presentation, likely be-cause of the inclusion of cases with combined LCDD & MCN(2–5,15). Three of our 23 patients (13%) and 11 of 64 previ-ously reported cases (17%) had no evidence of a monoclonalspike on either SPEP or UPEP (2–5,15), which indicates thatthe absence of such laboratory findings does not exclude adiagnosis of MIDD.
Analysis of treatment and outcome data is hampered by thesmall patient numbers, failure of some series to separate thesubgroup with concurrent MCN, lack of standardized therapy,and limited follow-up. The majority of patients received ste-
roids plus melphalan or a cytotoxic agent. It is general practiceto treat patients who have pure MIDD or LCDD & MCN withsimilar regimens, irrespective of whether they meet diagnosticcriteria for myeloma. Although renal prognosis is poor, patientsurvival can be considerable, with 70% and 37% 5-yr patientand renal survivals reported in one series (14). In our cohort,among patients with pure MIDD, 35% presented with ESRD,22% had worsening of renal function or progression to ESRD,and 43% had stable or improved renal function. No patient whopresented with ESRD or requiring dialysis improved; however,10 of the remaining 15 patients (67%) with pure MIDD hadstable or improved renal function at the end of a mean fol-low-up of 23.7 mo. Unfortunately, no conclusions could bedrawn regarding the relative efficacy of the varied treatmentregimens. Of interest, the presence of MM did not influencerenal or patient survival. This is consistent with our observa-tion that the only predictor of renal and patient survival was theinitial serum creatinine at the time of biopsy, which under-scores the paramount prognostic importance of renal MIDD.
An interesting and previously unreported finding was that ofa positive HCV antibody test with undetectable HCV by PCRin four of five patients with HCDD studied. None of thepatients had elevated bilirubin or transaminase to suggest ac-tive hepatitis. These findings may represent a false-positiveHCV antibody test because of interference by the abnormaltruncated heavy chains with the HCV immunoassay; a distantHCV infection also is possible.
In summary, MIDD is defined by linear deposits of mono-clonal light-chain components in renal basement membranes,often producing a nodular sclerosing glomerulopathy. Becausefewer than half of patients with MIDD have clinical or labo-ratory features of MM, renal biopsy plays an essential role inthe diagnosis of dysproteinemia. Accurate histopathologic di-agnosis requires, above all, a systematic analysis of the immu-nofluorescence findings. Approximately one third of caseshave overlapping features of MCN; this subgroup is distin-guished by greater renal insufficiency and less proteinuria atpresentation, a renal biopsy picture typically dominated byMCN, and poorer renal and patient outcomes. Because thisgroup more closely resembles that of MCN, it should besegregated from future analyses of pure MIDD. Althoughpatient and renal survival rates remain poor, with early detec-tion and treatment, stable or improved renal function may beachieved (4,14,20). In the future, large multicenter studies ofMIDD will be needed to determine the optimal mode oftherapy.
AcknowledgmentJ.L. is the recipient of a clinical fellowship award from the National
Kidney Foundation of New York/New Jersey.
References1. Buxbaum J, Gallo G: Nonamyloidic monoclonal immunoglobu-
lin deposition disease.Hematol Oncol Clin N Am13: 1235–1248,1999
2. Confalonieri R, Barbiano di Belgiojoso G, Banfi G, Ferrario F,Bertani T, Pozzi C, Casanova S, Lupo A, De Ferrari G, Minetti
J Am Soc Nephrol 12: 1482–1492, 2001 Monoclonal Immunoglobulin Deposition Disease 1491
L: Light chain nephropathy: Histological and clinical aspects in15 cases.Nephrol Dial Transplant2: 150–156, 1988
3. Tubbs R, Gephardt G, McMahon J, Hall P, Valenzuela R, VidtD: Light chain nephropathy.Am J Med71: 263–269, 1981
4. Ganeval D, Mignon F, Preud’homme J, Noel L, Morel-MarogerL, Droz D, Brouet J, Mery J, Grunfeld JP: Visceral deposition ofmonoclonal light chains and immunoglobulins: A study of renaland immunopathologic abnormalities.Kidney Int26: 1–9, 1984
5. Buxbaum J, Chuba J, Hellman G, Solomon A, Gallo G: Mono-clonal immunoglobulin deposition disease: Light chain and lightand heavy chain deposition disease and their relation to lightchain amyloidosis.Ann Int Med112: 455–464, 1990
6. Dalani D, Weber D, Alexanian R: Light-heavy chain depositiondisease progressing to multiple myeloma.Am J Hematol50:296–298, 1995
7. Kambham N, Markowitz G, Appel G, Kleiner M, Aucouturier P,D’Agati V: Heavy chain deposition disease: The disease spec-trum. Am J Kidney Dis33: 954–962, 1999
8. Aucouturier P, Khamlichi A, Touchard G, Justrabo E, Cogne M,Chauffert B, Martin F, Preud’homme JL: Heavy-chain deposi-tion disease.N Engl J Med329: 1389–1393, 1993
9. Tubbs R, Berkely V, Valenzuela R, McMahon J, Gephardt G,Fishleder A, Nally J, Pohl M, Bukowski R, Lichtin A: Pseudo-heavy chain (IgG4) deposition disease.Mod Pathol5: 185–190,1992
10. Cheng I, Ho S, Chan D, Ng W, Chan K: Crescenteric nodularglomerulosclerosis secondary to truncated immunoglobulinadeposition.Am J Kidney Dis28: 283–288, 1996
11. Montseny J, Kleinknecht D, Meyrier A, Vanhille P, Simon P,Pruna A, Eladari D: Long-term outcome according to renalhistological lesions in 118 patients with monoclonal gammopa-thies.Nephrol Dial Transplant13: 1438–1445, 1998
12. Herzenberg A, Kiali M, Magil A: Heavy chain deposition dis-ease: Recurrence in a renal transplant and report of IgG2 subtype.Am J Kidney Dis35: 1–5, 2000
13. Moulin B, Deret S, Mariette X, Kourilsky O, Imai H, Dupouet L,Marcellin L, Kolb I, Aucouturier P, Brouet JC, Ronco P,Mougenot B: Nodular glomerulosclerosis with deposition ofmonoclonal immunoglobulin heavy chains lacking CH1.J AmSoc Nephrol10: 519–528, 1999
14. Khamlichi AA, Aucouturier P, Preud’Homme JL, Cogne M:Structure of abnormal heavy chains in human heavy-chain-dep-osition disease.Eur J Biochem229: 54–60, 1995
15. Heilman R, Velosa J, Holley K, Offord K, Kyle R: Long-termfollow-up and response to chemotherapy in patients with light-chain deposition disease.Am J Kidney Dis20: 34–41, 1992
16. Ozaki S, Abe M, Wolfenbarger D, Weiss D, Solomon A: Pref-erential expression of humanl-light-chain variable-region sub-groups in multiple myeloma, AL-amyloidosis, and Walden-strom’s macroglobulinemia.Clin Immunol Immunopathol71:183–189, 1994
17. Denoroy L, Deret S, Aucouturier P: Overrepresentation of theVkIV subgroup in light chain deposition disease.Immunol Lett42: 63–66, 1994
18. Khamlichi AA, Aucouturier P, Preud’homme JL, Cogne M:Structure of abnormal heavy chains in human heavy-chain-dep-osition disease.Eur J Biochem229: 54–60, 1995
19. Herrera G, Shultz J, Soong S, Sanders P: Growth factors inmonoclonal light-chain related renal disease.Hum Pathol25:883–892, 1994
20. Komatsuda A, Wakui H, Ohtani H, Kodama T, Miki K, Imai H,Miura A: Disappearance of nodular mesangial lesions in a patientwith light chain nephropathy after long-term chemotherapy.Am JKidney Dis35: 1–5, 2000
1492 Journal of the American Society of Nephrology J Am Soc Nephrol 12: 1482–1492, 2001
top related